To assess the neurodevelopmental consequences resulting from KDM5 loss, we examined MB morphology of animals that were homozygous for a kdm5 null allele, kdm5¹⁴⁰ (Drelon et al., 2018; Drelon et al., 2019). As homozygous kdm5¹⁴⁰ animals fail to eclose from their pupal cases, we performed our immunohistochemical analyses on pharate adults, which externally appear indistinguishable from wild-type animals (Drelon et al., 2018). Following a well-established classification scheme used by others (Gombos et al., 2015; Kelly et al., 2016; Michel et al., 2004), MB defects were categorized as impacting MB growth and/or guidance, with the former defined by a stunted, overextended, or absent lobe and the latter by a misprojected lobe. Staining using an antibody specific to the NCAM-like cell adhesion molecule fasciclin 2 (Fas2) revealed highly penetrant MB abnormalities, with ~70% of animals exhibiting growth or guidance defects of the α/β lobes (Figure 1A, B). Interestingly, the predominant phenotype observed in our analysis was overextension of the β lobes and/or stunting of the α lobes. Because animals specifically lacking KDM5 demethylase activity have phenotypically normal α/β lobes (Zamurrad et al., 2018), our data demonstrate that KDM5 is required for correct MB morphology, independent of its canonical enzymatic function.

To assist in our functional understanding of KDM5, we sought to investigate its expression throughout central nervous system (CNS) development. To facilitate these analyses, we used CRISPR/Cas9 to generate a strain containing a 3xHA-tag fused to the endogenous locus of kdm5 that results in wild-type levels of protein expression (Figure 1C). We also confirmed that KDM5:HA was expressed in adult MB Kenyon cells by co-expression of a nuclear membrane-localized GFP reporter, UNC-84:GFP, with the MB driver OK107-Gal4 (Figure 1D). Immunostaining of the of wandering third-instar larvae revealed that KDM5 was localized to cortical nuclei while absent from neuropil-rich regions marked by the ubiquitous presynaptic active zone marker Bruchpilot (Brp) (Figure 2A). KDM5:HA localized to cortical nuclei across a variety of cell types, including neurons (Figure 2B), neuroblasts (NBs), and presumptive GMCs (Figure 2C). We also examined KDM5:HA expression in the adult brain, where KDM5:HA appeared to be similarly localized to cortical nuclei while absent from neuropil-rich regions, such as the antennal lobes and both the dorsolateral and ventrolateral protocerebra (Figure 2D, E). Given the broad expression pattern of KDM5:HA within a variety of cell types, KDM5 may regulate a range of neural processes, from NPC division and axonal growth to neuronal maturation and function.

To define the functional requirements of KDM5 during MB development, we utilized an inducible kdm5 shRNA transgene that we and others have shown effectively reduces KDM5 levels (Chen et al., 2019; Liu et al., 2014; Navarro-Costa et al., 2016). We first knocked down kdm5 broadly within all MBNBs, MB-GMCs, and Kenyon cells throughout development using OK107-Gal4 (Figure 3A, B). This resulted in significant neuromorphological defects of the α/β MB lobes, with the major phenotype being an overextension of the β lobes across the midline (Figure 3C, D). We next assessed the consequences of kdm5 depletion within distinct and overlapping subsets of mature Kenyon cells. Knocking down kdm5 using the mature Kenyon cell-specific Gal4 drivers C708a-, c305a-, H24-, and 201Y-Gal4 reduced KDM5 levels, but failed to produce significant gross morphological defects of the α/β lobes (Figure 3D, Figure 3—figure supplement 1). KDM5 is therefore unlikely to be required exclusively within mature, post-mitotic Kenyon cells for proper α/β lobe development.

In contrast, knockdown of kdm5 using two independent NB-restricted Gal4 driver lines, worniu-Gal4 (wor-Gal4) and inscuteable-Gal4 (insc-Gal4), resulted in profound α/β lobe defects (Figure 3C, D). Notably, a significant proportion of these brains simultaneously displayed both growth and guidance defects of the α/β lobes. Although wor-Gal4 expression is NB specific, KDM5 depletion and GFP expression continued to be observed in presumptive GMCs and post-mitotic cells surrounding each NB (Figure 3E). This is likely attributed to perdurance of the Gal4 activator protein and/or the kdm5 shRNA. KDM5 could therefore be required within the NB, GMC, or even post-mitotically within the immature neuron for proper α/β Kenyon cell neurodevelopment.

