Logo of geneticsGeneticsCurrent IssueInformation for AuthorsEditorial BoardSubscribeSubmit a Manuscript
Genetics. Jul 2010; 185(3): 811–822.
PMCID: PMC2907203

Molecular Genetic Analysis of Chd3 and Polytene Chromosome Region 76B-D in Drosophila melanogaster


The Drosophila melanogaster Chd3 gene encodes a member of the CHD group of SNF2/RAD54 ATPases. CHD proteins are conserved from yeast to man and many are subunits of chromatin-remodeling complexes that facilitate transcription. Drosophila CHD3 proteins are not found in protein complexes, but as monomers that remodel chromatin in vitro. CHD3 colocalize with elongating RNA polymerase II on salivary gland polytene chromosomes. Since the role of Chd3 in development was unknown, we isolated and characterized the essential genes within the 640-kb region of the third chromosome (polytene chromosome region 76B-D) that includes Chd3. We recovered mutations in 24 genes that are essential for zygotic viability. We found that transposon-insertion mutants for 46% of the essential genes are included in the Drosophila Gene Disruption Project collection. None of the essential genes that we identified are in a 200-kb region that includes Chd3. We generated a deletion of Chd3 by targeted gene replacement. This deletion had no effect on either viability or fertility.

THE SNF2/RAD54 family proteins are components of ATP-dependent chromatin-remodeling complexes involved in transcription and/or DNA repair (Eisen et al. 1995; Pollard and Peterson 1998). Within the SNF2/RAD54 family, the CHD (chromodomain-helicase-DNA-binding) group all contain CHROMO (chromatin organization modifier) domains, which interact with methylated lysines in the N-terminal tails of histones (Jacobs and Khorasanizadeh 2002; Nielsen et al. 2002). The CHD group can be further divided into at least three classes on the basis of the presence or absence of two other protein domains. The CHD1 class proteins have only the SNF2/RAD54 ATPase domain and CHROMO domains. The sole CHD group protein in Saccharomyces cerevisiae is a CHD1 class protein (Sc CHD1) that is not essential for viability (Woodage et al. 1997). In addition to the SNF2/RAD54 and CHROMO domains, the KISMET class proteins also have a BRK (brahma and kismet) domain, while the MI-2 (or CHD3/CHD4) class proteins have one or more PHD (plant homeo domain) zinc fingers near the amino terminus. Drosophila melanogaster has a single CHD1 class gene, Chd1. While zygotically expressed Chd1 is not required for viability, maternally encoded Chd1 gene products are required for early embryogenesis (i.e., Chd1 mutants are female sterile) (Konev et al. 2007; McDaniel et al. 2008). The sole KISMET class gene in D. melanogaster, kismet, is required for zygotic viability and the function of homeotic genes (Daubresse et al. 1999). Mi-2 and Chd3, the two MI-2 class genes in D. melanogaster, are both within the Df(3L)kto2 region of the genome. Mi-2 is required for zygotic viability (Kehle et al. 1998). The expression of Chd3 has been shown to be developmentally regulated (Murawska et al. 2008), and CHD3 proteins colocalize with elongating RNA polymerase II on Drosophila salivary gland polytene chromosomes (Murawska et al. 2008). What is the role of Chd3 in development? As part of a genetic investigation of the chromosomal region that spans polytene chromosomal bands 76B1 to 76D5, we have identified 24 essential genes around Chd3. Since Chd3 was not among the essential genes that we identified, we generated a targeted gene replacement allele of Chd3. Deletion of the majority of the open reading frame of Chd3 had no effects on either viability or fertility.


Flies were raised on a yeast/cornmeal/molasses/Tegosept medium at 25°. All mutations and chromosome aberrations are described in Lindsley and Zimm (1992) or FlyBase (http://flybase.org/) except Df(3L)A23, Df(3L)XS917, and Df(3L)kto22. Df(3L)A23 and Df(3L)XS917 were generated in the laboratories of David Stein and Gerry Rubin, respectively and provided by Jürg Müller. We recovered Df(3L)kto22 as a flanking deletion from P{lacW}trcS066917a. Putative deletions were detected by a change in eye color after crossing to the balancer TMS (which carries a transposon that expresses the P transposase). Because the P{lacW} transposon is still present in Df(3L)kto22, we used inverse PCR (Huang et al. 2000) to determine the molecular endpoints of the deletion, which spans from the site of insertion of P{lacW}trcS066917a to midway between the 3′ ends of Papss and Rab8.

Males were fed ethylmethane sulfonate (EMS) as described (Lewis and Bacher 1968; Kennison 1983). The mutagenized males were mated to virgin females and discarded after 4 days; the inseminated females were returned to new cultures for subsequent brooding. Mutagenized males were homozygous for either an unmarked chromosome from the iso-1 strain (Brizuela et al. 1994) or a third chromosome carrying red1 and e4. Both third chromosomes were isogenized prior to mutagenesis. We recovered mutations that failed to complement Df(3L)kto2 from three different experiments. Following the nomenclature suggested by Lindsley and Zimm (1992), we named these mutations l(3)76BD. In the first experiment, single sons with a mutagenized red1 e4 third chromosome were mated to three to five Df(3L)kto2/TM6C virgins. If few or no progeny carrying both the mutagenized third chromosome and Df(3L)kto2 were recovered, a balanced stock of l(3)76BD red1 e4/TM6C was selected and retested. From 5576 fertile sons, we recovered 124 l(3)76BD mutations. In the second and third experiments, balanced lines with a mutagenized third chromosome and TM6B were first generated. Only those lines in which few or no flies homozygous for the mutagenized third chromosome survived were subsequently tested by crossing to Df(3L)kto2/TM6C virgins. In the second experiment, from ~200 EMS-treated iso-1 third chromosome lines, we recovered four l(3)76BD mutations. In the third experiment, from 1938 EMS-treated red1 e4 third chromosome lines (which we estimated from the Poisson distribution to contain 3342 lethal mutations on the third chromosome), we recovered 67 l(3)76BD mutations.

