doi:10.1016/j.cub.2005.04.002 How to Cite or Link Using DOI (Opens New Window)
Copyright © 2005 Elsevier Ltd All rights reserved.

Dispatch

Axon Regeneration: It’s Getting Crowded at the Gates of TROY

Wim J. MandemakersCorresponding Author Contact Information, E-mail The Corresponding Author and Ben A. Barres
Stanford University School of Medicine, Neurobiology Department, Sherman Fairchild Science Building, Room D129, 299 Campus Drive, Stanford, California 94305-5125, USA.

Available online 26 April 2005.



A novel neuronal receptor complex that mediates myelin’s inhibitory action on nerve fiber regeneration has at last been identified. This discovery could be an important step towards promoting nerve regeneration after stroke or spinal cord injury.


Most tissues in our bodies have an astonishing ability to repair themselves after injury, but a notable exception is the adult mammalian central nervous system (CNS) — the brain and spinal cord — which has little innate capacity for repair. When the axons of CNS neurons are severed they are unable to regenerate, accounting for the devastating impairment induced by spinal cord injury, stroke and many other neurological conditions. What accounts for the failure of the CNS to regenerate? One crucial factor is that the adult CNS environment is strongly inhibitory to regenerating axons. In particular, much recent attention has been focused on the powerful inhibition of degenerating myelin.

How does myelin inhibit the growth of CNS axons? Three different myelin proteins have been identified that are strongly inhibitory to most types of CNS neurons: Nogo-A, myelin associated glycoprotein (MAG) and oligodendrocyte myelin glycoprotein (OMgp) [1]. Remarkably, all three of these myelin-associated inhibitor factors are able to bind to and activate the same axonal multi-protein receptor complex. The proteins that generate this receptor complex are the ligand binding Nogo-66 receptor (NgR1) and two signal transducing binding partners, the p75 neurotrophin receptor (p75) and the LRR and Ig domain-containing, Nogo receptor-interacting protein (LINGO-1) [2 and 3]. Interaction of myelin-associated inhibitors with the p75/NgR1/LINGO-1 complex activates the small GTPase RhoA, which rigidifies the actin cytoskeleton, thereby causing growth cone collapse [4]. The potential contribution of this signaling pathway to CNS regenerative failure is clearly shown by the ability of RhoA blockade to promote axon regeneration in vivo after injury [5 and 6].

Though the identification of the p75/NgR1/LINGO-1 receptor complex has been an exciting advance, a perplexing mystery has remained. Although NgR1 and LINGO-1 are widely expressed throughout the CNS, p75 is for the most part down-regulated during development so that it is not widely expressed in the adult CNS. Many CNS neuron types that are inhibited by myelin, such as retinal ganglion cells, do not express p75 in the adult CNS [7]. Furthermore, although neurons isolated from p75 mutant mice are less inhibited by myelin inhibitors in vitro, a significant inhibitory activity remains [8], and axon regeneration after spinal cord injury is not enhanced in these mice [9].

Might there be a p75 homologue that is more widely expressed in the adult CNS that could substitute for p75 in this receptor complex? As p75 is a member of the extensive TNF receptor family, two research groups [10 and 11] independently set out to test this hypothesis. In fact, several members of the TNF receptor family have already been reported to be widely expressed in the CNS including TROY, CD40, DR6, Fn14, TNFR1 and TNFR2. Using either ELISA or cell binding assays, both groups [10 and 11] found that, of these proteins, only TROY could interact with NgR1 and that its interaction was nearly eight-fold stronger than the interaction between p75 and NgR1. And by immunoprecipitation, both groups demonstrated that TROY is able to form a complex with NgR1 and LINGO-1 [10 and 11].

So is TROY a functional homolog of p75? To find out, both groups [10 and 11] next tested whether expression of the TROY/NgR1/LINGO-1 complex in non-neural cells can activate RhoA in response to myelin-associated inhibitor factor binding. They found strong RhoA activation, confirming that TROY can substitute for p75 in the RhoA signaling pathway.

In elegant experiments, both groups [10 and 11] also showed that TROY regulates neurite outgrowth of dorsal root ganglion neurons and cerebellar granule neurons grown on myelin, OMgp and Nogo-66 in vitro. Increasing neuronal TROY levels by expression of full-length TROY from a viral vector enhanced myelin inhibitor-mediated inhibition of neurite outgrowth, whereas expression of a truncated dominant negative TROY (DN-TROY) significantly lessened this inhibition. Moreover, interfering with the interaction between myelin-associated inhibitors and the TROY/NgR1/LINGO-1 complex by addition of soluble TROY fusion proteins to the culture media greatly promoted neurite outgrowth.

