BIG-2 Mediates Olfactory Axon
Convergence to Target Glomeruli
Tomomi Kaneko-Goto,1Sei-ichi Yoshihara,1,3Haruko Miyazaki,1,4and Yoshihiro Yoshihara1,2,*
1Laboratory for Neurobiology of Synapse, RIKEN Brain Science Institute, Saitama 351-0198, Japan
2Core Research for Evolutional Science and Technology (CREST), Japan Science and Technology Agency, Osaka 560-0082, Japan
3Present address: Department of Molecular Biology for Neural Systems, Research Institute for Frontier Medicine,
Nara Medical University, Nara 634-8521, Japan.
4Present address: Laboratory for Structural Neuropathology, RIKEN Brain Science Institute, Saitama 351-0198, Japan.
Olfactory sensory neurons expressing a given odor-
ant receptor converge axons onto a few topographi-
cally fixed glomeruli in the olfactory bulb, leading to
establishment of the odor map. Here, we report that
BIG-2/contactin-4, an axonal glycoprotein belonging
to the immunoglobulin superfamily, is expressed in
a subpopulation of mouse olfactory sensory neu-
rons. A mosaic pattern of glomerular arrangement
is observed with strongly BIG-2-positive, weakly
positive, and negative axon terminals in the olfactory
bulb, which is overlapping but not identical with
those ofKirrel2 and ephrin-A5.There is aclose corre-
lation between the BIG-2 expression level and the
odorant receptor choice in individual sensory neu-
rons. In BIG-2-deficient mice, olfactory sensory neu-
rons expressing a given odorant receptor frequently
innervate multiple glomeruli at ectopic locations.
These results suggest that BIG-2 is one of the axon
guidance molecules crucial for the formation and
maintenance of functional odor map in the olfactory
The olfactory system furnishes sophisticated molecular and
cellular mechanisms for reception, discrimination, and internal
representation of information carried by an immense number of
odor molecules in the external world. To achieve this feat, olfac-
tory sensory neurons (OSNs) in the olfactory epithelium (OE) are
destined to obey two basic principles during their unique differ-
for expression from a repertoire of ?1000 genes in mice, being
tuned to a range of odor ligands with shared structural features
that can bind and activate the expressed OR (Chess et al.,
1994; Malnic et al., 1999; Serizawa et al., 2003; Lewcock and
Reed, 2004; Shykind et al., 2004). Second, OSNs expressing
a given OR project and converge their axons onto a few topo-
graphically fixed glomeruli among ?2000 glomeruli spatially
arranged on the surface of the olfactory bulb (OB), resulting in
establishment of the ‘‘odor map’’ in OB (Vassar et al., 1994;
Ressler et al., 1994; Mombaerts et al., 1996; Mori et al., 1999,
Thus, we currently have a wealth of knowledge on the pattern
of olfactory axon wiring and the representation of odor informa-
tion on OB odor map. So far, several guidance systems have
been identified that play distinct roles in step-wise processes
including Cxcl12/Cxcr4, Robo/Slit, neuropilin/semaphorin, Eph/
ephrin, and Kirrel2/3 (Miyasaka et al., 2005, 2007; Schwarting
et al., 2000; Walz et al., 2002; Taniguchi et al., 2003; Imai et al.,
2006; Cutforth et al., 2003; Serizawa et al., 2006). However,
the precise molecular mechanisms underlying the formation
and maintenance of glomerular targeting still remain to be
BIG-2 (contactin-4) is a glycosylphosphatidylinositol (GPI)-
anchored axonal glycoprotein with six immunoglobulin (Ig)-like
domains and four fibronectin-type III (FnIII) repeats (Figure 1A),
belonging to the contactin subgroup of the Ig superfamily (Yosh-
ihara et al., 1995), also referred to contactin-4 (Zeng et al., 2002;
Hansford et al., 2003). BIG-2 shows a unique expression pattern
in neuron type- and developmental stage-specific manners in
various regions of the nervous system (Yoshihara et al., 1995).
