Glia-derived axonal growth inhibitory proteins limit functional repair following damage
to the adult central nervous system (CNS). Nogo proteins, myelin-associated glycoprotein
(MAG), oligodendrocyte myelin glycoprotein (OMgp) and B lymphocyte stimulator (BLyS),
are 4 inhibitors that commonly interact with the neuronal receptor, Nogo receptor-1
(NgR1), leading to inhibition of axonal growth. Here, we demonstrate that lateral
olfactory tract usher substance (LOTUS) binds to NgR1 and blocks the binding of all
four ligands to NgR1, resulting in the suppression of axonal growth inhibition induced
by these NgR1 ligands. LOTUS allows neurons to overcome NgR1-mediated axonal growth
inhibition, raising the possibility that LOTUS may be useful in future therapeutic
approaches as an endogenous potent inhibitor of NgR1 for promoting neuronal regeneration.
Damaged axons in the adult CNS are unable to regrow to their original targets. This
inability has been attributed to the non-permissive CNS environment that includes
myelin-associated inhibitory molecules such as Nogo proteins, myelin-associated glycoprotein
(MAG) and oligodendrocyte myelin glycoprotein (OMgp) (Schwab, 2010). These axonal
growth inhibitors commonly bind to Nogo receptor-1 (NgR1) which is expressed in many
types of CNS neurons and their axons (Schwab, 2010). B lymphocyte stimulator (BLyS)
which is expressed in astrocytes (Krumbholz et al., 2005) has also been identified
as a functional ligand for NgR1 (Schwab, 2010). The interaction of these four glial
components with NgR1 induces growth cone collapse and neurite outgrowth inhibition,
which limits the capability of injured neurons to be functionally restored in the
CNS (Schwab, 2010). Numerous studies have indicated that inhibition of NgR1 or its
ligands improves histological and behavioral recovery after CNS lesion (Schwab, 2010).
Moreover, triple deletion of Nogo, MAG and OMgp results in a higher degree of histological
and behavioral regeneration of injured CNS axons, as compared to single deletion of
Nogo (Schwab, 2010). This report suggests that suppression of multiple NgR1 ligands
is more effective in promoting neuronal regeneration of damaged axons in the CNS.
LOTUS antagonizes Nogo, MAG, OMgp and BLyS- activated NgR1: We identified lateral
olfactory tract usher substance (LOTUS) as a novel protein which contributes to axonal
bundle formation in lateral olfactory tract development by antagonizing NgR1 activation
by Nogo (Sato et al., 2011). We further examined whether LOTUS exerts a similar NgR1
antagonism with regards to MAG, OMgp and BLyS. The following observations were made:
(1) Overexpression of LOTUS with NgR1 in COS7 cells blocked the binding of these three
NgR1 ligands to NgR1. (2) In cultured dorsal root ganglion neurons where endogenous
LOTUS is only weakly expressed, LOTUS overexpression suppressed the growth cone collapse
and neurite outgrowth inhibition normally induced by these NgR1 ligands. (3) In cultured
olfactory bulb neurons which endogenously express LOTUS, LOTUS suppressed the growth
cone collapse normally induced by NgR1 ligands. Conversely, growth cone collapse was
induced by NgR1 ligands in lotus-deficient mice (Kurihara et al., 2014). Collectively,
our data suggest that LOTUS suppresses NgR1-mediated axonal growth inhibition by blocking
the interaction of NgR1 with its four ligands (
Figure 1
).
Figure 1
Schematic drawing of molecular mechanism associated with Nogo signaling.
NogoA, myelin-associated glycoprotein (MAG) and oligodendrocyte myelin glycoprotein
(OMgp) are axonal growth inhibitors derived from oligodendrocytes. B lymphocyte stimulator
(BLyS) is also an axonal growth inhibitor derived from astrocytes. These four glial
inhibitors bind to neuronal receptor Nogo receptor-1 (NgR1), which activates the intracellular
molecule RhoA via the NgR1 co-receptors, leucine-rich repeat and immunoglobulin domain-containing
Nogo receptor-interacting protein 1 (LINGO-1) and either the 75-kDa neurotrophin re-ceptor
(p75NTR) or tumor necrosis factor receptor superfamily member 19 (TROY), resulting
in induction of growth cone collapse and neurite outgrowth inhibition. Lateral olfactory
tractusher substance (LOTUS) interacts with NgR1 and interrupts the binding of Nogo66,
which is the functional ectodomain of NogoA that binds to NgR1, thereby inducing axonal
growth inhibition. Similarly, LOTUS interrupts the binding of MAG, OMgp and BLyS to
NgR1 and in this way, suppresses all of the four NgR1 ligands-induced axonal growth
inhibition.
