E93 Integrates Neuroblast Intrinsic State with Developmental Time to Terminate MB Neurogenesis via Autophagy

Abstract

Most neurogenesis occurs during development, driven by the cell divisions of neural stem cells (NSCs). We use Drosophila to understand how neurogenesis terminates once development is complete, a process critical for neural circuit formation. We identified E93, a steroid-hormone-induced transcription factor that downregulates phosphatidylinositol 3-kinase (PI3K) levels to activate autophagy for elimination of mushroom body (MB) neuroblasts. MB neuroblasts are a subset of Drosophila NSCs that generate neurons important for memory and learning. MB neurogenesis extends into adulthood when E93 is reduced and terminates prematurely when E93 is overexpressed. E93 is expressed in MB neuroblasts during later stages of pupal development only, which includes the time when MB neuroblasts normally terminate their divisions. Cell intrinsic Imp and Syp temporal factors regulate timing of E93 expression in MB neuroblasts, and extrinsic steroid hormone receptor (EcR) activation boosts E93 levels high for termination. Imp inhibits premature expression of E93 in a Syp-dependent manner, and Syp positively regulates E93 to promote neurogenesis termination. Imp and Syp together with E93 form a temporal cassette, which consequently links early developmental neurogenesis with termination. Altogether, E93 functions as a late-acting temporal factor integrating extrinsic hormonal cues linked to developmental timing with neuroblast intrinsic temporal cues to precisely time neurogenesis ending during development.

Keywords: E93; PI3-kinase; autophagy; ecdysone; mushroom body; neural stem cell; neuroblast; neurogenesis; steroid hormone; temporal factors.

Figure 1:. E93 is necessary and sufficient to eliminate MB neuroblasts and terminate MB neurogenesis.

(A) Schematic, timing of MB versus non-MB neuroblast (NB) elimination. (B) Top schematic, highlighting position of the MB calyx, used as a landmark in locating MB neuroblasts. Below, maximum intensity projection of the region outlined. Arrows indicate MB neuroblasts. (C) Average number of MB neuroblasts per brain hemisphere in 1-day-old adults. Numbers on bars indicate number of brain hemispheres scored for each of the indicated genotypes listed below. Error bars s.e.m. (D) Times of heat-shock treatments (arrow) for GAL4 flip out experiments with GAL4 flip out cassette below (see text). (E) Left, a MB neuroblast E93 RNAi clone in a 1-day-old adult after heat shock at 0 hrs. APF. Right, a control MB neuroblast, heat shocked at 0 hours APF, fails to express E93RNAi. Control imaged at an earlier time to identify a Dpn-positive MB neuroblast, that is normally absent in adulthood, but present when E93 is knocked down. White brackets mark neuroblasts in this and all subsequent figures. (F) Percentage of MB neuroblast E93 RNAi clones with a Dpn-positive neuroblast in one-day-old animals. Time of heat shock treatments indicated below and number of clones scored in columns. (G,H) Maximum intensity projections of adult MB neuropil. Scale bar (B) 20μm and (E) 10μm. See also Figure S1 and Table S1.

Figure 2:. MB neuroblasts express E93 after non-MB neuroblasts, during later stages of pupal development.

(A) Dorsal view of a wild type larval brain hemisphere, anterior up. Below, greyscale image of same brain hemisphere, labeled with Scrib to outline NB membranes. Arrows indicate MB neuroblasts, asterisks indicate some non-MB neuroblasts. (B-G) Wild type neuroblasts from indicated time points (above) stained with markers listed within panel B. (H-I) E93 is reduced in a MB neuroblast E93RNAi clone (H) compared to a control MB neuroblast clone (I). Scale bar (A) 20μm and (B) 10μm. See also Table S1.

Figure 3:. Failure to downregulate PI3-kinase activity on time allows E93RNAi MB neuroblasts to persist into adulthood.

(A,J,O) Average number of MB neuroblasts per brain hemisphere over time. Column numbers indicate number of hemispheres scored. Error bars indicate standard deviation. White columns (A) within colored columns indicate average number of mitotic MB neuroblasts. (B,P) Box plots of MB neuroblast diameters. Numbers at bottom indicate number MB neuroblasts analyzed. *p values<.001, two-tailed Student’s t-tests. (C-G) Top, colored overlay with single channel greyscale image below of MB neuroblasts. Markers listed within panels, genotypes and time above. (H,I) Quantification of MB neuroblast Foxo fluorescence intensities. Column numbers equal number of MB neuroblasts assayed. (H) *p values<.001, two-tailed Student’s t-tests. (I) *p value=.0003, one-way ANOVA. (K-N) Colored overlay of MB neuroblasts. Scale bar (C,K) 10μm. See also Figure S2 and Table S1.

Figure 4:. E93 regulates autophagy in MB neuroblasts.

(A-I) Top, colored overlay of MB neuroblasts. Below, colored overlay of cropped, maximum intensity projection of MB neuroblast above, single channel greyscale images below. White arrows indicate autophagomes (mCh,GFP double positive) and arrowheads indicate autolysosomes (mCh). (A-H) Co-express UAS-GFP-mCh-Atg8, (I) UAS-GFP-Atg8. (J) Quantification of autophagosomes and autolysosomes or autophagosomes only (L) over time. Black tics represent individual MB neuroblasts. Total number of MB neuroblasts assayed at top of column, red lines denote mean. (K) Distribution of percentages of autolysosomes relative to total puncta in MB neuroblasts over time (M) Schematic of autophagy flux reporter. *p values<.001, two-tailed Student’s t-tests. Scale bar (A) 10μm. See also Figure S3, Table S1, and Video S1.

Figure 5:. MB neuroblasts persist long-term in the absence of E93 and inhibition of apoptosis.

(A-C) Colored overlay of MB neuroblasts. (D) Average number of MB neuroblasts per brain hemisphere. Column numbers indicate number of hemispheres scored. Error bars indicate S.E.M. Scale bar 10μm. See also Table S1 and Video S1.

Figure 6:. Imp, Syp, and EcR regulate E93 expression in MB neuroblasts.

(A,C) Schematic summarizing changing 20-hydroxyecdysone levels (A) and timing of Imp and Syp expression in relation to E93 (C). (B) Quantification of E93 nuclear fluorescence intensities. Numbers in columns indicate number of clones scored. (D,E,G,H,J,K,M,N,P,Q) Top, colored overlay with single channel greyscale images below of MB neuroblasts. (F,I,L,O,R) Quantification of E93 nuclear fluorescence intensities. Column numbers equal number of MB neuroblasts assayed or number of clones. (L,O,R) E93 nuclear fluorescence intensities normalized to control E93 nuclear fluorescence intensities at 48 APF (B). Error bars equal S.E.M. *p values<.001, two-tailed Student’s t-tests. Scale bar (D) 10μm. See also Figure S4 and Table S1.

Figure 7:. E93 terminates growth of SypRNAi MB neuroblasts.

(A) Average number of MB neuroblasts per brain hemisphere. Column numbers equal number of brain hemispheres scored. (B) Percentage of MB neuroblast clones with a MB neuroblast. Column numbers indicate number of clones scored and times below indicate time of heat shock treatment. (C) Box plots of MB neuroblast diameters. Numbers at bottom indicate number MB neuroblasts analyzed. (D,E) Colored overlay of MB neuroblast with greyscale image below. (F) Model summary, see text for details. *p values<.001, two-tailed Student’s t-tests. See also Table S1.