*Department of Molecular Biology and Biochemistry, Okayama University Graduate School of Medicine, Dentistry and
Pharmaceutical Sciences, Okayama, Japan
Vessel Wall Biology, Institute for Experimental Medical Sciences, Lund University, Lund, Sweden
The axons of myelinated nerves in the adult CNS are
subdivided into several distinct structural and functional
domains, each of which differs in its molecular composition
(e.g., nodes, paranodes, juxtaparanodes) (Peles and Salzer
2000). In recent years, the molecular composition of each of
these domains has begun to emerge. The nodes of Ranvier
are highly enriched with voltage-gated sodium channels
(Nav) and they are exposed to the extracellular environment,
thereby enabling action potential regeneration and saltatory
conduction. Several cell adhesion molecules have been
defined to specifically localize on the axonal membrane in
the nodes (for review, see Poliak and Peles 2003; Salzer
2003). In comparison to such cell adhesion molecules and
ion channels, the presence of extracellular constituents at the
nodes has just begun to be appreciated. Several secreted
proteoglycans (e.g. versican, phosphacan, brevican) and
matrix proteins [e.g. tenascin-R (TN-R), Bral1] were found
to localize at the nodes of Ranvier (Xiao et al. 1997; Weber
et al. 1999; Hedstrom et al. 2007; Oohashi and Bekku
2007).
Brevican and versican V2 are CNS-specific member of a
hyaluronan-binding chondroitin sulfate proteoglycan (CSPG)
family, which are also called ‘lecticans’ because of the
presence of a homologous C-type lectin motif in their C-
terminal domain. Via this domain, lecticans can bind to
various extracellular matrix (ECM) proteins such as tenasc-
ins. Yamaguchi has proposed that the hyaluronan-lectican-
TN-R complex constitutes the core assembly of the adult
brain ECM (Yamaguchi 2000). In fact, electron microscopy
demonstrated the ability of tenascins to act as cross-linkers
between lectican-hyaluronan complexes (Lundell et al.
2004). In addition, in the macromolecular organization of
perineuronal nets (PNNs), TN-R appears to be an
Received October 16, 2008; revised manuscript received December 15,
2008; accepted December 16, 2008.
Address correspondence and reprint requests to Toshitaka Oohashi,
Department of Molecular Biology and Biochemistry, Okayama Univer-
sity Graduate School of Medicine, Dentistry and Pharmaceutical
Sciences, 2-5-1 Shikata-cho, Okayama 700-8558, Japan.
E-mail: oohashi@cc.okayama-u.ac.jp
Abbreviations used: CSPG, chondroitin sulfate proteoglycan; ECM,
extracellular matrix; Nav, voltage-gated sodium channels; PB, phosphate
buffer; PBS, phosphate-buffered saline; PNN, perineuronal net; TN-R,
tenascin-R.
Abstract
Brevican is known to be an abundant extracellular matrix
component in the adult brain and a structural constituent of
perineuronal nets. We herein show that brevican, tenascin-
R (TN-R) and phosphacan are present at the nodes of
Ranvier on myelinated axons with a particularly large
diameter in the central nervous system. A brevican defi-
ciency resulted in a reorganization of the nodal matrices,
which was characterized by the shift of TN-R, and con-
comitantly phosphacan, from an axonal diameter-dependent
association with nodes to an axonal diameter independent
association. Supported by the co-immunoprecipitation re-
sults, these observations indicate that the presence of TN-R
and phosphacan at nodes is normally brevican-dependent,
while in the absence of brevican these molecules can also
be recruited by versican V2. The versican V2 and Bral1
distribution was not affected, thus indicating a brevican-
independent role of these two molecules for establishing
hyaluronan-binding matrices at the nodes. Our results re-
vealed that brevican plays a crucial role in determining the
specialization of the hyaluronan-binding nodal matrix
assemblies in large diameter nodes.
Keywords: action potential, extracellular matrix, node of
Ranvier, proteoglycan, scaffolding proteins, tenascin-R.
J. Neurochem. (2009) 108, 1266–1276.
JOURNAL OF NEUROCHEMISTRY | 2009 | 108 | 1266–1276 doi: 10.1111/j.1471-4159.2009.05873.x
1266 Journal Compilation  2009 International Society for Neurochemistry, J. Neurochem. (2009) 108, 1266–1276
 2009 The Authors

important molecular cross-linker. Phosphacan is a different,
non-hyaluronan-binding type of CSPG, which has been
reported to show remarkable colocalization and high affinity
binding to TN-R at the nodes of Ranvier in the optic nerve
(Xiao et al. 1997). In comparison to wild type mice in the
optic nerves of TN-R knockout mice nodal immunoreactivity
of phosphacan was weakened and diffusely distributed
(Weber et al. 1999). Bral1 is a brain specific link protein, a
family of proteins, which bind to hyaluronan and to
the hyaluronan binding part of lecticans stabilizing this
interaction.
Thus, these molecules together with hyaluronan are able
to assemble an organized, PNNs like matrix. Their
collective presence indicates that, in addition to PNNs, a
second kind of dense organized hyaluronan and chondroi-
tin sulfate proteoglycans (CSPG) containing matrix exists
at the nodes of Ranvier. Versican V2 and Bral1 have been
observed at almost all nodes of myelinated nerves within
the CNS (Oohashi et al. 2002), thus indicating a possibly
ubiquitous presence of nodal matrices with a macromo-
lecular organization similar to PNNs. However, there is a
current lack of knowledge regarding the molecular details
and physiological roles of nodal matrices in the CNS. We
herein report that the nodes of the axon fibers of a
particularly large diameter have a more elaborate ECM
assembly than the nodes of smaller axons. Brevican plays
a pivotal role in determining the specialization of the
hyaluronan-binding nodal matrix assemblies in large
diameter nodes.