In the present study, we investigated the expression of BIG-2 in
the mouse olfactory system and found the mosaic pattern of
BIG-2 protein expression among glomeruli. The expression level
Furthermore, in BIG-2-deficient mice, olfactory axons frequently
role of BIG-2 in the establishment of precise odor map in OB.
Expression of BIG-2 in a Subset of OSNs
In the course of immunohistochemical staining of mouse brain
sections with antibodies against various cell recognition and
axon guidance molecules, we noticed the strong expression of
BIG-2 protein in the olfactory nerve layer and glomerular layer
of OB. This preliminary finding prompted us to investigate de-
tailed localization and physiological function of BIG-2 in the
primary olfactory system.
In situ hybridization analysis of coronal sections of 6-week-old
mouse heads at different anteroposterior levels showed that
834 Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc.
in the vomeronasal organ (Figures 1B–1D). A higher magnifica-
tion of the OE section revealed that BIG-2 mRNA was present
cells (Figure 1E). BIG-2-negative OSNs (black asterisks in Fig-
ure 1E) were intermingled with BIG-2-positive OSNs (white
asterisks in Figure 1E) in OE.
nohistochemistry with anti-BIG-2 antibody. BIG-2 protein was
abundantly present in axons of OSNs but was almost absent
from their dendrites and cell bodies (Figure 1F), which is a char-
acteristic pattern of subcellular localization for most of GPI-
anchored neuronal proteins (Faivre-Sarrailh and Rougon, 1997).
BIG-2-positive axon bundles were observed throughout OE,
similar to NCAM-positive axons (Figures 1F–1H). A higher mag-
nification of triple-labeled axon bundles with antibodies against
BIG-2, NCAM, and OCAM revealed that BIG-2-positive fibers
(Figures 1I and 1L–1N) constitute a subpopulation, but not all,
of axon bundles of OSNs, which partially overlaps but is distinct
from a population of NCAM-positive fibers (Figures 1J, 1L, and
1N). The expression pattern of BIG-2 did not correspond to the
olfactory zones whose boundary (between zone 1 and 2) can
be evidently delineated by OCAM-negative and -positive axons
(Figures 1K, 1M, and 1N; Yoshihara et al., 1997). These results
indicate that BIG-2 is expressed by a subset of OSNs with no
relation to the zonal organization of OE.
Mosaic Pattern of BIG-2 Protein Expression
among OB Glomeruli
To investigate the glomerular targeting patterns of BIG-2-posi-
tive and -negative axons, we performed immunohistochemical
analysis of BIG-2 expression in OB sections from 6-week-old
mice. A parasagittal section containing the glomerular sheet
of the medial OB showed that individual glomeruli are labeled
at different intensities with anti-BIG-2 antibody. Strongly BIG-
2-positive glomeruli were intermingled with weakly BIG-2-
positive and BIG-2-negative glomeruli on the surface of OB,
whereas almost all the glomeruli displayed a similar level of
NCAM expression (Figures 2A–2C). The distribution pattern of
glomeruli positive for BIG-2 was completely different from
those for OCAM and neuropilin-1 (NP-1) (Figures 2D–2K), indi-
cating that BIG-2 expression does not relate to the dorsoventral
Figure 1. Expression of BIG-2 mRNA and Protein in a Subset of OSNs
(A) A schematic diagram depicting the structure of BIG-2 protein. BIG-2 consists of N-terminal signal peptide (SP), six Ig-like domains (Roman numerals), four
FnIII-like domains (Arabic numerals), and C-terminal GPI tail. S-S: a disulfide bond in each Ig-like domain.
a subset of OSNs. White asterisks: OSNs with a high level of BIG-2 mRNA expression. Black asterisks: OSNs devoid of BIG-2 mRNA.
