The histone demethylase Kdm5 controls Hid-induced cell death in Drosophila

Abstract

We conducted an EMS mutagenesis screen on chromosome arm 2L to identify recessive suppressors of GMR-hid-induced apoptosis in the Drosophila eye. Through this screen, we recovered three alleles of the lysine demethylase gene Kdm5. Kdm5, a member of the JmjC-domain-containing protein family, possesses histone demethylase activity towards H3K4me3. Our data suggest that Kdm5 specifically regulates Hid-induced cell death during development, as we did not observe control of Reaper- or Grim-induced cell death by Kdm5. Interestingly, GMR-hid-induced apoptosis is suppressed independently of Kdm5’s demethylase activity. Our findings indicate that Rbf and dMyc are necessary for Kdm5 mosaics to suppress GMR-hid-induced cell death. Moreover, Kdm5 mosaics failed to suppress apoptosis induced by a mutant form of Hid that is resistant to inhibition by Erk-type MAPK activity. Additionally, Kdm5 dominantly enhances the wing phenotype of an activated MAPK mutant. These results collectively suggest that Kdm5 controls Hid-induced apoptosis by regulating the Rbf, dMyc, and MAPK pathways.

Introduction

Programmed cell death is an essential biological process during development and homeostasis of multi-cellular organisms. Apoptosis is the major form of cell death and accounts for more than 90% of all cell death (Newton et al., 2024). Apoptosis is evolutionarily conserved and is critical for the health of the organism. Alterations in the rate of apoptosis are linked to cancer, auto-immune diseases and neurodegeneration (Kayagaki et al., 2024). Therefore, a detailed understanding of the control of apoptosis is necessary to develop strategies for treatment of these diseases.

Control of apoptosis comes down to the control of caspases, a class of highly specialized Cys proteases which execute apoptosis when activated (Shalini et al., 2015). In living cells, caspases are inhibited by inhibitor-of-apoptosis proteins (IAPs) which directly bind to caspases and inhibit their activity (Gyrd-Hansen and Meier, 2010). In response to apoptosis-inducing signals, IAPs are degraded and release caspases for further activation. In Drosophila, the IAP-antagonists Reaper, Hid and Grim stimulate the ubiquitylation and degradation of Drosophila IAP1 (DIAP1) resulting in the release of the initiator caspase Dronc (Caspase-9-ortholog in Drosophila) from DIAP1 inhibition (Orme and Meier, 2009). Free Dronc is activated by incorporation into the apoptosome by the Apaf-1-like protein Dark (Dorstyn et al., 2018). Subsequently, the apoptosome activates effector caspases such as DrICE and Dcp-1 (caspase-3-orthologs in Drosophila) (Kumar, 2007) which execute the apoptotic process.

Expression of the IAP-antagonist Hid in the Drosophila compound eye using the GMR promoter (GMR-hid) causes a strong eye ablation phenotype due to massive apoptosis induced by Hid in the eye imaginal disc during larval development (Figure 1A) (Fan and Bergmann, 2008; Grether et al., 1995). We have used the strong eye ablation phenotype of GMR-hid in chemical (EMS) mutagenesis screens to identify genes involved in the control of Hid-induced apoptosis. Initially, in dominant modifier screens of the GMR-hid eye ablation phenotype, heterozygous mutants of diap1 (Goyal et al., 2000; Wilson et al., 2002) and of negative regulators of the EGFR/Ras/MAPK pathway such as argos and sprouty were recovered (Bergmann et al., 1998). Recovery of regulators of the EGFR/Ras/MAPK pathway revealed that Hid is negatively controlled both by MAPK phosphorylation and transcriptional repression (Bergmann et al., 1998; Bergmann et al., 2002; Kurada and White, 1998). However, mutations in the core components of the apoptotic pathway, dronc, dark and drICE were not recovered in the dominant/heterozygous modifier screens implying that they are not rate-limiting for the strong eye ablation phenotype of GMR-hid [reviewed in Xu et al. (2009)].

FIGURE 1

Therefore, in a second-generation mutagenesis effort, we screened homozygous mutants for suppression of GMR-hid. Because we assumed that mutants of the core apoptotic components would be homozygous lethal (which was later confirmed for dronc and dark mutations), we performed the mutagenesis screens in genetic mosaics obtained by ey-FLP/FRT-induced mitotic recombination (Newsome et al., 2000; Xu and Rubin, 1993). We referred to this procedure as GheF (GMR-hid ey-FLP) screening (Xu et al., 2005) (see Supplementary Figure S1). We performed GheF screening for all four autosomal chromosome arms (2L, 2R, 3L and 3R). Using GheF screening, we recovered mutations of dronc (located on 3L), dark (2R) and drICE (3R) as strong suppressors of the GMR-hid-induced eye ablation phenotype (Srivastava et al., 2007; Xu et al., 2005; Xu et al., 2006); reviewed in Xu et al. (2009). Furthermore, we recovered mutants in additional genes such as D-cbl (3L), Uba1, vps25, ptc and cos2 (all 2R) as moderately strong suppressors of GMR-hid using GheF screening (Christiansen et al., 2012; Christiansen et al., 2013; Herz et al., 2006; Herz et al., 2009; Lee et al., 2008; Wang et al., 2008). These genes do not encode core components of the apoptotic pathway but can modify the outcome of the apoptotic process.

Here, we report the results of GheF screening for chromosome arm 2L. We recovered three mutant alleles of the gene lysine demethylase 5 (Kdm5), also known as little imaginal discs (lid), as moderately strong suppressors of GMR-hid in genetic mosaics. Kdm5 was initially identified as a Trithorax Group (TrxG) gene and encodes the Drosophila homolog of human Retinoblastoma-binding protein 2 (Rbp2) (Gildea et al., 2000). Rbp2 was identified as a Rb binding protein in a yeast two-hybrid screen and was shown to control Rb-mediated gene expression (Benevolenskaya et al., 2005; Defeo-Jones et al., 1991; Fattaey et al., 1993; Kim et al., 1994). After identification of Rbp2 as a JmjC-domain-containing histone demethylase, it was grouped as a member of the JARID1 family of histone demethylases (Klose et al., 2006). Kdm5 is the only JARID1 ortholog in Drosophila, while the human genome contains four, with Rbp2 classified as JARID1A.

Kdm5 encodes a protein with multiple domains (Figure 2A). Most notable is the JmjC domain which has lysine demethylase activity and was found to specifically demethylate trimethylated lysine 4 of histone H3 (H3K4me3) (Eissenberg et al., 2007; Lee et al., 2007; Lloret-Llinares et al., 2008; Secombe et al., 2007). In addition, Kdm5 contains a JmjN domain, an ARID (A/T-rich interaction domain), a Zn finger (ZF) and three PHDs (plant homeobox domains) (Figure 2A). The ARID is implicated in binding to A/T-rich DNA sequences, while the ZF and the PHDs are involved in DNA/chromatin interactions. The JmjN, ARID, ZF and PHD1 domains are required for the demethylase activity of Kdm5 (Li et al., 2010). PHD3 can bind to H3K4me3 (Li et al., 2010; Liu and Secombe, 2015) while PHD1 binds to unmethylated Lys4 of histone 3 (H3K4me0) (Li et al., 2010; Torres et al., 2015).