We used another approach to try to tag additional genes in this region with P-element insertions. During P-element transposition, it has been reported that new insertion sites are more frequent than expected near the original insertion site (Tower et al. 1993; Zhang and Spradling 1993, 1994; Tower and Kurapati 1994; Matsubayashi and Yamamoto 1998). This is referred to as local P-element transposition. We hoped that this would be useful in isolating additional lethal P-insertion mutations in this region of the genome. We should point out that this is different from the generation of flanking deletions by exposure of a P-transposon insertion to P-element transposase, which is also sometimes referred to as local P-element transposition. To minimize the number of flanking deletions recovered, we started with P{lacW}Mi-2j3D4 Su(Tpl)j3D4, which is located in the first introns of both the Mi-2 and Su(Tpl) genes. This insertion mutant completely fails to complement Mi-2 alleles, but complements Su(Tpl) alleles well (at least 80% of the expected flies eclose). We selected new P{lacW} insertions that complement Mi-2. The crossing scheme is shown in Figure 1.

Figure 1.
Crossing scheme for local P-element transposition. Transpositions of the P-element insertion P{lacW}Mi-2j3D4 were generated in flies heterozygous to the TMS balancer chromosome, which carries a nonmobilizable transgene that produces the ...

To construct the vector for targeted gene replacement, we started with the “ends-out” arm-GFP vector constructed for the mre11 gene replacement (Min et al. 2005). For the downstream flanking DNA fragment, we used the primers 5′-TACGGCACTAGTCGACGAGTGTAATTGACTCC-3′ and 5′-CATCCACAATGATAGTCTAGGTAGC-3′ to PCR amplify a genomic fragment. We cut the PCR product with SpeI and cloned the resulting 3.4-kb fragment into the SpeI site in the arm-GFP vector. For the upstream flanking DNA, we used the primers 5′-TGACTCGGTACCGACTGTTGACGGTCTTGCCGCTG-3′ and 5′-CGTAGCTGGTCAGCATTACG-3′ to PCR amplify a genomic fragment. We cut the PCR product with KpnI and HindIII and cloned the resulting 2.2-kb fragment into the KpnI HindIII sites in the arm-GFP vector. A KpnI–NotI fragment with both flanking DNA fragments and the arm-GFP fragment was cloned into pW30 (Gong and Golic 2003). This construct was co-injected with P transposase into white mutant embryos and stable transformants were selected using the mini-white marker in pW30. The targeted replacement was generated and selected as a GFP-positive embryo as described (Min et al. 2005). The targeted replacement was verified by the failure to amplify by PCR Chd3 genomic sequences in Chd31 homozygotes, using the primers CHD3-2 (5′-GGCGATTGGATCTGCCCGCG-3′), CHD3-5 (5′-TCTCTGTGGACGCAGCCTTTC-3′), CHD3-6 (5′-CGGATGTATTGAAGAGCATGC-3′), CHD3-7 (5′-ATCACAGGGTCCTTTTATTC-3′), CHD3-9 (5′-CGAGGAGCATGATGACTACAA-3′), CHD3-10 (5′-CTCTTCGTTTTGTCATTTGTC-3′), CHD3-11 (5′-GTACCCAATGGCGGTTATTCC-3′), CHD3-12 (5′-CCCAAGACTCCTTCTCTTCA-3′), and CHD3-19 (5′-CTATGATTGATAACCCGCTGG-3′).

For the evolutionary comparisons, phylograms and cladograms were generated using the ClusalW2, TreeView32, and TreeViewX programs. Only the 11 Drosophila species that can be searched with the BLAT program on the UCSC Genome Browser website (http://genome.ucsc.edu/; Kent 2002) were examined for most of the species comparisons. The two exceptions were MI-2/CHD3 and lodestar/CG10445, for which NCBI BLAST (http://blast.ncbi.nlm.nih.gov/) was used to identify and characterize the D. willistoni genes. For the determination of evolutionarily conserved amino acids, we used the EvoPrinter version 1.1 (http://evoprinter.ninds.nih.gov/index11.html; Odenwald et al. 2005). Since the program will only compare a maximum of nine species, we used D. melanogaster and the eight most distantly related Drosophila species (D. erecta, D. yakuba, D. annanassae, D. pseudoobscura, D. persimilis, D. virilis, D. mojavensis, and D. grimshawi) with available BLAT files. Only those amino acid residues that were identical in all nine Drosophila species were counted as evolutionarily conserved. The sequences used to construct the cladogram of the OAK cluster genes are in supporting informaton, File S1.

The following primers were used for the PCR analysis of the 31.4-kb tandem duplication that includes Gyc76C: primer 1, 5′-TGGCAACGAACTCTAGGGACT-3′; primer 2, 5′-GCAGAGGAACTACTTAGCAGTC-3′; primer 3, 5′-ACCAACCACCTCCAACTAGAG-3′; and primer 4, 5′-GGGCACGTTTTTAGATGAGCTTC-3′.


A cladogram of the SNF2/RAD54 family proteins from D. melanogaster, S. cerevisiae, and Homo sapiens is shown in Figure 2. There are four CHD proteins in D. melanogaster. The MI-2 class of CHD proteins is distinguished by the presence of one or more PHD domains. The two Mi-2 class genes in D. melanogaster, Mi-2 and Chd3, are both within the chromosomal region deleted in Df(3L)kto2. The ~640 kb of genomic DNA deleted by Df(3L)kto2 is shown in Figure 3, with the genes annotated by the Drosophila Genome Project. We first present our efforts to identify all of the essential genes within Df(3L)kto2 and then our efforts to generate and characterize a Chd3 mutant.

