RNA Localization Meets Wingless Signaling

Science's STKE  24 Jul 2001:
Vol. 2001, Issue 92, pp. pe1
DOI: 10.1126/stke.2001.92.pe1


In many tissues in many organisms, messenger RNA (mRNA) is not randomly distributed throughout a cell, but is targeted and accumulated in specific subcellular locations. Signal transduction pathways can be extremely sensitive to gradients of signaling proteins and represent a cellular phenomenon where RNA localization may be important. Manseau discusses how targeting of the RNA for a secreted ligand, Wingless (Wg), is important in promoting proper Wg signaling in early Drosophila embryos and the mechanism for achieving this subcellular targeting. The possible mechanistic models for how restricted wg mRNA distribution influences Wg signaling are also detailed.

It is becoming evident that RNA localization is important to intercellular signaling. A number of RNAs that encode signaling molecules are asymmetrically localized. These include Vg1, which encodes a transforming growth factor-β (TGF-β)-like molecule (1), and sevenless, a receptor tyrosine kinase (2-4). Although it is unknown whether the asymmetric localization of these RNAs affects the signaling process, there is one signaling pathway in which it has been clearly demonstrated that asymmetric RNA distribution is important. Gurken mRNA, which encodes a TGF-α molecule, is asymmetrically localized within the developing Drosophila oocyte. This asymmetric localization is required to signal to the surrounding follicle cells to establish the anterior-posterior and dorsal-ventral axes of the egg and embryo (5-7). Two recent papers (8, 9) demonstrate that RNA localization is important in a second signaling pathway, the Wingless (Wg) signaling pathway.This second example of RNA asymmetry contributes to the establishment of apical and basal polarity of polarized epithelia, which, until now, has been generally accepted to be generated through protein sorting in the trans-Golgi network (10).

Wg is the prototypical member of the Wnt proto-oncogene family. In the early Drosophila embryo, the Wg signaling pathway establishes the pattern within a segment. Wg is expressed in segmentally repeated single-cell-wide stripes in the cellular blastoderm. The encoded protein is a secreted molecule that is a ligand for the Frizzled and Frizzled2 seven-pass transmembrane receptors (11-13). This signal is then conveyed to the nucleus of the receiving cells through a signal transduction cascade involving the novel protein Dishevelled, Zeste-white 3, a serine/threonine kinase, and Armadillo. Armadillo activates expression of Wg-responsive genes by complexing with the high-mobility group transcription factor Drosophila T cell factor [reviewed in (14); see the Connections Map of the Drosophila Wg Pathway (15)]. The mechanism used to transport Wg to neighboring cells is somewhat controversial and may involve either restricted extracellular diffusion, cell proliferation and movement, or planar transcytosis [for a review, see (16)].

The recent paper by Simmonds et al. demonstrates that apical localization of wg mRNA is important to the Wg signaling pathway (8). Previous work had demonstrated that other components of the Wg pathway, including Wg (17, 18), are apically localized, but this is the first work to show that this apical localization requires an apically localized RNA and that this RNA targeting is important for function. Using a new, highly sensitive fluorescent in situ hybridization technique, Simmonds et al. found that wg mRNA is localized apically in the cellular blastoderm of the Drosophila embryo. Sequences responsible for apical localization of wg mRNA are found in two regions of the 3′ untranslated region (UTR). By replacing the 3′ UTR of wg with that of partner of paired, which encodes a basally directed mRNA, or with that of SV40, which allows unlocalized mRNA, the authors were able to direct the wg mRNA to different regions of the cell (Fig. 1). They found that apically targeted wg mRNA would substantially phenotypically rescue wg mutations, whereas basally targeted or unlocalized mRNA's provided for less efficient rescue. They found that wg protein distributions were similarly affected (Fig. 1). In the uniformly distributed transcript, Wg protein was (surprisingly) still enriched apically, but more extracellular protein was found in the middle of the segment. In the case of the basally targeted wg mRNA, even more extracellular wg protein was seen over the middle of the segment, again apically enriched. Why does basally secreted Wg have a different distribution than apically secreted Wg? One possible explanation is that Wg protein secretion might occur at a faster rate from the basal surface, leading to higher levels of extracellular Wg. Alternatively, Wg protein might diffuse more rapidly from the basal surface into the extracellular matrix or it might be less efficiently endocytosed basally.

Fig. 1.

Different targeting of wg transcripts affects Wg protein distribution. (A) Wild-type targeting of wg RNA results in Wg protein being found apically in a single-cell-wide stripe. (B) When wg RNA is targeted apically, the Wg protein distribution resembles that seen in the wild type. (C) When wg RNA is distributed randomly, Wg protein is found both apically and basally. (D) When wg RNA is directed basolaterally, Wg protein is found apically, but is more broadly distributed across the segment. This figure based on results from Simmonds et al. (8).