FIGURE 2

H3K4me3 is enriched at promoter regions of transcriptionally active genes (Barski et al., 2007; Bernstein et al., 2005; Heintzman et al., 2007; Ng et al., 2003; Wirbelauer et al., 2005), and Kdm5 proteins were shown to colocalize with the H3K4me3 mark in those promoter regions (Liu and Secombe, 2015; Lloret-Llinares et al., 2012; Lopez-Bigas et al., 2008). Because the JmjC-domain of Kdm5 demethylates H3K4me3, a marker for gene activation, Kdm5 would be predicted to function as a silencer of gene expression by demethylating H3K4me3. However, this silencing function would be inconsistent with its original identification as a member of the TrxG gene family which is implicated in maintaining gene expression, for example, of homeotic genes. Consistently, Kdm5 has been shown to act as a transcriptional activator by maintaining Ubx expression in the haltere (Lloret-Llinares et al., 2008) and as co-activator of Drosophila Myc- (dMyc-) induced cell growth (Secombe et al., 2007). In the latter case, dMyc directly binds to the JmjC domain of Kdm5 and inhibits its catalytic demethylase activity. This enables the PHD3 of Kdm5 to bind to H3K4me3 and to recruit the Kdm5/dMyc complex to promoter regions of dMyc target genes such as Nop60B to stimulate dMyc-dependent transcription (Li et al., 2010; Liu and Secombe, 2015; Secombe et al., 2007). The interaction between Kdm5 and dMyc is also evolutionarily conserved for the mammalian proteins (Secombe et al., 2007). Another mechanism by which Kdm5 can act as an activator of gene expression and thus as a TrxG protein is through inhibition of the histone deacetylase Rpd3 (Lee et al., 2009). Therefore, Drosophila Kdm5 can act both as a transcriptional activator and transcriptional silencer. Consistent with this notion is also the observation that mammalian Rbp2/JARID1A can modulate both transcriptional activity and repression of Rb (Benevolenskaya et al., 2005). These opposite effects on control on transcription by Kdm5 are also confirmed by gene expression profiling analysis of Kdm5 mutants which revealed that of the deregulated genes about half are downregulated while the other half are upregulated (Drelon et al., 2018; Liu et al., 2014; Liu and Secombe, 2015).

As demonstrated first for dMyc-induced cell growth (Secombe et al., 2007) and consistent with its role as transcriptional activator, many of Kdm5’s functions are independent of the demethylase activity of the JmjC-domain of Kdm5 (Drelon et al., 2018; Hatch et al., 2021; Liu et al., 2014; Liu and Secombe, 2015). In fact, the demethylase activity of Kdm5 is not essential for development of Drosophila animals (Li et al., 2010). Flies defective for the JmjC-demethylase function of Kdm5 are phenotypically normal, although males, but not females, are short lived (Li et al., 2010). The survival of these demethylase-deficient animals can be potentially explained by a redundancy with another H3K4 demethylase gene in Drosophila, dKdm2, but nevertheless, global H3K4me3 levels are increased in demethylase-deficient Kdm5 flies (Li et al., 2010) suggesting that increased levels of H3K4me3 can be tolerated during development and also for normal survival of females. In contrast, a null allele of Kdm5 is 100% lethal (Drelon et al., 2018) suggesting that Kdm5 has essential functions during Drosophila development independently of its demethylase activity. The lethality of the null allele of Kdm5 was attributed to its role in larval growth and was mostly due to reduced proliferation of imaginal discs resulting in pupal lethality (Drelon et al., 2018). Apoptosis was also found to be increased in Kdm5 null mutant imaginal discs (Drelon et al., 2018), but the signal for control of apoptosis might be of non-autonomous origin coming from the prothoracic gland (Drelon et al., 2019). Kdm5 also regulates the expression of genes involved in mitochondrial morphology and physiology (Rogers et al., 2023) which is also a contributing factor for survival of the animals during development. Interestingly, restoring MAPK activity specifically in the prothoracic gland can restore viability in otherwise Kdm5 null mutant animals (Rogers et al., 2023) suggesting that Kdm5 also controls MAPK activation.

Here, we report the recovery of three EMS-induced alleles of Kdm5 as moderately strong suppressors of GMR-hid-induced eye ablation in genetic mosaics. Kdm5 specifically controls Hid-induced apoptosis and does not affect Reaper- and Grim-induced apoptosis. This function of Kdm5 is independent of the JmjC-demethylase activity. Furthermore, our data reveal that Kdm5 exerts its control over Hid-induced apoptosis through the activities of Rbf, dMyc, and MAPK. We present a model which incorporates these findings for the control of Hid-induced apoptosis.

Materials and methods

Identification of Kdm5 mutant alleles by GheF screening

The EMS mutagenesis screen for suppressors of GMR-hid in ey-FLP/FRT-induced mosaics for chromosome arm 2L is outlined in Supplementary Figure S1. Males of genotype y w ey-FLP; y⁺FRT40A were incubated on tissue wipes soaked in 5% sucrose solution containing 25 mM Ethyl methanesulfonate (EMS) for 24 h. After that, they were allowed to recover for 3 h before being mated to virgin females of genotype y w ey-FLP; w⁺FRT40A/CyO; GMR-hid at 25°C. 21,871 F1 offspring from this cross were screened for suppression of the GMR-hid-induced eye ablation phenotype. All dominant suppressors were discarded. 78 recessive suppressors were recovered, retested and balanced. One complementation group composed of three moderately strong suppressors of GMR-hid was identified by complementation crosses. Using several chromosomal deficiency stocks, the map position of this complementation group was determined at cytological interval 26A/B on chromosome arm 2L. Complementation crosses with existing Kdm5 mutants Kdm5¹⁰⁴²⁴ and Kdm5^k06801 (two P-element insertions in the first intron of the Kdm5 gene (Gildea et al., 2000)) identified the newly isolated complementation group as mutants of Kdm5. This was further confirmed by sequencing, rescue crosses and phenotypic analysis.

Drosophila husbandry and genetics

All crosses were performed on standard cornmeal-molasses medium (60 g/L cornmeal, 60 mL/L molasses, 23.5 g/L baker’s yeast, 6.5 g/L agar, 4 mL/L acid mix and 0.13% Tegosept). Genetic mosaics were induced in eye-antennal imaginal discs using the FLP/FRT mitotic recombination system (Xu and Rubin, 1993) using ey-FLP (Newsome et al., 2000) with GFP as genetic marker. Mutant clones are marked by loss of GFP. In the case of the MARCM (Lee and Luo, 2001) crosses in Figure 4G, mutant Kdm5 clones expressing the UAS-Kdm5^JmjC* transgene are positively marked by GFP. The wing in Figure 3G is taken from the F1 offspring of the following cross: y w; Kdm5^Q19FRT40/CyO x y w Ubx-FLP; w⁺FRT40/CyO. Non-CyO offspring was selected for imaging.

FIGURE 3

Immunohistochemistry

Eye-antennal imaginal discs from third instar larvae or 26–30 h APF pupal eye discs were dissected using standard protocols and labeled with antibodies raised against the following antigens: anti-Kdm5 (a kind gift of Julie Secombe); cleaved Caspase-3 (cC3) (Cell Signaling Technology); H3K4me3, H3K4me2, H3K4me1 (all Abcam); anti-Diap1 (a kind gift of H.D. Ryoo), anti-Dronc, anti-DrICE (all kind gifts of P. Meier); anti-Hid (a kind gift of H. Steller); anti-Rbf (a kind gift of N. Dyson); anti-β-Gal, anti-dMyc, anti-p53 (all DSHB); anti-JNK (Cell Signaling Technology). Cy3 fluorescently-conjugated secondary antibodies were obtained from Jackson ImmunoResearch. In each experiment, multiple clones in 10–20 eye imaginal discs were analyzed. Images were captured using an Olympus Optical FV500 confocal microscope.

Results

Isolation of three Kdm5 alleles as recessive suppressors of GMR-hid in genetic mosaics

We performed an EMS mutagenesis screen on chromosome arm 2L to recover recessive suppressors of the GMR-hid-induced eye ablation phenotype in genetic mosaics obtained by ey-FLP/FRT-mediated mitotic recombination. The scheme of the GheF (GMR-hid ey-FLP) screen is shown in Supplementary Figure S1. We screened approximately 22,000 F1 offspring of EMS-treated males and GheF-bearing females, and recovered 78 suppressors in genetic mosaics. All of these suppressors are recessive as they do not suppress GMR-hid in a heterozygous, i.e., dominant manner.

By complementation analysis, we identified one complementation group consisting of three alleles, initially termed I1, Q8 and Q19, which suppressed GMR-hid moderately strong in ey-FLP/FRT-induced mosaics (Figures 1A–D). Deficiency mapping with overlapping deficiencies covering chromosome arm 2L identified cytological interval 26B1/2 as the chromosomal location of this complementation group. Complementation crosses with existing mutants of genes in this chromosomal location identified Kdm5 (formerly known as little imaginal discs (lid)) as the underlying gene of this complementation group. Consistently, a transgene encoding wild-type Kdm5 can restore the strong eye ablation phenotype of GMR-hid in Kdm5 mosaics (Figures 4A–C). Therefore, we refer to the newly isolated Kdm5 alleles as Kdm5^I1, Kdm5^Q8 and Kdm5^Q19.

