Using the P{wHy} Hybrid Transposable Element to Disrupt Genes in Region 54D-55B in Drosophila melanogaster

Corresponding author: William M. Gelbart, Cambridge, MA 02138., gelbart{at}morgan.harvard.edu (E-mail)

Communicating editor: T. SCHÜPBACH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Understanding the function of each gene in the genome of a model organism such as Drosophila melanogaster is an important goal. The development of improved methods for uncovering the mutant phenotypes of specific genes can accelerate achievement of this goal. The P{wHy} hybrid transposable element can be used to generate nested sets of precisely mapped deletions in a given region of the Drosophila genome. Here we use the P{wHy} method to generate overlapping, molecularly defined deletions from a set of three P{wHy} insertions in the 54E-F region of chromosome 2. Deletions that span a total of 0.5 Mb were identified and molecularly mapped precisely. Using overlapping deletions, the mutant phenotypes of nine previously uncharacterized genes in a 101-kb region were determined, including identification of new loci required for viability and female fertility. In addition, the deletions were used to molecularly map previously isolated lethal mutations. Thus, the P{wHy} method provides an efficient method for systematically determining the phenotypes of genes in a given region of the fly genome.

THE availability of the genome sequence of Drosophila melanogaster has given researchers a glimpse of the complete set of Drosophila genes. The annotation of these genes continues to improve with computational and experimental methods, making the sequence of the 13,600 Drosophila genes available from any web terminal on the globe (ADAMS et al. 2000 ). Even so, understanding the function of these genes, alone and in concert with one another, is a daunting task.

Classic genetic approaches have yielded invaluable insight into gene function at the levels of biochemical function, function in a particular pathway, and function in a particular cell type, tissue, or organ. Indeed, the Berkeley Drosophila Genome Project (BDGP) has set a goal of obtaining a mutation in every gene in the fly genome to help elucidate gene function (SPRADLING et al. 1999 ). The BDGP gene disruption project and others have focused on genes that mutate to lethality, sterility, or other readily definable phenotypes (COOLEY et al. 1988 ; SPRADLING et al. 1999 ). A similar, smaller-scale effort was made to obtain P-element insertions in the 2.9-Mb Adh region. In this case, all insertions (not just those with definable phenotypes) were mapped, yielding insertions in a larger proportion of genes (ASHBURNER et al. 1999 ).

A major advantage of P-element disruption over chemical or high-energy radiation mutagenesis is that P elements provide a molecular foothold from which researchers can identify the site of insertion by molecular methods such as inverse PCR (iPCR; OCHMAN et al. 1988 ; TRIGLIA et al. 1988 ) or the recently reported universal fast-walking technique (UFW; MYRICK and GELBART 2002 ). Another advantage is that transposon insertions in or near genes can be used to generate new mutant alleles and local deletions by mobilization of P-element insertions (ROBERTSON et al. 1988 ) or by P-induced male recombination (PRESTON et al. 1996 ). On the other hand, a disadvantage of this approach is that P-element insertions are not random. Thus, as efforts to obtain P-element insertions into genes increase in scale, the majority of insertions will fall into regions in which a P element has already been identified and/or into intergenic regions.

Clearly, new methods could accelerate the pace of understanding gene function by facilitating the study of mutant phenotypes. Although both RNA-mediated interference and homologous recombination approaches have been reported (KENNERDELL and CARTHEW 1998 ; RONG and GOLIC 2000 , RONG and GOLIC 2001 ), it remains to be determined if these techniques can be used for a large-scale survey of mutant phenotypes of all genes in Drosophila.

Another approach is the P{wHy} hybrid transposable-element-based method for creating genomic deletions (HUET et al. 2002 ). With this method, the P{wHy} element is used to make nested sets of deletions in a particular region of the fly genome. The P{wHy} element (Fig 1A) consists of a transposase-deficient hobo element flanked by the white and yellow marker genes and by P-element ends that enable the entire P{wHy} hybrid element to be mobilized by a source of P-element transposase (note that P and hobo transposases do not cross-react; EGGLESTON et al. 1988 ; CALVI 1993 ). Subsequently, mobilization of hobo can be used to create nested sets of deletions that extend in one direction or the other from a given starting insertion of the P{wHy} element (Fig 1B and Fig C; HUET et al. 2002 ). Deletion of flanking sequence is thought to be caused by insertion of the hobo element into a nearby site followed by recombination between the resulting adjacent hobo insertions, thereby excising the intervening sequence (BLACKMAN et al. 1987 ; SHEEN et al. 1993 ; LIM and SIMMONS 1994 ). Because the hobo element remains intact in the deletion strains, iPCR or other techniques can be used to molecularly identify the deletion endpoints, resulting in a deletion mapped with to-the-base-pair accuracy (OCHMAN et al. 1988 ; TRIGLIA et al. 1988 ; HUET et al. 2002 ; MYRICK and GELBART 2002 ).

View larger version (19K): In this window In a new window Download PPT slide   Figure 1. The P{wHy} element. (A) The P{wHy} element is made up of P-element ends sufficient for P-transposase-mediated transposition (solid arrowheads) enclosing the white+ gene (solid box), a transposase-deficient Hobo transposon (boxed diagonals), and the yellow+ gene (open box). The shaded box indicates genomic DNA into which the P{wHy} element is inserted. After transposition with Hobo transposase, insertion of hobo into a new location followed by recombination between the original and newly inserted hobo transposons can result in B, the P{5'wHy} element, or C, the P{3'wHy} element. (D) The orientations of P{wHy}14F06, 14H10, and 02B10 relative to the previously identified and genetically characterized genes in the region (see Fig 2 for a complete map of transcripts in the region). Proximal is to the left and distal is to the right.

