Role of Elongator Subunit Elp3 in Drosophila Melanogaster Larval Development and Immunity

Role of Elongator Subunit Elp3 in Drosophila melanogaster Larval Development and Immunity

Jane Walker,*,1 So Yeon Kwon,† Paul Badenhorst,† Phil East,‡ Helen McNeill‡,2 and Jesper Q. Svejstrup* *Clare Hall Laboratories, Cancer Research UK London Research Institute, South Mimms EN6 3LD, United Kingdom; †Institute of Biomedical Research, University of Birmingham, Edgbaston B15 2TT, United Kingdom and ‡Lincoln’s Inn Fields Laboratories, Cancer Research UK London Research Institute, London WC2A 3PX, United Kingdom Manuscript received October 7, 2010 Accepted for publication January 12, 2011

ABSTRACT The Elongator complex has been implicated in several cellular processes, including gene expression and tRNA modification. We investigated the biological importance of the Elp3 gene in Drosophila melanogaster. Deletion of Elp3 results in larval lethality at the pupal stage. During early development, larval growth is dramatically impaired, with progression to the third instar delayed for 24 hr, and pupariation occurring only at day 14 after egg laying. Melanotic nodules appear after 4 days. Microarray analysis shows that stress response genes are induced and ecdysone-induced transcription factors are severely repressed in the mutant. Interestingly, the phenotypes of Elp3 flies are similar to those of flies lacking the domino gene, encoding a SWI/SNF-like ATP-dependent chromatin-remodeling enzyme. Indeed, the gene expression profiles of these mutants are also remarkably similar. Together, these data demonstrate that Drosophila Elp3 is essential for viability, normal development, and hematopoiesis and suggest a functional overlap with the chromatin remodeler Domino.

HE Elongator complex was first identified in yeast Furthermore, induction of genes such as INO1, PHO5, T as a protein complex associated with elongating and GAL10 isdelayedcomparedtowildtype(Otero et al. RNA polymerase II (RNAPII) (Otero et al. 1999) and 1999; Wittschieben et al. 1999). RNA immunoprecip- was later shown to consist of six subunits (Krogan and itation has shown that Elp3 binds the nascent mRNA of Greenblatt 2001; Li et al. 2001; Winkler et al. 2001). active genes in yeast (Gilbert et al. 2004), and using Particular attention has been paid to the function of the chromatin immunoprecipitation (ChIP) Elongator has 60-kDa subunit, Elp3, which contains a histone acetyl- also beenshown to associate with active loci in human cells, transferase domain (Wittschieben et al. 1999, 2000), as especially toward the 39-end of genes (Metivier etal. 2003; well as potential SAM-binding and iron-sulfur cluster Kouskouti and Talianidis 2005; Close et al. 2006). domains (Chinenov 2002; Paraskevopoulou et al. Together, these features suggest that Elongator partic- 2006; Greenwood et al. 2009). Elp3 possesses histone H3 ipates in RNAPII transcript elongation. acetylation activity in vitro (Wittschieben et al. 1999; Interestingly, other data have shown that Elongator is Hawkes et al. 2002; Kim et al. 2002; Winkler et al. 2002) also required for modification of the wobble position in and affects histone acetylation levels in vivo in yeast tRNA (tRNALys and tRNAGlu) in yeast (Huang et al. (Kristjuhan et al. 2002; Winkler et al. 2002) and 2005; Esberg et al. 2006) and for tubulin acetylation in humans (Close et al. 2006; Chen et al. 2009). Yeast elp higher eukaryotes (Creppe et al. 2009; Solinger et al. deletion mutants initially grow slowly on a variety of 2010). Indeed, a high proportion of Elongator is found in media,butthenreach growth rates thataresimilartowild the cytoplasm of most cell types (Hawkes et al. 2002; Kim type. They also show a delay in responding to stress, such et al. 2002; Pokholok et al. 2002; Rahl et al. 2005), all in as growth at 37° and ahigh salt environment(Otero et al. all suggesting multiple roles for the complex (reviewed by 1999; Fellows et al. 2000; Winkler et al. 2001). Svejstrup 2007). Other recent data suggest a role for Elongator also in DNA demethylation (Okada et al. 2010). Supporting information is available online at http://www.genetics.org/ cgi/content/full/genetics.110.123893/DC1. The proteins of the Elongator complex are conserved awkes im Available freely online through the author-supported open access from yeast to humans (H et al. 2002; K et al. option. 2002). We generated Drosophila strains that lack the 1Corresponding author: Clare Hall Laboratories, Blanche Lane, South highly conserved Elp3 gene (Elp3). Here, we report that Mimms EN6 3LD, United Kingdom. E-mail: [email protected] Drosophila Elp3 is required for viability. Deletion of 2Present address: Samuel Lunenfeld Research Institute, 600 University the gene thus has a profound effect on larval growth, Ave., Toronto, Ontario M5G 1X5, Canada. and Elp3 mutants die at the pupal stage. Moreover,

