Targeted gene expression as a means of altering cell fates and generating dominant phenotypes

Parts used

Key findings

• A two-part GAL4/UAS system restricts expression of any cloned gene to the cells that express yeast GAL4 in Drosophila. The target gene is subcloned downstream of five tandemly arrayed, optimized GAL4 binding sites (the UAS) and stays silent in the absence of GAL4, while the GAL4 transgene alone has no target to activate, so both parental lines remain viable and the target is switched on only in the progeny of a cross. Because GAL4 has no endogenous targets in Drosophila, only the introduced gene is misexpressed, and a single target line can be crossed to a library of GAL4 lines to produce many expression patterns.
• GAL4 expressed from the Drosophila Rh2 opsin promoter (Rh2-GAL4, built in the pGaTB vector with the hsp70 terminator) activates a UAS-lacZ reporter in the photoreceptor cells of the ocelli. Ocelli of flies carrying both Rh2-GAL4 and UAS-lacZ stain dark blue after a 15 minute X-Gal incubation, whereas ocelli of flies carrying only UAS-lacZ do not stain, and approximately 50% of the progeny of a heterozygous Rh2-GAL4 by homozygous UAS-lacZ cross express beta-galactosidase in the ocelli.
• The enhancer detection vector pGawB expresses GAL4 from the weak P-transposase promoter, so that a randomly integrated, enhancerless GAL4 gene comes under the control of nearby genomic enhancers. pGawB differs from the lacZ enhancer detection vector plwB in expressing GAL4 from its own AUG rather than as a P-transposase fusion, because the amino terminus of GAL4 supplies both nuclear localization and DNA binding. The vector can be remobilized to new genomic sites by P-transposition (shown with the constitutively active transposase source delta 2-3), so one transformant can generate many GAL4 lines with different expression patterns.
• In a screen of 220 independent pGawB insertion lines crossed to UAS-lacZ, 61% of GAL4 insertions activate UAS-lacZ in a specific embryonic pattern, 10.45% give no staining, 51% also activate in the salivary glands, and 28% activate in the salivary glands alone. The salivary gland bias suggests that the GAL4 vector construction created a position-dependent salivary gland enhancer.
• GAL4 activates UAS-lacZ in cells of all three embryonic germ layers, including endoderm (midgut), ectoderm (nervous system and epidermis) and mesoderm, and in the amnioserosa, foregut, hindgut, pharynx, malpighian tubules, vitellophages, oenocytes and glia. Representative pGawB lines are 1J3, which drives seven pair-rule stripes and head expression resembling hairy, 24B, which drives expression first in presumptive mesoderm and then in muscle, and 31-1, which drives expression in neuroblasts and then in neurons of the central and peripheral nervous systems. The pGawB insertion in line 1J3 lies at the hairy locus, since homozygotes, and heterozygotes in trans to the hairy allele h5HO7, show extra scutellar bristles.
• GAL4-directed expression has two limitations in Drosophila embryos. GAL4-dependent target expression is first detectable only after gastrulation, between 3.5 and 4 hours after fertilization, even though GAL4 mRNA is present at the cellular blastoderm stage, and none of the pGawB GAL4 lines screened shows appreciable germ line expression, in contrast to about one third of enhancer detection lacZ insertion lines.
• Misexpressing even-skipped with GAL4 from the pGawB lines 1J3 (hairy pattern), 24B (muscle) and 31-1 (throughout the central nervous system) each causes embryonic lethality. Line 1J3 sustains even-skipped stripes at stage 9, after endogenous even-skipped expression has normally faded, and line 31-1 expresses even-skipped across the whole nervous system instead of in a subset of neurons.
• Misexpressing even-skipped in the hairy pattern (UAS-even-skipped driven by the hairy-locus GAL4 line 1J3) expands even-skipped expression into the wingless-expressing cells and causes extra denticles in the naked cuticle between alternating abdominal segments, giving fusions of segments A3-A4, A5-A6 and A7-A8. In one experiment 68 of 103 embryos showed these fusions, 15 showed more extensive fusions and the rest looked wild-type, and the variable expressivity reflects the hairy promoter, because paired-GAL4 driving UAS-even-skipped gives alternating segmental fusions along the whole cuticle in 75% of embryos.
• Ectopic even-skipped represses wingless transcription in the cells where it is expressed. UAS-even-skipped driven by paired-GAL4 turns wingless off in alternating segments along the whole embryo, and UAS-even-skipped driven by the hairy-locus GAL4 line 1J3 represses alternate wingless stripes mainly in the posterior abdomen. Because wingless is required to instruct naked cuticle secretion between 6 and 9.5 hours of development, this repression converts naked-cuticle cells into denticle-secreting cells without causing central nervous system defects, so the epidermal and CNS roles of even-skipped can be studied separately.
• A lethal activated Ras can be kept in viable stocks as a silent UAS transgene and expressed only in the progeny of a GAL4 cross. Several independent UAS-Dras2Val14 lines (Dras2 with glycine 14 changed to valine) are viable and show no visible phenotype without GAL4, whereas Dras2Val14 expressed from a heat shock promoter during third instar larval development kills 60-80% of the population.
• The level and pattern of GAL4 expression in the imaginal discs sets the severity of the Dras2Val14 phenotype in adults. In crosses of UAS-Dras2Val14 line R52-a to pGawB lines, GAL4 insertions 71B and 30A, which are active in the wing disc but only weakly in the eye disc, give mild opaque wings with no obvious eye defect. Insertion 32B, which is highly active in both eye and wing discs, gives viable adults with a rough eye and wing vein defects (bifurcated longitudinal veins and extra vein material), fully penetrant in a sample of 100 flies. Insertion 69B, which is also highly active in both discs, gives a rough eye, reduced and disorganized wings, and mostly pupal lethality, and insertions 1J3 and 24B give pupal arrest.
• Targeted Dras2Val14 expression gives stable, reproducible dominant eye and wing phenotypes with normal viability and fertility, in contrast to the variable phenotypes from heat shock promoter or Dras2 promoter expression of Dras2Val14. These phenotypes provide a basis for genetic screens for second-site enhancers and suppressors of the Dras2 signal transduction pathway, and a rough eye without a wing defect can be generated by driving UAS-Dras2Val14 with sevenless-GAL4.

Abstract

We have designed a system for targeted gene expression that allows the selective activation of any cloned gene in a wide variety of tissue- and cell-specific patterns. The gene encoding the yeast transcriptional activator GAL4 is inserted randomly into the Drosophila genome to drive GAL4 expression from one of a diverse array of genomic enhancers. It is then possible to introduce a gene containing GAL4 binding sites within its promoter, to activate it in those cells where GAL4 is expressed, and to observe the effect of this directed misexpression on development. We have used GAL4-directed transcription to expand the domain of embryonic expression of the homeobox protein even-skipped. We show that even-skipped represses wingless and transforms cells that would normally secrete naked cuticle into denticle secreting cells. The GAL4 system can thus be used to study regulatory interactions during embryonic development. In adults, targeted expression can be used to generate dominant phenotypes for use in genetic screens. We have directed expression of an activated form of the Dras2 protein, resulting in dominant eye and wing defects that can be used in screens to identify other members of the Dras2 signal transduction pathway.