To increase the resolution of our analysis, we leveraged the restricted expression pattern of R71C09-Gal4, which drives expression within GMCs and early-born neurons of the Drosophila larval CNS (Figure 4A; Li et al., 2014; Marshall and Brand, 2017; Aughey et al., 2018). R71C09-Gal4 did not appear to drive expression within NBs, including MBNBs (Figure 4A). When kdm5 was knocked down using R71C09-Gal4, KDM5 was dramatically depleted within presumptive GMCs and immature neurons (Figure 4B). We additionally noted that the axons of newly born Kenyon cells, which traverse through the core fibers of the MB pedunculus, were also labeled by an mVenus reporter when driven by R71C09-Gal4 (Figure 4C). Consistent with these observations, the labeled neurons were of α/β Kenyon cell origin, the last MB cell subtype to be born prior to eclosion (Figure 4C). Depletion of kdm5 using this driver resulted in profound α/β lobe defects, indicating that KDM5 is functionally required within GMCs and immature MB neurons for proper axonal development (Figure 4D). To assess sufficiency, we used R71C090-Gal4 to re-express kdm5 in GMCs and immature neurons of kdm5¹⁴⁰ animals and found a significant recue of the defects we had previously observed (Figure 4E).

To understand how KDM5 functions within GMCs and immature neurons to regulate MB development, we used TaDa to carry out in vivo transcriptional profiling in a cell-type-specific and temporally controlled manner (Marshall et al., 2016a; Southall et al., 2013). By expressing Escherichia coli-derived DNA adenine methyltransferase (Dam) fused to Drosophila RNA polymerase II (Pol II) in MB-GMCs and immature Kenyon cells via the Gal4/UAS system, we surveyed genomic regions with altered Pol II occupancy upon depletion of KDM5. Dam-Pol II methylates adenine residues (m6A) at GATC motifs in close proximity to Pol II-occupied DNA, providing a surrogate for actively transcribed loci when normalized to expression of Dam alone (Doupé et al., 2018; Southall et al., 2013). As such, we refer to genomic loci with altered Pol II occupancy as differentially expressed genes (DEGs).

We expressed Dam-Pol II or Dam in GMCs and immature neurons of late third-instar larvae and pupae of homozygous kdm5¹⁴⁰ animals using the R71C09-Gal4 driver (Figure 5A). Induction of the TaDa system during this developmental window predominantly resulted in the transcriptomic profiling of MB-GMCs and immature α/β Kenyon cells as these belong to one of the few cell lineages undergoing extensive neurogenesis during this time period and express R71C09-Gal4 (Technau and Heisenberg, 1982; Ito and Hotta, 1992; Truman and Bate, 1988). Importantly, the temperature shifts required to induce the TaDa system in a kdm5¹⁴⁰ animals did not alter the frequency or type of structural MB defects observed (Figure 5—figure supplement 1).

Carrying out TaDa from five biological replicates, we identified a total of 636 DEGs using a statistical cutoff of FDR < 0.05, 438 of which were downregulated and 198 of which were upregulated (Figure 5B; Supplementary file 1). Consistent with previously published mRNA-seq data in Drosophila and other organisms, loss of KDM5 resulted in moderate changes to gene expression, with an average 2.4-fold decrease among downregulated genes and 2.1-fold increase among upregulated genes (Drelon et al., 2018; Iwase et al., 2016; Liu et al., 2014; Liu and Secombe, 2015; Lloret-Llinares et al., 2012; Lopez-Bigas et al., 2008; Lussi et al., 2016; Zamurrad et al., 2018). Thus, KDM5 likely functions within this cellular subpopulation to fine-tune gene expression across a number of pathways that are critical for MB development.