Materials and methods
Animals
The genotypes of brevican-deficient mice were determined by PCR
using allele-specific primer sets as previously described (Brak-
ebusch et al. 2002). Age matched adult mice (five months to one
year of age) were used in this study. All animal experiments
and animal care were carried out in accordance with the guidelines
of the Animal Care and Experimentation Committee of Okayama
University.
Antibodies
The rabbit polyclonal antibodies against Bral1, brevican (Ab1058),
neurocan, aggrecan (Millipore, Billerica, MA, USA), and the mouse
monoclonal antibody against Caspr have been described previously
(Zhou et al. 2001; Oohashi et al. 2002; Bekku et al. 2003). The
following antibodies were purchased and used: goat anti-TN-R
polyclonal antibody (1 : 50 for immunohistochemistry and 1 : 100
for immunoblotting; Santa Cruz Biotechnology, Santa Cruz, CA,
USA), rabbit anti-versican GAG-a polyclonal antibody (1 : 50 for
immunohistochemistry and 1 : 100 for immunoblotting; Millipore),
mouse anti-panNav monoclonal antibody (1 : 100; Sigma-Aldrich,
St. Louis, MO, USA), mouse anti-phosphacan antibody (6B4)
(1 : 50 for immunohistochemistry and 1 : 100 for immunoblotting;
Seikagaku, Tokyo, Japan).
Immunohistochemistry
The mice were deeply anesthetized with diethyl ether and then were
transcardially perfused with 50 mM phosphate-buffered saline
(PBS, pH 7.4) followed by 2% paraformaldehyde in PBS. The
brains and optic nerves were removed and cryoprotected with 30%
sucrose in PBS at 4C overnight. For some antibodies, the brains
were directly frozen in OCT compound (SAKURA, Tokyo, Japan)
and then the sectioned samples were immediately fixed in acetone
for 10 min at )20C. The cryostat sections (10 lm) were cut and
collected on MAS coated microscope glass slides (Matsunami,
Osaka, Japan) and then were allowed to air dry. The sections were
dipped in methanol containing 6% H
2
O
2
for 20 min, permeabilized
for 1 h in 0.1 M phosphate buffer (PB) containing 0.3% Triton X-
100 and 10% goat serum, pH 7.4, or containing 0.3% Triton X-100
and 5% skim milk, pH 7.4. For the double-labeling experiments, the
sections were incubated overnight at 4C with primary antibodies
diluted to the appropriate concentration in PB containing 0.3%
Triton X-100 and 10% goat serum, PB containing 0.3% Triton X-
100 and 5% skim milk or M.O.M. immunodetection kit (Vector
Laboratories, Burlingame, CA, USA). The sections were then
washed, and incubated with the fluorescent-labeled secondary
antibodies for 2 h at 20C. The secondary antibodies used in this
study were Alexa Fluor 488 (1 : 200) or 555 (1 : 500) donkey anti-
goat IgG (Invitrogen, Carlsbad, CA, USA), Alexa Fluor 488 goat or
chicken anti-rabbit IgG (1 : 500; Invitrogen), Alexa Fluor 488
chicken anti-mouse IgG (1 : 500; Invitrogen), Alexa Fluor 488 goat
anti-mouse IgM (1 : 500; Invitrogen), Alexa Fluor 555 goat anti-
mouse IgG (1 : 500; Invitrogen), Cy3-labelled goat anti-mouse IgG
(1 : 500; GE Healthcare, Buckinghamshire, UK), and Cy3-conju-
gated goat anti-mouse IgM (1 : 500; Jackson ImmunoResearch,
West Grove, PA, USA). The images were captured using either a
laser-scanning microscope LSM510 (Carl Zeiss, Oberkochen,
Germany) or an Olympus BX50 light microscope (Olympus, Tokyo,
Japan) equipped with an AxioCam digital camera (Carl Zeiss),
followed by image manipulation with Adobe Photoshop (Adobe
Systems, San Jose, CA, USA).
Measurement of the axon diameters
The method used to measure the axon diameter was adapted from
a previously described method (Devaux et al. 2003). Briefly,
transverse sections of the brains from three mice, from five to six
months of age, for each measurement were immunolabeled for
brevican, TN-R or phosphacan and panNav. Cy3-labelled goat
anti-mouse IgG or Alexa Fluor 555 goat anti-mouse IgG F(ab’)
2
fragment was used for each double immunostaining as the
secondary antibody for panNav. The nodes of the well aligned
facial nerve axons were less dense than those of the optic nerves,
which thus made the nodes of the facial nerve axons more suitable
to measure the nodal diameter. These images were captured with a
laser-scanning microscope LSM510 focused on the expression of
panNav. The diameter of each Nav channel cluster was measured
by a digital linear gauge perpendicularly-placed on the axon using
the LSM510 software program (Carl Zeiss). We measured the axon
diameter within the nodal region from one lateral surface to the
opposite lateral surface of the axon, which were recognized by
panNav. The measurements were always performed in a single
channel image of panNav in a blind manner, and each node was
then scored as TN-R-, brevican- or phosphacan-positive, or TN-R-,
 2009 The Authors
Journal Compilation  2009 International Society for Neurochemistry, J. Neurochem. (2009) 108, 1266–1276
Brevican at the nodes of Ranvier | 1267