Axonal mRNA translation: an unexpected link to axon survival and the mitochondrion

Abstract

Localized mRNA translation plays roles in dendrites and axons, but the regulatory mechanisms and downstream pathways are not well understood. Yoon et al. show that lamin B2, well known as a nuclear protein, undergoes regulated synthesis in axons, promoting mitochondrial function and axon survival.

Eukaryotic cells are compartmented into functionally distinct subcellular regions, and proper localization of proteins is usually essential for function. While many proteins possess amino acid sequences that target them to specific locations, it is becoming increasingly clear that mRNA transport and local translation plays a widespread role in protein localization (Holt and Bullock, 2009; Swanger and Bassell, 2011).

Neurons present an extreme example of cell compartmentation where protein synthesis can differ not only between axons and dendrites, but also between different regions of a dendrite or axon. In dendrites, local translation is regulated by synaptic activity and plays a role in plasticity. In axons, protein synthesis can be regulated in the growth cone in response to guidance cues and this can contribute to growth cone turning, collapse, or change in responsiveness. Local protein synthesis has also been implicated in axon regeneration (Holt and Bullock, 2009; Swanger and Bassell, 2011).

Although a large number of mRNAs have been found to localize within the axon, we still have limited knowledge about the roles of individual locally translated mRNAs: either the axonal functions of specific mRNAs, or which of them may be regulated in response to extracellular cues. In the xxx February issue of Cell, Christine Holt and colleagues report the unexpected discovery that a major protein subject to translational regulation within the axon is a member of the lamin B family – proteins known for decades as major structural components of the lamina which lines the inner face of the nuclear membrane (Yoon et al., 2012).

While unexpected findings can be difficult to pursue they often lead to informative breakthroughs. Following up on their surprising discovery that lamin B2 is translated in the axon, Yoon et al. showed that preventing its synthesis leads to axon degeneration, revealing a new role for local translation in axon survival. A further unanticipated finding was that lamin B2 accumulates in axonal mitochondria and regulates their size and activity. Although their observations were unexpected, they seem to fit well with an array of previous observations in biology and disease. For example, neurodegenerative diseases have been linked both to mitochondrial dysfunction and to lamin mutations (laminopathies), and indeed Type 2 Charcot-Marie-Tooth disease, a neuropathy involving axon degeneration, is known to result from inherited mutations in either mitochondrial proteins or lamins (Dauer and Worman, 2009; Lu et al., 2009). The mechanistic links uncovered by this study illustrate how the study of local translation can not only benefit our understanding of this widespread posttranscriptional regulatory mechanism, but can also help more generally to uncover unexpected functions of molecules within specific compartments of the cell.

Yoon et al. (2012) came across this unsuspected role of lamin B2 in a proteome-wide screen for proteins synthesized in axons in response to the extracellular cue Engrailed. Engrailed is a homeodomain protein, long known as a nuclear transcription factor. Although at first sight Engrailed might not seem like an obvious molecule to use as an extracellular cue, work most notably by the group of Alain Prochiantz has shown that homeodomain proteins can cross the cell membrane, and previous studies by the Prochiantz and Holt groups showed that Engrailed can act as a guidance cue for retinal ganglion cell (RGC) axons (Brunet et al., 2005). These are the axons that transmit information from the retina to the tectum, the primary visual center of the brain in non-mammalian vertebrates. Connections between the retina and the tectum are highly organized topographically to produce an accurate representation of the outside world in the tectum. To generate these orderly connections during development, RGC axons are guided within the tectum by gradients of cues, including ephrins and Engrailed (Luo and Flanagan, 2007). Yoon et al. (2012) chose to study RGC axons and Engrailed because the turning response is translation dependent, and Engrailed strongly upregulates axonal protein synthesis.

To screen for proteins synthesized in axons after Engrailed stimulation, Yoon et al. (2012) ingeniously combined a metabolic labeling technique with 2D gel analysis. Axons were isolated in culture, stimulated with Engrailed and newly synthesized proteins labeled by incorporation of a modified amino-acid (AHA) that can be subsequently fluorescently tagged (Dieterich et al., 2010). Newly synthesized proteins from Engrailed stimulated and unstimulated axons were labeled with differently colored fluorophores and run together on a 2D gel, where proteins whose synthesis was upregulated, downregulated or unchanged could be identified as green, red or yellow spots respectively. By mass spectrometry of these protein spots, they identified twelve proteins increased by Engrailed in the axon, and surprisingly lamin B2, a protein known for its nuclear functions, was induced the most strongly (Figure 1).

Figure 1. Axonal translation of Lamin B2.

In response to extracellular Engrailed, a major upregulated protein is Lamin B2, which localizes to mitochondria, regulates their size and activity, and promotes axon survival.

Extraordinary claims tend to require extraordinary evidence and Yoon et al. (2012) used an impressive series of experiments to provide evidence that lamin B2 is synthesized and functions within the axon. This included experiments to demonstrate the presence of lamin B2 protein and mRNA in RGC axons in vitro and in vivo. In an interesting additional approach, they use the recently developed TRAP technique (Heiman et al., 2008) to show that lamin B2 mRNA is associated with ribosomes in RGC axons in vivo. In this method, animals are generated expressing GFP-tagged ribosomes, then the ribosomes are immunoprecipitated using anti-GFP antibodies, and associated RNAs are identified. Here, the authors created Xenopus tadpoles in which GFP-tagged ribosomes were only expressed in a transplanted eye. These RGC axons innervated the host tectum, ensuring that tectal lysates would contain GFP-tagged ribosomes originating exclusively from RGC axons. Anti-GFP immunoprecipitation demonstrated the association of lamin B2 mRNA with ribosomes in retinal axons, providing evidence that lamin B2 is locally translated in these axons in vivo.