Here we use the P{wHy} method to disrupt and characterize genes by generating deletions from adjacent starting insertion sites and then placing overlapping deletions in trans to one another. The deletions cover an 0.5-Mb region in cytological region 54D-55B of chromosome 2 between the cytogenetically defined loci abero (54E; KANIA et al. 1995 ) and staufen (stau; 55B; ST. JOHNSTON et al. 1991 ). Although the region contains many genes based on cDNA evidence and computational prediction (ADAMS et al. 2000 ; RUBIN et al. 2000 ), few mutations were known to disrupt genes in the region. We show that the method proves effective for the 54D-55B region: complementation analyses with overlapping deletions uncover the mutant phenotypes of several loci, including loci required for viability and female fertility, and the deletions were also used to molecularly map anonymous, EMS-induced mutations in the region.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Fly stocks and culturing:
Fly strains were maintained at 25° using standard protocols for media preparation and culturing. Strains carrying the following mutations were obtained from the National Drosophila Stock Center (Bloomington, IL): grh^s2140, grh^IM, grh⁰⁶⁸⁵⁰, hal^DB48, stau¹, stau^ry9, thr¹, thr^k07805b, Df(2R) Pcl7B, l(2)k11505, l(2)k11311, l(2)k09924, and l(2)PC4-A¹³⁹, l(2)PC4-B¹¹⁰, l(2)PC4-D²⁰², l(2)PC4-E¹¹⁹, l(2)PC4-F¹⁹⁸, l(2)PC4-G²²³, and l(2)PC4-M⁴²⁰ (DUNCAN 1982 ; DYNLACHT et al. 1989 ; SCHUPBACH and WIESCHAUS 1991 ; ST. JOHNSTON et al. 1991 ; D'ANDREA et al. 1993 ; FLYBASE 1999 ; SPRADLING et al. 1999 ). The EP(2)616 insertion (RORTH et al. 1998 ) was provided by Exelixis (San Francisco). The adrift¹ (ENGLUND et al. 1999 ) mutation and the Df(2R)30W deletion were provided by C. Samakovlis. The P{wHy}14F06, 14H10, and 02B10 insertion strains have unique insertions of the P{wHy} element in a y¹ Df(1)67c background, hereafter referred to as y^- w^-. All crosses were performed at 25° and stocks were maintained at 22° or 25°.

P{wHy} insertion site identification:
Autosomal insertions of P{wHy} were molecularly mapped on the 3' end using iPCR (OCHMAN et al. 1988 ; TRIGLIA et al. 1988 ). Genomic DNA was digested with Sau3AI or HinPI. PCR was performed using the following primers: for 5' end mapping, Plac1, cacccaaggctctgctcccacaat and Plac4, gactgtgcgttaggtcctgttca; for 3' end mapping, P4-2, caatcatatcgctgtctccactcagact and Pye1, gttgcgatttcgggagctacaatcgg. The iPCR amplicons were sequenced using the Plac1 or P4-2 oligonucleotide primers.

The P{wHy}14F06, 14H10, and 02B10 insertion sites were confirmed by using genomic DNA-specific primers 5' or 3' of the insert site (below) with Pendout2 (cgacgggaccaccttatgtt) in PCR assays, and PCR products were sequenced with Pendout2 to confirm the insertion sites. The genomic DNA-specific primers were as follows: for 14F06, 5'-atctcctcctcgctggactcggact and 5'-cttttgcacgcaagcgcagc; for 14H10, 5'-agtttaaataggatctcgcc and 5'-gatgtagtagtgttgaaaggtga; for 02B10, 5'-gatgtggaaattgtatgagagtag and 5'-tagtagtctttttcgaagctctg.

Generation and identification of genomic disruptions:
To mobilize hobo in P{wHy}14F06, 14H10, and 02B10, homozygous or heterozygous males carrying an insertion were crossed to virgin females carrying a source of hobo transposase (y^- w^-;Gla/CyO-P{hsH\\T-2}). The resulting larvae were heat-shocked for 30 min at 37° on the second, fourth, and sixth days after crossing the adults. In the F₁ cross, individual y^- w^-;P{wHy}/CyO-P{hsH\\T-2} males were crossed to y^- w^-; Gla/SM6a virgin females. Phenotypically w⁺ y^- and w^- y⁺ lines were recovered from the F₂ and crossed to y^- w^-; Gla/SM6a virgin females to establish balanced stocks. The 5' and 3' genomic DNA-specific primers that flank the insertion site described above were used with the Pendout2 primer in a one-tube PCR assay to test for the presence of 5' and 3' ends. Lines in which both the 5' and 3' products were present were discarded (these represent deletions of the marker genes without associated disruption of flanking genomic DNA). Lines in which only the predicted P end was detected were retained (see Fig 1B and Fig C).

Mapping of hobo insertion sites:
The 5' hobo insertion sites (w⁺ y^- lines) and 3' hobo insertion sites (w^- y⁺) were mapped by iPCR (OCHMAN et al. 1988 ; TRIGLIA et al. 1988 ) or UFW (MYRICK and GELBART 2002 ). For iPCR, genomic DNA was digested with EcoRI, AluI, or MseI. PCR was performed with the following primers: for 5' hobo flanking DNA, Ph5-1, acgcaaaacaccgtattattcgg and Ph5-2, cgtaggtagtcgagtcaaatggc; for 3' hobo flanking DNA, Ph-EA1, gggcataatctatttcgcttttct and Ph3-2, cgagtattttgtgtgccgcaagt. PCR amplicons were sequenced with Ph5-1 or Ph3-2.