(F–H)DoublelabelingofacoronalOEsectionwithantibodiesagainstBIG-2andNCAM.BIG-2 protein(F)isexpressedinOSNaxonsextending fromOEaswellas
NCAM (G). (H) A merged image showing BIG-2 (red) and NCAM (green).
(I–N) Triple labeling of OSN axon bundles projecting from OE to OB with antibodies against BIG-2, NCAM, and OCAM. BIG-2 protein (I) is present in a subset of
OSN axons that are overlapping but different from NCAM-positive (J) and OCAM-positive axons (K). (L–N) Merged images of BIG-2 (red), NCAM (green), and
OCAM (blue) signals.
Scale bars: (B–D) and (F–H) 1 mm, (E) 25 mm, (I–N) 100 mm.
BIG-2 Mediates Olfactory Axon Targeting
Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc. 835
axis defined by OCAM and the anteroposterior axis defined by
NP-1 (Yoshihara et al., 1997; Nagao et al., 2000; Imai et al.,
We next examined the overall distribution of BIG-2-positive
and -negative glomeruli in OB. Double-immunofluorescence la-
beling of OB coronal sections with anti-BIG-2 and anti-vesicular
glutamate transporter 2 (VGluT2) antibodies clearly delineated
BIG-2-positive and -negative glomeruli, revealing that the mo-
saic pattern of BIG-2 expression was observed at all antero-
posterior levels of OB (Figures 2L–2Q). A three-dimensionally
reconstructed OB map (Figures 2R–2U) showed a wide distribu-
tion of strongly BIG-2-positive glomeruli (Figures 2R–2U, red)
intermingled with weakly positive (Figures 2R–2U, green) and
negative glomeruli (Figures 2R–2U, blue). A careful observation,
however, revealed that BIG-2-positive glomeruli were somehow
clustered in several regions of OB: at the anterior tip, on the
dorsal surface, on the lateral surface, and in the ventroposterior
region. Also in the accessory olfactory bulb, BIG-2 displayed
a mosaic pattern of expression among glomeruli with variable
signal intensities (Figures 2P and 2Q). In addition, BIG-2 expres-
sion in a subpopulation of olfactory axons was already observed
during embryonic development (see Figures S1A–S1L available
It has been reported that homophilic adhesion molecules,
Kirrel2 and Kirrel3, and repulsive guidance partners, ephrin-A5
and EphA5, are expressed in complementary subsets of OSNs
respectively and show distinct and mosaic patterns of expres-
sion among OB glomeruli (Cutforth et al., 2003; Serizawa et al.,
2006). To clarify their relationship with BIG-2, we performed
triple-immunofluorescence labeling of OB sections with anti-
bodies against BIG-2, Kirrel2, and ephrin-A5. All the three
molecules showed mosaic staining of OB glomeruli (Figure 3).
BIG-2-positive glomeruli partly overlap with Kirrel2- or ephrin-
A5-positive glomeruli, but the overall distribution of positive glo-
meruli appeared different among the three molecules (Figures
3A–3D). Higher magnification revealed that individual glomeruli
express different levels of BIG-2, Kirrel2, and ephrin-A5 (Figures
3E–3H). We quantified signal intensities of BIG-2, Kirrel2, and
ephrin-A5 in glomeruli (n = 506 from three mice), plotted on scat-
ter graphs (Figures 3I and 3J) and calculated the correlation
coefficients (r values). The expression of BIG-2 in individual glo-
meruli showed no correlation with that of Kirrel2 (r = 0.12 ± 0.04;
Figure 3I). In contrast, there was a weak but significant positive
correlation between BIG-2 and ephrin-A5 (r = 0.28 ± 0.12;
Figure 3J). Taking it into account that ephrin-A5 expression
tends to be higher in OSNs with reduced neural activity (Seri-
zawa et al., 2006), it is likely that BIG-2 undergoes a similar
regulation. These results imply a combinatorial code of axon
guidance molecules that specifies the identities of individual
glomeruli in OB.