FIGURE 4

Interestingly, the eye ablation phenotype obtained by GMR-reaper and GMR-grim transgenes is not suppressed by Kdm5 mosaics (Figures 1E–H). This observation suggests that Kdm5 specifically controls Hid-induced apoptosis.

Phenotypic characterization of the new Kdm5 alleles

The three Kdm5 alleles isolated in this study have point mutations in the residues indicated in Figure 2A. Kdm5^I1 has an early STOP codon at codon 45 (Figure 2A) and likely encodes a very strong, if not a null, loss-of-function mutant. Consistently, the suppression of GMR-hid by Kdm5^I1 is the strongest of the three alleles. Immunolabeling of mosaic Kdm5^I1 eye imaginal discs with anti-Kdm5 antibody confirms that no Kdm5 protein is produced in Kdm5^I1 mutant clones (Figures 2B, B’). Kdm5^Q8 has a point mutation changing Cys863 in the ZF domain to Tyr, and Kdm^Q19 has a premature STOP codon at codon 1,526 (Figure 2A). Using the anti-Kdm5-specific antibody, we found that the mutant Kdm5^Q8 and Kdm5^Q19 genes encode unstable proteins (Figures 2C, D’). In the case of Kdm5^Q19, this may be due to nonsense-mediated RNA decay, while the point mutation in the ZF domain of Kdm5^Q8 appears to render the protein unstable.

Previous work has shown that the JmjC-demethylase domain of Kdm5 demethylates H3K4me3 (Eissenberg et al., 2007; Lee et al., 2007; Lloret-Llinares et al., 2008; Secombe et al., 2007). Using our newly generated Kdm5 alleles, we confirmed that the global levels of H3K4me3 are increased in Kdm5 mutant clones of larval mosaic eye-antennal imaginal discs (Supplementary Figures S2A–A’’) suggesting that Kdm5 indeed possesses H3K4me3 demethylase activity and further confirming that the newly recovered alleles are Kdm5 mutants. Consistent with previous observations (Eissenberg et al., 2007; Lee et al., 2007; Lloret-Llinares et al., 2008; Secombe et al., 2007), our genetic analysis of Kdm5 mosaics does not demonstrate any demethylase activity towards H3K4me2 and H3K4me1 (Supplementary Figures S2B–C’’). These findings confirm that Kdm5 encodes a H3K4me3 demethylase consistent with previous reports (Eissenberg et al., 2007; Lee et al., 2007; Lloret-Llinares et al., 2008; Secombe et al., 2007).

Cell death phenotypes of Kdm5 mutants

Because we identified the Kdm5 alleles as suppressors of the GMR-hid-induced eye ablation phenotype which is caused by massive apoptosis in the larval eye imaginal disc (Figure 3A (Fan and Bergmann, 2008; 2014), we examined whether Kdm5 affects apoptosis. First, the GMR-hid-induced apoptosis in larval eye imaginal discs is strongly suppressed autonomously in Kdm5 mutant clones (Figures 3B, B’), explaining the recovery of eye tissue in GMR-hid/Kdm5 mosaic flies.

We also examined whether Kdm5 has a role for normal developmentally occurring programmed cell death. In the pupal retina, all unspecified cells are removed by Hid-dependent apoptosis (Kurada and White, 1998). In Kdm5 mutant clones, this developmental apoptosis is strongly reduced (Figure 3C). Furthermore, while wild-type flies contain 4 scutellar bristles (macrochaetae) on the scutellum (Figure 3D), homozygous hypomorphic Kdm5^I1/Kdm5¹⁰⁴²⁴ survivors carry 1 to 2 additional macrochaetae (Figure 3E), a phenotype associated with reduced apoptosis (Kanuka et al., 2005) and also observed for dronc, dark and cytochrome c mutants (Lindblad et al., 2021; Mendes et al., 2006; Rodriguez et al., 1999). Finally, a wave of apoptosis occurs during wing maturation in newly eclosed flies (Kimura et al., 2004). When this apoptosis is blocked, a wing blemishing phenotype appears due to incomplete apoptosis as observed for H99 (deleting reaper, hid and grim), dronc, dark and drICE mutants (Chew et al., 2004; Link et al., 2007). In Kdm5 mosaic wings, a wing blemishing phenotype occurs in all animals tested (Figures 3F, G). Together, these data illustrate that Kdm5 is an important gene for the control of apoptosis during Drosophila development.

Kdm5 controls apoptosis independently of its demethylase activity

Although Kdm5 is not a transcription factor, its best characterized function is control of transcription, either as an activator or as a silencer (Secombe and Eisenman, 2007). The distinguishing feature between these two opposing functions is the requirement of the demethylase activity of the JmjC domain of Kdm5 (Secombe and Eisenman, 2007). Therefore, we examined if the JmjC demethylase activity of Kdm5 is required for the control of apoptosis. As expected, expression of a UAS-Kdm5^wt rescue transgene can revert the suppression of GMR-hid by Kdm5 mosaics (Figures 4A–C). Interestingly, a UAS-Kdm5^JmjC* transgene in which two critical residues in the demethylase domain have been mutated and thus generate a catalytic mutant (Secombe et al., 2007), can also revert the suppression of GMR-hid by Kdm5 mosaics (Figure 4D) suggesting that the JmjC demethylase activity of Kdm5 is not required for the suppression of GMR-hid. Expression of the UAS-Kdm5^wt and UAS-Kdm5^JmjC* transgenes under control of act-Gal4 [the same Gal4 driver used in Figures 4C, D)] does not cause an eye ablation phenotype (Figures 4E, F). This suggests that the reversal of GMR-hid suppression by Kdm5 mosaics observed with these UAS transgenes (Figures 4C, D) is not due to eye ablation caused by the expression of the UAS-Kdm5^wt and UAS-Kdm5^JmjC* alone.

To verify that the UAS-Kdm5^JmjC* transgene indeed encodes a demethylase-defective variant, we found that the elevated H3K4me3 levels in Kdm5 mutant clones are not normalized by expression of the UAS-Kdm5^JmjC* transgene (Figure 4G”) confirming that the Kdm5^JmjC* transgene is indeed demethylase-deficient. These results suggest that Kdm5 controls Hid-induced apoptosis independently of its demethylase activity. The demethylase-independent function of Kdm5 implies that it may act as a transcriptional activator for control of Hid-induced apoptosis.

Kdm5 does not control the expression of apoptotic genes

To examine the mechanism by which Kdm5 controls Hid-induced apoptosis, we tested the expression levels of genes involved in apoptosis. First, although we already found that Kdm5 mosaics do not affect the GMR-reaper- and GMR-grim-induced eye ablation phenotypes (Figure 1), we nevertheless tested whether Kdm5 can act directly on the GMR promoter and thus controls hid expression from GMR-hid. However, expression of a GMR-lacZ reporter transgene is not affected in Kdm5 mosaics (Figure 4H”) suggesting that Kdm5 does not transcriptionally control the GMR promoter. Furthermore, the levels of all genes involved in and required for apoptosis (dronc, dark, drICE, hid, etc.) were not changed in Kdm5 mutant clones (Table 1).

Although the lack of a requirement of the demethylase domain indicates an activator function of Kdm5 for control of Hid-induced apoptosis, we also examined genes encoding negative regulators of apoptosis whose deregulation would require a silencing function of Kdm5 to suppress GMR-hid when Kdm5 is mutant. However, DIAP1 protein levels and expression of β-Gal from a diap1-lacZ reporter transgene are normal in Kdm5 mutant clones (Table 1). A Hid-specific inhibitor of apoptosis is the miRNA bantam, targeting the 3′UTR of the hid mRNA (Brennecke et al., 2003). However, a bantam sensor is unchanged in Kdm5 mutant clones (Table 1). These data suggest that Kdm5 is not directly controlling the gene expression or protein levels of apoptotic genes for control of Hid-induced apoptosis.