For UFW 5' hobo mapping, the following primers were used (listed in the order in which they were used): h5-1, 5'-actacctacgagaccactcg; h5-2, 5'-tttaggcactgtgtgagcgg(n₁₀); h5-3, 5'-taacggtatacccacaagtg; h5-4, 5'-acgcaaacacctattgattcgg; and h5-5, 5'-gatgtgcgtggcgagtagcaccc. For UFW 3' hobo mapping, the following primers were used: h3-1, 5'-ccgaatcaatacggtgttttgcgt; h3-2, 5'-cgagtggtctcgtaggtact(n₁₀); h3-3, cacttgtgggtataccgtta; h3-4, 5'-gatcgttgactgtgcgtccactca; h3-5, 5'-acacaacgtcggtaaaacactcga.

Complementation analysis and fertility tests:
Complementation analysis was performed by crossing heterozygous virgin females of one genotype to heterozygous males of the other genotype. All genotypes were balanced over SM6a or CyO second chromosome balancers, and the Cy⁺ phenotype was used to identify trans-heterozygous individuals resulting from the cross (n > 100 adults scored in all cases). Whenever possible, additional markers were scored. For example, for P{3'wHy}02B10 crosses to P{5'wHy}14H10 lines (balanced with SM6a in all cases), it was possible to identify trans-heretozygotes as the w⁺, y⁺, Cy⁺ class. All viable trans-heterozygous combinations were tested for female fertility by crossing trans-heterozygous females to wild-type males; all homozygous viable genomic disruption strains were also tested for female fertility.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Three insertions define a 101-kb region in polytene region 54:
Using iPCR, several independent insertions of the P{wHy} hybrid transposable element (Fig 1A) were molecularly mapped from a collection of autosomal insertions. Sequencing of flanking DNA revealed that three inserts, P{wHy}14F06, P{wHy}14H10, and P{wHy}02B10 (hereafter, 14F06, 14H10, and 02B10), are in a region in which little genetic data are connected with molecularly defined genes in the region (Fig 1D; GenBank accession nos. BH836458, BH836459, BH836460, BH836461, BH836462, BH836463, BH836464, BH836465, BH836466, BH836467, BH836468, BH836469). The three inserts have the same orientation of insertion and are within 101 kb of one another between the cytogenetically defined loci abero (54E; KANIA et al. 1995 ) and grainyhead (54F1-5; DYNLACHT et al. 1989 ) and, at the molecular level, between rhino (VOLPE et al. 2001 ) and CG6370 (Fig 1D; Fig 2; ADAMS et al. 2000 ). We compared the genomic sequence of the region to cDNAs, expressed sequence tags, and computationally predicted genes to obtain a map of the genes in the region (ADAMS et al. 2000 ; RUBIN et al. 2000 ). Twenty-five protein-coding transcription units are predicted to be present between 14F06 and 02B10 (Fig 2).

View larger version (13K): In this window In a new window Download PPT slide   Figure 2. P{wHy}-generated deletions and transcripts in the 54D-55A region. Proximal is to the left and distal is to the right in this and all subsequent figures. Inverted triangles, sites of insertion of 14F06, 14H10, and 02B10. Shaded line, genomic DNA. Arrows show the location of known and predicted transcripts and indicate the direction of transcription; exon/intron boundaries are not shown. The names of a subset of transcripts are indicated for reference. Solid lines, the region deleted in each P{wHy}-generated deletion that was molecularly mapped.

Each of the three P{wHy} inserts is predicted to disrupt a transcript (Fig 2). Whereas the 14F06 insertion chromosome is homozygous lethal, the 14H10 and 02B10 insertion chromosomes are homozygous viable. The 14F06 insertion disrupts the eIF3-S8 gene, which encodes a highly conserved translation factor (LASKO 2000 ). The transcript disrupted by 14H10, CG4996, encodes a conserved protein of unknown function (Fig 3). The transcript disrupted by 02B10, CG6370, encodes a homolog of the human ribophorin II protein, a dolichyl-diphosphooligosaccharide-protein glycosyltransferase (Fig 3; CRIMAUDO et al. 1987 ). The inserts represent new alleles of the genes and thus can be designated eIF-3S8^14F06, CG4996^14H10, and CG6370^02B10; for convenience, we will continue to refer to the insertions as 14F06, 14H10, and 02B10.

View larger version (22K): In this window In a new window Download PPT slide   Figure 3. Overlapping deletions between 14H10 and 02B10. (A) Map of the 14H10-02B10 region. All known and predicted transcripts in the region are indicated with arrows. For HLH54F, the exon/intron structure is indicated. Solid lines, the region removed in P{wHy}-generated deletions extending from 14H10 and 02B10. An arrowhead on a solid line indicates that the deletion extends in the direction of the arrow beyond the region depicted in the figure. (B) Complementation analysis between P{5'wHy}14H10 deletions and P{3'wHy}02B10 deletions. +, viable and fertile; -, inviable.

Mobilization of hobo in 14F06, 14H10, and 02B10 results in disruption of genomic DNA:
To generate and isolate unidirectional deletions from each of the three starting P{wHy} insertions, we introduced hobo transposase to generate deletions (see MATERIALS AND METHODS; HUET et al. 2002 ). After introduction of hobo transposase, local transposition followed by recombination is thought to result in deletion of flanking DNA. Deletions of the marker genes or of the marker genes and flanking genomic DNA are detected by loss of the white⁺ or yellow⁺ marker genes. Next, disruption of only the marker gene (both P ends retained) is distinguished from disruption of flanking DNA (one P end retained; Fig 1B and Fig C) by assaying for the presence or absence of the P-element ends. Phenotypically w⁺ y^- strains or w^- y⁺ strains in which only the appropriate P-element end is present were retained for further study. We refer to this class of events as "genomic disruptions" and term the newly formed P{wHy} derivatives P{5'wHy} (phenotypically w⁺ y^-) and P{3'wHy} (phenotypically w^- y⁺; Fig 1B and Fig C).

