Preference For Transposition Research Articles

Use of mini-white as a reporter gene to screen for GAL4 insertions with spatially restricted expression pattern in the developing eye in drosophila.

We have previously used mini-white as a reporter gene in an enhancer trap screen to identify enhancers/silencers that specify spatially restricted expression patterns during Drosophila eye development (Sun et al., 1995). The mini-white gene carries its own eye-specific enhancer (Pirrotta et al., 1988), therefore causing uniform eye pigmentation. In rare cases, the P[mini-white] insertions cause a spatially restricted pigmentation pattern in the adult eye (Sun et al., 1995), suggesting that the mini-white enhancer activity can be suppressed by local silencer activities. The spatial expression pattern is likely a net effect of local enhancers and silencers. The screen yielded more than 20 loci with a position-specific expression (POSE) pattern in the developing eye (Sun et al., 1995). In this study, we employed the same strategy to screen for GAL4 insertion lines with a POSE pattern in the developing eye. We induced P[GawB] or P[hs-GAL4] (Brand and Perrimon, 1993) transposition from several independent starting chromosomal locations to avoid bias in transposition preference. A total of about 420,000 flies were screened and 22 lines with nonuniform eye pigmentation patterns were isolated. One line (Notch) was identified by a notched wing phenotype. Three of the lines showed no GAL4 expression pattern. Nineteen lines showed specific expression patterns in the eye disc (Fig. 1), indicating that the mini-white-based screen is an efficient screening method. The GAL4 lines were named based on their pattern of eye pigmentation (Table 1, Fig. 1). The site of insertions was mapped by segregation analysis, followed by plasmid rescue to clone the flanking genomic fragment and sequencing the ends of the rescued fragments (Table 1). The nearest gene is indicated (Table 1). In four lines (AG2, PDM1, Cir2, and Cwh1), plasmid rescue using multiple restriction enzymes failed to recover any genomic fragment, suggesting that the P-element has undergone some deletion or rearrangement. Eleven of the lines are homozygous lethal or semilethal. Two lines have dominant phenotypes: VG1 has a grooved notum (Fig. 4a), and Notch has notched wings (not shown). Insertions into Notch (N), four jointed ( fj), wingless (wg), optomototr-blind (omb), homothorax (hth), and patched (ptc) that caused mutant phenotypes were confirmed by complementation tests with known mutant alleles (see Table legend). Kruppel (Kr), hth, and omb each have two insertion lines, indicating that these are hot spots for P insertions. Multiple insertions in hth and omb were also isolated in our previous screen (Sun et al., 1995). The P5 line has an insertion within the hth locus, but the lethality was complemented by hth and hth, and the lethality may thus be caused by a second site mutation. The two insertions into Kr are located in the same position (Table 1), but one (AG1) caused no phenotype, while the other (AM1) is homozygous semilethal, suggesting that the two lines may differ in some rearrangement or deletion in the region flanking the insertion. Similarly, omb1 and omb3 are also inserted in the same position upstream of omb but caused different phenotypes (Table 1). The GAL4s were crossed to UAS-lacZ and the expression patterns in embryos, third instar larval imaginal discs (Fig. 1, 2), and in adult testes and ovaries (Fig. 3) were examined. When the GAL4 is inserted in or near a gene with a published expression pattern, the expression patterns were compared. In AG1 and AM1, the GAL4 is inserted in a region near the telomere that has a very low gene density. The closest gene is Kr, which is about 30 kb away. The AG1 and AM1 lines show similar expression patterns. The embryonic expression is similar to the expression of Kr in the Bolwig organ (Schmucker et al., 1992). In Notch, the insertion is within the N gene, but the expression pattern is dissimilar to that of N (Kidd et al., 1989). In VG1, the GAL4 is inserted within the fj gene, and the expression pattern is similar to that of fj (Villano and Katz, 1995; Brodsky and Steller, 1996). In A4, the GAL4 insertion is 749 bp upstream of tiptop, which has no reported expression pattern. The expression pattern in wing and leg discs is similar to that of teashirt (tsh; Erkner et al., 1999;

RAG transposase can capture and commit to target DNA before or after donor cleavage.

The discovery that the V(D)J recombinase functions as a transposase in vitro suggests that transposition by this system might be a potent source of genomic instability. To gain insight into the mechanisms that regulate transposition, we investigated a phenomenon termed target commitment that reflects a functional association between the RAG transposase and the target DNA. We found that the V(D)J recombinase is quite promiscuous, forming productive complexes with target DNA both before and after donor cleavage, and our data indicate that the rate-limiting step for transposition occurs after target capture. Formation of stable target capture complexes depends upon the presence of active-site metal binding residues (the DDE motif), suggesting that active-site amino acids in RAG-1 are critical for target capture. The ability of the RAG transposase to commit to target prior to cleavage may result in a preference for transposition into nearby targets, such as immunoglobulin and T-cell receptor loci. This could bias transposition toward relatively "safe" regions of the genome. A preference for localized transposition may also have influenced the evolution of the antigen receptor loci.