Figure 2.
Evolutionary relationships of SNF2/RAD54 family proteins. A cladogram of the SNF2/RAD54 family proteins from Homo sapiens (31 Hs proteins in black), Drosophila melanogaster (17 Dm proteins in red), and Saccharomyces cerevisiae (17 Sc proteins in blue). ...
Figure 3.
Molecular map of the genomic region deleted in Df(3L)kto2. The ~640 kb of genomic DNA (from 3L: 19291k to 19927k, release 5.23) is broken into three parts (A, B, and C), and is represented by the horizontal black arrows at the top of each part. ...

After EMS mutagenesis, we recovered 195 l(3)76BD mutations that failed to complement Df(3L)kto2. These mutations define 24 complementation groups, which we provisionally designated l(3)76BDa through l(3)76BDx. These complementation groups and the number of alleles that we recovered for each are shown in Table 1. We recovered an average of 8.1 alleles per complementation group, with only one complementation group [l(3)76BDp] represented by a single allele. To further localize our complementation groups, we also crossed representatives of each complementation group to the 26 chromosomal deletions in Figure 4 that overlap Df(3L)kto2 to differing extents. Thirteen of the deletions (those indicated by the red bars in Figure 4) have molecularly defined breakpoints, which are useful in integrating the genetic and molecular maps.

Figure 4.
Complementation of essential genes with a set of deletions that overlap Df(3L)kto2. The essential genes are listed at the top of the map in blue, with a horizontal bar for each deletion indicating which genes failed to complement that deletion. Not shown ...
Complementation groups represented by the mutations within the Df(3L)kto2 region of the genome

We also tested all of the putative lethal transposon insertion mutants in this 640-kb region that the Drosophila Gene Disruption Project has made available from the Bloomington Drosophila Stock Center. These transposon insertion mutants are listed in Table 2, and include P- (P), piggyBac- (PBac), and Minos- (Mi) transposable element insertions. Four of the transposon insertion mutants complement Df(3L)kto2 for viability and fertility, indicating that the lethality of the insertion chromosome is not due to disruption of the associated gene. The remaining 19 transposon insertion mutants fail to complement one or more of our complementation groups. The complementation groups that failed to complement each transposon insertion mutant are shown in Table 2. For all of the transposon insertion mutants from the Drosophila Gene Disruption Project, the location from the deletion mapping coincides with the location of the transposon insertion. Eleven of our 24 complementation groups (46%) were represented by transposon insertion mutants from the Drosophila Gene Disruption Project. This is only slightly higher than the 40% frequency reported by the Gene Disruption Project for the entire autosomal collection 6 years ago (Bellen et al. 2004). We used this information to assign 9 of our complementation groups in Table 1 to molecularly identified genes. Because of the overlapping nature of the Mi-2 and Su(Tpl) transcription units, the locations of the transposon insertion mutants were insufficient to comfortably assign the l(3)76BDe and l(3)76BDs complementation groups.

Transposon insertion mutants identified by the Drosophila Gene Disruption Project and maintained by the Bloomington Drosophila Stock Center

Since more than half of the essential genes in this region were not represented in the Drosophila Gene Disruption Project collection, we wanted to isolate additional P-transposon insertion mutants within this region of the genome to correlate more of our genetic complementation groups with molecularly identified genes. We chose to use local P transposition to isolate additional lethal transposon mutations within Df(3L)kto2. We started with a P{lacW} insertion mutant of Mi-2. Among ~68,000 progeny, we recovered and tested 475 putative local P-element transpositions (see materials and methods). Twenty-five of the local transpositions failed to complement Df(3L)kto2 for viability. By crossing to representative deletions and mutations, we found that in all 25 putative local transpositions, the failure to complement Df(3L)kto2 mapped to Su(Tpl). The majority (22/25) of the new insertion chromosomes also showed variegation of the mini-white gene in P{lacW}. This variegation is suppressed when heterozygous to Su(z)12 alleles, suggesting that the variegation is due to repeat-induced silencing (Dorer and Henikoff 1994). We characterized 18 of the local transpositions by inverse PCR. The original P{lacW} insertion appeared to be present in all 18. We characterized six of these chromosomes in more detail. We could not detect any changes in the flanking DNA on either side of the original P{lacW} insertion and inverse PCR gave multiple bands for most of the local transpositions. Sequencing of the PCR products showed that one product from each transposition was usually from the original insertion, and the additional bands were from insertions of multiple P{lacW} elements into the original P{lacW} insertion. Repeat-induced silencing caused by the insertion of multiple P{lacW} elements appears to restore at least some function for the Mi-2 gene, while reducing function of the Su(Tpl) gene. The Su(Tpl) transcription unit is entirely within the first large intron of the Mi-2 transcription unit (and transcribed from the same DNA strand). Both genes are ubiquitously expressed during embryogenesis. The ability of repeat-induced silencing of a transposon insertion to alter function of both genes in opposite ways is intriguing. However, the important point for this study is the failure of local transposition to identify any new insertion mutants in the surrounding genes in the region. As in the original studies (Tower et al. 1993; Zhang and Spradling 1993), the vast majority of local transpositions are within the original P-transposon insertion. While these data do not help to identify additional transcription units with essential functions within this genomic region, they do suggest that local transposition does not occur at a sufficiently high frequency to make it a useful practical approach to tagging new genes.