Having demonstrated that lamin B2 is produced in RGC axons, Yoon et al. then examined its function there in vivo. First, they knocked down lamin B2 in the eye by antisense morpholino and found that RGC axons degenerated after they reach the tectum while their cell bodies survived. This shows that lamin B2 is essential for RGC axon maintenance in vivo.

They then asked whether its role in axon maintenance relied on its presence in the axon rather than in the nucleus. Consistent with this hypothesis, the lamin B2 knockdown phenotype could be rescued by overexpressing a mutant form of lamin B2 that cannot enter the nucleus. Moreover, knockdown of lamin B2 by applying morpholino specifically to the tectum, presumably affecting RGC axons but not their cell bodies, led to axon degeneration without altering lamin B2 levels in the soma. These results indicate that lamin B2 functions in the cytoplasm to promote axon survival independently from its role in the nucleus.

By what mechanism, then, might lamin B2 play a role in axon survival? A first indication of its function in axons came from its localization: in axons, lamin B2 immunolabeling was found to coincide well with mitochondria. Furthermore, lamin B2 knockdown led to elongated mitochondrial morphology and reduced the mitochondrial potential necessary to drive ATP synthesis. This was also accompanied by reduced bidirectional movement of vesicles in the axon, a process heavily dependent on the ATP produced by mitochondria. These data show that lamin B2 promotes mitochondrial function, providing a plausible explanation for its effects on axon survival.

Taken together, these results clearly show that lamin B2 is synthesized within the axon, and provide impressive evidence that this local synthesis is required for maintenance of RGC axons in vivo within the tectum. Regarding the precise developmental roles of this mechanism, these interesting results suggest more than one potential model. One model would be based on the precedent of target-derived survival factors, such as the neurotrophins, which maintain the survival of neurons and their axons that have reached the correct target (Vanderhaeghen and Cheng, 2010). According to this model, tectal Engrailed protein could provide a survival signal for RGC axons that have correctly reached the tectum. Although a survival-promoting role for Engrailed has not yet been shown directly in axons, such a role seems very plausible in view of a recent study showing that Engrailed promotes the survival of dopaminergic neurons by a mechanism involving translation of nuclear-encoded mitochondrial proteins (Alvarez-Fischer et al., 2011). An alternative model, not mutually exclusive, could involve roles of Engrailed in topographic mapping within the tectum. In support of this model, Engrailed is known to cause topographically-specific attraction and repulsion of RGC axons from the nasal and temporal sides of the retina, respectively, in a mechanism involving translational regulation (Brunet et al., 2005). A role for Engrailed in topographic guidance is also supported by in vivo evidence (Wizenmann et al., 2009), and the mechanism appears to involve regulation of sensitivity to ephrins, which again involves translation of mitochondrial proteins (Stettler et al., 2012). Together, these studies suggest parallel roles for Engrailed in survival of neurons and axons, as well as in guidance. Whether these functions all operate through a single pathway downstream of Engrailed, or through multiple pathways, and whether they all involve mitochondrial function, or lamin B2, will be interesting questions for the future.

Interestingly, the guidance and survival responses shown for Engrailed all involve protein synthesis, raising the question of why local protein synthesis might provide particular advantages as a mechanism. In growth cone guidance, asymmetric protein synthesis of cytoskeletal proteins on one side of the growth cone is thought to mediate the asymmetric turning response to directional cues such as netrin (Holt and Bullock, 2009; Swanger and Bassell, 2011). In axon survival, local control of protein synthesis could offer a simple mechanism to selectively promote the survival of a subset of axon segments or branches where translation is occurring. Finally, the case of lamin B2 illustrates another potential advantage of local translation: synthesis far from the nucleus may help prevent nuclear entry directed by the nuclear localization sequence in the protein, permitting a separate non-nuclear function in the mitochondrion, thus providing a way for a single protein to have distinct functions in different compartments of the cell.

Another fascinating question raised by this study is how, mechanistically, does lamin B affect mitochondrial shape and function? Nuclear lamins act as a structural scaffold important for nuclear membrane integrity, and their phosphorylation leads to nuclear membrane fragmentation during cell division (Dauer and Worman, 2009). By analogy, we might speculate that lamins could control mitochondrial membrane cleavage during fission. This could explain why lamin B2 knockdown leads to highly elongated mitochondria, a phenotype typically seen in mutations affecting cleavage. However, beyond their structural role, lamins play a variety of other functions in the nucleus such as regulation of transcription and DNA repair (Dauer and Worman, 2009), and novel mechanisms of action in mitochondria cannot be excluded. Interestingly, lamin has been found to have a role in linking the nucleus to the microtubule cytoskeleton, regulating nuclear translocation (Coffinier et al., 2010) and neuronal migration (Patterson et al., 2004). It would be interesting to know whether lamin B2 might have analogous functions in linking the mitochondrion to the cytoskeleton, in the axon and perhaps also in migrating cells.

The unexpected links among axonal mRNA translation, lamin B2, and the mitochondrion promise to shed new light on the future understanding of laminopathies and mitochondrial dysfunction in neurodegenerative disease. They also provide new insight into the basic biology of axons, as well as the fundamental mechanisms that produce the localized organization of cellular components.