Figure 2. A MosaicPattern ofBIG-2 Expres-
sion among OB Glomeruli
(A–C) Double labeling of a glomerular sheet on the
medial side of a parasagittal OB section with anti-
bodies BIG-2and NCAM.StronglyBIG-2-positive,
weakly positive, and negative glomeruli are inter-
mingled on the OB surface (A), while NCAM is
presentevenlyinall theglomeruli (B).(C) A merged
image showing BIG-2 (red) and NCAM (green).
Anterior is to the left and dorsal is to the top.
(D–K) Triple labeling of OB coronal sections with
antibodies against BIG-2 (D and H), OCAM (E
and I), and NP-1 (F and J). (D–G) Low-power mi-
crographs of a representative OB section. (H–K)
Higher magnifications of the glomerular layer of
OB. (G and K) Merged images of BIG-2 (red),
OCAM (green), and NP-1 (blue) signals. A mosaic
expression pattern of BIG-2 is different from clus-
tered patterns of OCAM and NP-1.
(L–Q) Double labeling of OB coronal sections at
different anteroposterior levels with antibodies
against BIG-2 (red) and VGluT2 (green). VGluT2
signals in presynaptic terminals of OSNs clearly
delineate the glomerular layer of OB. A mosaic
pattern of BIG-2 expression among glomeruli is
observed at all the anteroposterior levels of OB.
(R–U) Three-dimensionally reconstructed glomer-
ular maps of strongly BIG-2-positive (red), weakly
positive (green), and negative (blue) glomeruli
viewed from dorsal (R), ventral (S), lateral (T), and
medial (U) sides.
Scale bars: (A–C) and (H–K) 200 mm, (D–G) and
(L–Q) 400 mm.
BIG-2 Mediates Olfactory Axon Targeting
836 Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc.
Correlation of BIG-2 Expression Level and OR Gene
Choice in OSNs
Based on the above results indicating the different levels of BIG-
2 expression on OSN axons and their terminations onto distinct
glomeruli, we hypothesized that BIG-2 expression may correlate
with OR gene choice by individual OSNs. To examine this possi-
bility, we used specific antibodies against four well-character-
ized ORs (MOR28, mOR-EG, OR-I7, and mOR256-17; Araneda
et al., 2000; Serizawa et al., 2000; Strotmann et al., 2004; Oka
et al., 2006) for double-immunofluorescence labeling with anti-
BIG-2 antibody on OB sections from 6-week-old mice.
MOR28 is expressed by OSNs in the ventral area of OE, which
project their axons to two relatively large glomeruli in the ventro-
labeled for BIG-2, the MOR28-positive glomeruli on both the
medial and lateral sides showed a markedly high level of BIG-2
expression (Figures 4A–4G and 4K). A similar result was repro-
ducibly obtained from six adult mice. In addition, a high level of
BIG-2 expression was detected in MOR28-positive glomeruli
even at postnatal day 7 (Figures S1M–S1R). These results
suggest that the MOR28-positive OSNs always express BIG-2
protein at a high level.