Kdm5 controls hid-induced cell death through Rbf and dMyc

To gain further insight into the role of Kdm5 for the control of Hid-induced apoptosis in Drosophila, we considered that the mammalian homolog of Kdm5, Rbp2 (JARID1A), was identified as a Rb binding protein (Defeo-Jones et al., 1991; Fattaey et al., 1993; Kim et al., 1994). The Drosophila homolog of Rb, Rbf, has been implicated in the control of Hid-induced apoptosis through inhibition of E2F1-dependent transcription (Moon et al., 2006). Therefore, we tested for a genetic interaction between Rbf and the ability of Kdm5 mosaics to suppress the GMR-hid eye ablation phenotype (Figures 5A, B). Indeed, heterozygosity of Rbf using two different Rbf alleles can revert the suppression of GMR-hid by Kdm5 mosaics back to the unmodified eye ablation phenotype (Figures 5A–D) suggesting that the suppression of the GMR-hid eye ablation phenotype by Kdm5 mosaics requires Rbf. The Rbf mutants to not modify the GMR-hid eye ablation phenotype on their own (i.e., in a Kdm5 wild-type background) (Figure 5H) suggesting that Rbf is rate-limiting for the suppression of GMR-hid by Kdm5. One function of Rbf is to negatively control the transcription factor E2F1 (Moon et al., 2006). Therefore, we also tested for a genetic interaction between E2F1 and Kdm5. Consistently, the suppressed eye ablation phenotype of GMR-hid by Kdm5 is even more strongly suppressed if E2F1 is heterozygous in this genetic background (Figure 5E). Heterozygous E2F1 mutants alone do not modify the GMR-hid phenotype (Figure 5I). These genetic interactions suggest that Kdm5 controls GMR-hid-induced apoptosis through negative regulation of Rbf.

FIGURE 5

Another factor which was shown to interact with Kdm5 both genetically and physically is Drosophila Myc (dMyc) (Secombe et al., 2007). In this context, it is interesting to note that this interaction results in inhibition of the demethylase activity of Kdm5 (Secombe et al., 2007) which is not required for the suppression of GMR-hid (Figure 4D). Therefore, we performed similar genetic interaction experiments as with Rbf and E2F1. Indeed, heterozygously, dMyc mutants revert the suppression of GMR-hid by Kdm5 back to the original eye ablation phenotype (Figures 5F, G) suggesting that dMyc is required for the suppression of GMR-hid by Kdm5. Heterozygous dMyc mutants alone do not suppress the GMR-hid eye ablation phenotype (Figure 5J) suggesting that dMyc is rate-limiting for the suppression of GMR-hid by Kdm5. This genetic interaction implies that Kdm5 and dMyc have an antagonistic relationship in this context.

Finally, to examine if Rbf/E2F1 and dMyc cooperate for the control of GMR-hid induced cell death, we took advantage of the observation that overexpression of dMyc can suppress the GMR-hid eye ablation phenotype (Figures 5K, L). This suppression is reverted back to the original GMR-hid eye ablation phenotype by heterozygosity of Rbf (Figure 5M) suggesting that dMyc requires Rbf to suppress GMR-hid. Together, these genetic interaction data demonstrate that Kdm5 regulates GMR-hid through inhibition of Rbf and dMyc, and that Rbf and dMyc cooperate in this context.

Kdm5 regulates GMR-hid-induced eye ablation through control of MAPK signaling

While E2F1 and potentially dMyc can directly control endogenous hid expression (Moon et al., 2005), this type of regulation does not explain the suppression of GMR-hid by Kdm5 as hid expression from GMR-hid is independent of the control of endogenous hid. This consideration would imply a post-translational control of Hid directly or indirectly by Kdm5. We have previously shown that Erk-type MAPK, encoded by the rolled (rl) gene, can post-translationally control Hid function by inhibitory phosphorylation (Bergmann et al., 1998). This type of regulation by MAPK on GMR-hid would also explain the inability of Kdm5 mosaics to suppress GMR-reaper and GMR-grim (Figure 1) which are not subject to control by MAPK (Bergmann et al., 1998; Kurada and White, 1998).

Therefore, we examined if Kdm5 mosaics can suppress a GMR-hid mutant in which the five MAPK phosphorylation sites of Hid have been mutated (GMR-hid^Ala5) and is thus unresponsive to inhibitory MAPK phosphorylation (Bergmann et al., 1998). Indeed, Kdm5 mosaics are not able to suppress the eye ablation phenotype of GMR-hid^Ala5 (Figures 6A–C). As a positive control, to demonstrate that the very strong GMR-hid^Ala5 eye ablation phenotype is in fact suppressible, we tested another mutant, S2, which was recovered in the same GheF screen of chromosome arm 2L and which displays a similar moderate suppression of GMR-hid^wt in mosaics as Kdm5 (Supplementary Figure S1). S2 is a single allele recovered in the GheF 2L screen and the underlying gene mutated in S2 is unknown. Indeed, S2 mosaics can suppress the GMR-hid^Ala5 eye ablation phenotype (Figure 6D) suggesting that the GMR-hid^Ala5 eye ablation phenotype is suppressible and that the inability of Kdm5 to suppress GMR-hid^Ala5 might be because of its unresponsive nature to MAPK phosphorylation. The MAPK-dependent suppression of GMR-hid^wt by Kdm5 suggests that Kdm5 negatively regulates MAPK activity and therefore that Kdm5 mutants have increased MAPK activity.

FIGURE 6

To further characterize the genetic interaction between rl/MAPK and Kdm5, we took advantage of the wing vein phenotype caused by a weak gain-of-function mutant of MAPK called rl^Sevenmaker (rl^Sem) (Brunner et al., 1994) (Figures 6E, F). The wing vein phenotype of rl^Sem is strongly enhanced by heterozygosity of Kdm5 (Figure 6H) further supporting that Kdm5 negatively regulates MAPK. In strong cases, we also observed a wing blistering phenotype which was not observed in rl^Sem wings alone (Figure 6I). Heterozygous Kdm5 wings do not show any specific wing vein phenotype (data not shown). Interestingly, heterozygosity of hid also strongly enhances the rl^Sem wing vein phenotype (Figure 6G) demonstrating the strong genetic link between Kdm5 and hid through MAPK signaling. In summary, these genetic interaction studies demonstrate that Kdm5 regulates Hid through control of MAPK activity.

Discussion

In this paper, we report the recovery of three alleles of the histone demethylase gene Kdm5 as moderately strong suppressors of GMR-hid-induced apoptosis. To our knowledge, these are the first EMS-induced alleles of Kdm5. We identified the mutations of these alleles and found that they either produce truncated proteins or unstable proteins, or both (Figure 2). Kdm5 is a nuclear protein and although it is not a transcription factor, all evidence points to an essential role of Kdm5 in transcriptional control, both as an activator and a silencer. The silencing function of Kdm5 is mediated through demethylation of H3K4me3 by the JmjC domain (Secombe and Eisenman, 2007). However, the demethylase activity of the JmjC domain of Kdm5 is not involved in the control of Hid-induced apoptosis (Figure 4D). Therefore, other domains of Kdm5 are involved in the control of Hid-induced apoptosis. A good candidate for such a domain is the ZF domain, as the Kdm5^Q8 allele changes a conserved Cys residue to Tyr. However, this mutation causes protein instability (Figure 2) and the general instability of Kdm5^Q8 protein might cause the suppression, and not the specific mutation in the ZF domain.