Genetics 187: 1067–1075 (April 2011) 1068 J. Walker et al. melanotic nodules are formed in third instar larvae, generated by PCR using the following primers: Br-C 59 (gag implicating Elp3 in Drosophila innate immune function. ctaacagaggtgcccgt), Br-C 39 (ttggcttggcgctctgtgag), Actin 59 9 Interestingly, we observe that many of the characteristics (cttaccctgaagtaccccatt), and Actin 3 (ggcctccatgccgaggaacga). For antimicrobial peptide analysis, RNA was prepared from of elp3 flies are shared with flies lacking the ATP- wandering stage larvae. Probes were synthesized using the dependent chromatin remodeler Domino, suggesting following primers: TEP 59 (ctacatttgatactgcagaaaatatagaac), functional overlap between Elongator and Domino- TEP 39 (ccgagctcctgtagatatttatattcgcat), DIPT 59 (ttcttcaattga mediated chromatin remodeling. gaacaactgagatgc), DIPT 39 (gaagtctgcctcaatgttccgggttaa), DRS 59 (atgatgcagatcaagtacttgttcgcc), and DRS 39 (ttagcatccttcgcca ccagcacttcag). Immunofluorescence: Drosophila hemocytes were isolated from five wandering-stage third instar larvae of each genotype MATERIALS AND METHODS and stained for immunofluorescence as described in Kwon inenko Drosophila strains: The EP insertion line EP2367 was et al. (2008). Lamellocyte-specific mouse MAb L1b (S obtained from the Szeged Stock Center. The GAL4 driver et al. 2004) and the pan-hemocyte mouse MAb anti-Hemese urucz lines, Act5C-GAL4 (y1 w*; P{Act5C6 GAL4}17bFO1/TM6B, Tb1) (K et al. 2003) were used at 1:60 dilution. FITC- and hs-GAL4 (w*; P{GAL4-Hsp70.PB}2/CyO), were obtained conjugated secondary antibodies ( Jackson ImmunoResearch) from the Bloomington Drosophila Stock Center.The hemocyte- were used at 1:1000. Samples were mounted in Vectashield specific GAL4 driver lines He-GAL4 (Zettervall et al. 2004) containing DAPI (Vector Laboratories) and observed using a and Pxn-GAL4 (Stramer et al. 2005) were used. UAS-Elp3 RNAi confocal microscope (Zeiss Axiovert 100M). lines, CG15433R-2 and CG15433R-3, were provided by the Microarray analysis: Wandering-stage wild type, Elp3, and raun National Institute of Genetics (NIG) stock center (Mishima, Japan). domino (B et al. 1997, 1998) larvae were frozen in liquid The helper transposase line used was P{ry[1t7.2] ¼ D2-3}99B. nitrogen and then homogenized in Trizol (Invitrogen). RNA Excision of Elp3: The P-element-containing line EP2367 was purified on RNeasy columns (Qiagen) and analyzed on contains a P element inserted 65 bp upstream of the transcrip- GeneChip Drosophila Genome 2.0 arrays. The domino and tion start site of Elp3 (CG15433). Excision was achieved by Elp3 data sets were processed separately, performing back- crossing to the helper transposase line D2-3. Individual males ground correction and quartile normalization and calculating were crossed to w1118;CyO/nocSco balancer females, and prog- signal intensity estimates using robust multichip analysis rizarry eny were assessed for homozygous lethality. Stocks of mutant (RMA) (I et al. 2003). Prior to statistical analysis, we lines were analyzed by PCR and sequence analysis. preselected informative genes (>0.05 coefficient of variance). Rescue: Full-length Elp3 cDNA from the clone RE35395 Using ANOVA, we selected Elp3-dependent and Domino- (ResGen) was cloned into the vector p(UAS-HA) (Parker dependent changes against wild-type controls. We selected genes et al. 2001), and transgenic UAS-Elp3 lines were generated by using a 0.05 false discovery rate (fdr) threshold. Analysis was entleman P-element-mediated transformation. Elp3 expression was carried out using Bioconductor 2.5 (G et al. 2004). achieved by crossing to the Act5C-GAL4 driver line. Generation of HAT-defective Elp3 strain: Elp3 cDNA was RESULTS mutated using the Stratagene QuikChange II site-directed mutagenesis kit to generate a cDNA encoding full-length Elp3 Isolation and characterization of Drosophila Elp3 with an Y531F amino acid substitution that inactivates the HAT mutants: The Drosophila homolog of yeast Elp3 was ittschieben domain (W et al. 2000). The resultant cDNA was identified by sequence homology searching. CG15433 cloned into p(UAS-HA), and transgenic UAS-Elp3Y531F lines were generated as above. (here named Drosophila Elp3) shares a 92% amino acid RT-PCR: Wild-type and Elp3 homozygous mutant third similarity with human Elp3, diverging mainly at the N instar larvae were homogenized in Trizol buffer, and chloro- terminus (Wittschieben et al. 1999). Elp3 is located at form extracted. An equal volume of 70% ethanol was added to cytological position 24F, and its promoter overlaps that the supernatant, which was then loaded onto an RNAeasy mini of the oppositely transcribed morgue gene. As a first step column (Qiagen), and total RNA was purified according to the manufacturer’s instructions. DNAse treatment with Turbo in generating an Elp3 mutant strain, we searched for DNAse (Ambion) was followed by cDNA synthesis using a P-element insertions in the vicinity of Elp3 and identified Taqman reverse transcription kit (Applied Biosystems). Real- EP2367, which contains an EP element inserted 65 bp time PCR was performed on either an ABI 7000 or a Bio-Rad IQ upstream of the Elp3 transcription start site. The real-time PCR machine. RT-PCR primers used were the homozygous viable EP2367 line was used as the basis following: Actin5C 59 (agcgcggttactctttcacc), Actin5C 39 (ggcca tctcctgctcaaagtc), morgue 59 (caagtggcagcagcagaaac), morgue 39 for generating additional imprecise excision deletions (ttcactagccatccacagcg), E74A 59 (agatcgagagcctgctgtcc), E74A of Elp3. Three independent homozygous lethal excision 39 (gacgctggcgaagtcatcat), Fbp1 59 (atgctgttggtggcctttg), Fbp1 alleles were isolated: Elp3Ex1, Elp3Ex6, and Elp3Ex9. Elp3Ex1 39 (agatcctccaaggacatgcg), Dp110 59 (cgaaccttacaccgacgaaac), deletes Elp3 from the P-element insertion site to the Dp110 39 (cgctgctaaagctcgttgtg), Myc 59 (caaatcagcagggagctt triplet encoding K277, while both Elp3Ex6 and Elp3Ex9 cag), Myc 39 (ttggaatgctggttgagcac), E2F 59 (gttgataagcgcaagc tgca), and E2F 39 (ccagtcggaatttccagctct). delete Elp3 to the triplet encoding S377, as well as a sub- Northern analysis: Total RNA was prepared every day as stantial part of the flanking morgue gene (Figure 1A). described above from homozygous mutant Elp3 larvae from Western blot analysis confirmed that Elp3 protein was 5 days after egg laying (AEL) to 14 days AEL. Thirty larvae of not detectable in imaginal disc and brain extracts of the 5- to 10-day-old larvae, and 15 of the 10- to 14-day-old larvae these mutants (Figure 1B). Importantly, expression of were harvested. RNA was separated on a 1% formaldehyde Ex1 agarose gel and blotted onto a Hybond N1 membrane (GE the flanking gene, morgue, was unaffected in the Elp3 Ex1 Healthcare). The membrane was probed with a Broad mutant (Figure 3C). For this reason, Elp3 was used in Complex core region probe and Actin42A probe. Probes were all subsequent experiments. Functions of Drosophila Elongator 1069