Since the 14H10 insertion is between 02B10 and 14F06, events in both directions are useful for obtaining overlapping deletions. Therefore, about twice as many crosses were performed with 14H10 as with the others. With 14F06, 273 fertile crosses resulted in 78 w⁺ y^- and w^- y⁺ strains; 35 of these are genomic disruptions. With 14H10, 559 fertile crosses resulted in identification of 174 w⁺ y^- and w^- y⁺ strains; 78 of these are genomic disruptions. With 02B10, 204 fertile crosses resulted in identification of 55 w⁺ y^- and w^- y⁺ strains; 25 of these are genomic disruptions.

To determine the extent of the putative deletions in each genomic disruption, the new 5' or 3' hobo insertion points in P{5'wHy} and P{3'wHy} strains were sequenced by iPCR or UFW. Of the 96 genomic disruptions analyzed, 67 unambiguously map to the 54D-55B region and carry single insertions of hobo. Genetic data with the unmapped lines are consistent with disruption of adjacent genomic DNA, but these lines may have multiple hobo insertions or rearrangements that preclude molecular mapping, and thus were excluded from the analysis presented below. The GenBank accession numbers corresponding to the flanking DNA sequence from hobo mapping are provided for each genomic disruption in Table 1 (14F06), Table 2 (14H10), and Table 3 (02B10).

Two observations based on the sequence data support the idea that the genomic disruptions are deletions. First, the hobo flanking sequence is in the orientation expected for a simple deletion. Second, the positions of the putative deletion breakpoints relative to the starting insertion sites are consistent with simple deletions in the direction predicted from the marker gene phenotype (w⁺ y^- or w^- y⁺). Thus, together with the original insertion point of the P{wHy} element, the hobo mapping data indicate the predicted extent of the deletions. The putative deletions range in size from 412 to 377,752 bp (Table 1, Table 2, and Table 3; Fig 2). Together, the deletions extend over >0.5 Mb between CG10936 (ADAMS et al. 2000 ) and stau (ST. JOHNSTON et al. 1991 ) in cytogenetic region 54D-55B (Fig 2).

Genomic disruptions fail to complement previously identified loci in 54D-55B:
If the genomic disruptions are deletions, then they should fail to complement existing mutations in the 54D-55B region in a predictable pattern. Therefore, we tested whether the genomic disruption strains have complementation patterns consistent with deletions of the predicted sizes. Mutations in several previously identified loci were useful for this analysis—namely, the recessive lethal loci grainyhead (grh; DYNLACHT et al. 1989 ), three rows (thr; D'ANDREA et al. 1993 ), and eIF3-S8 (this work; LASKO 2000 ) and the female sterile locus stau (ST. JOHNSTON et al. 1991 ).

Some of the genomic disruptions extending distally from 02B10 and 14H10 are predicted to delete grh and thr (Fig 4A). We observe the expected complementation pattern between deletions of specific predicted sizes and mutant alleles of these genes (Fig 4B). Similarly, three deletions extending proximally from 14H10 that are predicted to remove eIF3-S8 (P{3'wHy}14H10Y-51, P{3'wHy}14H10Y-52, and P{3'wHy}14H10Y-53; Table 2) fail to complement the lethality of eIF3-S8^14F06. The distal breakpoint of the longest deletion, P{5'wHy}14H10W-35, is 1424 bp 3' of the end of the stau coding region and thus is not expected to disrupt stau function (Fig 4A). As expected, the P{5'wHy}14H10W-35 deletion complements stau and eIF3-S8^14F06 but fails to complement grh, thr, and Df(2R)Pcl7B (DUNCAN 1982 ; DYNLACHT et al. 1989 ; ST. JOHNSTON et al. 1991 ; D'ANDREA et al. 1993 ; ENGLUND et al. 1999 ). Lastly, Df(2R)30W, a deletion that disrupts EG:52C10.5 but not adrift (C. ENGLUND and C. SAMAKOVLIS, personal communication), complements P{5'wHy}02B10W-06 but fails to complement longer deletions. Therefore, the molecular and genetic data support the idea that the genomic disruptions are deletions, making them useful for further genetic analysis.

View larger version (21K): In this window In a new window Download PPT slide   Figure 4. Complementation between deletions in the grainyhead region and grainyhead and three rows. (A) Map of the grainyhead region. Transcripts distal of the 02B10 insertion are indicated with arrows. Solid lines, the region removed in P{wHy}-generated deletions extending distal of 02B10 (see Fig 3 for transcripts and deletions between 14H10 and 02B10). An arrowhead on a solid line indicates that the deletion extends in the direction of the arrow beyond the region. (B) Complementation analysis between the deletions shown in A and grainyhead, three rows, and staufen. +, viable and fertile; -, inviable.