The analysis was next expanded to other three ORs (Figures
4H–4J and 4L–4N). Both the medial and lateral glomeruli positive
for mOR-EG in the dorsal region of OB showed a moderate level
of BIG-2 expression was relatively low in glomeruli positive for
mOR256-17 and OR-I7 on both the medial and lateral sides (Fig-
ures 4I, 4J, 4M, and 4N). To correlate the BIG-2 and OR expres-
sion in OSNs, we quantified the expression levels of BIG-2 in in-
glomeruli on the graph (Figure 4O). Consistently, MOR28 glo-
meruli were positive for BIG-2 very strongly, mOR-EG glomeruli
moderately, and mOR256-17 and OR-I7 very weakly. Intensities
of BIG-2 expression in individual OR-positive glomeruli were
almost the same between the medial and lateral sides of OB
(Figure 4P). These results demonstrate a close correlation be-
tween the BIG-2 expression level and the OR gene choice in
To further confirm the correlation between BIG-2 and OR ex-
pressions in OSNs, we used H-MOR28-ires-tauECFP transgenic
mice in which MOR28 and tauEYFP are ectopically expressed
under the control of H elements in a large population of OSNs
located in ventrolateral region of OE (Serizawa et al., 2003,
2006). In the H-MOR28-ires-tauECFP transgenic mice, MOR28-
positive olfactory axons innervated numerous ectopic glomeruli
located in OB, all of which showed a very high level of BIG-2 ex-
pression (Figure S2) similar to the endogenous MOR28 glomeruli
(Figure 4). This result clearly demonstrates that MOR28-positive
OSNs express the high level of BIG-2 consistently.
Neural Activity-Regulated Expression of BIG-2
Neural activity of OSNs regulates the expression levels of axon
guidance molecules, Kirrel2/Kirrel3, and ephrin-A5/EphA5 (Ser-
izawa et al., 2006). To examine a possible involvement of neural
activity in the regulation of BIG-2 expression, we carried out two
gene (Brunet et al., 1996). Because the CNGA2 gene is located
on the X chromosome, the heterozygous female mice show
mosaic expression of CNGA2 in OSNs through the process of
random X inactivation (Zhao and Reed, 2001). As a result, each
OR-defined glomerulus is segregated into two compartments
(or two glomeruli) that are innervated by either CNGA2-positive
or -negative OSNs (Zheng et al., 2000). We examined BIG-2,
Kirrel2, and ephrin-A5 expression in the CNGA2 mutant mice
at postnatal day 14 (Figures 5A–5E) and quantified the expres-
sion levels in individual glomeruli (Figures 5F–5H). As reported
previously (Serizawa et al., 2006), CNGA2-mediated neural ac-
tivity regulated the expression of Kirrel2 positively and ephrin-
A5 negatively (Figures 5B–5E, 5G, and 5H). BIG-2 expression
Figure 3. Overlapping but Distinctive Glo-
merular Expression of BIG-2, Kirrel2, and
(A–H) Triple labeling of OB coronal sections with
antibodies against BIG-2 (A and E), Kirrel2 (B and
F), and ephrin-A5 (C and G). (A–D) Low-power mi-
crographs of a representative OB section. (E–H)
Higher magnifications of the glomerular layer of
OB. (D and H) Merged images of BIG-2 (red),
Kirrel2 (green), and ephrin-A5 (blue) signals. Note
that individual glomeruli are labeled with distinct
combinations of BIG-2, Kirrel2, and ephrin-A5.
Scale bars: (A–D) 400 mm, (E–H) 200 mm.
(I and J) Scatter plots of staining intensities (arbi-
trary unit) for BIG-2 versus Kirrel2 (I) and BIG-2
versus ephrin-A5 (J) in glomeruli (n = 506). Signal
intensities of each glomerulus triple-labeled for
anti-BIG-2, anti-Kirrel2, and anti-ephrin-A5 anti-
bodies were measured using the NIH ImageJ pro-
gram. ‘‘r’’ values indicate correlation coefficients
(average ± SEM from three mice).
BIG-2 Mediates Olfactory Axon Targeting
Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc. 837
Figure 4. Correlation of BIG-2 Expression and OR Gene Choice in OSNs
(A–F) Double labeling of OB coronal sections with antibodies against BIG-2 (A and D) and MOR28 (B and E). Blue signals (B and E) show counterstaining of cell
nuclei with DAPI. (C and F) Merged images of BIG-2 (green) and MOR28 (red) signals. Arrowheads in (A) and (D) indicate the strongly BIG-2-positive MOR28
glomeruli in the medial and lateral hemisphere of OB, respectively.