Figure 1.—Gene deletions in Elp3Ex1 and Elp3Ex6. (A) Map of the 24F region of chromosome 2 and regions deleted in different Elp3 lines. Lines with double arrows designate deletions. The position of the ELP3 gene (solid bar), as well as that of other genes in the region (shaded bars), is indicated. (B) Western blot of larval brain and imaginal disc extracts probed with afﬁnity-puriﬁed yeast Elp3 antibody ( JSV59).

Rescue of Elp3 lethality: To verify that the lethal of some of the activated genes, such as Broad Complex phenotype of the ElpEx1 mutant strain was indeed caused (BR-C) and E74, leads to a loss of viability at the pupal by disruption of Elp3, we attempted to rescue lethality by stage and to abnormal development of imaginal discs expression of a wild-type Elp3 cDNA using the GAL4-UAS (Kiss et al. 1988; Fletcher and Thummel 1995). Since system (Brand and Perrimon 1993). Full-length Dro- the Elp3 mutants died at the pupal stage and displayed a sophila Elp3 cDNA was cloned into the pUASHA vector failure of disc development, the induction of these (Parker et al. 2001), and UAS-Elp3 transgenic lines were ecdysone-induced transcripts was examined. RT-PCR generated. Ubiquitous ectopic expression of UAS-Elp3 analysis of ecdysone target genes revealed that many under the control of the Act5C-GAL4 strain rescued the were not induced correctly in Elp3 larvae. No induction lethality of Elp3Ex1. However, while males showed normal was apparent until 8–9 days after egg laying, and even at fertility, the resulting female ﬂies were sterile. It is known maximal levels, induction was 50- to 100-fold lower than that the GAL4-UAS driver used here is not expressed in in control larvae (Figure 3A: see right panel inset for the female germline (Brand and Perrimon 1993). The detail of delayed and reduced Fbp1 induction in the sterility of the rescued female ﬂies thus suggests that Elp3 mutant).Northern analysis of the Broad Complex (BR-C), plays a role in the development of the female reproduc- one of the primary responders to ecdysone, showed that tive system. expression of this gene (maximally expressed after 5 days Deletion of Elp3 leads to a pronounced delay in in normal cells) was also delayed and greatly diminished larval development: Homozygous Elp3 larvae showed a in Elp3 larvae (Figure 3B: compare lane 9 with lane 1). In profound developmental delay and died during pupar- contrast to the ecdysone-induced genes, many other iation. By day 5 AEL, a marked difference in size between tested genes were expressed at levels comparable to mutant and heterozygous larvae was apparent (Figure controls (examples shown in Figure 3C; see also below). 2A). Progression to the third instar was delayed by 24 Elp3 larvae develop melanotic nodules: One partic- hr, and larvae remained slow growing. However, by 10 ularly obvious feature of Elp3 mutant larvae is the day AEL, a normal size was reached, and at day 14, appearance of melanotic nodules (Figure 2D). Mela- pupariation occurred. Growth of the larval discs and wing notic nodules are often caused by dysregulated activa- discs was severely impaired (Figure 2, B and C, and data tion of either the Toll pathway (Qiu et al. 1998) or the not shown). Salivary glands remained small, and squashes JAK/STAT pathway, as in the Hopscotch mutant hopTum-1 of polytene chromosomes showed that the chromatin was (Hanratty and Dearolf 1993; Harrison et al. 1995). fragmented and fragile, effectively precluding analysis by A number of cellular and molecular features accompany polytene chromosome spreading. the production of melanotic nodules promoted by these Ecdysone-induced transcription is delayed and pathways. First, elevated numbers of circulating hemo- reduced: The development and progression of larvae cytes are detected. Second, many of these hemocytes to pupae are controlled by pulses of the steroid differentiate into an otherwise rare hemocyte type, the hormone ecdysone. During the third larval instar, a lamellocyte. Finally, increases in the expression of major ecdysone pulse that initiates transcription of a set antimicrobial peptides are detected, for example, Dro- of genes required for metamorphosis occurs. Mutation somycin (Drs) and Diptericin (Dipt) in the case of the 1070 J. Walker et al.