In another approach to correlate transcription units with our complementation groups, we crossed our new mutants to mutants in previously identified genes. We assigned 4 of the complementation groups [l(3)76BDb, l(3)76BDc, l(3)76BDi, and l(3)76BDp] to previously identified genes by their failure to complement the mutations ash16 (Tripoulas et al. 1994), kto1 (Kennison and Tamkun 1988; Treisman 2001), trpml1(Venkatachalam et al. 2008), and pipC14 (Zhu et al. 2005), respectively. In addition, mutant alleles from six of our complementation groups were sequenced in earlier reports by ourselves and others [l(3)76BDa (Geng et al. 2000), l(3)76BDe (Kehle et al. 1998), l(3)76BDo (Birve et al. 2001), l(3)76BDq (Moshkin et al. 2002), l(3)76BDs (Eissenberg et al. 2002), and l(3)76BDv (Veraksa et al. 2002)] and assigned to the transcription units listed in Table 1. Finally, HLH106 has been reported to be essential (Horn et al. 2003), but we were unable to obtain the original allele to test for complementation with our mutants. Our data from the deficiency mapping suggested that l(3)76BDw was the best candidate. We sequenced the HLH106 gene from three alleles of our l(3)76BDw complementation group. All three alleles were isolated on the red1 e4 chromosome, which differs from the iso-1 sequence at amino acid 224 (F in iso-1 and L in the red1 e4 marked chromosome). In addition, all three mutant alleles have single GC to AT bp transitions that introduce stop codons and prematurely truncate the HLH106 protein. HLH106 alleles 1, 2, and 3 change amino acid residues Q715, W508, and W719, respectively, to stop codons. Using all of this information, we have been able to assign 19 of our 24 complementation groups to identified genes.

Chd3 is in a 200-kb region in which we have not identified any lethal complementation groups (Figure 3A). The CHD3 protein is 892 amino acids in length compared to 1982 amino acids for MI-2. Since we recovered 16 alleles of Mi-2, we might have expected to recover at least seven alleles of Chd3, if gene mutability is related to the number of amino acids in the encoded proteins. However, it was possible that Chd3 is essential for zygotic viability and that we had failed to isolate mutant alleles by chance. It was also possible that Chd3 is not required for zygotic viability, but is required for fertility of males and/or females. We decided to use ends-out, or replacement, gene targeting techniques (Gong and Golic 2003; Min et al. 2005) to make a deletion of most of the Chd3 open reading frame. The scheme for generating the targeted deletion is shown in Figure 5. Although the CG33062 gene model has been withdrawn by the FlyBase Genome Annotators (http://flybase.org/reports/FBgn0053062.html), at the time that we designed our targeted deletion, the close proximity of the CG33062 and Chd3 transcription start sites was a concern. Another factor that influenced the design of our deletion was the presence of several HindIII restriction sites in close proximity (HindIII is one of the restriction sites in the vector into which DNA flanking the targeted deletion is inserted). To avoid possible effects on transcription of CG33062, we decided to leave the Chd3 promoter intact and to replace 1898 bp of the Chd3 transcription unit (and an additional 1175 bp downstream) with a transgene that expresses the GFP protein from an armadillo gene promoter (Min et al. 2005). Our mutant allele, which we call Chd31, lacks the final 529 amino acids (~60% of the protein), including ~80% of the SNF2 domain. We verified the presence of the deletion in the Chd31 mutant chromosome by the failure to amplify by PCR Chd3 genomic sequences in the Chd31 homozygotes (shown in Figure 5B). We examined transcript levels of the remaining 5′ region of Chd3 in Chd31 homozygotes using quantitative RT–PCR. Surprisingly, the Chd31 mutant transcripts are present at higher levels than Chd3 transcripts in wild type (compared to RpL32 transcript levels) in both embryos and first instar larvae (data not shown). Chd31 homozygotes and hemizygotes show no reduction in viability, nor any discernible effects on fertility of either males or females. Thus, we conclude that Chd3 is dispensable for viability and fertility. Although we have not noted any phenotypes associated with loss of Chd3 function, we have not carefully examined the Chd31 mutant flies for changes in behavior, learning, or responses to environmental challenges.

Figure 5.
Scheme for generating the Chd31 allele. (A) The P-element transposon for generating the Chd31 deletion allele is shown at the top, integrated into genomic DNA at a random location. The P-element terminal repeats are indicated by upward pointed black triangles. ...

There are several features of the 76B-D genomic region that we believe merit discussion. The first feature is the presence of a 31.4 -kb tandem duplication (present at the left of Figure 3C) that involves three transcription units. Two of the transcription units are completely duplicated (trpml/CG42638 and CG42529/CG14101). Only the coding exons of the third gene (CG42636/CG42637, also known as Gyc76C) are duplicated, with several 5′ noncoding exons proximal to, and outside of, the tandem duplication. The duplication is flanked by Doc transposable elements. Tandem duplications appear to be generated during meiosis at appreciable frequencies (Green 1959, 1961; Gelbart and Chovnick 1979). One mechanism for the generation of tandem duplications is the unequal exchange between different insertions of the same transposable element (Goldberg et al. 1983). In the present example, unequal exchange between the Doc transposable elements could have generated the duplication in the sequenced iso-1 strain. We isolated EMS-induced mutants for both trpml and Gyc76C in the red1 e4 strain, which would be unlikely if the coding regions of both genes are duplicated in this strain. It is possible that the duplication is present only in the iso-1 strain. To test for the presence of this tandem duplication, we designed primers flanking each of the Doc transposable elements in the genomic sequence and used them for PCR analysis. Each of these primers was ~200 bp from the ends of the Doc elements. The locations of these primers are shown in Figure 3C. We confirmed the presence of the distal and proximal Doc insertions in the iso-1 strain. These Doc insertions were not present in any of the other strains that we examined (Oregon R, Canton-S, and the red1 e4 strain in which our mutants were induced). We were unable to confirm the presence of the middle Doc element that forms the junction between the two duplicated segments in our copy of the iso-1 strain, even though the primers that should amplify this middle element are two of the same primers that were used to amplify the distal and proximal Doc elements under the same PCR conditions. Thus, we believe that this duplication arose after we constructed the iso-1 strain in 1986, but before the BAC and WGS genomic libraries were made in 1998 and 1999, respectively (Celniker et al. 2002). Another polymorphic tandem duplication of >30 kb was found by the Drosophila Genome Project in the iso-1 strain (Celniker et al. 2002). This duplication is flanked by hobo and cruiser transposable elements. Polymorphic duplications are common in human populations and may be a significant cause of human disease (Iafrate et al. 2004; Sebat et al. 2004; De Smith et al. 2008; Wain et al. 2009; Stankiewicz and Lupski 2010).