The control of apoptosis is very specific to the IAP antagonist Hid, as both the GMR-reaper- and GMR-grim-induced eye ablation phenotypes are not affected by Kdm5 mosaics (Figure 1). Consistently, Kdm5 does not control expression of transgenes from the GMR promoter and of core components of the apoptotic pathway such as dronc, dark and diap1 which would otherwise also affect Reaper- and Grim-induced apoptosis. The phenotypic characterization of the new Kdm5 alleles supports the notion that Kdm5 is at least partially required for apoptosis. Consistently, it was reported that overexpression of Kdm5 can ectopically induce apoptosis in a dMyc-expressing context (Secombe et al., 2007) and that JARID1A^−/− mice show a decrease in apoptosis of hematopoietic stem cells (Klose et al., 2007). However, it has also been reported that wing imaginal discs of homozygous Kdm5 mutant larvae have actually increased levels of apoptosis (Drelon et al., 2018). Nevertheless, this increased apoptosis in wing imaginal discs is caused non-autonomously by signaling defects in the mutant prothoracic gland (Drelon et al., 2019). We would also like to emphasize that Kdm5 only controls Hid-induced apoptosis. Thus, other types of apoptosis, such as Reaper- or Grim-induced apoptosis which are not under Kdm5 control, may be the cause of this ectopic apoptosis seen in Kdm5 mutants.

While we were unable to identify a specific target gene of Kdm5 that is involved in the control of apoptosis (see Table 1), we established genetic interactions between Kdm5 on one hand and Rbf/E2F1 and dMyc on the other hand for the control of Hid-induced apoptosis (Figure 5). The mammalian ortholog of Drosophila Kdm5, Rbp2/JARID1A was originally identified as an interacting protein with Rb, and that interaction is conserved in Drosophila (Benevolenskaya et al., 2005; Defeo-Jones et al., 1991; Fattaey et al., 1993; Kim et al., 1994; Secombe et al., 2007). Both Drosophila Kdm5 and mammalian JARID1A can form a protein complex with the corresponding Myc orthologs, which in Drosophila controls Myc-induced growth (Drelon et al., 2018; Secombe et al., 2007). Interestingly, dMyc binds to the JmjC domain of Kdm5 and inhibits its demethylase activity (Secombe et al., 2007). This inhibition converts the silencer function of Kdm5 into a transcriptional (co-) activator (Secombe and Eisenman, 2007; Secombe et al., 2007). Given that the demethylase function of Kdm5 is not required for the control of Hid-induced apoptosis (Figure 4) and that dMyc is rate-limiting for the control of Hid-induced apoptosis by Kdm5 (Figure 5), it is possible that a similar inhibition of Kdm5’s demethylase activity by dMyc occurs for the control of Hid-induced apoptosis by Kdm5.

Putting all these data together, we propose the following model of Kdm5 action for the control of Hid-induced apoptosis. Kdm5 forms protein complexes with both Rbf and dMyc (Figure 7). dMyc inhibits the demethylase function of Kdm5 and thus maintains high levels of H3K4me3 at promoter regions of dMyc target genes. At the same time, the interaction between Rbf and Kdm5 blocks Rbf’s ability to inhibit E2F1 which can now also bind to enhancer/promoter regions of target genes (Figure 7). Our genetic analysis suggests that dMyc and E2F1 synergistically control Hid-induced apoptosis (Figure 5) and thus may act on the promoter region of the same genes. However, while E2F1 has been shown to be present at the promoter region of endogenous hid (Moon et al., 2005), it is unlikely that the cooperative control by Rbf and dMyc occurs on the endogenous hid gene. First, all our assays were performed with the heterologous GMR promoter which is not controlled by Rbf and dMyc. Second, the GMR promoter is not affected by Kdm5 (Figure 4). Third, expression of endogenous hid is not altered in Kdm5 mutant clones (Table 1). These considerations suggest that Kdm5 regulates the expression of another gene(s) for the control of the apoptotic activity of Hid.

FIGURE 7

While we cannot exclude that another apoptotic gene is regulated by Kdm5, E2F1 and dMyc, our genetic analysis points towards control of MAPK activity (Figure 6). We showed previously that in contrast to Reaper and Grim, Hid is negatively controlled by phosphorylation of five Ser or Thr residues by Erk-type MAPK (Bergmann et al., 1998). A phosphorylation-defective mutant of Hid under GMR-control (GMR-hid^Ala5) cannot be suppressed by Kdm5 mosaics (Figure 6) strongly suggesting that Kdm5 controls MAPK phosphorylation activity for regulation of Hid. This may occur through regulation of a gene which directly or indirectly controls MAPK activity (Figure 7). Regulation of MAPK activity by Kdm5 is not unprecedented. Drelon et al. (2019) showed that Kdm5 regulates MAPK activity through transcriptional control of the receptor tyrosine kinase (RTK) torso in the prothoracic gland (Drelon et al., 2019). While there is no known role of Torso for development of imaginal discs in Drosophila larvae, other RTKs such as the Drosophila EGFR receptor (DER) have very established roles, especially for eye development (Malartre, 2016). However, while Drelon et al. (2019) demonstrated that Kdm5 is required for MAPK activity in prothoracic glands, our data suggest that Kdm5 negatively controls MAPK activity for the control of Hid-induced apoptosis in imaginal discs. This discrepancy may be caused by tissue-specific differences for the control of MAPK activity by Kdm5 and may be the result of the control of different target genes in these tissues. Therefore, instead of torso in the prothoracic gland, genes that negatively control DER and thus MAPK activity such as argos or sprouty may be controlled by Kdm5 for control of Hid-induced apoptosis. Future work will reveal the target gene(s) of Kdm5 that control MAPK activity for regulation of Hid-induced apoptosis in Drosophila.

Kdm5 is not the only JmjC-domain-containing protein that regulates Hid activity. The Drosophila ortholog of the phosphatidylserine receptor, dPSR, has also been implicated in controlling Hid activity (Krieser et al., 2007). dPSR is a JmjC-domain-only protein, and its mammalian ortholog JMJD6 demethylates histone H3 at Arg2 and histone H4 at Arg3 (Chang et al., 2007). Initially, PSR was classified as the phosphatidylserine receptor for engulfment of apoptotic cells (Fadok et al., 2000). However, the function of dPSR in engulfment in Drosophila is uncertain. dPSR mutant macrophages (hemocytes) can engulf apoptotic cells and dPSR protein is nuclear (Krieser et al., 2007) which is inconsistent with the predicted localization of an engulfment receptor at the cell surface. Interestingly, further characterization of dPSR revealed that it inhibits the ability of Hid to induce apoptosis (Krieser et al., 2007) suggesting that Kdm5 and dPSR may have antagonistic activities for the control of the apoptotic activity of Hid. Whether Kdm5 and dPSR directly interact to exert this antagonistic function is unknown. It is also unknown whether dPSR requires a functional JmjC domain for this activity.

Given the irreversible effect of apoptosis, it is not surprising that apoptosis is a highly regulated process. Apoptosis is controlled at transcriptional, post-transcriptional (alternative splicing) and post-translational (phosphorylation, ubiquitylation, etc.) levels. Here, we also add epigenetic control as an additional control mechanism of apoptosis. Resistance to apoptosis is a hallmark of many cancers and is often accompanied by epigenetic deregulation (Chakraborty et al., 2024; Ozyerli-Goknar and Bagci-Onder, 2021). Thus, it is crucial to better understand the epigenetic mechanisms that suppress apoptosis in various cancers and other diseases to help advance the development of targeted therapies.

We have performed recessive mutagenesis screens for the control of Hid-induced apoptosis in genetic mosaics using GheF screening for all autosomal chromosome arms. These screens yielded mutations in the core components of the apoptotic pathway, Dronc, Dark and Drice (Wu et al., 2016; Xu et al., 2005; Xu et al., 2006; Xu et al., 2009). In addition, we recovered mutations in genes that modify the outcome of the apoptotic pathway such as Uba1 affecting ubiquitylation and D-cbl, regulating MAPK activity (Lee et al., 2008; Wang et al., 2008). We also recovered mutations in vps25, cos2 and ptc which affect apoptosis non-autonomously through control of the Notch and Hedgehog signaling pathways (Christiansen et al., 2012; Christiansen et al., 2013; Herz et al., 2006; Herz et al., 2009). Here, we added epigenetic control of apoptosis by the recovery of Kdm5 mutants as an additional layer of regulation. In summary, these screens underscore the importance of unbiased mutagenesis screens to understand the control of a biological process, in this case apoptosis, at all possible levels. This detailed understanding will also have important implications for understanding and treatment of diseases in human patients.