Toll pathway and the complement-like protein Tep1 in the case of the JAK/STAT pathway (Lagueux et al. 2000). Strikingly, staining of hemocytes in Elp3 larvae with lamellocyte- and hemocyte-specific antibodies showed that neither lamellocyte nor hemocyte numbers were increased. Instead, hemocyte numbers were reduced, and the remaining hemocytes aggregated to form micro- nodules, which were in circulation (Figure 4, A–D). CrystalcellnumberswerealsounchangedinElp3 larvae (our unpublished observations). It seems likely that these micro-nodules nucleate the large melanotic nodules observed in the larvae. Antimicrobial peptides do not appear to be strongly induced in Elp3 mutants (Figure 4E), suggesting that the Toll and JAK/STAT pathways are not the primary pathways induced in this case. Expression of Elp3 RNA interference in hemocytes: To investigate whether melanotic nodules were caused by disruption of Elp3 in hemocytes or by the loss of Elp3 in other tissues, we specifically ablated Elp3 function in hemocytes by inducible RNA interference (RNAi). UAS- RNAi lines that target Elp3 were obtained from the NIG stock center and crossed with either of two hemocyte- specific GAL4 drivers, Pxn-GAL4 and He-GAL4. Elp3 RNAi reduced hemocyte numbers between 4- and 15- fold (Figure 5D), and resulted in the aggregation of hemocytes (Figure 5, A–C), resembling the hemocyte phenotype observed in ElpEx1 homozygous mutant larvae. This suggests that Elp3 is required autonomously for hemocyte development. However, visible melanotic nodules were not observed in the progeny of the RNAi crosses. The inability of Elp3 RNAi to trigger melanotic nodule formation may be due either to incomplete knockdown of Elp3 or to a lack of expression of the RNAi in all cells involved in the formation of the nodules. Expression of a HAT-defective Elp3: Yeast Elp3 mutated at Y540 has dramatically reduced histone acetyltransferase activity (Wittschieben et al. 2000). To assess the contribution of the HAT activity to the Elp3 phenotype in flies, we created a mutant at the corresponding site in the Drosophila protein (Y531F). Induction of the resulting PUAS-Elp3HAT strain in a wild-type Elp3 background with embryonic GAL4 drivers leads to embryonic lethality (Table 1), while induction by the wing-specific driver MS1096-GAL4 produces abnormal wing phenotypes (supporting information, Figure S1). Induction by a heat-shock driver resulted in delayed development and pupal lethality as in the deletion mutants. The HAT mutant Elp3 therefore acts as a dominant negative. These results indicate that the histone acetyltransferase activity of Elp3 is crucial to its Figure 2.—Phenotype of Elp3 mutant larvae. (A) Larvae at role in development and survival. 5 days AEL. Heterozygotes are at the wandering stage, but Whole-genome expression profiling of Elp3Ex1 Elp3 larvae remain in the food. Note the presence of melanotic mutant larvae: To identify pathways that may be affected nodules (blackspots)already atthis stage. (Band C) Wandering- stage Elp3 larvae show poorly developed leg and wing discs. in Elp3 mutant larvae, microarray analysis of wandering- (D) Comparison of 5 days AEL wild-type and 12 days AEL stage mutant larvae was performed. Although several of Elp3 larvae, showing melanotic nodules in the mutant larvae. the changes observed in the mutant relate to the for- Functions of Drosophila Elongator 1071

Figure 3.—Ecdysone-induced transcription is delayed and reduced in Elp3 mutants. Total RNA was prepared from whole larvae 4 days AEL and daily until pupariation occurred. cDNA either was synthesized for quantitative RT-PCR analysis or RNA (10 mg) was subjected to Northern analysis. (A) E74A transcription is not induced to detectable levels in Elp3 larvae (left), while Fbp1 is induced slowly from 8 days until maximal levels are reached at 12 days AEL (right). The level of transcription is very low compared to the control. Numbers on y-axes (and above bars) are in arbitrary units from RT-PCR. nd, not deter- mined. (B) Northern analysis of Broad Complex transcript. A probe from the core region of BR-C was used. Expression is induced to a high level at 5 days AEL in wild type (lane 1), while very weak induction is observed at 12 days AEL in the mutant (lane 9). (C) Other tested genes are expressed at normal levels in the Elp3 mutant. All numbers obtained from RT-PCRs were first nor- malized to actin. The expression in Elp3 mutant is shown relative to the level in heterozygotes (dashed line). E74A and Fbp1 (from A) are shown for comparison. mation of melanotic nodules, expression of many stress- sensory, sympathetic, and some parasympathetic neu- induced proteins, transcription factors, and neuronal rons in the autonomic and sensory nervous system genes was also affected (Table 2; a complete list of (Anderson et al. 2001; Cuajungco et al. 2001). significantly affected genes can be found in Table S1). Comparison of Elp3 and domino microarrays: The Highly overexpressed transcripts of relevance to the lack of antimicrobial peptide induction in Elp3 flies was melanotic nodules include the integrins aPS4 and aPS5. also confirmed by the microarray analysis. Lack of These lamellocyte-specific integrins are also overex- antimicrobial peptide induction is rare in larvae that pressed in hopTum-l and Tl10b larvae (Irving et al. 2005) display a melanotic nodule phenotype. It is, however, and are thought to contribute to the adhesion of also seen in domino (dom) mutants, in which a DNA- lamellocytes in the encapsulation event. Similarly, the dependent ATPase of the SWI/SW2 family is inactivated prophenoloxidase enzyme DoxA3 and serine proteases (Braun et al. 1998). Some other characteristics of the are highly overexpressed, reflecting the activation of the dom mutants are also similar to those of Elp3 mutants, melanization cascade. such as a prolonged larval stage, pupal lethality, poor Several of the stress-induced Turandot peptides are formation of imaginal discs, and the formation of larval also strongly induced in Elp3 mutants. These peptides melanotic nodules. Both Elp3 and domino mutants are are induced by the MEKK pathway (Brun et al. 2006) and lethal, and the genes are on the same chromosome, are also induced in hopTum-l and Nurf301 mutants (Kwon complicating genetic interaction experiments. Elp3 et al. 2008). Turandot peptides are secreted into the and dom mutants do not show dominant interactions as hemolymph after stress induction such as heat shock or Elp3/dom trans-heterozygous (Elp3/1, domino/1) flies oxidative stress and may act in a similar way to heat-shock are viable and indistinguishable from wild type. However, proteins (Ekengren and Hultmark 2001). Similarly, both Elp3 and dom mutants act as dominant enhancers many glutathione transferase genes are highly induced, of hopTum-l melanotic tumor mutations (data not shown). further implying an activation of oxidative stress path- To confirm interactions between Elp3 and dom,wepre- ways. Members of the immune induced molecule family pared mRNA from dom mutants, analyzed it on micro- are also highly induced. These transcripts have been arrays, and then compared it to the expression pattern of found to be associated with the activation of Toll and Elp3 mutants. JAK/STAT pathways of immunity (Kwon et al. 2008) and The overlap in overall gene expression levels between with salt-stressed flies (Stergiopoulos et al. 2009). the two data sets was striking, both in upregulated and Conversely, it is interesting that some neuronal genes downregulated genes (P , 0.05 and fold change relative were downregulated in Elp3 mutants since neuronal to wild type .2.5) (Figure 6A). Transcription of 357 defects are evident in the human disorder familial genes was elevated .2.5-fold in both mutants, which dysautonomia, a genetic disorder caused by a tissue- represents 66% of upregulated transcripts in Elp3 and specific reduction in the amount of the Elongator Elp1 80% of upregulated genes in the dom mutant. Of the subunit. This affects the development and survival of transcripts downregulated .2.5-fold, 246 were repressed 1072 J. Walker et al.