To determine if these dysregulated genes were enriched for functional categories, we used the program GeneOntology (The Gene Ontology Consortium, 2019; The Gene Ontology Consortium, 2000; The Gene Ontology Consortium, 2019), which utilizes the PANTHER Classification System (Mi et al., 2013) to mine gene ontology (GO) information. This revealed significant enrichment for categories involved in GMC fate determination (e.g., pros, brat, mira, cas), neural precursor cell proliferation (e.g., pros, wor, ase, esg, sna, E(spl)mγ-HLH, E(spl)mβ-HLH), axon guidance (e.g., pros, dac, brat, Hr51, chic, tap, pdm3, nerfin-1, lola), and cytoplasmic translation, among others (Figure 5C; Supplementary file 2). The finding that ribosomal protein genes, such as RpS2, RpS24, RpS28b, RpL3, Rpl23, RpL39, and RpL41, were altered in our kdm5¹⁴⁰ TaDa data is consistent with our previously reported RNA-seq data from adult heads of demethylase-dead kdm5 strains showing that genes required for cytoplasmic translation were affected (Zamurrad et al., 2018). Interestingly, however, knockdown of the transcription factor myc, which regulates the expression of most ribosomal protein genes (Bellosta and Gallant, 2010; Grewal et al., 2005), in MBNBs, GMCs, and Kenyon cells using the OK107-Gal4 driver, did not result in any gross morphological defects of the α/β lobes (Figure 5—figure supplement 2). These data show that our observed MB phenotypes are not due to the regulation of cytoplasmic translation by KDM5.

H3K4me3 marks have been shown to be associated with regions of accessible chromatin, which often contain regulatory DNA sequences such as promoters and enhancers (Bhaumik et al., 2007; Li et al., 2012; Park et al., 2020; Thurman et al., 2012). If KDM5 functions independent of its H3K4me3 demethylase activity within GMCs and immature neurons to regulate MB development, loss of KDM5 may have only minimal effects on chromatin accessibility within this cellular subpopulation. To measure changes to chromatin landscape upon KDM5 loss, we utilized chromatin accessibility profiling using targeted DamID (CATaDa) (Aughey et al., 2018; Aughey et al., 2019). Dam methylates regions of highly accessible chromatin, thus providing an in vivo surrogate for chromatin accessibility (Aughey et al., 2018; Aughey et al., 2019). Comparing Dam methylation levels in R71C09-Gal4 expressing cells between kdm5¹⁴⁰ and wild-type animals revealed minor changes to chromatin accessibility at a cutoff of FDR < 0.01 (Figure 5—figure supplement 3A, B). Overlapping reduced accessibility regions with the downregulated genes from our TaDa analysis revealed only 17 genes associated with reduced chromatin accessibility in GMCs and immature neurons at a cutoff of FDR < 0.01 (Figure 5—figure supplement 3C). Although this represented a significantly enriched proportion of the total number of genes associated with changes in chromatin accessibility (p=4.787E-09), they were not enriched via GO analysis for any biological categories. Because animals lacking demethylase activity do not present with gross MB morphological defects (Zamurrad et al., 2018), KDM5 functions to regulate MB morphology largely via demethylase-independent transcriptional mechanisms that do not dramatically alter chromatin accessibility.

Since kdm5¹⁴⁰ animals have chronic loss of kdm5 in all tissues, the gene expression changes we observed may be the result of both cell-autonomous and non-cell-autonomous effects. To look directly at the KDM5-regulated transcriptome in GMCs and early-born neurons, we performed TaDa in animals expressing a kdm5 shRNA transgene under the control of R71C09-Gal4. In addition to utilizing a Dam-only control, we also accounted for activation of the RNAi pathway by expressing a scramble shRNA transgene under identical conditions. To ensure that changes to gene expression were reflective of the severe MB phenotypes we had previously observed, we induced the expression of kdm5 shRNA and dam-pol II for 6 days, beginning during the early second-instar larval (L2) stage, to allow for sufficient depletion of KDM5 within MB-GMCs and immature α/β Kenyon cells (Figure 5D). This temporally targeted knockdown strategy recapitulated the adult α/β lobe phenotypes we had observed for constitutive kdm5 knockdown with R71C09-Gal4 (Figure 5—figure supplement 4). From three biological replicates and a statistical cutoff of FDR < 0.05, we identified 1069 DEGs, 659 of which were downregulated and 410 of which were upregulated (Figure 5E; Supplementary file 3). Compared to the kdm5¹⁴⁰ TaDa, the greater number of DEGs from the kdm5 RNAi TaDa could be attributed to the extended duration of the kdm5 knockdown and induction of Dam-Pol II. Additionally, we observed larger changes to gene expression, with an average 5.5-fold decrease among downregulated genes and a 4.3-fold increase among upregulated genes (Figure 5E; Supplementary file 3). This could be ascribed to the acute loss of kdm5 expression caused by RNAi-mediated depletion, which would decrease the likelihood of compensatory changes occurring.