Overlapping deletions define regions required for viability:
We were particularly interested in the region between 14F06 and 02B10, since no previously identified mutations have been associated with the genes in this interval. To test the requirement for specific regions of DNA between 14F06 and 02B10, strains with deletions extending distally from 14F06 and 14H10 were crossed to those with deletions extending proximally from 02B10 and 14H10 (Fig 3A). The gene proximal to the 14H10 insertion (CG6401) is essential, since deletions that extend proximally from 14H10 that disrupt this gene or larger regions (for example, P{3'wHy}14H10Y-39) are inviable in trans to deletions that remove the entire region (Fig 3). No further phenotypic information about the genes between eIF3-S8 and CG6401 can be uncovered because shorter deletions from 14F06 and 14H10 fail to overlap, and longer deletions from 14H10 delete the essential gene eIF3-S8 (Fig 2). A larger sample size of deletions extending from 14F06 toward 14H10 would increase the chance that deletions that overlap to remove genes between CG6401 and eIF3-S8 could be identified.

The most informative analysis was in the region between the 14H10 and 02B10 insertions. As shown in Fig 3, deletions that disrupt HLH54F (GEORGIAS et al. 1997 ) are semilethal (<30% of expected, n > 500) in trans to one another, including deletions that disrupt exons 1 and 2 in trans to a deletion that disrupts exon 3 (class C vs. classes II and III). However, deletions that extend distally from 14H10 beyond HLH54F (class D deletions) are inviable in trans to class II and III deletions. Class D deletions are viable in trans to the 02B10 insertion in CG6370 and to a short deletion that disrupts CG6370 (P{3'wHy}02B10Y-10; Fig 3). Together these data suggest that CG6370 is dispensable, but that BcDNA: GH07485 (GH07485; RUBIN et al. 2000 ) is required for full viability. Finally, complementation between class B and class II defines a region dispensable for viability, fertility, and gross morphology (Fig 3).

Overlapping deletions define a region required for female fertility:
In addition to uncovering essential loci, the analysis shows that a locus is required for female fertility in the region between 14H10 and 02B10 (Fig 3). Females with class B deletions are sterile in trans to class III deletions, as are homozygous P{5'wHy}14H10W-15 females and trans-heterozygous P{5'wHy}14H10W-15/P{5'wHy}14H10W-16 females. These females fail to lay eggs, and their ovaries are smaller than those of wild type (data not shown). The locus is limited proximally by P{5'wHy}14H10W-14, the largest class A deletion tested that complements class III deletions for fertility, and distally by P{3'wHy}02B10Y-12, the largest class II deletion that complements class B deletions for fertility (Fig 3). The deletions complement a mutation in haulted, which maps genetically to the 54-55 region (SCHUPBACH and WIESCHAUS 1991 ).

Mapping anonymous EMS-induced mutations with P{wHy} deletions:
In addition to helping to uncover knockout phenotypes of genes between 14F06 and 02B10, the deletions in our collection have proven useful for molecular mapping of existing lesions that map genetically to the 54D-55B region. We performed complementation analysis between distally extending deletions from 14H10 and 02B10 and strains from a collection of lethal mutations that were isolated in a screen for EMS-induced mutations that fail to complement Df(2R)PC4 [55A1; 55F1-2] (FLYBASE 1999 ). A subset of these mutations that map to the proximal end of the deletion in Df(2R)PC4 were tested [l(2)PC4-A, -B, -D, -E, -F, -G, and -M].

The l(2)PC4-B, -E, -F, -G, and -M mutations that were tested complement the largest deletion in the region, P{5'wHy}14H10W-35, suggesting that the lesions are distal to stau in 55B. Both l(2)PC4-D²⁰² and l(2)PC4-A¹³⁹ fail to complement a subset of the P{wHy}-generated deletions (Table 4). The l(2)PC4-A locus is within an 239-kb region, limited proximally by P{5'wHy} 14H10W-34, the longest deletion that complements l(2)PC4-A¹³⁹, and distally by P{5'wHy}14H10W-35, which fails to complement l(2)PC4-A¹³⁹ (Table 4; Fig 4). At the level of known and predicted transcripts, this region extends from Black cells to stau and includes many candidate transcription units for the l(2)PC4-A locus. The l(2)PC4-D locus maps to a 24-kb region limited proximally by P{5'wHy}14H10W-25, which complements l(2)PC4-D²⁰², and distally by P{5'wHy}14H10W-27, which fails to complement l(2)PC4-D²⁰² (Table 4; Fig 5). Three transcripts map to this interval: UbcD10, EG:52C10.1, and EG:52C10.2 (ADAMS et al. 2000 ; AGUILERA et al. 2000 ). These data suggest that l(2)PC4-D²⁰² affects UbcD10, EG52C10.1, or EG:52C10.2.

View larger version (13K): In this window In a new window Download PPT slide   Figure 5. The l(2)PC4-D mutation maps to the Ubc10, EG:52C10.1, and EG52C10.2 regions. (A) Map of the 12-kb adrift to EG:52C10.2 region (exon/intron boundaries not shown). Solid lines, the adrift gene; shaded lines, candidate transcription units for the l(2)PC4-D locus. (B) A subset of the deletions tested for complementation with l(2)PC4-D202 (see Table 4). Solid lines, the longest deletion that complements l(2)PC4-D202; shaded lines, the shortest deletions that fail to complement the mutation.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

P{wHy}-generated genomic disruptions behave as deletions:
We mobilized the hobo transposon in P{wHy} insertions in 54E/F to generate genomic disruptions and screened for events that remove flanking genomic DNA proximally or distally from the starting insertion. By mapping the new hobo insertion point, we were able to identify the breakpoint ends of putative deletions. Molecular evidence and results of complementation analyses with eIF3-S8^14F06 and previously isolated mutations of grh and thr (DYNLACHT et al. 1989 ; D'ANDREA et al. 1993 ) are consistent with the idea that the majority of the genomic disruptions in the region are deletions. Thus, in a relatively small-scale screen (1000 crosses), deletions that extend >0.5 Mb can be identified and recovered.