Figure 5.—Hemocyte-specific expression of Elp3 RNAi re- duces hemocyte number and induces cell aggregation. (A) Wild-type hemocytes and hemocytes after (B) Pxn-GAL4 and (C) He-GAL4 directed Elp3 RNAi. (D) Number of hemocytes in hemolymph of control and Elp3 RNAi larvae. Hemocyte numbers are unchanged in backgrounds that contain only the He-GAL4, Pxn-GAL4,orUAS-Elp3-RNAi constructs alone (Zettervall et al. 2004; Stramer et al. 2005; data not shown). Figure 4.—Hemocytes in Elp3 mutants. Hemocytes from control and Elp3 larvae were stained with (A and B) pan- hemocytic (anti-Hemese) and (C and D) lamellocyte-specific stress stimuli (Otero et al. 1999). In Arabidopsis, (MAb L1b) antibodies. Antibody staining is revealed in green; deletion of Elongator subunits results in a delay in DNA stained with DAPI is shown in purple. (E) Northern germination and seedling growth (Nelissen et al. 2005). analysis of expression of thiolester containing protein 1 Deletion of an Elongator subunit is lethal in the mouse (Tep1) and Diptericin (Dipt) in control (wild type), Elp3, hen and hopTum-1 larvae. (C et al. 2009). We found that deletion of the gene encoding the Elongator subunit Elp3 in the fly results in both Elp3 and dom mutants, representing 72% of genes in larval lethality at the pupal stage. Moreover, larval repressed in Elp3 and 49% of genes repressed in dom. growth is dramatically impaired during early develop- Linear regression analysis (Figure 6B) as well as the heat ment, with progression to the third instar significantly map of genes whose expression was changed .3-fold in delayed and pupariation occurring only at day 14 after mutant relative to wild-type genes (Figure 6C) also egg laying. Interestingly, melanotic nodules appear in showed that the fold changes in expression correlate larvae after 4 days. extremely well between the two mutants. The overlap in To help define which pathways were altered in Elp3 gene expression profiles thus mirrors the observed larvae, a genome-wide expression experiment was car- striking similarity of phenotypes in the dom and Elp3 ried out. The microarray analysis of late-stage Elp3 mutants. mutant larvae shows that many stress-induced genes are upregulated. For example, members of the Turan- dot and glutathione S-transferase gene families are DISCUSSION highly induced. These proteins are often associated with Previous experiments on Elongator complex mostly the oxidative stress response in cells (Ekengren et al. focused on its role in transcription or tRNA modifica- 2001; Li et al. 2008). Recent reports suggest that the Elp3 tion. In this report, we examined the biological con- mutation is associated with amylotrophic lateral sclerosis sequences of a lack of Elp3 function in Drosophila (Simpson et al. 2009), a neurodegenerative disorder, melanogaster and identified a striking similarity to those which may be associated with oxidative damage induced observed upon deleting the gene encoding the SWI/ by free radicals. Elp3 contains an oxidation-sensitive SNF family chromatin-remodeling enzyme Domino. In iron-sulfur cluster (Greenwood et al. 2009), so it is an yeast, Elongator mutants are viable, but characterized intriguing possibility that Elp3 itself somehow plays by their slow-growth phenotype and delayed response to a role in detecting the redox state of cells. Functions of Drosophila Elongator 1073

TABLE 1 TABLE 2 Expression of an Elp3 HAT-defective mutant by embryonic Examples of altered gene expression in Elp3 GAL4 drivers wandering-stage larvae

