L-bodies are novel RNA-protein condensates driving RNA transport in Xenopus oocytes

Background: Phase separated bodies are all the rage to localize RNAs

Recently, there’s been a burst of new discoveries around cellular phase separated bodies as translational hubs, a way to spatially organize the cell’s cytoplasm. These foci, called ribonucleoprotein (RNP) granules, take advantage of RNA and protein binding features to collect the relevant biomolecules into the granule. In addition to their unique biophysical properties as liquid-like phase separated droplets, RNP granules also serve a biological function to regulate mRNA localization and distribution throughout the cytoplasm in order to generate cellular asymmetry and polarity that benefits cellular function. However, it’s unclear how these combinations of RNAs, proteins, and other elements may form RNPs and mediate their physical nature.

That’s where the Xenopus oocyte comes in handy. mRNAs must be localized in developing oocytes to pattern the embryo after fertilization, and these are specifically transported to the bottom of the oocyte, called the vegetal pole. These specific mRNAs rely on their RNA binding proteins (RBPs) to create an RNP structure that can be transported by microtubule motors down to the vegetal pole. Neil and Jeschonek et al have looked into the process more carefully to discover Localization bodies (L-bodies), which contain these RNP granule components and turn out to be multiphasic based on the biomolecular identities of L-body components. They unveil a new organization for cytoplasmic granules, opening new doors for investigation into their function in oocyte biology.

Key findings: It’s all about localization signals

Key mRNAs encoding germ layer formation, such as vg1, vegT, and trim36, are well known to localize to the vegetal pole in Xenopus oocytes via interactions with their RBPs. They appear in large clusters at the vegetal oocyte cytoplasm, enriched with multiple foci of each vegetal mRNA and its associated RBP, while other non-localized RNAs such as gapdh are not specifically enriched though included in the cluster. The composition of these clusters identified them as RNP granules, so the authors call these structures Localization bodies, or L-bodies.

L-bodies are encased in microtubule baskets with both kinesin and dynein motors, in line with previous observations that vegetal mRNAs are transported in a microtubule motor-dependent fashion. The microtubule baskets suggest that the L-bodies, rather than individual mRNAs, are interacting with this transport machinery.

Importantly, mRNA localization signals provide specificity for L-body composition. The authors microinjected a short RNA that recapitulates the vegetal localization of vg1, demonstrating that it can be packaged into L-bodies with endogenous vg1. However, if a mutant version that lacks RBP binding (mutLE) is microinjected, the RNA is not localized to the vegetal pole and is not incorporated into L-bodies. Thus, enrichment into L-bodies is mediated by sequence-specific features of target RNAs, which is then required for subsequent vegetal localization.

 

The authors then biochemically purified L-bodies and used mass spectrometry to identify candidate proteins. They found 86 L-body protein components, including all known RBPs, dynein, and kinesin. Excitingly, they found novel components that may provide functional specificity to L-bodies, but they also discovered that the proteome is filled with proteins that have been identified in other types of cytoplasmic granules, furthering their connection to previous studies of RNP granules. The majority of proteins have intrinsically disordered regions, and prion-like domains in particular are overrepresented in L-bodies.

Interestingly, thioflavin staining revealed a mesh-like structure in L-bodies, indicative of a more gel- or solid-like state, rather than liquid-like. This also fits in neatly with the overrepresentation of prion-like domains, which tend to lead to gel- or amyloid-like properties. Ex vivo, L-bodies are insensitive to hexanediol, suggesting that they are not liquid-like at all.

This hypothesis led the authors to perform FRAP experiments to probe the dynamic properties of L-bodies, and this led to an intriguing discovery. L-body proteins are dynamic and recover quickly after photobleaching, but localized RNAs are relatively immobile. Non-localized RNAs, such as gapdh or mutLE, are dynamic, similar to proteins. This suggests that L-bodies are multiphasic, where the important localized RNAs are immobile and create a gel-like center, while proteins are more dynamic and envelope this structure.

 

The authors propose a model where the localized RNAs are an organizational scaffold, and the interactions with specific RBPs promote local enrichment and incorporation of mRNAs into L-bodies. High RNA concentration promotes more RNA/RNA interactions, thus forming a solid gel like state and creating multiphasic behavior. Proteins can become more enriched with the RNA scaffold, but non-localized RNAs lack the ability to interact with the original scaffold, so they freely diffuse in and out. These new insights into mRNA localization through multiphasic L-bodies leads to interesting hypotheses about their function in oocytes and how they may mediate oocyte biology, including a role in packaging maternal material and silencing translation for very long periods of time.

 

Questions for the authors

How are L-bodies spaced throughout the cytoplasm? The images of the microtubule basket make it seem like L-bodies are connected to each other via the microtubule network as well. Apart from the role in transporting L-bodies, could the microtubules help to space out the bodies to maintain local foci rather than coalescing into a very large single L-body?

Do the novel proteins found in the L-body proteome provide any hints to their specific role in oocytes, especially since they are not found in other types of cytoplasmic granules? Relatedly, are specific families/classes of protein functions enriched in the proteome (rather than overrepresented domains)?

The role of L-bodies in oocytes in particularly intriguing to speculate on. For example, what would happen if L-bodies were disrupted after migration to the vegetal pole? How might that affect development/patterning/translation in the oocyte?

Tags: biomolecular condensates, frogs, oocytes, phase separation, rna, rnp granules, xenopus laevis

doi: https://doi.org/10.1242/prelights.21239

Read preprint (No Ratings Yet)