There are also several genes in this 640-kb region of the genome that appear to have arisen by tandem gene duplication before the divergence of the Drosophila genus. Approximately 40 kb proximal to Chd3 there is a cluster of three adjacent genes (CG9449, CG9451, and CG9452), which encode predicted acid phosphatases of ~400 amino acids each. These three putative acid phosphatases are more related to each other (~50% identical over the entire length) than to any other proteins in the genome. This cluster is conserved among all of the Drosophila species that we examined; however, we were unable to identify a homolog in the more distantly related dipteran, the mosquito Anopheles gambiae. Just proximal to the acid phosphatase cluster is another cluster of genes related to cuticular proteins. Only the first two genes (Cpr76Ba and Cpr76Bb) appear to have arisen by tandem gene duplication. Both predicted proteins are ~200 amino acids in length and are 44% identical. The other two genes, (Cpr76Bc and Cpr76Bd) are both predicted to encode much larger proteins with only limited homology to each other and to the Cpr76Ba and Cpr76Bb proteins. While all four of these genes are conserved in all of the sequenced Drosophila species that we examined, Cpr76Bb is the only annotated gene within Df(3L)kto2 for which no cDNA or EST has been recovered. In the mosquito A. gambiae, this cluster has expanded to include at least nine genes related to Cpr76Ba and Cpr76Bb. Finally, the most proximal pair of adjacent genes within the Df(3L)kto2 genomic region, which appear to have arisen by gene duplication, vermiform (verm) and serpentine (serp), both encode predicted proteins of just over 540 amino acids that are 57% identical along their entire lengths. Both genes are present in all of the Drosophila species examined, as well as in the mosquito A. gambiae. Both verm and serp are essential for zygotic viability.

Surrounding Chd3, there is an unusual cluster of related genes that we call the OAK cluster (after the 825-Oak gene). The D. melanogaster OAK cluster genes are the dark green transcription units in Figure 3A. There are 3 OAK cluster genes in A. gambiae and between 4 and 12 OAK cluster genes in each of the other Drosophila species examined. The 3 OAK cluster genes in A. gambiae are adjacent to each other. Most of the OAK cluster genes in Drosophila are between CG32206 and CG9449 (both of which are conserved between all 11 Drosophila species examined). The remainder of the OAK cluster genes (1 in D. persimilis, 2 in D. sechellia, and 5 in D. yakuba) are on small unlocalized contigs and could also be from the genomic regions between CG32206 and CG9449. The sequence comparisons of the OAK cluster proteins (the cladogram in Figure 6) suggest either that many of the duplications occurred after the divergence of each species or that there is concerted evolution actively constraining sequence divergence within a particular species. For example, 7 of the 8 genes in D. ananassae are more related to each other than to any other genes in the Drosophila genus. Similarly, all 6 of the D. grimshawi genes are more closely related to each other than to any of the other Drosophila genes. Another odd feature of the OAK cluster is that the distal and proximal genes in D. melanogaster overlap duplicated noncoding RNAs (CR32205 and pncr009:3L, the light green transcription units in Figure 3A) that are transcribed from the opposite DNA strand. The CR32205 and pncr009:3L RNA sequences are ~75% identical over their entire lengths. In addition to the OAK cluster in 76B, there are more distantly related clusters in the same chromosome arm (Muller element D) of all Drosophila species. In D. melanogaster, these include a cluster of 3 genes (CG13679, CG13678, and CG13674) in 66B12-66C1 and a cluster of 4 genes (CG13051, CG13066, CG13067, and CG13069) in 72D10.

Figure 6.
Cladogram of the OAK cluster proteins. Only the proteins from A. gambiae (Agam) and D. melanogaster (Dmel) are encoded by annotated genes. The proteins from D. grimshawi (Dgri), D. mojavensis (Dmoj), D. virilis (Dvir), D. pseudoobscura (Dpse), D. persimilis ...


D. melanogaster has two MI-2 type genes, but only one (Mi-2) appears to be essential for viability. Chd3, the second MI-2 type gene, is most highly expressed maternally (Murawska et al. 2008), but the maternal expression does not appear to be required for oogenesis or for the survival of the resulting progeny. Both MI-2 and CHD3 proteins are recruited to the same sites on polytene chromosomes, sites that also bind elongating RNA polymerase II (Murawska et al. 2008). At least one MI-2 type gene can be found in all multicellular eukaryotes, including A. gambiae and all Drosophila species examined. Only in the melanogaster subgroup, is there a second MI-2 type gene, Chd3 (Figure 7A). Mi-2 and Chd3 can be easily distinguished, as the MI-2 protein is almost twice the size of the CHD3 protein. Mi-2 has also been more highly conserved during evolution of the genus Drosophila (Figure 7B).