Gal4 driver of Gene Ontology Fold change Elp3 HAT mutant Phenotype Defense PtcGal4 Embryonic lethal Diphenol OxidaseA3 32 EngrailedGal4 Embryonic lethal IM2 28 RhomboidGal4 Embryonic lethal IM10 22 Actin5CGal4 Embryonic lethal IM1 17 HairyGal4 Normal development Scavenger Receptor Class C 9 KruppelGal4 Normal development Spheroide 9 ElavGal4 Normal development IM23 7 MS1096Gal4 Deformed wings CG13422 6 Heat shock Delayed development and pupal death Proteolysis l-Try 26 In a wild-type background, expression of an Elp3 HAT CG30098 9 point-mutant typically results in lethality. Wing development CG16997 8 is severely impaired when the mutant protein is expressed CG31265 7 from a wing-specific driver (MS1096-GAL4) (see also Figure S1). Heat-shock induction of the mutant Elp3 at first and sec- Serine protease immune response integrator 3 ond instar stages caused a delay in development and death at Serine protease inhibitor 16 late larval stages. Stress Turandot C 94 Victoria 59 The appearance of melanotic nodules indicates dys- Turandot B 14 regulation of the innate immune system. This is con- Glutathione transferase D5 11 firmed by the whole-genome expression profile, which Glutathione transferase E6 10 revealed the induction of many genes known to be Peroxiredoxin2540 5 involved in the aberrant activation of the larval immune Methuselah-like 2 8 Integrins system (Irving et al. 2005). However, it is not clear which aPS4 115 signal transduction pathways have been activated in this aPS5 10 case, as antimicrobial peptides such as drosomycin (Toll Ecdysone-induced pathway) and TEP ( JAK/STAT pathway) do not appear ImpE2 34 to be induced. It is possible that the effect on the Sgs1 27 immune system stems from the activation of the stress Eig71Ee 12 response described above, underlining the close con- Sgs4 10 Broad Complex 7 nection between the stress and immune responses in Sgs5 6 Drosophila. Eip74EF 5 Our analysis also revealed that ecdysone-induced Eip78C 3 transcription is severely impaired in Elp3 mutant larvae. Axonal guidance The delayed development of the mutant larvae might Nervous fingers 1 3 potentially postpone the ecdysone burst. However, the Smooth 3 size of the mutant was similar to that of the wild type at SoxNeuro 3 the time of the microarray analysis, arguing against this Genes are grouped into Gene Ontology functions. Whole- possibility. Northern analysis of Broad Complex con- number fold changes compared to wild type are shown. firms that the induction is delayed and reduced, suggesting a requirement for Elp3 in the regulation of reflected in several aspects of similarity in the mutant transcription. Interestingly, a failure of ecdysone in- phenotypes, such as delayed growth and progression to duction is also observed in other known chromatin third instar, poor disc formation, pupal lethality, and modifier mutants such as Nurf301 (Badenhorst et al. melanotic nodule formation. Domino is a member of 2005) and dom (this report). The absence of these the SWI2/SNF2 family of DNA-dependent ATPases and chromatin-remodeling factors results in lethality and is involved in several aspects of chromatin remodelling. the production of melanotic nodules, supporting the We suggest that the similar phenotypes arise from idea that, like these factors, Elongator also affects the coordinate regulation of similar sets of target genes ability of the chromatin to respond to ecdysone and and imply functional collaboration between Elongator- other environmental and developmental cues in an and Domino-mediated chromatin remodeling. appropriate manner. Indeed, comparison of Elp3 and Since Elongator is also known to be present in the dom gene expression profiles shows a compelling simi- coding region of certain human genes (Kouskouti and larity in overall effects on gene expression. This is also Talianidis 2005; Close et al. 2006), it seems possible 1074 J. Walker et al.

We thank Jay Uhler, Nic Tapon, Sally Leevers, and Christophe Antoniewski for discussions and reagents; Terence Gilbank and Stephen Murray for their help with micro-injections; and Luke Alphey for supplying the pUASHA plasmid. This work was supported by grants from Cancer Research UK to J.Q.S. and H.M.