(G–N) BIG-2 expression in individual OR-defined glomeruli. (G–J) OR-defined glomeruli in the medial hemisphere of OB. (K–N) OR-defined glomeruli in the lateral
hemisphere of OB. BIG-2 immunoreactivity is strongly positive in MOR28 glomeruli (G and K), moderately positive for mOR-EG glomeruli (H and L), and very
weakly positive for mOR256-17 glomeruli (I and M) and OR-I7 glomeruli (J and N). Asterisks in (G) represent the MOR28-positive medial glomerulus that was
used as the relative standard (100%) of BIG-2 signal intensity.
(O) Glomerulus ranking for BIG-2 expression levels and its correlation with OR expression. Intensities of BIG-2 expression were measured for 2842 glomeruli on
immunostained OB sections (n = 12) from six mice and plotted in a descending rank order. Glomeruli positive for MOR28 (n = 12), mOR-EG (n = 12), mOR256-17
(n = 17), and OR-I7 (n = 13) are shown in pink, blue, yellow, and green, respectively.
(P) Average relative intensities of BIG-2 expression in OR-defined glomeruli. Intensities of BIG-2 expression in individual OR-defined glomeruli are almost the
same between the medial (M) and lateral (L) hemispheres of OB. The numbers of glomeruli analyzed are shown in the columns. Error bars indicate SEM.
Scale bars: (A–F) 400 mm, (G–N) 100 mm.
BIG-2 Mediates Olfactory Axon Targeting
838 Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc.
tended to be negatively influenced by the neural activity (Figures
5A, 5B, and 5F), although the extent of regulation was much
smaller than that of ephrin-A5 (Figure 5H). Next, we compared
BIG-2 expression levels between CNGA2-positive and -negative
mOR-EG glomeruli. We expected that either up- or down-
regulation of BIG-2 could be easily detected in mOR-EG glo-
type mice (Figure 4). In the CNGA2 mutant mice, two adja-
cent CNGA2-positive and -negative mOR-EG glomeruli were
observed on both the medial (Figures 5J–5L) and lateral
Figure 5. Neural Activity-Dependent Expression of BIG-2
(A–P)Mosaic analysis intheCNGA2 heterozygous female mice.TheCNGA2 gene located on Xchromosome undergoes thestochasticmonoallelic inactivation in
the female mice, resulting in generation of two populations of OSNs (CNGA2-positive and -negative). In OB, CNGA2-positive and -negative OSN axons are seg-
regated into distinct glomeruli. (A–E) Representative images of three consecutive coronal sections (18 mm each) of OB from the CNGA2 heterozygous female
mouse: double labeling with anti-BIG-2 (A) and anti-CNGA2 (B), single labeling with anti-Kirrel2 (C), and double labeling with anti-CNGA2 (D) and anti-ephrin-
A5 (E). Dotted circles indicate CNGA2-negative glomeruli. Individual glomeruli (#1–5) display distinct patterns of molecular expression. For example, glomeruli
#1 and #5 are CNGA2 positive, Kirrel2 positive, and ephrin-A5 negative, while BIG-2 is positive in #1 but negative in #5. Glomeruli #2, #3, and #4 are CNGA2
negative, Kirrel2 negative, and ephrin-A5 positive, while BIG-2 shows different levels of expression. (F–H) Signal intensities of BIG-2 (F), Kirrel2 (G), and
ephrin-A5 (H) in each glomerulus were quantified from 12 OB sections from two CNGA2 heterozygous female mice and shown in box plots. Means, red circles;
medians, middle line; 75thand 25thquartiles, top and bottom lines, respectively; whiskers show range. *p < 0.001 (two-tailed t test). The numbers of glomeruli
analyzed are shown in parentheses. (I–P) Triple labeling with antibodiesagainst BIG-2 ([I] and [M]; green in [L] and [P]), mOR-EG ([J] and [N]; red in [L] and [P]), and
CNGA2 ([K]and [O]; bluein[L] and[P]) wereperformedon OBsections ofCNGA2 heterozygous female mouse.Representative images ofsplitmOR-EG glomeruli
with different intensities of BIG-2 expression on the medial (I–L) and the lateral side (M–P) of OB. Split mOR-EG-positive glomeruli were observed in 8 out of
16 half-bulbs analyzed. In all the cases of split glomeruli, the level of BIG-2 expression was consistently higher in the CNGA-2-negative mOR-EG glomeruli
(arrowheads) than the CNGA2-positive mOR-EG glomeruli (arrows).