NogoA extracellular peptide residues 1–40 (NEP1–40), leucine-rich glioma inactivated
1 and soluble NgR1 peptide can also interrupt the interaction of NgR1 with its ligand(s)
(Schwab, 2010). NEP1–40 has been shown to specifically block the binding of Nogo66
to NgR1, where Nogo66 is the functional domain of NogoA that binds to NgR1 and induces
axonal growth inhibition. NEP1–40 does not block the binding of MAG or OMgp. leucine-rich
glioma inactivated 1 has been shown to interfere with Nogo66-binding to NgR1. It remains
unknown whether NEP1–40 interrupts BLyS-binding to NgR1 and whether leucine-rich glioma
inactivated 1 interrupts the binding of MAG, OMgp or BLyS to NgR1. However, LOTUS
does inhibit the interaction of NgR1 with all of the four ligands (Sato et al., 2011;
Kurihara et al., 2014). It has been known that Nogo66 competes with MAG and OMgp for
the binding to NgR1, indicating that the Nogo66 binding site on NgR1 overlaps those
of MAG and OMgp. From these studies, it is suspected that a single soluble NgR1 peptide
cannot simultaneously bind to Nogo66, MAG and OMgp. LOTUS may have the benefit of
simultaneously antagonizing NgR1 activation by all four ligands although whether LOTUS
antagonizes NgR1 in equal stoichiometry is still undetermined. The overlap of Nogo66,
MAG or OMgp binding sites on NgR1 raises the possibility that LOTUS may interfere
with the binding of Nogo66, MAG and OMgp to NgR1 in a competitive manner. It is unknown
which region of NgR1 interacts with BLyS and whether BLyS competes with the other
ligands for the binding to NgR1. It is also possible that LOTUS may exert its inhibitory
effect via an allosteric mechanism that only interferes with BLyS-binding toNgR1,
or this allosteric inhibition may extend to all four ligands. To elucidate the molecular
mechanism by which LOTUS antagonizes NgR1, further investigations in structural biology
are required.
Identification of a novel NgR1 ligand, chondroitin sulfate proteoglycans: Recently,
chondroitin sulfate proteoglycans, which are abundant in reactive astrocytes derived
from glial scars, have been identified as a functional ligand for NgR1 and Nogo receptor-3,
an NgR1 homologue (Dickendesher et al., 2012). Genetic deletion of chondroitin sulfate-synthesizing
enzyme (Takeuchi et al., 2013) or NgR1 (Schwab, 2010) promotes the ability of damaged
CNS axons to be re-elongated, suggesting that the binding of chondroitin sulfate proteoglycans
to NgR1 may be involved in the failure of damaged CNS axons to regenerate. Moreover,
double administration of antibodies neutralizing NogoA and chondroitinase ABC, which
catalyzes the degradation of the glycosaminoglycan chains on the chondroitin sulfate
proteoglycans, is more effective in enhancing the histological and behavioral recovery
after CNS injuries, compared with single administration (Zhao et al., 2013). This
report suggests that concurrent inhibition of NgR1 ligands may be more effective in
overcoming the failure of damaged CNS axons to regenerate. To clarify whether LOTUS
is also able to inhibit chondroitin sulfate proteoglycans-mediated activation of NgR1,
further investigation is required to ascertain whether LOTUS can also suppress chondroitin
sulfate proteoglycans-binding to NgR1 and chondroitin sulfate proteoglycans-induced
axonal growth inhibition as shown for the other four ligands. We previously revealed
that the carboxyl-terminal region of LOTUS antagonizes NgR1 activation by Nogo66 (Kurihara
et al., 2012). It will be interesting to explore whether this region would exert similar
antagonistic effects on NgR1 with regards to MAG, OMgp and BLyS and which region of
LOTUS is necessary and sufficient to exert the antagonistic activity on NgR1 with
regards to all of the four ligands.
Future perspectives and challenges: It has often been established in animal models
of spinal cord injury that negatively regulating NgR1 or its ligands enhances the
histological and behavioral repair after CNS lesion in vivo. In addition, these molecules
are associated with the neurological disorder multiple sclerosis. NogoA, BLyS and
NgR1 are up-regulated in the lesioned brains of multiple sclerosis patients (Krumbholz
et al., 2005; Satoh et al., 2005). Antibodies neutralizing NogoA and genetic deletion
of NgR1 alleviate the symptoms of experimental autoimmune encephalomyelitis, an animal
model of multiple sclerosis(Petratos et al., 2012). Consequently, Nogo or BLyS function
through binding to NgR1 may impede relief of multiple sclerosis symptoms. We have
recently found that LOTUS is down-regulated in the cerebral spinal fluid of multiple
sclerosis patients (Takahashi et al., 2014). This finding suggests that decrease of
LOTUS concentration is associated with the pathological conditions of multiple sclerosis.
The binding of Nogo, MAG, OMgp and BLyS to NgR1 transduces signals through the NgR1
co-receptors, leucine-rich repeat and immunoglobulin domain-containing Nogo receptor-interacting
protein 1 (LINGO-1) and either the 75-kDa neurotrophin receptor (p75NTR) or tumor
necrosis factor receptor superfamily member 19, to intracellular molecules, RhoA and
its effect or, Rho-associated, coiled-coil containing protein kinase (Schwab, 2010).
Soluble LINGO-1 peptides, a RhoA inactivator and a coiled-coil containing protein
kinase inhibitor also improve the histological and behavioral recovery following CNS
lesion (Schwab, 2010). However, the interaction of NgR1 with its ligands can also
mediate signal transduction through p75NTR and tumor necrosis factor receptor superfamily
member 19. Furthermore, NgR1 ligands also activate protein kinase C independently
of RhoA-mediated signaling, which induces growth cone collapse and neurite outgrowth
inhibition (Hasegawa et al., 2004). Therefore, it is likely that these inhibitors
may be not sufficient to completely inhibit the signal transduction induced by the
interaction of NgR1 with its ligands. LOTUS can completely block the interaction of
NgR1 with all four ligands and therefore LOTUS can completely shut down NgR1-mediated
axonal growth inhibition. Therefore, further studies are required to elucidate whether
therapeutic approaches using LOTUS (for example, implantation of LOTUS-overexpressing
cells and/or administration of LOTUS recombinant protein) can promote neuronal regeneration
in neurological disorders such as spinal cord injury and multiple sclerosis.
This work was supported by a grant-in-aid from the Ministry of Education, Culture,
Sports, Science and Technology of Japan and by grants for Research and Development
project of Yokohama City University.