LITERATURE CITED Anderson, S. L., R. Coli,I.W.Daly,E.A.Kichula,M.J.Rork et al., 2001 Familial dysautonomia is caused by mutations of the IKAP gene. Am. J. Hum. Genet. 68: 753–758. Badenhorst, P., H. Xiao,L.Cherbas,S.Y.Kwon,M.Voas et al., 2005 The Drosophila nucleosome remodeling factor NURF is required for Ecdysteroid signaling and metamorphosis. Genes Dev. 19: 2540–2545. Brand, A. H., and N. Perrimon, 1993 Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118: 401–415. Braun, A., B. Lemaitre,R.Lanot,D.Zachary and M. Meister, 1997 Drosophila immunity: analysis of larval hemocytes by P-element-mediated enhancer trap. Genetics 147: 623–634. Braun, A., J. A. Hoffmann and M. Meister, 1998 Analysis of the Drosophila host defense in domino mutant larvae, which are de- void of hemocytes. Proc. Natl. Acad. Sci. USA 95: 14337–14342. Brun, S., S. Vidal,P.Spellman,K.Takahashi,H.Tricoire et al., 2006 The MAPKKK Mekk1 regulates the expression of Turan- dot stress genes in response to septic injury in Drosophila. Genes Cells 11: 397–407. Chen, Y. T., M. M. Hims,R.S.Shetty,J.Mull,L.Liu et al., 2009 Loss of mouse Ikbkap, a subunit of elongator, leads to transcriptional deficits and embryonic lethality that can be rescued by human IKBKAP. Mol. Cell. Biol. 29: 736–744. Chinenov, Y., 2002 A second catalytic domain in the Elp3 histone acetyltransferases: A candidate for histone demethylase activity? Trends Biochem. Sci. 27: 115–117. Close, P., N. Hawkes,I.Cornez,C.Creppe,C.A.Lambert et al., 2006 Transcription impairment and cell migration defects in elongator-depleted cells: implication for familial dysautonomia. Mol. Cell 22: 521–531. Creppe, C., L. Malinouskaya,M.L.Volvert,M.Gillard,P.Close et al., 2009 Elongator controls the migration and differentia- tion of cortical neurons through acetylation of alpha-tubulin. Cell 136: 551–564. Cuajungco, M. P., M. Leyne,J.Mull,S.P.Gill,J.F.Gusella et al., Figure 6.—Microarray analysis of Elp3 and dom mutants 2001 Cloning, characterization, and genomic structure of the shows considerable overlap. (A) A Venn diagram of tran- mouse Ikbkap gene. DNA Cell Biol. 20: 579–586. scripts, the expression of which is increased or decreased Ekengren, S., and D. Hultmark, 2001 A family of Turandot-related .2.5-fold in Elp3 and dom mutants compared to wild type genes in the humoral stress response of Drosophila. Biochem. (wandering-stage larvae). (B) Scatter plot showing the con- Biophys. Res. Commun. 284: 998–1003. Ekengren,S.,Y.Tryselius,M.S.Dushay,G.Liu,H.Steiner et al., cordance in fold change of the overlapping genes between 2001 A humoral stress response in Drosophila. Curr. Biol. 11: the dom and Elp3 comparisons. The red line shows a linear re- 714–718. 2 gression fit across the data (r ¼ 0.89). (C) Heat map showing Esberg, A., B. Huang,M.J.Johansson and A. S. Bystrom, overlap between the Elp3 and dom differential gene lists (ex- 2006 Elevated levels of two tRNA species bypass the require- pression changing by more than 3-fold). Genes have been ment for elongator complex in transcription and exocytosis. clustered using euclidean distance and a complete hierarchi- Mol. Cell 24: 139–148. cal method. Fellows, J., H. Erdjument-Bromage,P.Tempst and J. Q. Svejstrup, 2000 The Elp2 subunit of elongator and elongating RNA polymerase II holoenzyme is a WD40 repeat protein. J. Biol. Chem. 275: 12896–12899. that the effect of Elp3 deletion in the fly could also be Fletcher, J. C., and C. S. Thummel, 1995 The ecdysone-inducible due to alterations in chromatin regulation. In general, Broad-complex and E74 early genes interact to regulate target the Drosophila genome-wide expression data presented gene transcription and Drosophila metamorphosis. Genetics 141: 1025–1035. here complement gene expression analysis in Arabidop- Gentleman, R. C., V. J. Carey,D.M.Bates,B.Bolstad,M.Dettling sis, mouse, and humans, further supporting a role for et al., 2004 Bioconductor: open software development for com- Elongator during transcript elongation and chromatin putational biology and bioinformatics. Genome Biol. 5: R80. lose hen Gilbert, C., A. Kristjuhan,G.S.Winkler and J. Q. Svejstrup, remodeling (C et al. 2006; C et al. 2009; 2004 Elongator interactions with nascent mRNA revealed by Nelissen et al. 2010). Our data also reveal important RNA immunoprecipitation. Mol. Cell 14: 457–464. roles for Elongator in ecdysone regulation, morphogen- Greenwood, C., L. A. Selth,A.B.Dirac-Svejstrup and J. Q. Svejstrup, 2009 An iron-sulfur cluster domain in Elp3 impor- esis, and the suppression of melanotic nodules during tant for the structural integrity of elongator. J. Biol. Chem. 284: development. 141–149. Functions of Drosophila Elongator 1075