Figure 7.
Evolution of the Mi-2/Chd3 and lds/CG10445 gene pairs in the Drosophila lineage. (A) The phylogeny of the 12 sequenced Drosophila species. Mi-2 and lds are present in all species, while Chd3 is present only in the melanogaster subgroup and CG10445 is ...

It has been suggested that Chd3 originated as a duplication in the melanogaster subgroup from the integration of a truncated, reverse-transcribed Mi-2 mRNA, based on the lack of introns in Chd3 (Murawska et al. 2008). We would point out that the homology between Mi-2 and Chd3 in the melanogaster subgroup is entirely within a single exon of the Mi-2 genes. Thus, regardless of the mechanism of gene duplication, Chd3 would be expected to lack introns. The phylogram of MI-2 and CHD3 proteins in the genus Drosophila (Figure 7B) does not fit the proposal that Chd3 arose only in the melanogaster subgroup, but suggests that the original duplication occurred much earlier and might have been present in all of the Sophophora, with subsequent deletion in some species. Chd3 is within the OAK cluster of related genes. Unequal recombination between OAK cluster genes that flank Chd3 would delete Chd3 (Green 1959, 1961; Goldberg et al. 1983). Unequal recombination would also help to explain the concerted evolution of the OAK cluster genes (Williams et al. 1989). The transcription unit just distal to Chd3, CG33062, is found in all of the Sophophora, but not in the subgenus Drosophila, suggesting that it might have arisen during the same event that generated Chd3. Another of the SNF2/RAD54 related genes shown in Figure 2, the lodestar (lds) gene, also appears to be duplicated in the Sophophora (Figure 7, A and C). The duplicated gene, CG10445, is adjacent to lds in all species except D. willistoni, with lds and CG10445 divergently transcribed.

While there are large differences in gene mutability in D. melanogaster (Lefevre and Watkins 1986), the basis has not been clear. There are several factors that could contribute, such as differing lengths of transcription units, differing lengths of proteins, and differing degrees in the ability of proteins to tolerate single amino acid changes and still retain function. We have identified the transcription units for 19 of our essential complementation groups, and since the mutations in Table 1 were all selected in the same sets of experiments, we can use the numbers of alleles for each of the 19 complementation groups to assess the role of various parameters in causing differences in gene mutability. Of the parameters above, we find that the best correlations are between mutability and length of protein products. Figure 8 shows comparisons of EMS mutability with both the numbers of core amino acids (those present in all protein isoforms for a given gene) and with the numbers of evolutionarily conserved core amino acids. With the exceptions of serp and trpml, which are more mutable than expected, and CG32210, which is less mutable than expected, the correlations are striking. From the linear regressions, we would estimate that we have isolated an average of one mutation for every 123 core amino acids, or one mutation for every 73 evolutionarily conserved core amino acids.

Figure 8.
Correlations between the numbers of alleles recovered after EMS mutagenesis and the numbers of amino acids. (A) The correlation between the numbers of alleles and the number of core amino acids (the amino acid residues present in all protein isoforms ...

Another important conclusion from the work presented here is that the Drosophila Gene Disruption Project has been very successful in producing mutants for almost half of the transcription units, but that additional methods will be necessary to generate mutants for the other half of the transcription units. The more traditional approach that we have presented here, i.e., to saturate specific regions of the genome with radiation or chemically induced mutations, can identify the essential functions, but is not very effective in identifying mutations in nonessential genes. While targeted gene replacement (Rong and Golic 2000) can be used for any gene regardless of function, it is currently not efficient enough for large-scale production of mutants. Koundakjian et al. (2004) described a very powerful approach that could be used to identify both essential and nonessential genes. They produced a large number of stocks with highly mutagenized autosomes. Even heterozygotes from such a collection could be screened for mutations in specific genes using high-throughput methods, such as TILLING (targeting induced local lesions in genomes) (McCallum et al. 2000). Another approach is the generation of stocks with transgenes for conditional RNA interference (Dietzl et al. 2007; Ni et al. 2008). These conditional mutants are useful for characterizing the functions of both essential and nonessential genes. While tremendous progress has been made in the past 100 years, much work still remains before we completely understand the functions encoded by the D. melanogaster genome.

We thank David Stein, Jürg Müller, and Gerry Rubin for sharing unpublished information and mutant strains. We also thank Craig Montell and the Bloomington Drosophila Stock Center for providing mutant strains. Finally, we are grateful to Judy Kassis for helpful comments on both the experiments and the manuscript. This research was supported by the Intramural Research Program of the National Institutes of Health, National Institute of Child Health and Human Development.


Supporting information is available online at http://www.genetics.org/cgi/content/full/genetics.110.115121/DC1.


