Macdonald Lab

Translational regulation of oskar mRNA

The Osk protein is tightly restricted to the posterior pole of the oocyte, where it recruits factors required for provision of determinants that direct posterior body patterning and germ cell formation. If Osk is allowed to spread outside this region, body patterning is reprogrammed (a lethal event) and ectopic germ cells form (Smith et al. 1992). To achieve this restriction of Osk protein, osk mRNA is very effectively localized to the posterior of the oocyte (Kim-Ha et al. 1991; Kim-Ha et al. 1993), and translation of osk mRNA is tightly regulated (Kim-Ha et al. 1995).

Our work on osk mRNA translational regulation began with the discovery of Bruno, which binds to multiple sites in the osk mRNA 3' UTR. Mutation of these sites led to precocious accumulation of Osk protein, implicating Bru as a translational repressor (Kim-Ha et al. 1995). We identified arrest as the gene encoding Bruno (Webster et al. 1997), and developed in vitro translation systems from Drosophila ovaries and embryos to prove that Bru acts as a repressor (Lie and Macdonald 1999b).

Our picture of Bru function has expanded in the last few years. Using in vitro selection methods we found that Bru has multiple classes of binding sites: the BREs we originally identified, as well as several other types. Some of these appear in the osk mRNA 3' UTR, clustered close to the BREs (Reveal et al. 2011). Each of the three types of sites in osk (BREs, type II sites and type III sites) contribute to translational control. Surprisingly, in experiments in which we mutate only subsets of Bru binding sites, we found an additional role for them in translational activation (Reveal et al. 2010). It appears that Bru mediates both functions, serving both as a repressor and an activator.

Current work is aimed at understanding how osk mRNA translation is activated, and how activation is coordinated with mRNA localization. Our evidence does not support existing models, and instead points to a program with multiple forms of activation. We are testing this model, including predictions for how Bru activity is controlled to release osk mRNA from repression.

Our analysis of the function of the Bru binding sites revealed a surprising phenomenon: osk mRNA translation can be regulated in trans (Reveal et al. 2010). Specifically, copies of osk mRNA with the normal regulatory elements (but with a mutation in the coding region that prevents them from making Osk protein) can restore proper regulation to copies of osk mRNA with mutations in Bru binding sites. Defects in both repression and activation can be restored in this manner. We suspect that trans regulation is made possible by the assembly of osk mRNA in particles, with the close physical association allowing regulators bound to one mRNA to act on other mRNAs.

We plan to evaluate the importance of assembly of osk mRNA in particles for trans regulation. Many mRNAs are found in cytoplasmic particles, raising the possibility that trans regulation may be relatively common but not previously detected. We plan to determine if this phenomenon is widespread.

Cytoplasmic RNPs

Studies on transcriptional regulation began with a focus on cis-regulatory elements (promoters and, later, enhancers) and trans-acting factors (RNA polymerase and transcription factors). Elucidation of the details of the basic transcriptional apparatus and its operations using naked DNA and purified proteins was followed by a growing interest in the cellular substrate for transcription: chromatin. We now know that chromatin has a dynamic structure and exerts substantial control on transcription. Similarly, it seems very likely that subcellular context will be an important part of post-transcriptional regulatory events in which the participants are well defined cis-acting elements (such as BREs) and individual regulators (such as Bruno).

There are several large RNPs in the ovary: nuage, sponge bodies, and polar granules. Nuage occupies a perinuclear position in the nurse cells, sponge bodies are more cytoplasmic, and polar granules are assembled only at the posterior pole of the oocyte.

We are studying several of these RNPs, characterizing their makeup, relationship to one another, and functions. One notable feature is the overlap in components, with many proteins present in more than one type of RNP. This overlap is particularly striking iun a comparison of nuage and polar granules, leading us to ask if nuage particles are precursors to polar granules. Using a combination of live imaging and photobleaching methods we found that polar granules are assembled de novo (Snee and Macdonald, 2004). Current efforts focus on sponge bodies and nuage, as well as the RNPs containing mRNAs under control of miRNAs.

Ectopic germ cells from overexpression of Oskar

Altered nuclear body morphology from D. immigrans Oskar expression in D. melanogaster germ cells.

Germ cell formation and function

Polar granules, described in the previous section, are implicated in germ cell formation. Early transplantation experiments demonstrated that transfer of polar cytoplasm to a different position in the embryo led to ectopic germ cell formation at that site. More recently, mutants lacking polar granules have been shown to be defective in germ cell formation.

Oskar protein is a key player in polar granule formation. Oskar appears to function by nucleating formation of the granules, and recruits other granule components. To ask if Oskar acts only in the initial formation of polar granules during oogenesis, or if it has a continuing role, we took advantage of morphological differences in the polar granules of D. melanogater and D. immigrans. Although polar granules of both species are initially similar in morphology, during early stages of embryogenesis the D. melanogaster granlules adopt a spherical shape while the D. immigrans graules become rod-like. We expressed the D. immigrans oskar gene in D. melanogaster flies, and found that the granules were converted to the rod-like morphology. As part of this work we also found that Oskar is present in nuclear bodies, structures similar in morphology to the polar granules but found in the nuclei of embryonic germ cells (polar granules are cytoplasmic), and that the D. immigrans Oskar protein converts the nuclear bodies to the D. immigrans morphology. Thus, the function of Oskar is not limited to the well known role of nucleating polar granules formaiton during oogenesis, but also has a continuing role in embryonic germ cells (Jones and Macdonald, 2007).