To obtain a list of high-confidence genes regulated by KDM5, we compared the DEGs found in our kdm5¹⁴⁰ and kdm5 knockdown TaDa datasets. This revealed a total of 335 overlapping dysregulated genes, 319 of which were up- or downregulated in both datasets (r = 0.68, p<0.0001) (Figure 5F, G; Supplementary file 4). GO analysis of the 319 similarly dysregulated genes revealed an even greater enrichment in categories related to GMC fate determination (pros, mira, cas), neural precursor cell proliferation (wor, esg, pros, mira, sna), and axon guidance (pros, chic, lola, Oli, tap, Nrt, nerfin-1, Hsc70-4), among others (Figure 5H; Supplementary file 4). Interestingly, only 17 of these genes overlapped with our previously published kdm5¹⁴⁰ wing disc RNA-seq data (Drelon et al., 2018; Figure 5—figure supplement 5). This nonsignificant overlap suggests that KDM5 functions in a predominantly tissue- and cell-specific manner to regulate the expression of downstream targets.

To identify biologically relevant functional networks for the 319 KDM5-regulated, high-confidence genes, we performed gene network and community clustering analyses (Morris et al., 2011; Shannon et al., 2003). These analyses revealed seven discrete networks with greater than two nodes (Figure 6A). Of these, two networks were highly enriched for genes implicated in MB development (toy and tap), neural precursor cell proliferation (pros, ase, wor, mira, sna, esg, E(spl)mγ-HLH, and E(spl)mβ-HLH), and axon growth and guidance (pros, nerfin-1, tap, emc, Nrt, and elav).

To determine which genes affected by loss of KDM5 were likely to be direct targets, we generated a transgenic Drosophila strain expressing Dam fused to KDM5 under the control of UAS (UAS-dam-kdm5). To match the developmental window utilized for our RNA pol II TaDa analyses, we expressed Dam-KDM5 or Dam in late third-instar larvae and pupae using the R71C09-Gal4 driver. From six biological replicates, we identified a total of 4399 genes with KDM5 peaks using a statistical cutoff of FDR < 0.01 (Supplementary file 5). Of these, 120 genes overlapped with the 319 genes we previously identified as being KDM5-regulated and represented a significant enrichment (Figure 6B, C; p=6.396E-07). Of particular interest was one gene, prospero (pros), which encodes a homeodomain-domain-containing transcription factor that is a critical regulator of axon pathfinding and growth (Doe et al., 1991; Froldi et al., 2015; Tea et al., 2010; Vaessin et al., 1991). Our Dam-KDM5 occupancy data showed that KDM5 binds to TSS within the pros locus (Figure 7A) and, together with our Pol II TaDa data, suggest that KDM5 directly regulates pros expression.

If KDM5 functions upstream of Pros to regulate its expression, it is likely that KDM5 may also indirectly regulate the expression of Pros targets. To explore this possibility, we compared our analyses with published Dam-Pros occupancy data within GMCs and immature neurons (Liu et al., 2020; Gene Expression Omnibus [GEO] accession number GSE136413). Strikingly, 45% (142 of 319) of KDM5-regulated genes were bound by Pros (p=2.20E-16, Fisher’s exact test) (Figure 7B, C; Supplementary file 6). In addition, a majority of Pros-bound targets were not co-bound by KDM5 (Figure 7B). These data suggest that Pros could be a key mediator of KDM5 function in GMCs and immature neurons. Our data also implicate Pros in the activation of genes as the majority of Pros-bound genes were downregulated upon KDM5 depletion (Figure 7C).