  • Bellen, H. J., R. W. Levis, G. Liao, J. W. Carlson, G. Tsang et al., 2004. The BDGP gene disruption project: single transposon insertions associated with 40% of Drosophila genes. Genetics 167 761–781. [PMC free article] [PubMed]
  • Birve, A., A. K. Sengupta, D. Beuchle, J. Larsson, J. A. Kennison et al., 2001. Su(z)12, a novel Drosophila Polycomb group gene that is conserved in vertebrates and plants. Development 128 3371–3379. [PubMed]
  • Brizuela, B. J., L. Elfring, J. Ballard, J. W. Tamkun and J. A. Kennison, 1994. Genetic analysis of the brahma gene of Drosophila melanogaster and polytene chromosome subdivisions 72AB. Genetics 137 803–813. [PMC free article] [PubMed]
  • Celniker, S. E., D. A. Wheeler, B. Kronmiller, J. W. Carlson, A. Halpern et al., 2002. Finishing a whole-genome shotgun: release 3 of the Drosophila melanogaster euchromatic genome sequence. Genome Biol. 3 0079.1–0079.14. [PMC free article] [PubMed]
  • Daubresse, G., R. Deuring, L. Moore, O. Papoulas, I. Zakrajsek et al., 1999. The Drosophila kismet gene is related to chromatin-remodeling factors and is required for both segmentation and segment identity. Development 126 1175–1187. [PubMed]
  • de Smith, A. J., R. G. Walters, P. Froguel and A. I. Blakemore, 2008. Human genes involved in copy number variation: mechanisms of origin, functional effects and implications for disease. Cytogenet. Genome Res. 123 17–26. [PMC free article] [PubMed]
  • Dietzl, G., D. Chen, F. Schnorrer, K. C. Su, Y. Barinova et al., 2007. A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature 448 151–156. [PubMed]
  • Dorer, D. R., and S. Henikoff, 1994. Expansions of transgene repeats cause heterochromatin formation and gene silencing in Drosophila. Cell 77 993–1002. [PubMed]
  • Eisen, J. A., K. S. Sweder and P. C. Hanawalt, 1995. Evolution of the Snf2 family of proteins: subfamilies with distinct sequences and functions. Nucleic Acids Res. 23 2715–2723. [PMC free article] [PubMed]
  • Eissenberg, J. C., J. Ma, M. A. Gerber, A. Christensen, J. A. Kennison et al., 2002. dELL is an essential RNA polymerase II elongation factor with a general role in development. Proc. Natl. Acad. Sci. USA 99 9894–9899. [PMC free article] [PubMed]
  • Gelbart, W. M., and A. Chovnick, 1979. Spontaneous unequal exchange in the rosy region of Drosophila melanogaster. Genetics 92 849–859. [PMC free article] [PubMed]
  • Geng, W., B. He, M. Wang and P. N. Adler, 2000. The tricornered gene, which is required for the integrity of epidermal cell extensions, encodes the Drosophila nuclear DBF2-Related kinase. Genetics 156 1817–1828. [PMC free article] [PubMed]
  • Goldberg, M. L., J.-Y. Sheen, W. J. Gehring and M. M. Green, 1983. Unequal crossing-over associated with asymmetrical synapsis between nomadic elements in the Drosophila melanogaster genome. Proc. Natl. Acad. Sci. USA 80 5017–5021. [PMC free article] [PubMed]
  • Gong, W. J., and K. G. Golic, 2003. Ends-out, or replacement, gene targeting in Drosophila. Proc. Natl. Acad. Sci. USA 100 2556–2561. [PMC free article] [PubMed]
  • Green, M. M., 1959. Non-homologous pairing and crossing over in Drosophila melanogaster. Genetics 44 1243–1256. [PMC free article] [PubMed]
  • Green, M. M., 1961. Further data on non-homologous pairing and crossing over in Drosophila melanogaster. Genetics 46 1555–1560. [PMC free article] [PubMed]
  • Horn, C., N. Offen, S. Nystedt, U. Hacker and E. A. Wimmer, 2003. piggyBac-based insertional mutagenesis and enhancer detection as a tool for functional insect genomics. Genetics 163 647–661. [PMC free article] [PubMed]
  • Huang, A. M., E. J. Rehm and G. M. Rubin, 2000. Recovery of DNA sequences flanking P-element insertions: inverse PCR and plasmid rescue, pp. 429–437 in Drosophila Protocols, edited by W. A. Sullivan, M. Ashburner and R. S. Hawley, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
  • Iafrate, A. J., L. Feuk, M. N. Rivera, M. L. Listewnik, P. K. Donahoe et al., 2004. Detection of large-scale variation in the human genome. Nat. Genet. 36 949–951. [PubMed]
  • Jacobs, S. A., and S. Khorasanizadeh, 2002. Structure of HP1 chromodomain bound to a lysine 9-methylated histone H3 tail. Science 295 2080–2083. [PubMed]
  • Kehle, J., D. Beuchle, S. Treuheit, B. Christen, J. A. Kennison et al., 1998. dMi-2, a hunchback-interacting protein that functions in Polycomb repression. Science 282 1897–1900. [PubMed]
  • Kennison, J. A., 1983. Analysis of Y-linked mutations to male sterility in Drosophila melanogaster. Genetics 103 219–234. [PMC free article] [PubMed]
  • Kennison, J. A., and J. W. Tamkun, 1988. Dosage-dependent modifiers of Polycomb and Antennapedia mutations in Drosophila. Proc. Natl. Acad. Sci. USA 85 8136–8140. [PMC free article] [PubMed]
  • Kent, W. J., 2002. BLAT: the BLAST-like alignment tool. Genome Res. 12 656–664. [PMC free article] [PubMed]
  • Konev, A. Y., M. Tribus, S. Y. Park, V. Podhraski, C. Y. Lim et al., 2007. CHD1 motor protein is required for deposition of histone variant h3.3 into chromatin in vivo. Science 317 1087–1090. [PMC free article] [PubMed]
  • Koundakjian, E. J., D. M. Cowan, R. W. Hardy and A. H. Becker, 2004. The Zuker collection: a resource for the analysis of autosomal gene function in Drosophila melanogaster. Genetics 167 203–206. [PMC free article] [PubMed]
