| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Molecular Neuroscience Group, Division of Medical Sciences, University of Birmingham, Birmingham, UK
1Correspondence: Molecular Neuroscience Group, Division of Medical Sciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK. E-mail: a.logan{at}bham.ac.uk
SPECIFIC AIMS
We added exogenous tumor necrosis factor 
converting enzyme, TNF- converting enzyme (TACE), which induces regulated intramembraneous proteolysis of p75NTR and ectodomain shedding of NgR, to cultured dorsal root ganglia neurons (DRGN) to establish whether this strategy disinhibits DRGN neurite outgrowth and branching in the presence of inhibitory central nervous system (CNS) myelin extracts.
PRINCIPAL FINDINGS
1. TACE induces RIP of p75NTR
The failure of CNS axons to regenerate has been attributed to a combined effect of the lack of neurotrophins to support neuron survival and growth, the formation of a glial scar that is inhibitory to regenerating axons, and the presence of potent CNS myelin-derived inhibitory molecules in the distal nerve stump. Signaling of growth cone collapse by CNS myelin-derived inhibitory ligands such as myelin-associated glycoprotein (MAG), oligodendrocyte-derived myelin glycoprotein (OMgp), and Nogo-A is mediated through their binding to the Nogo receptor (NgR). NgR complexes with p75NTR/TROY and LINGO-1, and signals inhibition of axon growth cone advance, leading to sequential ROCK/LIM kinase/cofilin-mediated actin filament depolymerization and growth cone collapse. This is mediated through both a Rho-A- and an NgR/Ca2+-dependent phosphorylation of epidermal growth factor receptor (EGFR) -mediated pathway by an unknown mechanism.
Shedding of the extracellular domain (ECD) of membrane-anchored proteins is a post-translational mechanism for regulating receptor function. ECD shedding is essential for proper signaling of both EGFR ligands and notch-mediated lateral inhibition, and for both limiting inflammatory reactions and regulating axonal guidance. "Sheddases" proteolytically cleave membrane-bound proteins to release their ECD and include tumor necrosis factor- converting enzyme (TACE), which releases tumor necrosis factor (TNF-) from cells by liberating the ECD of membrane-bound pro-TNF. TACE also cleaves the ECD of other receptors; for example, ECD shedding of p75NTR initiates regulated intramembraneous proteolysis (RIP) by triggering 
-secretase cleavage of the intracellular domain (ICD). We therefore exogenously added TACE to DRGN cultures to initiate RIP of p75NTR and ECD shedding of NgR in an attempt to block the Rho-A- and EGFR-mediated inhibitory signaling pathways to promote fibroblast growth factor-2 (FGF2) -stimulated DRGN neurite outgrowth in the presence of CNS myelin.
The addition of FGF2 alone to cultured DRGN was able to initiate limited RIP of p75NTR (Fig. 1
) However, the addition of TACE and FGF2 initiated significantly greater levels of RIP of p75NTR, leading to the appearance of a 55 kDa (p75ECD) fragment, which was shed into the culture media, and the generation of an intracellular 25 kDa (p75ICD) fragment (Fig 1B
). RIP of p75NTR coincided with a significant attenuation of Rho-A activation and EGFR phosphorylation. RIP of p75NTR was blocked by the addition of tissue inhibitor of metalloprotease-3 (TIMP3), a potent inhibitor of TACE activation, while levels of activated Rho-A and phosphorylated EGFR were unaffected (Fig. 1)
.
|
2. TACE also cleaves the ECD of NgR
It has been reported that TACE is also able to cleave the ECD of NgR. Since NgR is the ligand binding receptor for CNS myelin-derived inhibitory ligands, we determined whether NgRECD shedding ocurred in our system. Significant NgRECD shedding into the media was observed in lysates from DRGN cultures treated with TACE and FGF2 while addition of TIMP-3 blocked NgRECD shedding.
3. No fragmentation of TROY
In the absence of p75NTR, a related receptor, TROY, functions to stimulate growth cone collapse. We therefore determined whether TROY is also subject to fragmentation in our culture system. We failed to detect fragmentation of TROY in either cell lysates, or the culture media. Our results suggest that TROY may not be the subject of fragmentation by TACE. However, it is quite possible that other sheddases may fragment TROY in a manner similar to p75NTR.