While roles for Pros in regulating NB and GMC asymmetric division are well-described (Choksi et al., 2006; Doe et al., 1991; Vaessin et al., 1991), its role in MB development remains unexplored. R71C09-Gal4-mediated knockdown of pros resulted in early larval lethality, preventing us from examining MB phenotypes. However, depleting Pros within the smaller cellular population of MB progenitor and Kenyon cells using OK107-Gal4 allowed for the formation of viable adults. Analyses of these adult brains revealed a complete disruption of MB morphology, with both growth and guidance defects present in all animals observed (Figure 7D, E).

As kdm5 or pros depletion within MB progenitor and Kenyon cells each independently resulted in significant axonal defects, we next assessed whether kdm5 and pros functioned within the same genetic pathway. To test this, we generated animals heterozygous for null alleles of both kdm5 and pros, thereby simultaneously reducing levels of both proteins. Animals heterozygous for either kdm5¹⁴⁰ or pros¹⁷ did not display significant MB defects. In contrast, adults that were transheterozygous for kdm5¹⁴⁰ and pros¹⁷ displayed MB abnormalities (Figure 7F, G). These data are consistent with KDM5 and Pros functioning synergistically to affect MB development. KDM5 is therefore likely to function within GMCs and immature neurons to modulate levels of pros and its downstream targets, thereby regulating MB development (Figure 7H).

A detailed list of the genotypes of the flies used in each figure is included in the Key Resources Table in the Appendix.

All GAL4 and GMR GAL4 lines were generated at the Janelia Research Campus/HHMI (Pfeiffer et al., 2008; Jenett et al., 2012) and were obtained from the Bloomington Drosophila Stock Center (BDSC) at Indiana University.

The following transgenes were used: UAS-kdm5-shRNA (RRID:BDSC_35706), 20XUAS-IVS-CsChrimson.mVenus (RRID:BDSC_55136), UAS-tub-GAL80^ts (RRID:BDSC_7108), UAS-pros-dsRNA (RRID:BDSC_42538), and UAS-myc-dsRNA (VDRC stock# KK106066). The UAS-LT3-dam and UAS-LT3-dam-pol II lines were kindly shared by Andrea Brand (U. Cambridge, Gurdon). The 5XUAS-unc84-2Xgfp line was kindly shared by Gilbert Henry and Todd Laverty (HHMI Janelia). The kdm5¹⁴⁰ mutant allele and UAS-kdm5 transgene have been previously described (Drelon et al., 2018; Secombe et al., 2007).

To tag the endogenous kdm5 locus with three in-frame HA epitope tags, we used CRISPR/Cas9-mediated knock-in. The HA epitopes and the homology arms, carrying a synonymous mutation for the PAM sequence, were PCR amplified from a clone containing the kdm5 locus from the wild-type strain w¹¹¹⁸ (pattB.gkdm5; Navarro-Costa et al., 2016). The donor DNA repair template consisted of three PCR fragments cloned, by In-Fusion HD (Takara, Bio), into the pHD-ScarlessDsRed vector (Addgene plasmid #51434). AarI and SapI enzymes were used to linearize the plasmid. The flyCRISPR Optimal Target Finder tool was used to select the target genomic cleavage and design the gRNA (http://targetfinder.flycrispr.neuro.brown.edu). The top and bottom oligos for the gRNA were phosphorylated and annealed using T4 polynucleotide kinase (NEB) and then cloned into the pU6 vector (Addgene plasmid #53062), which was linearized with BpiI (NEB). The gRNA and donor DNA were sent to the BestGene for injection into embryos expressing Cas9 in the germline (y¹, M{vas-Cas9.RFP-}ZH-2A, w¹¹¹⁸; RRID:BDSC_55821). To remove the DSRed cassette, transformed flies were balanced and then crossed with flies carrying piggyBac transposase (RRID:BDSC_32073). Flies lacking expression of RFP were recovered and homozygous stocks were sequenced for the kdm5 locus. The correct removal of the DSRed cassette and presence of the 3xHA tag were confirmed by PCR sequencing and western blot.