Overlapping P{wHy}-generated deletions reveal mutant phenotypes:
A subset of the genomic disruptions from 14F06, 14H10, and 02B10 are predicted to remove overlapping regions of DNA (Fig 2). The 14H10 and 02B10 insertions disrupt CG4996 and CG6370, respectively, but are fully viable and fertile. Deletion of CG4996 or CG6370 also has no discernible effect on viability, fertility, or gross morphology (Fig 3). Complementation of class B with class II deletions suggests that the following genes are similarly dispensable: CG10931 and CG5002, which are completely removed in class B/class II animals, and CG6385, which is disrupted in class B/class II animals (Fig 3).

Clearly, these five genes do not provide nonredundant information required for overall survival to adulthood, gross morphology, or fertility. Obviously, tests for more subtle phenotypes need to be applied. Moreover, there is great value in knowing that deletion of a specific gene has no obvious visible lethal or sterile phenotype.

Combining the complementation data with the molecular map also reveals that the regions of DNA required for viability correlate to the eIF3-S8, CG6401, and GH07485 genes (Fig 2 and Fig 3; ADAMS et al. 2000 ; LASKO 2000 ). The CG6401 gene encodes an N-acetyl glucosaminyl phosphatidylinositol synthesis protein similar to human PIG-A (MIYATA et al. 1993 ). The GH07485 gene encodes a member of the acyl Co-A oxidase family of proteins with the greatest similarity to the human peroxisomal acyl Co-A oxidase (FOURNIER et al. 1994 ).

A subset of deletions define a region of DNA required for female fertility. The region is limited proximally by the P{5'wHy}14H10W-14 breakpoint and distally by the P{3'wHy}02B10Y-12 breakpoint. Therefore, it is likely to result from the disruption of CG12298 and/or CG14487 (Fig 3). The CG12298 gene encodes a kinesin-like protein. Conceptual translation of CG14487 reveals no significant similarity to proteins outside of Drosophila, but does show significant similarity to a family of predicted proteins in the fly genome with the closest similarity to the CG12960, CG12961, CG17379, CG17381, CG13582, CG14584, and CG12525 proteins.

In the absence of genomic disruptions that separate these candidate genes, further experiments will be necessary to determine which gene(s) in the region are required for female fertility. Clearly, the set of 70 deletions described is not sufficient to achieve the goal of removing each transcript individually as was accomplished for HLH-54F. This appears to reflect the fact that our initial screen was fairly modest (1000 chromosomes), rather than reflecting a general limitation of the P{wHy} approach. HUET et al. 2002 suggest that 100 deletions from a single starting insertion would be needed to resolve every transcript in a 50-kb region. Thus, a larger-scale screen would likely provide useful reagents for targeting individual transcripts and enable us to limit the region required for fertility to a single transcription unit.

The P{wHy} method complements other methods for gene disruption:
We have shown that P{wHy}-generated deletions from adjacent insertions can be used to remove specific genes. The advantages of this technique include: (1) transposon insertions into intergenic regions are not disadvantageous; (2) a given phenotype can be precisely mapped to a small interval on the genomic DNA independently of gene annotation in the region; (3) conversely, even in the absence of a phenotype, one can be confident that a gene or genes have been deleted; and (4) data for the 54D-55B region suggest that the method provides a relatively efficient approach for determining the mutant phenotypes of all the genes in a specific region.

Like P-element mobilization or PMR (ROBERTSON et al. 1988 ; PRESTON et al. 1996 ), the P{wHy} method relies on having an insertion (or in this case, at least two adjacent insertions) in the region of interest. However, the "reach" of the P{wHy} technique is fairly large, as deletions of up to a few hundred kilobase pairs can be isolated. Moreover, the presence of hobo immediately adjacent to the genomic disruptions enables precise molecular mapping of deletion endpoints. We anticipate that once a large collection of molecularly mapped insertions is available, many regions of the genome can be surveyed using the method presented here. For regions in which the adjacent insertions are farther apart (i.e., farther than 50 kb apart), larger screens for genomic disruptions will be necessary, as longer disruptions are less common than shorter ones (this work; HUET et al. 2002 ).

Combining P{wHy} with other approaches:
The P{wHy} method seems particularly well suited for use in conjunction with EMS screens for mutations that fail to complement a deletion of the region of interest. In this case, we benefited from an EMS screen having been performed with Df(2R)PC4 (FLYBASE 1999 ). The lethal loci identified in the screen had been mapped with existing deficiencies to within a few lettered units on the cytogenetic map. Complementation analysis with P{wHy}-generated deletions limits the regions in which the l(2)PC4-A and l(2)PC4-D loci reside to 239 and 24 kb, respectively (Table 4; Fig 4).

X- and gamma-ray-induced deletions commonly used for EMS screens can often have >100 kb of genomic DNA deleted. This is beyond the 50- to 60-kb region well covered in a P{wHy} screen for deletions from a single starting insertion (HUET et al. 2002 ). Thus, it may be necessary to start with several P{wHy} insertions in the region of interest to map all of the EMS alleles in a region. Alternatively, the largest of a set of P{wHy} deletions can be used in a screen for noncomplementation, and smaller P{wHy} deletions can be used to molecularly map the resultant noncomplementing lesions. The availability of a set of starting insertions, then, will be one limiting factor in using the P{wHy} approach to map anonymous mutations in a given region.