A number of mechanisms for RNA localization have been proposed [reviewed in (19)]. One of these is vectorial transport out of the cell nucleus to the site of localization. Second, anchoring at the site of localization can lead to the accumulation of transcripts at the site of anchoring. A third mechanism for localization of transcripts is transport along microtubules or microfilaments to the site of localization. Finally, stabilization at the site of localization and degradation elsewhere can lead to the accumulation of transcripts at the site of stabilization.

In the second paper discussed here, Wilkie and Davis distinguish among these four mechanisms for RNA localization in the Wg signaling pathway (9). By carefully monitoring the spatial distribution of transcripts as they move through the nucleus, the authors found that apically localized transcripts (wg, runt, and fushi tarazo) and basally localized transcripts (string) all exit the nucleus uniformly. Thus, the model of vectorial transport out of the cell nucleus was eliminated. Next, in experiments in which AlexaFluor dyes were incorporated into in vitro transcribed mRNAs, the authors found that injected, apically targeted mRNAs assemble into bright particles that move directly to the apical cytoplasm. This suggests that apical localization of these transcripts involves an active transport mechanism. These bright particles are apical localization intermediates containing apically targeted RNAs, but not basally targeted or uniformly distributed RNAs. The microtubule inhibitor colcemid prevents apical localization of apically targeted transcripts, whereas the microfilament inhibitor cytochalasin B does not affect apical transcript localization. This implicates the microtubule cytoskeleton in the apical localization process (Fig. 2).

Fig. 2.

Model for apical transport of wg RNA. Wg is transcribed in the nucleus and assembles into RNP particles. The particles move randomly through the nucleus and exit the nucleus on all sides. In the cytoplasm, the particles move apically along microtubules by utilizing the dynein motor. This figure adapted from Wilkie and Davis (9).

Through a series of elegant experiments, Wilkie and Davis demonstrate that dynein is the motor protein responsible for the movement of the RNA particles along the microtubules. They found that injection of two different dynein heavy chain monoclonal antibodies inhibited apical localization of injected, apically targeted RNAs. In addition, two different hypomorphic allelic combinations of dynein heavy chain mutants, which cause slower transport of lipid droplets, result in 60 to 70% slower movement of RNA particles toward the apical side of the cell. p50 (also known asdynamitin) is a component of the dynactin complex, which regulates dynein (20). Overexpression of p50 disrupts the dynactin complex and can be used to demonstrate involvement of dynein in a particular cellular process (21). Injection of p50 slowed apical localization of injected RNAs, conclusively demonstrating that dynein is required for apical localization. Finally, injection of colcemid or antibodies against dynein also caused apically localized RNA particles to diffuse from the site of injection, indicating that microtubules and dynein are also required for anchoring RNA particles in the apical cytoplasm.

So, why are transcripts localized during signaling? In the case of gurken, it is thought that localization of gurken transcripts leads to local synthesis of protein and, thus, targeting of gurken protein to particular sets of neighboring follicle cells necessary to establish the anterior-posterior and dorsal-ventral axes (6, 7, 22). Simmonds et al. (8) have offered a number of possible explanations for why Wg might be apically targeted. Although not the case for Wg in the early embryo, in other tissues, the basolateral surface of wg-expressing cells is adjacent to cells that do not express wg. Thus, apical targeting of wg RNA could direct Wg protein away from these basally located cells and toward apically located cells. Alternatively, controlling where wg RNA is localized could affect translation efficiency, protein processing, cofactor association, or trafficking. Western analysis of Wg protein made from basally versus apically directed mRNAs suggests that translation efficiency and protein processing are similar for the differently targeted RNAs (8). Porcupine, a component of the Wg pathway, is an endoplasmic reticulum (ER) protein required for Wg secretion (23). If porcupine is restricted to an apical region of the ER, it could differentially affect the secretory pathway that Wg enters and, thus, could affect secretion or associated cofactors. Apical targeting of wg RNA may result in association of Wg protein with different extracellular matrix components. Wg associates with heparin sulfate proteoglycans. This association is required for Wg signaling either to facilitate binding of Wg to the receptor or to limit diffusion of Wg away from the receptor (24-26). This could explain why basally targeted wg RNA results in a broader distribution of Wg protein. Finally, apical targeting of wg RNA may be required for association of Wg protein with apically localized receptor complexes. A number of other components of the Wg signaling pathway are targeted apically. These include the Frizzled receptor, which is present on the apical surface of imaginal disc cells (27), and Armadillo, which functions as a structural component of the apically localized adherens junction and transmits the signal to the nucleus (28).

Where is this work headed? There are a number of remaining questions relating to the mechanism of apical RNA targeting. What are the other components of the apical transport particles? How are these particles targeted apically? How do they attach to the dynein motor? A major question with regards to mRNA localization and signaling is, why does apical and basal targeting of wg RNA result in different distributions of Wg protein? Some of the possible explanations discussed above need to be explored. Finally, what is the function of apical targeting of wg RNA. Could it be necessary to direct Wg away from basally situated cells? Answering these questions should yield many interesting results at the intersection of cell signaling and RNA localization.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
View Abstract

Navigate This Article