References

• 1

  AbbottM. K.LengyelJ. A. (1991). Embryonic head involution and rotation of male terminalia require the Drosophila locus head involution defective. Genetics129, 783–789. 10.1093/genetics/129.3.783

  • CrossRef
  • Google Scholar

• 2

  BarskiA.CuddapahS.CuiK.RohT. Y.SchonesD. E.WangZ.et al (2007). High-resolution profiling of histone methylations in the human genome. Cell129, 823–837. 10.1016/j.cell.2007.05.009

  • CrossRef
  • Google Scholar

• 3

  BenevolenskayaE. V.MurrayH. L.BrantonP.YoungR. A.KaelinW. G.Jr. (2005). Binding of pRB to the PHD protein RBP2 promotes cellular differentiation. Mol. Cell18, 623–635. 10.1016/j.molcel.2005.05.012

  • CrossRef
  • Google Scholar

• 4

  BergmannA.AgapiteJ.McCallK.StellerH. (1998). The Drosophila gene hid is a direct molecular target of Ras-dependent survival signaling. Cell95, 331–341. 10.1016/s0092-8674(00)81765-1

  • CrossRef
  • Google Scholar

• 5

  BergmannA.TugentmanM.ShiloB. Z.StellerH. (2002). Regulation of cell number by MAPK-dependent control of apoptosis: a mechanism for trophic survival signaling. Dev. Cell2, 159–170. 10.1016/s1534-5807(02)00116-8

  • CrossRef
  • Google Scholar

• 6

  BernsteinB. E.KamalM.Lindblad-TohK.BekiranovS.BaileyD. K.HuebertD. J.et al (2005). Genomic maps and comparative analysis of histone modifications in human and mouse. Cell120, 169–181. 10.1016/j.cell.2005.01.001

  • CrossRef
  • Google Scholar

• 7

  BrenneckeJ.HipfnerD. R.StarkA.RussellR. B.CohenS. M. (2003). Bantam encodes a developmentally regulated microRNA that controls cell proliferation and regulates the proapoptotic gene hid in Drosophila. Cell113, 25–36. 10.1016/s0092-8674(03)00231-9

  • CrossRef
  • Google Scholar

• 8

  BrunnerD.OellersN.SzabadJ.BiggsW. H.ZipurskyS. L.HafenE. (1994). A gain-of-function mutation in Drosophila MAP kinase activates multiple receptor tyrosine kinase signaling pathways. Cell76, 875–888. 10.1016/0092-8674(94)90362-x

  • CrossRef
  • Google Scholar

• 9

  ChakrabortyS.NandiP.MishraJ.NiharikaRoyA.MannaS.et al (2024). Molecular mechanisms in regulation of autophagy and apoptosis in view of epigenetic regulation of genes and involvement of liquid-liquid phase separation. Cancer Lett.587, 216779. 10.1016/j.canlet.2024.216779

  • CrossRef
  • Google Scholar

• 10

  ChangB.ChenY.ZhaoY.BruickR. K. (2007). JMJD6 is a histone arginine demethylase. Science318, 444–447. 10.1126/science.1145801

  • CrossRef
  • Google Scholar

• 11

  ChenP.NordstromW.GishB.AbramsJ. M. (1996). grim, a novel cell death gene in Drosophila. Genes Dev.10, 1773–1782. 10.1101/gad.10.14.1773

  • CrossRef
  • Google Scholar

• 12

  ChewS. K.AkdemirF.ChenP.LuW. J.MillsK.DaishT.et al (2004). The apical caspase dronc governs programmed and unprogrammed cell death in Drosophila. Dev. Cell7, 897–907. 10.1016/j.devcel.2004.09.016

  • CrossRef
  • Google Scholar

• 13

  ChristiansenA. E.DingT.BergmannA. (2012). Ligand-independent activation of the Hedgehog pathway displays non-cell autonomous proliferation during eye development in Drosophila. Mech. Dev.129, 98–108. 10.1016/j.mod.2012.05.009

  • CrossRef
  • Google Scholar

• 14

  ChristiansenA. E.DingT.FanY.GravesH. K.HerzH. M.LindbladJ. L.et al (2013). Non-cell autonomous control of apoptosis by ligand-independent Hedgehog signaling in Drosophila. Cell death Differ.20, 302–311. 10.1038/cdd.2012.126

  • CrossRef
  • Google Scholar

• 15

  Defeo-JonesD.HuangP. S.JonesR. E.HaskellK. M.VuocoloG. A.HanobikM. G.et al (1991). Cloning of cDNAs for cellular proteins that bind to the retinoblastoma gene product. Nature352, 251–254. 10.1038/352251a0

  • CrossRef
  • Google Scholar

• 16

  DorstynL.AkeyC. W.KumarS. (2018). New insights into apoptosome structure and function. Cell death Differ.25, 1194–1208. 10.1038/s41418-017-0025-z

  • CrossRef
  • Google Scholar

• 17

  DrelonC.BelalcazarH. M.SecombeJ. (2018). The histone demethylase KDM5 is essential for larval growth in Drosophila. Genetics209, 773–787. 10.1534/genetics.118.301004

  • CrossRef
  • Google Scholar

• 18

  DrelonC.RogersM. F.BelalcazarH. M.SecombeJ. (2019). The histone demethylase KDM5 controls developmental timing in Drosophila by promoting prothoracic gland endocycles. Development146, dev182568. 10.1242/dev.182568

  • CrossRef
  • Google Scholar

• 19

  DuW.VidalM.XieJ. E.DysonN. (1996). RBF, a novel RB-related gene that regulates E2F activity and interacts with cyclin E in Drosophila. Genes Dev.10, 1206–1218. 10.1101/gad.10.10.1206

  • CrossRef
  • Google Scholar

• 20

  EissenbergJ. C.LeeM. G.SchneiderJ.IlvarsonnA.ShiekhattarR.ShilatifardA. (2007). The trithorax-group gene in Drosophila little imaginal discs encodes a trimethylated histone H3 Lys4 demethylase. Nat. Struct. Mol. Biol.14, 344–346. 10.1038/nsmb1217

  • CrossRef
  • Google Scholar

• 21

  FadokV. A.BrattonD. L.RoseD. M.PearsonA.EzekewitzR. A.HensonP. M. (2000). A receptor for phosphatidylserine-specific clearance of apoptotic cells. Nature405, 85–90. 10.1038/35011084

  • CrossRef
  • Google Scholar

• 22

  FanY.BergmannA. (2008). Distinct mechanisms of apoptosis-induced compensatory proliferation in proliferating and differentiating tissues in the Drosophila eye. Dev. Cell14, 399–410. 10.1016/j.devcel.2008.01.003

  • CrossRef
  • Google Scholar

• 23

  FanY.BergmannA. (2014). Multiple mechanisms modulate distinct cellular susceptibilities toward apoptosis in the developing Drosophila eye. Dev. Cell30, 48–60. 10.1016/j.devcel.2014.05.007

  • CrossRef
  • Google Scholar

• 24

  FattaeyA. R.HelinK.DembskiM. S.DysonN.HarlowE.VuocoloG. A.et al (1993). Characterization of the retinoblastoma binding proteins RBP1 and RBP2. Oncogene8, 3149–3156.