4. RIP of p75NTR significantly enhances FGF2-stimulated DRGN neurite outgrowth in the presence of inhibitory CNS myelin extracts
Immunocytochemistry, followed by image analysis of DRGN cultured in the absence of CNS myelin, showed significant DRGN neurite outgrowth when FGF2 alone was added. The addition of either FGF2 plus PMA (a known activator of TACE) or FGF2 plus TACE plus TIMP3 did not affect neurite outgrowth. The addition of either TACE alone or PMA alone did not promote neurite outgrowth (Fig. 2
). After the addition of a predetermined concentration of CNS myelin protein extract, neurite outgrowth was completely blocked in the presence of FGF2 (Fig. 2)
. With the addition of CNS myelin ligands to cultures, TACE significantly disinhibited FGF2-stimulated DRGN neurite outgrowth compared to that seen 1) without CNS myelin; 2) with TACE alone; and 3) with PMA (P<0.0001, Fig. 2
). Addition of TIMP3 blocked TACE activity in FGF-2-stimulated DRG cultures and restored the inhibition of DRGN neurite outgrowth mediated by CNS myelin (Fig. 2)
.
|
5. Addition of TACE significantly enhances DRGN neurite branching and correlates with microtubule-associated protein-1B (MAP1B) levels
TACE added to FGF2-stimulated DRGN, cultured in the presence of CNS myelin, increased the number of DRGN primary neurites and their branching frequency. After treatment with FGF2 without CNS myelin, the majority of DRGN had 3–4 primary neurites, each with 1–4 branches. After the addition of TACE and FGF2 in the presence of CNS myelin, the majority of DRGN grew >5 primary neurites, each with >10 branches. Treatment of DRGN with PMA plus FGF2 similarly increased the numbers of primary neurites and their branches compared to DRGN treated with PMA alone. TIMP3 restored the inhibitory potential of CNS myelin extracts in the presence of TACE and blocked neurite outgrowth and branching in FGF2-stimulated cultured DRGN.
Microtubule-associated protein-1B (MAP1B), a member of the MAP family, is expressed in regenerating axons and their growth cones, and promotes branching of DRGN neurites. When lysates of DRG cultures were probed for MAP1B, there was a significant up-regulation of MAP1B protein after treatment with TACE plus FGF2 in both the presence and absence of CNS myelin. MAP1B was also up-regulated in DRGN treated with PMA plus FGF2 in the presence of CNS myelin compared to PMA alone, but MAP1B levels were significantly less than those observed with TACE plus FGF2 (P<0.001 TACE+FGF2 vs. PMA+FGF2). In TIMP3-treated cultures, levels of MAP1B were similar to those in control DRGN.
CONCLUSIONS AND SIGNIFICANCE
The failure of CNS axons to regenerate is a significant problem, which has devastating effects on affected patients. Many strategies focused on neurotrophic factor treatments have been used to promote CNS axon growth after injury but have so far yielded disappointing results. Disinhibition of CNS axon regeneration may be induced by several experimental approaches. For example, blocking the activation of Rho-A and ROCK with either the antagonists C3 transferase or Y-27632 enhances axonal outgrowth on myelin substrates in vitro and in vivo. However, the modes of delivery of these antagonists may determine their potency, since C3 was not effective in all in vivo studies. Single gene ligand knockout (e.g., of Nogo-A) does not promote CNS axon regeneration probably because multiple alternative ligands remain active. We have already shown that against Rho-A, siRNA, p75NTR and NgR disinhibit FGF2-stimulated DRGN neurite outgrowth in the presence of CNS myelin ligands. Similarly, inhibition of NgR, either using a dominant negative mutant or after inhibiting NgR transgenically with a soluble function-blocking NgR fragment, enhances axon regeneration.
Our current study utilizing exogenous addition of TACE to initiate RIP of p75NTR (Fig. 3
1)and NgR ectodomain shedding (Fig. 3
2) represents yet another potential mechanism to disinhibit CNS axon regeneration. Exogenous addition of TACE induces RIP of p75NTR and fragments NgR, which coincides with a failure of activation of downstream Rho-A (Fig. 3
4) and reduced phosphorylation of EGFR, thereby paralyzing growth cone collapse mediated by CNS myelin-derived inhibitory ligand binding (Fig. 3
3). Blocking fragmentation of p75NTR and NgR with specific inhibitors restores the inhibitory potential of CNS myelin, suggesting that fragmentation of p75NTR and NgR may be key to the success of CNS axon regeneration. The advantage of this approach is that TACE may be easy to administer therapeutically. ECD shedding of NgR would contribute to the overall success of this approach, since TROY may replace the fragmented p75NTR.
|
TACE treatment almost completely fragmented p75NTR in DRGN cultures, but there was no neurite outgrowth in the presence of CNS myelin unless FGF2 was added. Therefore, neurons require stimulation into an "activated growth state" through, for example, either NTF, or cAMP administration, to drive axon growth and promote neuron survival. The significant enhancement in NTF-stimulated neurite outgrowth in disinhibited DRGN in the presence of CNS myelin suggests that neurite outgrowth is held in check in nondisinhibited DRGN. Since TROY and NgR were not completely fragmented in our experiments, they are still likely to be active in mediating growth cone collapse, particularly in a proportion of DRGN, which do not express full-length p75NTR. Whether p75NTR and TROY are differentially expressed in different DRGN remains to be investigated.
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-5339fje
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