(Q–Y) Unilateral naris occlusion of wild-type mice. Mice were subjected to the unilateral naris occlusion on postnatal day 2 and analyzed after 6 months. (Q–V)
Representative images of two coronal sections of OB from the unilaterally naris occluded mouse: triple labeling with antibodiesagainst BIG-2 (Qand T), Kirrel2 (R
and T), and ephrin-A5 (S and T) and double labeling with antibodies against TH (U) and OMP (V). A merged image of BIG-2 (red), Kirrel2 (blue), and ephrin-A5
(green) is shown in (T). (W–Y) Box plots comparing signal intensities between the open- and closed-side glomeruli of BIG-2 (W), Kirrel2 (X), and ephrin-A5 (Y).
Quantification was made from six pairs of OB sections from three naris-occluded mice. Means, red circles; medians, middle line; 75thand 25thquartiles, top
and bottom lines, respectively; whiskers show range. *p < 0.001 (two-tailed t test). The numbers of glomeruli analyzed are shown in parentheses.
Scale bars: 200 mm.
BIG-2 Mediates Olfactory Axon Targeting
Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc. 839
(Figures 5N–5P) sides of OB. The level of BIG-2 expression was
significantly higher in the CNGA2-negative mOR-EG glomeruli
(arrowheads) than the CNGA2-positive glomeruli (arrows)
In the second experiment, wild-type mice at postnatal day 2
were subjected to unilateral naris occlusion and analyzed by
immunohistochemistry after 6 months to examine whether the
reduced neural activityof OSNs mayinfluence the BIG-2expres-
sion level. Efficiency of the naris occlusion was validated by the
loss of tyrosine hydroxylase (TH) immunoreactivity in the OB glo-
merular layer on the closed side (Figure 5U; Stone et al., 1990),
despite apparently normal glomerular innervation of olfactory
marker protein (OMP)-positive sensory axons (Figure 5V). In
agreement with the previous finding by Serizawa et al. (2006),
Kirrel2 and ephrin-A5 were dramatically downregulated and
upregulated, respectively, on the olfactory axons, particularly
in the olfactory nerve layer, on the closed side (Figures 5R–5T,
5X, and 5Y). The open- and closed-side difference in BIG-2 ex-
pression levels on olfactory axons (Figures 5Q and 5T) appeared
smaller than those of Kirrel2 and ephrin-A5 but proved to be
significant by a statistical analysis for quantification of signal
intensities in individual glomeruli (Figure 5W). These results indi-
cate that the neural activity negatively regulates the expression
Abnormal Glomerular Targeting of Olfactory Axons
in BIG-2-Deficient Mice
A variety of proteins belonging to the Ig superfamily play impor-
tant roles as axon guidance molecules in the formation and
maintenance of neural circuits in the nervous system (Yoshihara
et al., 1991; Maness and Schachner, 2007). By analogy, it is con-
ceivable that BIG-2 protein expressed on a subset of OSN axons
in an OR type-specific fashion may be involved in axonal guid-
ance and targeting to specific glomeruli in OB. To investigate
this possibility, we generated BIG-2-deficient mice by disrupting
the exon 2 that encodes the N-terminal signal peptide of BIG-2
protein (Figures 6A and 6B). The absence of BIG-2 protein in
the mutant mice was confirmed by both Western blot and immu-
nohistochemical analyses (Figure 6C; data not shown). A gross
anatomy and a layer organization of OE in BIG-2-deficient
mice appeared normal, as assessed by Nissl staining (Figures