To generate UAS-LT3-dam-kdm5 flies, kdm5 cDNA was PCR amplified from pB-Lid and cloned into the pUAST-attB-LT3-NDam vector (Marshall et al., 2016a). pUAST-attB-LT3-NDam was linearized with Notl and Xbal, and the cut plasmid and PCR insert were assembled by In-Fusion HD (Takara, Bio). The construct was sequenced and sent to BestGene for injection into embryos carrying the attP2 landing site (RRID:BDSC_8622).

For fixation of late third-instar larval brain tissue, the CNS was first removed via dissection in cold phosphate buffered saline (PBS, pH 7.4) and allowed to incubate in fixation buffer (4% paraformaldehyde in PBS) at room temperature (RT) for 40 min. After washing for three cycles of 15 min in PBS with 0.2% Triton X-100 (0.2% PBT), brains were incubated with blocking buffer (5% normal goat serum in 0.2% PBT) for 30 min at RT. The brains were then incubated with primary antibodies for 2 days at 4°C and washed for three cycles of 15 min in 0.2% PBT. Secondary antibodies were then added and brains were incubated for 2 days at 4°C. After being washed in PBS for three cycles of 15 min at RT, brains were incubated in a drop of Vectashield mounting medium (Vector Laboratories, H-1000) or DAPI Fluoromount G (SouthernBiotech, OB010020) overnight at 4°C. Brains were then mounted on glass slides (Superfrost Plus, Fisherbrand), flanked by glass spacers and covered with a final glass coverslip before being used for image analysis.

A similar protocol was followed for immunostaining of pharate adult and 3- to 5-day-old adult brain tissue with the following exceptions. Pharate adult heads or whole adult animals were allowed to incubate in fixation buffer (4% paraformaldehyde in 0.2% PBT) at 4°C for 3 hr. Heads or whole animals were then washed in 0.2% PBT for three cycles of 15 min at RT. Brains were then dissected from the fixed tissue and blocked for 30 min at RT. Antibody incubation and mounting was identical to that described above.

The following primary antibodies were used: mouse anti-brp (1:50, DSHB cat# nc82, RRID:AB_2314866), mouse anti-Fas2 (1:25, DSHB cat# 1D4 anti-fasciclin II, RRID:AB_528235), rabbit anti-HA (1:100, Cell Signaling Technology cat# 3724; RRID:AB_1549585), mouse anti-HA (1:100, Cell Signaling Technology cat# 2367, RRID:AB_10691311), and rat anti-Dpn (1:100, Abcam cat# ab195173; RRID:AB_2687586). Primary antibodies were prepared in 5% NDS/0.2% PBT with 0.02% NaN₃. The following secondary antibodies were used: goat anti-mouse Alexa-488 (1:500, Thermo Fisher Scientific cat# A32723; RRID:AB_2633275), goat anti-mouse Alexa-568 (1:500, Thermo Fisher Scientific cat# A11004; RRID:AB_2534072), goat anti-rabbit Alexa-488 (1:500, Thermo Fisher Scientific cat# A11034; RRID:AB_2576217), goat anti-rabbit Alexa-568 (1:500, Thermo Fisher Scientific cat# A11004, RRID:AB_2534072), and goat anti-rat Alexa 568 (1:500, Thermo Fisher Scientific cat# A11077; RRID:AB_2534121). All secondary antibodies were diluted in 5% NDS/0.2% PBT.

Western analysis was carried out as previously described (Drelon et al., 2019). Briefly, 3- to 5-day-old adult fly heads were homogenized in 2× NuPAGE LDS sample buffer, sonicated for 10 min, treated with DTT, run on a 4–12% Bis-Tris 1 mm gel, and transferred to a PVDF membrane. The following primary antibodies were used: rabbit anti-KDM5 (1:1000, Secombe J; Genes Dev. 2007; cat# lid, RRID:AB_2569502) and mouse anti-HA (1:1000, Cell Signaling Technology cat# 2367, RRID:AB_10691311). The following secondary antibodies were used: IRDye 680RD donkey anti-mouse IgG (1:8000; LI-COR Biosciences cat# 925-68072, RRID:AB_2814912) and IRDye 800CW donkey anti-rabbit IgG (1:8000; LI-COR Biosciences cat# 926-32213, RRID:AB_621848). Blots were scanned and processed using a LI-COR Odyssey Infrared scanner.

