The nature of the wimp mutation.

    
        wimp is a dominant maternal-effect mutation, allelic to mutations in the second largest (140 kD) subunit of RNA polymerase II, that interacts with a specific subset of early genes, including many segmentation genes such as the pair-rule gene hairy. wimp is a change-of-function (antimorphic) mutation that reduces, but does not eliminate, transcription of these interacting genes, such that one wild-type copy of the interacting gene is no longer sufficient for normal development.

    Embryos heterozygous for an interacting mutation, such as hairy, die when derived from a wimp/+ mother:

The two classes of progeny that inherit the hairy mutant allele die because wimp reduces transcription of hairy such that embryos with only one wild-type copy of hairy no longer make sufficient hairy product to support normal development. The cuticle phenotype of the dead embryos resembles that of hairy mutant embryos.	While the same classes of progeny are produced when the genotypes of the parents are reversed, the two classes that inherit the hairy mutant allele are now viable.

    The reduction in interacting gene product due to the wimp maternal effect is reflected in both protein levels (shown below for the segmentation protein, Ftz) and at the level of transcription as assayed by in situ hybridization.

wildtype
wimp +/++
wimp +/+ftz (trans-heterozygous)

    Embryos derived from wimp/+ mothers crossed to wildtype fathers develop normally, despite the reduction of transcript levels of all interacting genes. In addition to its conditional maternal effect summarized above, wimp is also a recessive lethal mutation: homozygous wimp embryos die with severe cuticle pattern disruptions, suggesting that wimp may affect processes such as cellularization and gastrulation.

    We have looked at the interaction of the wimp maternal effect with a variety of different loci, including mutations in maternally- and zygotically-acting loci, by scoring for embryonic lethality.

    Based on our analysis of the interaction of the wimp maternal effect with a number of available early developmental mutations, we concluded that:

• The wimp maternal effect reduces transcription of the interacting gene. However, this is not necessarily a direct effect. wimp could be reducing the transcription of a specific gene or affecting a function regulating that specific gene.
• The cuticle phenotype of the resulting dead embryos resembles that of the interacting mutant being tested.
• Loci that do not interact with the wimp maternal effect are expressed at normal levels, hence wimp does not globally reduce transcription.
• The wimp maternal effect interacts with both maternal and zygotic mutations implicated in a wide range of early processes including segmentation, sex-determination, and cellularization.
• The wimp maternal effect only interacts with mutations whose defects are manifested early - either maternally or during the earliest zygotic stages (prior to the action of the segment polarity class) - presumably because at later times zygotically-encoded RNA polymerase is available. Later developmental phenotypes characteristic of interacting genes are not affected by wimp.
• The wimp maternal effect only interacts with a subset of the mutants tested. For any given category of mutants, some interact while others do not. We do not know the basis for this selectivity. Since wimp is a change-of-function mutation, it may reflect promoter-specific mechanisms of transcriptional activation, consistent with the proposal that different factors recognize different faces of the multi-subunit RNA polymerase.

Using the wimp mutation to identify maternal effect lethals

    Based on our survey of interactions of early developmental genes with the wimp maternal effect, we reasoned that genetic screens using wimp would identify mutations in a variety of early developmental processes including fertilization, nuclear cleavage divisions, nuclear migration, pattern formation, sex determination, cellularization and gastrulation. Although wimp exhibits interaction specificity such that it does not interact with all mutations, it renders many interacting recessive mutations pseudodominant and greatly eases the identification of novel maternal loci. Importantly, wimp allows recovery of mutations in genes that are required both maternally and zygotically, a class of mutation that is lost in conventional screens because of recessive lethality. wimp also allows identification of maternally-acting mutations that are required for cell viability (mutations where germline clones cannot be recovered).

    With this in mind, we set out to identify wimp-interacting mutations with crucial roles in Drosophila oogenesis or early embryogenesis that define new pathways of regulation. We have screened approximately 3500 lethal P-element lines (lines from Bloomington, Berkeley, and I. Kiss) and have isolated 98 mutations representing 92 complementation groups that interact maternally with wimp and that are essential for early development. We have put these mutations into different classes on the basis of their cuticle phenotypes, examples of which are shown below:

    To get a better idea of the kinds of gene products/processes being disrupted by these mutations, we rescued the DNA sequences flanking the P-element for 85 of the mutant lines. Using primers specific to the P-element ends, we obtained roughly one kilobase of sequence flanking each side of the P-element and then searched for sequence homologies. Roughly 30% of the sequences showed homology to known genes or protein motifs.

   Our initial characterizations indicate that we have pulled out mutations with crucial roles in early development that define new pathways of regulation, bridge steps in existing pathways, or define a new role during oogenesis for a known gene. The recovered mutations encode genes with roles in transcription, the cell cycle, signal transduction, the cytoskeleton, and RNA stability and localization -- all important processes in the proper establishment of the body plan.

   Importantly, these are mutations with maternal roles that could not have been detected by other means. 90% of the loci we identified in this screen are required for cell viability such that the only way to look at their maternal requirements is in the sensitized wimp background. These mutants provide useful entry points into specific developmental pathways, and their subsequent analysis is expected to complement those of existing mutations, and to help in understanding mechanisms used in various early events required for proper embryonic development.

   We are focusing our efforts on characterizing mutations in the roadkill class since many of these are likely to reside in previously uncharacterized loci with the promise of yielding important new information about early developmental events in general, as well as the specific mechanisms in which they are involved. In addition, we are characterizing three other mutations, dCtBP, dSir2, and Rho1 GTPase, that complement our studies on transcriptional repression and morphogenesis/signaling.

Reference:

Parkhurst, S.M. and Ish-Horowicz, D.
wimp, a dominant maternal-effect mutation, reduces transcription of a specific subset of segmentation genes in Drosophila.
Genes & Dev. 5, 341-357 (1991)

The wimp allele is available from the Bloomington Stock Center (stock number 5874).

Last updated 10/22/08