Taken together, these results indicate that TROY transduces myelin-mediated inhibition of neurite outgrowth, provide evidence that TROY can functionally replace p75 in the NgR1 complex, and add TROY to the rapidly growing list of proteins that mediate axon inhibition by activating RhoA (Figure 1).


Display Full Size version of this image (49K)

Figure 1. TROY is a novel NgR1 coreceptor and mediates CNS myelin mediated axon outgrowth inhibition.

The myelin associated inhibitor factors Nogo-66, MAG and OMgp block regeneration of axons by binding to a shared receptor NgR1. In addition to an interaction between NgR1, LINGO-1 and p75, a newly identified receptor complex consisting of NgR1, LINGO-1 and TROY can transduce signaling upon binding of myelin associated inhibitor factor to NgR1, leading to RhoA activation and axon outgrowth inhibition. Blocking the formation of these receptor complexes by addition of dominant negative (DN) forms of either p75 or TROY (DN-p75, DN-TROY) antagonizes the axon outgrowth inhibitory effect of myelin associated inhibitor factor and myelin leading to greatly improved neurite outgrowth of dorsal root ganglion and cerebellar granule neurons in vitro.

Is TROY responsible for the residual inhibition of neurite outgrowth in response to myelin inhibitors observed in neurons lacking p75 function in vitro? To address this question, Park et al. [10] very elegantly demonstrated that dorsal root ganglion neurons can be subdivided into p75-expressing and non-expressing neurons, which can be distinguished by staining with isolectin B4 (IB4). Dorsal root ganglion neurons expressing IB4 do not express p75 and vice versa. Both populations are responsive to myelin-associated inhibitors, indicating that p75 is not essential for myelin-mediated inhibition of neurite outgrowth. But DN-TROY dramatically enhances neurite outgrowth from both IB4+ and IB4– dorsal root ganglion neurons cultured on myelin inhibitor substrates, and IB4– neurons isolated from p75-deficient mice still show a partial response to myelin inhibitors [10].

Interestingly, however, even in the absence of both TROY and p75 function, neurite lengths were still shorter than control, suggesting the possibility that additional, p75/TROY-independent, signaling pathways may yet exist that transduce myelin-mediated inhibition of neurite outgrowth. Nevertheless, introduction of full length TROY (FL-TROY) enhanced responsiveness of p75-deficient IB4– neurons, further substantiating that TROY is a functional homolog of p75 in transducing the inhibitory signals of myelin-associated inhibitors [10].

To demonstrate conclusively that TROY has a role in transducing inhibition of neurite outgrowth by myelin inhibitors, Shao et al.[11] generated TROY-deficient mice. These mice are viable and exhibit no obvious anomalies. As expected, dorsal root ganglion neurons isolated from TROY-deficient mice were less sensitive to myelin-associated neurite outgrowth inhibitors, but they were only able to completely overcome inhibition at low concentrations of Nogo-66, OMgp and myelin. Presumably much of the residual neurite outgrowth inhibition at higher myelin inhibitor concentrations is mediated by p75. From these exciting new results it can be concluded that TROY plays an important role in mediating the inhibitory effect of myelin inhibitors via the TROY/NgR1/LINGO-1 complex, and can functionally replace p75.

What implications does the discovery of this new receptor complex have for promoting CNS regeneration in vivo? The functional redundancy of p75 and TROY might very well explain the lack of improved regeneration after spinal cord injury in p75 knockout mice [9]. Thus, whether TROY-deficient mice will display enhanced regeneration is now a very exciting question. As p75 is widely upregulated after injury, however, p75 might actually be functionally redundant in TROY knockout animals [12]. This issue will no doubt soon be addressed by the generation of TROY/p75 double-knockout mice. That these mice will display enhanced regeneration is far from certain, as local administration of a p75 dominant-negative (DN-p75), which also blocks TROY, does not improve regeneration within the injured spinal cord [9 and 10]. Not only is it possible that additional TNF-receptor family members are involved, but also other NgR and LINGO protein family members (NgR2, NgR3, LINGO-2, LINGO-3 and LINGO-4) may play a role in myelin induced axon outgrowth inhibition [13, 14, 15, 16, 17 and 18]. Furthermore, additional myelin inhibitory proteins, such as Sema4D and the amino-terminal domain of Nogo-A, as well as unidentified myelin-associated inhibitors and their receptors, might mediate axon outgrowth inhibition independent of NgR1 receptor complexes [19 and 20].

Despite all these unanswered questions, the identification of TROY is a major step forward in understanding how myelin inhibits axon regeneration, providing new insight into the signaling pathways involved in regenerative failure and new avenues for how repair may be stimulated.

http://www.sciencedirect.com/science...65392b05f71bfd