All tissue images were taken on a Leica SP8 confocal microscope using either a ×20 air lens (N.A. = 0.75 air, W.D. = 0.64 mm) or a ×63 immersion lens (N.A. = 1.4 oil, W.D. = 0.14 mm). All MB confocal stacks were taken under either ×2 or ×2.5 zoom, in a 1024 × 1024 configuration and using 1 μm resolution. Image stacks were processed with Figi (ImageJ). Figures were composed using Microsoft Powerpoint.

To profile Pol II occupancy in GMCs and immature neurons of kdm5¹⁴⁰ pupae, kdm5¹⁴⁰/CyO:gfp; UAS-LT3-dam or kdm5¹⁴⁰/CyO:gfp; UAS-LT3-dam-pol II flies were crossed with kdm5¹⁴⁰/CyO:gfp; R71C09-Gal4 flies and were allowed to lay eggs overnight for 12–14 hr at 25°C. Embryos were then moved to 18°C for a period of 16 days, and GFP-negative wandering third-instar larvae were transferred to a restrictive temperature of 29°C for 3 days to induce the expression of the UAS-dam and UAS-dam-pol II transgenes. As a parallel control, kdm5 wild-type animals carrying UAS-LT3-dam or UAS-LT3-dam-pol II were crossed with R71C09-Gal4 and were also allowed to lay eggs for 24 hr at 25°C. Embryos were then moved to 18°C for 9 days and subsequently transferred to 29°C for 3 days.

For kdm5 knockdown experiments, flies carrying either UAS-LT3-dam; UAS-kdm5_shRNA or UAS-LT3-dam-pol II; UAS-kdm5_shRNA were crossed with flies carrying UAS-tubulin-Gal80^ts; R71C09-Gal4. Parallel crosses were also performed with flies bearing a scrambled shRNA sequence in place of kdm5 shRNA. All flies were allowed to lay eggs overnight for 12–14 hr at 25°C. Embryos were then transferred to 18°C for 4 days and then subsequently transferred to 29°C for 6 days.

To determine genomic regions directly bound by KDM5 within GMC and immature neurons, UAS-LT3-Dam or UAS-LT3-Dam-kdm5 animals were crossed with flies carrying tubulin-Gal80^ts; R71C09-Gal4 and allowed to lay eggs for 12–14 hr at 25°C. Embryos were transferred to 18°C for 9 days and subsequently moved to 29°C for 3 days.

Tissue processing for all TaDa experiments was performed as previously described in Marshall et al., 2016a with the following modifications. A total of 40 pupae of each genotype were homogenized in 500 mM EDTA followed by DNA extraction using the Zymo Quick-DNA Miniprep Plus Kit. DpnI digestion, PCR adaptor ligation, DpnII digestion, and PCR amplification were performed as described. DNA was sonicated using a Diagenode Bioruptor for 8–10 cycles (5 min at high power, 30 s on/30 s off) and analyzed using an Agilent Bioanalyzer. DamID adaptor removal and DNA cleanup were performed as previously described (Marshall et al., 2016a), and samples were submitted to BGI for sequencing.

Sequencing libraries were prepared at BGI Genomics following a ChIP-seq workflow. DNA fragments were first end-repaired and dA-tailed using End Repair and A-Tailing enzyme. Adaptors were then ligated for sequencing and ligated DNA purified using AMPure beads. DNA was then PCR amplified with BGI primers for eight cycles and PCR purified with AMPure beads. DNA was then homogenized, circularized, digested, and again purified. DNA was then prepared into proprietary DNA nanoballs (DNB) for sequencing on a BGISEQ-500 platform with 50 bp read length and 20 M clean reads.