  • Google Scholar

• 25

  GildeaJ. J.LopezR.ShearnA. (2000). A screen for new trithorax group genes identified little imaginal discs, the Drosophila melanogaster homologue of human retinoblastoma binding protein 2. Genetics156, 645–663. 10.1093/genetics/156.2.645

  • CrossRef
  • Google Scholar

• 26

  GoyalL.McCallK.AgapiteJ.HartwiegE.StellerH. (2000). Induction of apoptosis by Drosophila reaper, hid and grim through inhibition of IAP function. EMBO J.19, 589–597. 10.1093/emboj/19.4.589

  • CrossRef
  • Google Scholar

• 27

  GretherM. E.AbramsJ. M.AgapiteJ.WhiteK.StellerH. (1995). The head involution defective gene of Drosophila melanogaster functions in programmed cell death. Genes Dev.9, 1694–1708. 10.1101/gad.9.14.1694

  • CrossRef
  • Google Scholar

• 28

  Gyrd-HansenM.MeierP. (2010). IAPs: from caspase inhibitors to modulators of NF-kappaB, inflammation and cancer. Nat. Rev. Cancer10, 561–574. 10.1038/nrc2889

  • CrossRef
  • Google Scholar

• 29

  HatchH. A. M.BelalcazarH. M.MarshallO. J.SecombeJ. (2021). A KDM5-Prospero transcriptional axis functions during early neurodevelopment to regulate mushroom body formation. Elife10, e63886. 10.7554/eLife.63886

  • CrossRef
  • Google Scholar

• 30

  HeintzmanN. D.StuartR. K.HonG.FuY.ChingC. W.HawkinsR. D.et al (2007). Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nat. Genet.39, 311–318. 10.1038/ng1966

  • CrossRef
  • Google Scholar

• 31

  HerzH. M.ChenZ.ScherrH.LackeyM.BolducC.BergmannA. (2006). vps25 mosaics display non-autonomous cell survival and overgrowth, and autonomous apoptosis. Development133, 1871–1880. 10.1242/dev.02356

  • CrossRef
  • Google Scholar

• 32

  HerzH. M.WoodfieldS. E.ChenZ.BolducC.BergmannA. (2009). Common and distinct genetic properties of ESCRT-II components in Drosophila. PLoS One4, e4165. 10.1371/journal.pone.0004165

  • CrossRef
  • Google Scholar

• 33

  KanukaH.KuranagaE.TakemotoK.HiratouT.OkanoH.MiuraM. (2005). Drosophila caspase transduces Shaggy/GSK-3beta kinase activity in neural precursor development. EMBO J.24, 3793–3806. 10.1038/sj.emboj.7600822

  • CrossRef
  • Google Scholar

• 34

  KayagakiN.WebsterJ. D.NewtonK. (2024). Control of cell death in health and disease. Annu. Rev. Pathol.19, 157–180. 10.1146/annurev-pathmechdis-051022-014433

  • CrossRef
  • Google Scholar

• 35

  KimY. W.OttersonG. A.KratzkeR. A.CoxonA. B.KayeF. J. (1994). Differential specificity for binding of retinoblastoma binding protein 2 to RB, p107, and TATA-binding protein. Mol. Cell Biol.14, 7256–7264. 10.1128/mcb.14.11.7256

  • CrossRef
  • Google Scholar

• 36

  KimuraK.KodamaA.HayasakaY.OhtaT. (2004). Activation of the cAMP/PKA signaling pathway is required for post-ecdysial cell death in wing epidermal cells of Drosophila melanogaster. Development131, 1597–1606. 10.1242/dev.01049

  • CrossRef
  • Google Scholar

• 37

  KloseR. J.KallinE. M.ZhangY. (2006). JmjC-domain-containing proteins and histone demethylation. Nat. Rev. Genet.7, 715–727. 10.1038/nrg1945

  • CrossRef
  • Google Scholar

• 38

  KloseR. J.YanQ.TothovaZ.YamaneK.Erdjument-BromageH.TempstP.et al (2007). The retinoblastoma binding protein RBP2 is an H3K4 demethylase. Cell128, 889–900. 10.1016/j.cell.2007.02.013

  • CrossRef
  • Google Scholar

• 39

  KrieserR. J.MooreF. E.DresnekD.PellockB. J.PatelR.HuangA.et al (2007). The Drosophila homolog of the putative phosphatidylserine receptor functions to inhibit apoptosis. Development134, 2407–2414. 10.1242/dev.02860

  • CrossRef
  • Google Scholar

• 40

  KumarS. (2007). Caspase function in programmed cell death. Cell death Differ.14, 32–43. 10.1038/sj.cdd.4402060

  • CrossRef
  • Google Scholar

• 41

  KuradaP.WhiteK. (1998). Ras promotes cell survival in Drosophila by downregulating hid expression. Cell95, 319–329. 10.1016/s0092-8674(00)81764-x

  • CrossRef
  • Google Scholar

• 42

  LeeN.Erdjument-BromageH.TempstP.JonesR. S.ZhangY. (2009). The H3K4 demethylase lid associates with and inhibits histone deacetylase Rpd3. Mol. Cell Biol.29, 1401–1410. 10.1128/MCB.01643-08

  • CrossRef
  • Google Scholar

• 43

  LeeN.ZhangJ.KloseR. J.Erdjument-BromageH.TempstP.JonesR. S.et al (2007). The trithorax-group protein Lid is a histone H3 trimethyl-Lys4 demethylase. Nat. Struct. Mol. Biol.14, 341–343. 10.1038/nsmb1216

  • CrossRef
  • Google Scholar

• 44

  LeeT.LuoL. (2001). Mosaic analysis with a repressible cell marker (MARCM) for Drosophila neural development. Trends Neurosci.24, 251–254. 10.1016/s0166-2236(00)01791-4

  • CrossRef
  • Google Scholar

• 45

  LeeT. V.DingT.ChenZ.RajendranV.ScherrH.LackeyM.et al (2008). The E1 ubiquitin-activating enzyme Uba1 in Drosophila controls apoptosis autonomously and tissue growth non-autonomously. Development135, 43–52. 10.1242/dev.011288

  • CrossRef
  • Google Scholar

• 46

  LiL.GreerC.EisenmanR. N.SecombeJ. (2010). Essential functions of the histone demethylase lid. PLoS Genet.6, e1001221. 10.1371/journal.pgen.1001221

  • CrossRef
  • Google Scholar

• 47

  LindbladJ. L.TareM.AmcheslavskyA.ShieldsA.BergmannA. (2021). Non-apoptotic enteroblast-specific role of the initiator caspase Dronc for development and homeostasis of the Drosophila intestine. Sci. Rep.11, 2645. 10.1038/s41598-021-81261-0

  • CrossRef
  • Google Scholar

• 48

  LinkN.ChenP.LuW. J.PogueK.ChuongA.MataM.et al (2007). A collective form of cell death requires homeodomain interacting protein kinase. J. Cell Biol.178, 567–574. 10.1083/jcb.200702125

  • CrossRef
  • Google Scholar

• 49

  LiuX.GreerC.SecombeJ. (2014). KDM5 interacts with Foxo to modulate cellular levels of oxidative stress. PLoS Genet.10, e1004676. 10.1371/journal.pgen.1004676

  • CrossRef
  • Google Scholar

• 50

  LiuX.SecombeJ. (2015). The histone demethylase KDM5 activates gene expression by recognizing chromatin context through its PHD reader motif. Cell Rep.13, 2219–2231. 10.1016/j.celrep.2015.11.007

  • CrossRef
  • Google Scholar

• 51

  Lloret-LlinaresM.CarreC.VaqueroA.de OlanoN.AzorinF. (2008). Characterization of Drosophila melanogaster JmjC+N histone demethylases. Nucleic Acids Res.36, 2852–2863. 10.1093/nar/gkn098

  • CrossRef
  • Google Scholar

• 52

  Lloret-LlinaresM.Perez-LluchS.RossellD.MoranT.Ponsa-CobasJ.AuerH.et al (2012). dKDM5/LID regulates H3K4me3 dynamics at the transcription-start site (TSS) of actively transcribed developmental genes. Nucleic Acids Res.40, 9493–9505. 10.1093/nar/gks773

  • CrossRef
  • Google Scholar

• 53

  Lopez-BigasN.KisielT. A.DeWaalD. C.HolmesK. B.VolkertT. L.GuptaS.et al (2008). Genome-wide analysis of the H3K4 histone demethylase RBP2 reveals a transcriptional program controlling differentiation. Mol. Cell31, 520–530. 10.1016/j.molcel.2008.08.004

  • CrossRef
  • Google Scholar

• 54

  MalartreM. (2016). Regulatory mechanisms of EGFR signalling during Drosophila eye development. Cell Mol. Life Sci.73, 1825–1843. 10.1007/s00018-016-2153-x

  • CrossRef
  • Google Scholar

• 55

  MendesC. S.AramaE.BrownS.ScherrH.SrivastavaM.BergmannA.et al (2006). Cytochrome c-d regulates developmental apoptosis in the Drosophila retina. EMBO Rep.7, 933–939. 10.1038/sj.embor.7400773

