Coordination of insulin and Notch pathway activities by microRNA miR-305 mediates adaptive homeostasis in the intestinal stem cells of the Drosophila gut

Abstract

Homeostasis of the intestine is maintained by dynamic regulation of a pool of intestinal stem cells. The balance between stem cell self-renewal and differentiation is regulated by the Notch and insulin signaling pathways. Dependence on the insulin pathway places the stem cell pool under nutritional control, allowing gut homeostasis to adapt to environmental conditions. Here we present evidence that miR-305 is required for adaptive homeostasis of the gut. miR-305 regulates the Notch and insulin pathways in the intestinal stem cells. Notably, miR-305 expression in the stem cells is itself under nutritional control via the insulin pathway. This link places regulation of Notch pathway activity under nutritional control. These findings provide a mechanism through which the insulin pathway controls the balance between stem cell self-renewal and differentiation that is required for adaptive homeostasis in the gut in response to changing environmental conditions.

Keywords: Notch; insulin; microRNA; stem cell.

Figure 1.

miR-305 targets in ISC self -renewal and differentiation. (A) Schematic representation of the intestinal epithelium. EB, EC, and EE cells. (B) Predicted miR-305 targets in signaling pathways involved in ISC self-renewal and differentiation.

Figure 2.

The miR-275/miR-305 cluster controls ISC proliferation. (A) miR-275 and miR-305 miRNAs are located near the 3′ end of the noncoding transcript CR43857 in the interval between the protein-coding genes chm and CG5261. All three genes are transcribed in the same orientation. Black triangles indicate P-element insertion sites. Arrows indicate predicted transcription start sites. The interval deleted in the KO allele is indicated by parentheses. The genomic region containing the miRNAs was replaced by a mini-white cassette flanked by LoxP sites (details are in Supplemental Fig. S1A). miR-275 and miR-305 miRNAs were reduced to low levels in animals carrying the P-element insertion alleles cuc¹ and cuc¹⁰⁶⁰⁷ in trans to the KO allele (Supplemental Fig. S1B). The cuc P-element insertions appear to be alleles of miR-275–305. Expression of CG5261 was unaffected in the miR-275–305 deletion mutant, monitored by quantitative RT–PCR (qRT–PCR). (B) Posterior midguts were collected from 7-d-old adults and labeled to visualize ISCs (Dl-Gal4>UAS-RFP; red). esg-GFP labels both ISCs and EB cells (green). Nuclei were labeled with DAPI (blue). (Top panel) KO/+ control. (Bottom panel) KO/KO mutant. (Histogram) Dl-positive and esg-positive cells were counted from seven midgut samples of each genotype, represented as average percent of total cells. ANOVA: P = 0.008 comparing the number of Dl-Gal4 cells in KO/KO versus control; P = 0.024 comparing esg-Gal4 cells in KO/KO versus control. Persistence of Dl-Gal4-driven RFP expression might lead us to underestimate the proportion of esg-GFP-positive cells that are EBs but would not affect the comparison between genotypes. (C) EdU incorporation in KO/+ control and KO/KO mutant midguts. esg-Gal4 was used to label ISCs and EB cells (green). Anti-EdU (red) labels cells that underwent DNA synthesis during the 30-min exposure to EdU. (D,E) MARCM clones in the midgut labeled with GFP. (D, left) Control clones. (Right) KO/KO mutant clones. (E) Mutant clones contained more cells on average.

Figure 3.

miRNA sponges distinguish between miR-275 and miR-305 function. (A) esg-Gal4 was used to direct UAS-GFP alone or together with UAS sponge transgenes that deplete miR-275 or miR-305. Depletion of miR-305 increased the number of esg-Gal4-expressing ISCs and EB cells. ANOVA: P = 0.4 comparing control with miR-275 sponge and P < 0.0001 comparing control with miR-305 sponge. (B) Dl-Gal4 was used to express UAS-miR-275 or UAS-miR-305 in ISCs in the KO/KO mutant background. Dl-Gal4>UAS-GFP cells (green). DAPI (red). ANOVA: P = 0.5 comparing control with UAS-miR-275; P < 0.0001 comparing control with UAS-miR-305.

Figure 4.