  • Lefevre, Jr., G., and W. Watkins, 1986. The question of the total gene number in Drosophila melanogaster. Genetics 113 869–895. [PMC free article] [PubMed]
  • Lewis, E. B., and F. Bacher, 1968. Method of feeding ethyl methane sulfonate (EMS) to Drosophila males. Dros. Inf. Serv. 43 193.
  • Lindsley, D. L., and G. G. Zimm, 1992. The Genome of Drosophila melanogaster. Academic Press, San Diego.
  • Matsubayashi, H., and M. T. Yamamoto, 1998. Dissection of chromosome region 89A of Drosophila melanogaster by local transposition of P elements. Genes Genet. Syst. 73 95–103. [PubMed]
  • McCallum, C. M., L. Comai, E. A. Greene and S. Henikoff, 2000. Targeted screening for induced mutations. Nat. Biotechnol. 18 455–457. [PubMed]
  • McDaniel, I. E., J. M. Lee, M. S. Berger, C. K. Hanagami and J. A. Armstrong, 2008. Investigations of CHD1 function in transcription and development of Drosophila melanogaster. Genetics 178 583–587. [PMC free article] [PubMed]
  • Min, G., X. Bi and Y. S. Rong, 2005. Targeted mutagenesis of Drosophila atm and mre11 genes. Dros. Inf. Serv. 88 79–83.
  • Moshkin, Y. M., J. A. Armstrong, R. K. Maeda, J. W. Tamkun, C. P. Verrijzer et al., 2002. Histone chaperone ASF1 cooperates with the Brahma chromatin-remodelling machinery. Genes Dev. 16 2621–2626. [PMC free article] [PubMed]
  • Murawska, M., N. Kunert, J. van Vugt, G. Längst, E. Kremmer et al., 2008. dCHD3, a novel ATP-dependent chromatin remodeler associated with sites of active transcription. Mol. Cell. Biol. 28 2745–2757. [PMC free article] [PubMed]
  • Ni, J.-Q., M. Markstein, R. Binari, B. Pfeiffer, L.-P. Liu et al., 2008. Vector and parameters for targeted transgenic RNA interference in Drosophila melanogaster. Nat. Methods 15 49–51. [PMC free article] [PubMed]
  • Nielsen, P. R., D. Nietlispach, H. R. Mott, J. Callaghan, A. Bannister et al., 2002. Stucture of the HP1 chromodomain bound to histone H3 methylated at lysine 9. Nature 416 103–107. [PubMed]
  • Odenwald, W. F., W. Rasband, A. Kuzin and T. Brody, 2005. EVOPRINTER, a multigenomic comparative tool for rapid identification of functionally important DNA. Proc. Natl. Acad. Sci. USA 102 14700–14705. [PMC free article] [PubMed]
  • Pollard, K. J., and C. L. Peterson, 1998. Chromatin remodeling: A marriage between two families? Bioessays 20 771–780. [PubMed]
  • Rong, Y. S., and K. G. Golic, 2000. Gene targeting by homologous recombination in Drosophila. Science 288 2013–2018. [PubMed]
  • Sebat, J., B. Lakshmi, J. Troge, J. Alexander, J. Young et al., 2004. Large-scale copy number polymorphism in the human genome. Science 305 525–528. [PubMed]
  • Stankiewicz, P., and J. R. Lupski, 2010. Structural variation in the human genome and its role in disease. Annu. Rev. Med. 61 437–455. [PubMed]
  • Tower, J., and R. Kurapati, 1994. Preferential transposition of a Drosophila P element to the corresponding region of the homologous chromosome. Mol. Gen. Genet. 244 484–490. [PubMed]
  • Tower, J., G. H. Karpen, N. Craig and A. C. Spradling, 1993. Preferential transposition of Drosophila P elements to nearby chromosomal sites. Genetics 133 347–359. [PMC free article] [PubMed]
  • Treisman, J. E., 2001. Drosophila homologues of the transcriptional coactivation complex subunits TRAP240 and TRAP230 are required for identical processes in eye-antennal disc development. Development 128 603–615. [PubMed]
  • Tripoulas, N. A., E. Hersperger, D. La Jeunesse and A. Shearn, 1994. Molecular genetic analysis of the Drosophila melanogaster gene absent, small or homeotic discs1 (ash1). Genetics 137 1027–1038. [PMC free article] [PubMed]
  • Venkatachalam, K., A. A. Long, R. Elsaesser, D. Nikolaeva, K. Broadie et al., 2008. Motor deficit in a Drosophila model of mucolipidosis type IV due to defective clearance of apoptotic cells. Cell 135 838–851. [PMC free article] [PubMed]
  • Veraksa, A., X. Li, J. A. Kennison and W. McGinnis, 2002. DEAF-1 function is essential for the early embryonic development of Drosophila. Genesis 33 67–76. [PubMed]
  • Wain, L. V., J. A. L. Armour and M. D. Tobin, 2009. Genomic copy number variation, human health, and disease. Lancet 374 340–350. [PubMed]
  • Williams, S. M., J. A. Kennison, L. G. Robbins and C. Strobeck, 1989. Reciprocal recombination and the evolution of the ribosomal gene family of Drosophila melanogaster. Genetics 122 617–624. [PMC free article] [PubMed]
  • Woodage, T., M. A. Basrai, A. D. Baxevanis, P. Hieter and F. S. Collins, 1997. Characterization of the CHD family of proteins. Proc. Natl. Acad. Sci. USA 94 11472–11477. [PMC free article] [PubMed]
  • Zhang, P., and A. C. Spradling, 1993. Efficient and dispersed local P element transposition from Drosophila females. Genetics 133 361–373. [PMC free article] [PubMed]
  • Zhang, P., and A. C. Spradling, 1994. Insertional mutagenesis of Drosophila heterochromatin with single P elements. Proc. Natl. Acad. Sci. USA 91 3539–3543. [PMC free article] [PubMed]
  • Zhu, X. J., J. Sen, L. Stevens, J. S. Goltz and D. Stein, 2005. Drosophila Pipe protein activity in the ovary and the embryonic salivary gland does not require heparan sulfate glycosaminoglycans. Development 132 3813–3822. [PubMed]

Articles from Genetics are provided here courtesy of Genetics Society of America
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...