  • CrossRef
  • Google Scholar

• 56

  MoonN. S.Di StefanoL.DysonN. (2006). A gradient of epidermal growth factor receptor signaling determines the sensitivity of rbf1 mutant cells to E2F-dependent apoptosis. Mol. Cell Biol.26, 7601–7615. 10.1128/MCB.00836-06

  • CrossRef
  • Google Scholar

• 57

  MoonN. S.FrolovM. V.KwonE. J.Di StefanoL.DimovaD. K.MorrisE. J.et al (2005). Drosophila E2F1 has context-specific pro- and antiapoptotic properties during development. Dev. Cell9, 463–475. 10.1016/j.devcel.2005.08.015

  • CrossRef
  • Google Scholar

• 58

  NewsomeT. P.AslingB.DicksonB. J. (2000). Analysis of Drosophila photoreceptor axon guidance in eye-specific mosaics. Development127, 851–860. 10.1242/dev.127.4.851

  • CrossRef
  • Google Scholar

• 59

  NewtonK.StrasserA.KayagakiN.DixitV. M. (2024). Cell death. Cell187, 235–256. 10.1016/j.cell.2023.11.044

  • CrossRef
  • Google Scholar

• 60

  NgH. H.RobertF.YoungR. A.StruhlK. (2003). Targeted recruitment of Set1 histone methylase by elongating Pol II provides a localized mark and memory of recent transcriptional activity. Mol. Cell11, 709–719. 10.1016/s1097-2765(03)00092-3

  • CrossRef
  • Google Scholar

• 61

  OhS. W.KingsleyT.ShinH. H.ZhengZ.ChenH. W.ChenX.et al (2003). A P-element insertion screen identified mutations in 455 novel essential genes in Drosophila. Genetics163, 195–201. 10.1093/genetics/163.1.195

  • CrossRef
  • Google Scholar

• 62

  OrmeM.MeierP. (2009). Inhibitor of apoptosis proteins in Drosophila: gatekeepers of death. Apoptosis14, 950–960. 10.1007/s10495-009-0358-2

  • CrossRef
  • Google Scholar

• 63

  Ozyerli-GoknarE.Bagci-OnderT. (2021). Epigenetic deregulation of apoptosis in cancers. Cancers (Basel)13, 3210. 10.3390/cancers13133210

  • CrossRef
  • Google Scholar

• 64

  PierceS. B.YostC.BrittonJ. S.LooL. W.FlynnE. M.EdgarB. A.et al (2004). dMyc is required for larval growth and endoreplication in Drosophila. Development131, 2317–2327. 10.1242/dev.01108

  • CrossRef
  • Google Scholar

• 65

  RodriguezA.OliverH.ZouH.ChenP.WangX.AbramsJ. M. (1999). Dark is a Drosophila homologue of Apaf-1/CED-4 and functions in an evolutionarily conserved death pathway. Nat. Cell Biol.1, 272–279. 10.1038/12984

  • CrossRef
  • Google Scholar

• 66

  RogersM. F.MarshallO. J.SecombeJ. (2023). KDM5-mediated activation of genes required for mitochondrial biology is necessary for viability in Drosophila. Development150, dev202024. 10.1242/dev.202024

  • CrossRef
  • Google Scholar

• 67

  RyooH. D.BergmannA.GonenH.CiechanoverA.StellerH. (2002). Regulation of Drosophila IAP1 degradation and apoptosis by reaper and ubcD1. Nat. Cell Biol.4, 432–438. 10.1038/ncb795

  • CrossRef
  • Google Scholar

• 68

  Schreiber-AgusN.SteinD.ChenK.GoltzJ. S.StevensL.DePinhoR. A. (1997). Drosophila Myc is oncogenic in mammalian cells and plays a role in the diminutive phenotype. Proc. Natl. Acad. Sci. U. S. A.94, 1235–1240. 10.1073/pnas.94.4.1235

  • CrossRef
  • Google Scholar

• 69

  SecombeJ.EisenmanR. N. (2007). The function and regulation of the JARID1 family of histone H3 lysine 4 demethylases: the Myc connection. Cell Cycle6, 1324–1328. 10.4161/cc.6.11.4269

  • CrossRef
  • Google Scholar

• 70

  SecombeJ.LiL.CarlosL.EisenmanR. N. (2007). The Trithorax group protein Lid is a trimethyl histone H3K4 demethylase required for dMyc-induced cell growth. Genes Dev.21, 537–551. 10.1101/gad.1523007

  • CrossRef
  • Google Scholar

• 71

  ShaliniS.DorstynL.DawarS.KumarS. (2015). Old, new and emerging functions of caspases. Cell death Differ.22, 526–539. 10.1038/cdd.2014.216

  • CrossRef
  • Google Scholar

• 72

  SrivastavaM.ScherrH.LackeyM.XuD.ChenZ.LuJ.et al (2007). ARK, the Apaf-1 related killer in Drosophila, requires diverse domains for its apoptotic activity. Cell death Differ.14, 92–102. 10.1038/sj.cdd.4401931

  • CrossRef
  • Google Scholar

• 73

  TorresI. O.KuchenbeckerK. M.NnadiC. I.FletterickR. J.KellyM. J.FujimoriD. G. (2015). Histone demethylase KDM5A is regulated by its reader domain through a positive-feedback mechanism. Nat. Commun.6, 6204. 10.1038/ncomms7204

  • CrossRef
  • Google Scholar

• 74

  WangY.WerzC.XuD.ChenZ.LiY.HafenE.et al (2008). Drosophila cbl is essential for control of cell death and cell differentiation during eye development. PLoS One3, e1447. 10.1371/journal.pone.0001447

  • CrossRef
  • Google Scholar

• 75

  WhiteK.TahaogluE.StellerH. (1996). Cell killing by the Drosophila gene reaper. Science271, 805–807. 10.1126/science.271.5250.805

  • CrossRef
  • Google Scholar

• 76

  WilsonR.GoyalL.DitzelM.ZachariouA.BakerD. A.AgapiteJ.et al (2002). The DIAP1 RING finger mediates ubiquitination of Dronc and is indispensable for regulating apoptosis. Nat. Cell Biol.4, 445–450. 10.1038/ncb799

  • CrossRef
  • Google Scholar

• 77

  WirbelauerC.BellO.SchubelerD. (2005). Variant histone H3.3 is deposited at sites of nucleosomal displacement throughout transcribed genes while active histone modifications show a promoter-proximal bias. Genes Dev.19, 1761–1766. 10.1101/gad.347705

  • CrossRef
  • Google Scholar

• 78

  WuY.LindbladJ. L.GarnettJ.Kamber KayaH. E.XuD.ZhaoY.et al (2016). Genetic characterization of two gain-of-function alleles of the effector caspase DrICE in Drosophila. Cell death Differ.23, 723–732. 10.1038/cdd.2015.144

  • CrossRef
  • Google Scholar

• 79

  XuD.LiY.ArcaroM.LackeyM.BergmannA. (2005). The CARD-carrying caspase Dronc is essential for most, but not all, developmental cell death in Drosophila. Development132, 2125–2134. 10.1242/dev.01790

  • CrossRef
  • Google Scholar

• 80

  XuD.WangY.WilleckeR.ChenZ.DingT.BergmannA. (2006). The effector caspases drICE and dcp-1 have partially overlapping functions in the apoptotic pathway in Drosophila. Cell death Differ.13, 1697–1706. 10.1038/sj.cdd.4401920

  • CrossRef
  • Google Scholar

• 81

  XuD.WoodfieldS. E.LeeT. V.FanY.AntonioC.BergmannA. (2009). Genetic control of programmed cell death (apoptosis) in Drosophila. Fly. (Austin)3, 78–90. 10.4161/fly.3.1.7800

  • CrossRef
  • Google Scholar

• 82

  XuT.RubinG. M. (1993). Analysis of genetic mosaics in developing and adult Drosophila tissues. Development117, 1223–1237. 10.1242/dev.117.4.1223

  • CrossRef
  • Google Scholar