miR-305 acts in ISCs. (A–C) Images of posterior midguts expressing a miR-305 sensor transgene (green). Samples were labeled with anti-Dl to visualize ISCs (red) and with DAPI (blue). (A) Normal control, surface view. ISCs and EB cells are diploid and have small nuclei. (Arrows) A Dl-expressing ISC; (arrowheads) adjacent EBs. miR-305 sensor GFP levels were lower in the nuclei of the small cells. This difference disappeared in the miR-305 mutant background (Supplemental Fig. S2). (B) Optical cross-section showing an adjacent pair of basally located ISCs and EB cells. miR-305 sensor GFP levels were similar to background GFP levels in the ISCs but were detectable above background in the EB cell. (C) miR-305 sensor GFP levels in a 5-d-old KO/Df mutant midgut. The arrowhead indicates a Dl-expressing ISC. The level of GFP expression is comparable in the ISCs and other cells in the miR-275–305 KO/Df mutant combination. Note the dysplastic appearance of the gut, with mispositioning of the normally basally located ISCs and evidence of Dl expression in partially differentiated cells (partially endoreplicated), features that are more typically found in older flies (see also Supplemental Fig. S2). (D, top row) Summary of the cell type specificity of the Gal4 drivers. (Middle row) Gal4-driven expression of the UAS-miR-305 sponge transgene to selectively deplete miR-305 in ISCs and EB, EC, and EB cells. ISCs were labeled with anti-Dl (red). (Bottom row) Gal4 driver controls without the UAS-miR-305 sponge. The ratio of Dl⁺/total cells is shown in the bottom left corner for each genotype (average ± standard deviation [SD] from counts of eight midguts). The difference between miR-305 sponge-expressing and control was significant for Dl-Gal4 (P = 0.0011, Mann Whitney test). The differences were not significant with the other Gal4 drivers (see Supplemental Fig. S3).

Figure 5.

miR-305 acts via regulation of insulin and Notch pathways in ISCs. (A) Predicted target sites for miR-305 in the 3′ UTRs of the mRNAs encoding InR, PI3K, and Hairless. Residues changed in the target site mutant UTR reporters are shown in red. (B) Effect of miR-305 on luciferase reporters containing the 3′ UTRs of the indicated transcripts. (Ctl) Control luciferase reporter with an SV40 3′ UTR; (UTR) intact 3′ UTR reporters; (Mut) target site-mutated version of the UTR reporter. In all cases, cells were transfected to express miR-305 under the control of the tubulin promoter. Data are the average of four experiments ± SD. P < 0.01 comparing InR, PI3K, or Hairless UTRs with the control UTR; P < 0.025 comparing intact and mutant PI3K UTRs; P < 0.01 for InR and Hairless (Student’s t-test, two-tailed, unequal variance). (C) ISC RNA was TU-tagged by expression of UPRT under Dl-Gal4 in the adult midgut. Dl-Gal4 was combined with Gal80^ts. Animals were reared at 18°C to keep Gal4 inactive and shifted to 29°C at day 5 and aged for 2 d before TU incorporation. InR, PI3K, and Hairless transcript levels were measured by qRT–PCR. Additional replicates are provided in Supplemental Figure S5. (D) esg-Gal4 was used to direct transgene expression in the ISCs and EB cells. esg-Gal4 was combined with Gal80^ts. Animals were reared at 18°C to keep Gal4 inactive and shifted to 29°C at day 2 and aged for 5 d. esg-Gal4-expressing cells (green). (Red) Nuclei labeled with DAPI. (Scatter plot) Ratio of esg-Gal4 cells to total cells. Data were analyzed by ANOVA. (E) Dl-Gal was used to direct transgene expression in the ISCs. Dl-Gal4 was combined with Gal80^ts. Animals were reared at 18°C to keep Gal4 inactive and shifted to 29°C at day 2 and aged for 5 d. The ratio of Dl-Gal4 cells to total cells is shown. Data were analyzed by ANOVA. (F) Dl-Gal was used to direct transgene expression in the ISCs. (Panel 1) Dl-Gal4 + UAS-GFP. (Panel 2) Dl-Gal4 + UAS-GFP + UAS-PI3K. (Panel 3) Dl-Gal4 + UAS-GFP + UAS-Hairless. Dl-Gal4-expressing ISCs (green). Nuclei were labeled with DAPI (red). (Scatter plot) Ratio of Dl-Gal4-expressing to total cells. Data were analyzed by ANOVA.

Figure 6.

Regulation of miR-305 by the insulin pathway. (A) qRT–PCR to measure the unprocessed miR-305 primary transcript in RNA isolated from ISCs by TU tagging. The level of miR-305 primary transcript increased versus the fed control in two independent experiments. Flies were 7 d of age at the time of TU labeling. (B,C) miR-305 sensor transgene expression. Dl-Gal4 was used to direct UAS-RFP in the ISCs (arrows). (B) Control. (C) Dl-Gal4 + UAS-RFP + UAS-PI3K or UAS-PTEN^RNAi transgenes were used to activate the IIS pathway in the ISCs and were examined at 7 d of age. (D) Ratio of Dl⁺ ISCs that showed miR-305 sensor GFP expression versus those without detectable levels of GFP for the experiment in B and C. ANOVA: P < 0.0001 for UAS-PI3K and for UAS-PTEN-RNAi compared with the control.

Figure 7.

Regulation of miR-305 by the insulin pathway. (A) Dl-Gal4 was used to drive expression of the UAS-miR-305 sponge or a control UAS-GFP transgene. Dl-Gal4 was combined with Gal80^ts. Animals were raised at the permissive temperature to maintain low Gal4 activity, shifted to 29°C at 7 d of age to activate Dl-Gal4 in the ISCs, and aged under fed or starved conditions for 5 d. Representative images are shown. (B) Ratio of Dl-Gal4-expressing cells to total cells (n = 10 midguts for each genotype and treatment). Horizontal lines show median ± SD. Data were analyzed by ANOVA.