The advantages of combining overlapping P{wHy} deletion analysis with an EMS screen are threefold. First, P{wHy}-generated deletions can be used to determine loss-of-function phenotypes (or lack thereof), assisting a focused effort in screening for EMS-induced mutations. Second, P{wHy} deletions can be used to map EMS-generated mutations to particular regions of the DNA, as shown here for l(2)PC4-A and l(2)PC4-D. Third, the EMS alleles can then provide genetic reagents for analyses such as mosaic analysis that are not feasible with the overlapping P{wHy} deletion approach. Thus, both on its own and in conjunction with other approaches, the P{wHy} strategy for determining complete loss-of-function phenotypes with overlapping deletions can contribute to our understanding of gene function in Drosophila and other organisms.

	ACKNOWLEDGMENTS

The authors thank Adam Casterano, Phillip Chan, Madeline Crosby, Francois Huet, Yevgenya Kraytsberg, Todd Martin, Sara McKinley, Jaitra Murthy, and Kyl Myrick for technical support and many helpful discussions. We thank Kathy Matthews at the National Drosophila Stock Center and Christos Samakovlis at Umea University for providing us with mutant strains used for this study. Lastly, thanks go to Ines Alvarez-Garcia and Madeline Crosby for critical review of the manuscript. This work was supported by grant no. R37 GM28669 from the National Institutes of Health to W.M.G.

Manuscript received December 3, 2001; Accepted for publication June 6, 2002.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADAMS, M. D., S. E. CELNIKER, R. A. HOLT, C. A. EVANS, and J. D. GOCAYNE et al., 2000  The genome sequence of Drosophila melanogaster.. Science 287:2185-2195.[Abstract/Free Full Text]

AGUILERA, M., M. OLIVEROS, M. MARTINEZ-PADRON, J. A. BARBAS, and A. FERRUS, 2000  Ariadne-1: A vital Drosophila gene is required in development and defines a new conserved family of ring-finger proteins. Genetics 155:1231-1244.[Abstract/Free Full Text]

ASHBURNER, M., S. MISRA, J. ROOTE, S. LEWIS, and R. BLAZEJ et al., 1999  An exploration of the sequence of a 2.9-Mb region of the genome of Drosophila melanogaster: the Adh region. Genetics 153:179-219.[Abstract/Free Full Text]

BLACKMAN, R. K., R. GRIMAILA, M. M. D. KOEHLER, and W. M. GELBART, 1987  Mobilization of hobo elements residing within the decapentaplegic gene complex: suggestion of a new hybrid dysgenesis system in Drosophila melanogaster. Cell 49:497-505.[Medline]

CALVI, B. R., 1993 The hobo transposable element in Drosophila: evolution, regulation, and use as a genetic tool. Ph.D. Thesis, Harvard University, Cambridge, MA.

COOLEY, L., R. KELLEY, and A. SPRADLING, 1988  Insertional mutagenesis of the Drosophila genome with single P elements. Science 239:1121-1128.[Abstract/Free Full Text]

CRIMAUDO, C., M. HORTSCH, H. GAUSEPOHL, and D. I. MEYER, 1987  Human ribophorins I and II: the primary structure and membrane topology of two highly conserved rough endoplasmic reticulum-specific glycoproteins. EMBO J. 6:75-82.[Medline]

D'ANDREA, R. J., R. STRATMANN, C. F. LEHNER, U. P. JOHN, and R. SAINT, 1993  The three rows gene of Drosophila melanogaster encodes a novel protein that is required for chromosome disjunction during mitosis. Mol. Biol. Cell 4:1161-1174.[Abstract]

DUNCAN, I. M., 1982  Polycomblike: a gene that appears to be required for the normal expression of the Bithorax and Antennapedia gene complexes of Drosophila melanogaster.. Genetics 102:49-70.[Abstract/Free Full Text]

DYNLACHT, B. D., L. D. ATTARDI, A. ADMON, M. FREEMAN, and R. TJIAN, 1989  Functional analysis of NTF-1, a developmentally regulated Drosophila transcription factor that binds neuronal cis elements. Genes Dev. 3:1677-1688.[Abstract/Free Full Text]

EGGLESTON, W. B., D. M. JOHNSON-SCHLITZ, and W. R. ENGELS, 1988  P-M hybrid disgenesis does not mobilize other transposable element families in D. melanogaster.. Nature 331:368-370.[Medline]

ENGLUND, C., A. E. UV, R. CANTERA, L. D. MATHIES, and M. A. KRASNOW et al., 1999  adrift, a novel bnl-induced Drosophila gene, required for tracheal pathfinding into the CNS. Development 126:1505-1514.[Abstract]

FLYBASE,, 1999  The FlyBase database of the Drosophila genome projects and community literature. Nucleic Acids Res. 27:85-88. http://flybase.bio.indiana.edu/(.[Abstract/Free Full Text]

FOURNIER, B., J. M. SAUDUBRAY, B. BENICHOU, S. LYONNET, and A. MUNNICH et al., 1994  Large deletion of the peroxisomal acyl-CoA oxidase gene in pseudoneonatal adrenoleukodystrophy. J. Clin. Invest. 94:526-531.