Hanratty,W.P.,andC.R.Dearolf, 1993 The Drosophila Tumorous- Okada, Y., K. Yamagata,K.Hong,T.Wakayama and Y. Zhang, lethal hematopoietic oncogene is a dominant mutation in the 2010 A role for the elongator complex in zygotic paternal ge- hopscotch locus. Mol. Gen. Genet. 238: 33–37. nome demethylation. Nature 463: 554–558. Harrison, D. A., R. Binari,T.S.Nahreini,M.Gilman and N. Otero, G., J. Fellows,Y.Li,T.de Bizemont,A.M.G.Dirac et al., Perrimon, 1995 Activation of a Drosophila Janus kinase 1999 Elongator, a multisubunit component of a novel RNA ( JAK) causes hematopoietic neoplasia and developmental de- polymerase II holoenzyme for transcriptional elongation. Mol. fects. EMBO J. 14: 2857–2865. Cell 3: 109–118. Hawkes, N. A., G. Otero,G.S.Winkler,N.Marshall,M.E. Paraskevopoulou, C., S. A. Fairhurst,D.J.Lowe,P.Brick and S. Dahmus et al., 2002 Purification and characterization of the hu- Onesti, 2006 The Elongator subunit Elp3 contains a Fe4S4 clus- man elongator complex. J. Biol. Chem. 277: 3047–3052. ter and binds S-adenosylmethionine. Mol. Microbiol. 59: 795–806. Huang, B., M. J. Johansson and A. S. Bystrom, 2005 An early step Parker, L., S. Gross and L. Alphey, 2001 Vectors for the expres- in wobble uridine tRNA modification requires the Elongator sion of tagged proteins in Drosophila. Biotechniques 31: 1280– complex. RNA 11: 424–436. 1282, 1284, 1286. Irizarry, R. A., B. M. Bolstad,F.Collin,L.M.Cope,B.Hobbs et al., Pokholok, D. K., N. M. Hannett and R. A. Young, 2002 Exchange 2003 Summaries of Affymetrix GeneChip probe level data. Nu- of RNA polymerase II initiation and elongation factors during cleic Acids Res. 31: e15. gene expression in vivo. Mol. Cell 9: 799–809. Irving, P., J. M. Ubeda,D.Doucet,L.Troxler,M.Lagueux et al., Qiu, P., P. C. Pan and S. Govind, 1998 A role for the Drosophila 2005 New insights into Drosophila larval haemocyte functions Toll/Cactus pathway in larval hematopoiesis. Development through genome-wide analysis. Cell. Microbiol. 7: 335–350. 125: 1909–1920. Kim, J. H., W. S. Lane and D. Reinberg, 2002 Human Elongator fa- Rahl, P. B., C. Z. Chen and R. N. Collins, 2005 Elp1p, the yeast cilitates RNA polymerase II transcription through chromatin. homolog of the FD disease syndrome protein, negatively regu- Proc. Natl. Acad. Sci. USA 99: 1241–1246. lates exocytosis independently of transcriptional elongation. Kiss,I.,A.H.Beaton,J.Tardiff,D.Fristrom andJ.W.Fristrom, Mol. Cell 17: 841–853. 1988 Interactions and developmental effects of mutations in the Simpson, C. L., R. Lemmens,K.Miskiewicz,W.J.Broom,V.K. Broad-Complex of Drosophila melanogaster. Genetics 118: 247–259. Hansen et al., 2009 Variants of the elongator protein 3 (ELP3) Kouskouti, A., and I. Talianidis, 2005 Histone modifications de- gene are associated with motor neuron degeneration. Hum. fining active genes persist after transcriptional and mitotic inac- Mol. Genet. 18: 472–481. tivation. EMBO J. 24: 347–357. Sinenko, S. A., E. K. Kim,R.Wynn,P.Manfruelli,I.Ando et al., Kristjuhan, A., J. Walker,N.Suka,M.Grunstein,D.Roberts et al., 2004 Yantar, a conserved arginine-rich protein is involved in 2002 Transcriptional inhibition of genes with severe histone h3 Drosophila hemocyte development. Dev. Biol. 273: 48–62. hypoacetylation in the coding region. Mol. Cell 10: 925–933. Solinger,J.A.,R.Paolinelli,H.Kloss,F.B.Scorza,S.Marchesi Krogan, N. J., and J. F. Greenblatt, 2001 Characterization of a six- et al., 2010 The Caenorhabditis elegans Elongator complex regu- subunit Holo-Elongator complex required for the regulated ex- lates neuronal alpha-tubulin acetylation. PLoS Genet. 6: e1000820. pression of a group of genes in Saccharomyces cerevisiae. Mol. Stergiopoulos, K., P. Cabrero,S.A.Davies and J. A. Dow, Cell. Biol. 21: 8203–8212. 2009 Salty dog, an SLC5 symporter, modulates Drosophila re- Kurucz, E., C. J. Zettervall,R.Sinka,P.Vilmos,A.Pivarcsi et al., sponse to salt stress. Physiol. Genomics 37: 1–11. 2003 Hemese, a hemocyte-specific transmembrane protein, af- Stramer, B., W. Wood,M.J.Galko,M.J.Redd,A.Jacinto et al., fects the cellular immune response in Drosophila. Proc. Natl. 2005 Live imaging of wound inflammation in Drosophila em- Acad. Sci. USA 100: 2622–2627. bryos reveals key roles for small GTPases during in vivo cell mi- Kwon, S. Y., H. Xiao,B.P.Glover,R.Tjian,C.Wu et al., 2008 The gration. J. Cell Biol. 168: 567–573. nucleosome remodeling factor (NURF) regulates genes involved Svejstrup, J. Q., 2007 Elongator complex: How many roles does it in Drosophila innate immunity. Dev. Biol. 316: 538–547. play? Curr. Opin. Cell Biol. 19: 331–336. Lagueux, M., E. Perrodou,E.A.Levashina,M.Capovilla and J. A. Winkler, G. S., T. G. Petrakis,S.Ethelberg,M.Tokunaga,H. Hoffmann, 2000 Constitutive expression of a complement-like Erdjument-Bromage et al., 2001 RNA polymerase II Elongator protein in toll and JAK gain-of-function mutants of Drosophila. holoenzyme is composed of two discrete subcomplexes. J. Biol. Proc. Natl. Acad. Sci. USA 97: 11427–11432. Chem. 276: 32743–32749. Li, H. M., G. Buczkowski,O.Mittapalli,J.Xie,J.Wu et al., Winkler, G. S., A. Kristjuhan,H.Erdjument-Bromage,P.Tempst 2008 Transcriptomic profiles of Drosophila melanogaster third and J. Q. Svejstrup, 2002 Elongator is a histone H3 and H4 instar larval midgut and responses to oxidative stress. Insect Mol. acetyltransferase important for normal histone acetylation levels Biol. 17: 325–339. in vivo. Proc. Natl. Acad. Sci. USA 99: 3517–3522. Li, Y., Y. Takagi,Y.Jiang,M.Tokunaga,H.Erdjument-Bromage Wittschieben, B. O., G. Otero,T.de Bizemont,J.Fellows,H. et al., 2001 A multiprotein complex that interacts with RNA Erdjument-Bromage et al., 1999 A novel histone acetyltrans- polymerase II elongator. J. Biol. Chem. 276: 29628–29631. ferase is an integral subunit of elongating RNA polymerase II ho- Metivier, R., G. Penot,M.R.Hubner,G.Reid,H.Brand et al., loenzyme. Mol. Cell 4: 123–128. 2003 Estrogen receptor-alpha directs ordered, cyclical, and Wittschieben, B. O., J. Fellows,W.Du,D.J.Stillman and J. Q. combinatorial recruitment of cofactors on a natural target pro- Svejstrup, 2000 Overlapping roles for the histone acetyltrans- moter. Cell 115: 751–763. ferase activities of SAGA and Elongator in Vivo. EMBO J. 19: Nelissen, H., D. Fleury,L.Bruno,P.Robles,L.De Veylder et al., 3060–3068. 2005 The elongata mutants identify a functional Elongator Zettervall,C.J.,I.Anderl,M.J.Williams,R.Palmer,E.Kurucz complex in plants with a role in cell proliferation during organ et al., 2004 A directed screen for genes involved in Drosophila growth. Proc. Natl. Acad. Sci. USA 102: 7754–7759. blood cell activation. Proc. Natl. Acad. Sci. USA 101: 14192–14197. Nelissen, H., S. De Groeve,D.Fleury,P.Neyt,L.Bruno et al., 2010 Plant Elongator regulates auxin-related genes during RNA polymerase II transcription elongation. Proc. Natl. Acad. Sci. USA 107: 1678–1683. Communicating editor: J. Tamkun