GEORGIAS, C., M. WASSER, and U. HINZ, 1997  A basic-helix-loop-helix protein expressed in precursors of Drosophila longitudinal visceral muscles. Mech. Dev. 69:115-124.[Medline]

HUET, F., J. T. LU, K. V. MYRICK, L. R. BAUGH, and M. A. CROSBY et al., 2002  A deletion-generator compound element allows deletion saturation analysis for genomewide phenotypic annotation. Proc. Natl. Acad. Sci. USA 99:9948-9953.[Abstract/Free Full Text]

KANIA, A., A. SALZBERG, M. BHAT, D. D'EVELYN, and Y. HE et al., 1995  P-element mutations affecting embryonic peripheral nervous system development in Drosophila melanogaster. Genetics 139:1663-1678.[Abstract]

KENNERDELL, J. R. and R. W. CARTHEW, 1998  Use of dsRNA-mediated genetic interference to demonstrate that frizzled and frizzled2 act in the Wingless pathway. Cell 95:1017-1026.[Medline]

LASKO, P., 2000  The Drosophila melanogaster genome: translation factors and RNA binding proteins. J. Cell Biol. 150:F51-F56.[Abstract/Free Full Text]

LIM, J. K. and M. J. SIMMONS, 1994  Gross chromosome rearrangements mediated by transposable elements in Drosophila melanogaster.. Bioessays 16:269-275.[Medline]

MIYATA, T., J. TAKEDA, Y. IIDA, N. YAMADA, and N. INOUE et al., 1993  The cloning of PIG-A, a component in the early step of GPI-anchor biosynthesis. Science 259:1318-1320.[Abstract/Free Full Text]

MYRICK, K. V. and W. M. GELBART, 2002  Universal fast walking for direct and versatile method determination of flanking sequence. Gene 284:125-131.[Medline]

OCHMAN, H., A. S. GERBER, and D. L. HARTL, 1988  Genetic applications of an inverse polymerase chain reaction. Genetics 120:621-623.[Abstract/Free Full Text]

PRESTON, C. R., J. A. SVED, and W. R. ENGELS, 1996  Flanking duplications and deletions associated with P-induced male recombination in Drosophila. Genetics 144:1623-1638.[Abstract]

ROBERTSON, H. M., C. R. PRESTON, R. W. PHILLIS, D. M. JOHNSON-SCHLITZ, and W. K. BENZ et al., 1988  A stable genomic source of P-element transposase in Drosophila melanogaster. Genetics 118:461-470.[Abstract/Free Full Text]

RONG, Y. S. and K. G. GOLIC, 2000  Gene targeting by homologous recombination in Drosophila.. Science 288:2013-2018.[Abstract/Free Full Text]

RONG, Y. S. and K. G. GOLIC, 2001  A targeted gene knockout in Drosophila. Genetics 157:1307-1312.[Abstract/Free Full Text]

RØRTH, P., K. SZABO, A. BAILEY, T. LAVERTY, and J. REHM et al., 1998  Systematic gain-of-function genetics in Drosophila. Development 125:1049-1057.[Abstract]

RUBIN, G. M., L. HONG, P. BROKSTEIN, M. EVANS-HOLM, and E. FRISE et al., 2000  A Drosophila complementary DNA resource. Science 287:2222-2224.[Abstract/Free Full Text]

SCHÜPBACH, T. and E. WIESCHAUS, 1991  Female sterile mutations on the second chromosome of Drosophila melanogaster. II. Mutations blocking oogenesis or altering egg morphology. Genetics 129:1119-1136.[Abstract]

SHEEN, F., J. K. LIM, and M. J. SIMMONS, 1993  Genetic instability in Drosophila melanogaster mediated by hobo transposable elements. Genetics 133:315-334.[Abstract]

SPRADLING, A. C., D. STERN, A. BEATON, E. J. RHEM, and T. LAVERTY et al., 1999  The Berkeley Drosophila Genome Project gene disruption project: single P-element insertions mutating 25% of vital Drosophila genes. Genetics 153:135-177.[Abstract/Free Full Text]

ST. JOHNSTON, R. D., D. BEUCHLE, and C. NUSSLEIN-VOLHARD, 1991  staufen, a gene required to localize maternal RNAs in the Drosophila egg. Cell 66:51-63.[Medline]

TRIGLIA, T., M. G. PETERSON, and D. J. KEMP, 1988  A procedure for in vitro amplification of DNA segments that lie outside the boundaries of known sequences. Nucleic Acids Res. 16:8186.[Free Full Text]

VOLPE, A. M., H. HOROWITZ, C. M. GRAFER, S. M. JACKSON, and C. A. BERG, 2001  Drosophila rhino encodes a female-specific chromo-domain protein that affects chromosome structure and egg polarity. Genetics 159:1117-1134.[Abstract/Free Full Text]

This article has been cited by other articles:

				K. S. Weiler E(var)3-9 of Drosophila melanogaster Encodes a Zinc Finger Protein Genetics, September 1, 2007; 177(1): 167 - 178. [Abstract] [Full Text] [PDF]

				T. E. Rusten, L. M.W. Rodahl, K. Pattni, C. Englund, C. Samakovlis, S. Dove, A. Brech, and H. Stenmark Fab1 Phosphatidylinositol 3-Phosphate 5-Kinase Controls Trafficking but Not Silencing of Endocytosed Receptors Mol. Biol. Cell, September 1, 2006; 17(9): 3989 - 4001. [Abstract] [Full Text] [PDF]

				H. J. Bellen, R. W. Levis, G. Liao, Y. He, J. W. Carlson, G. Tsang, M. Evans-Holm, P. R. Hiesinger, K. L. Schulze, G. M. Rubin, et al. The BDGP Gene Disruption Project: Single Transposon Insertions Associated With 40% of Drosophila Genes Genetics, June 1, 2004; 167(2): 761 - 781. [Abstract] [Full Text] [PDF]

				E. Ryder, F. Blows, M. Ashburner, R. Bautista-Llacer, D. Coulson, J. Drummond, J. Webster, D. Gubb, N. Gunton, G. Johnson, et al. The DrosDel Collection: A Set of P-Element Insertions for Generating Custom Chromosomal Aberrations in Drosophila melanogaster Genetics, June 1, 2004; 167(2): 797 - 813. [Abstract] [Full Text] [PDF]