6D–6G). No apparent change was observed in the expression
levels and patterns of other cell surface molecules, such as
NCAM, OCAM, NP-1, NP-2, Kirrel2, and ephrin-A5, that are
present in either all or subsets of OSN axons (Figure 6C; data
Based on the close correlation between the BIG-2 expression
level and the OR gene choice (Figure 4), we assumed a putative
function of BIG-2 in convergence and glomerular targeting of
OSN axons expressing individual ORs. To corroborate this pos-
sibility, we investigated OB glomeruli representing individual
ORs in detail by immunohistochemistry with antibodies against
MOR28, mOR-EG, and OR-I7 in wild-type and BIG-2-deficient
littermates (6–8 weeks). In wild-type mice, MOR28-positive
OSNs innervated mostly a single glomerulus on both the medial
and lateral sides of OB (90% of half-bulbs), while a small glo-
merulus ectopically targeted by MOR28-positive axons was
observed in only 10% of half-bulbs (Figure 6R). In contrast, the
BIG-2 deficiency resulted in more frequent appearance of
ectopic MOR28 glomeruli (28% of half-bulbs; Figures 6H–6L
and 6R). The average number of MOR28 glomeruli per half-
bulb in BIG-2-deficient mice (1.3 ± 0.1) was significantly larger
than that in wild-type mice (1.1 ± 0.1; Figure 6S). Next, we exam-
ined glomerular targeting of OR-I7-positive OSNs and found
more severe defects with abnormally multiple glomeruli (Figures
6M–6Q). Ectopically located small glomeruli were detected in
83% of half-bulbs in BIG-2-deficient mice, compared to 25%
in wild-type mice (Figure 6R). The average numbers of OR-I7
glomeruli per half-bulb were 1.3 ± 0.2 in wild-type mice and
2.4 ± 0.3 in BIG-2-deficient mice (Figure 6S). Similarly, ectopi-
cally located mOR-EG glomeruli were more frequently observed
in BIG-2-deficient mice (71% of half-bulbs) than in wild-type
mice (42% of half-bulbs; Figure 6R). The average numbers of
mOR-EG glomeruli per half-bulb were 1.4 ± 0.1 in wild-type
mice and 1.8 ± 0.1 in BIG-2-deficient mice (Figure 6S). There
was no shift in the positions of OR-defined major glomeruli and
no tendency in the relative positions of ectopic glomeruli along
mice (data not shown). A similar ectopic innervation to multiple
glomeruli was observed for OR-I7-expressing OSNs in BIG-2-
deficient mice as early as postnatal day 7 (Figure S3). These
results indicate that BIG-2 is required for convergence of OSN
axons to proper target glomeruli in the OB.
Presence of Heterophilic BIG-2-Binding Partner
on Olfactory Axons
To gain insight into the mode of BIG-2 action on olfactory axon
convergence, we prepared a recombinant fusion protein con-
sisting of mouse BIG-2 extracellular region and human placental
alkaline phosphatase (BIG-2/AP) and performed an overlay as-
say on OB sections from adult wild-type and BIG-2-deficient
mice. BIG-2/AP strongly bound onto the nerve layer and glomer-
ular layer in OB of BIG-2-deficient mice in a mosaic fashion (Fig-
ure 7B), whereas the BIG-2/AP binding was very weak on OB
sections from wild-type mice (Figure 7A). This finding implies
the presence of a binding partner of BIG-2 in glomeruli, which
may be masked by endogenous BIG-2 in wild-type mice, but un-
covered in BIG-2-deficient mice to be detectable by exogenous
BIG-2/AP in the overlay assay. An alternative possibility would
be that the expression of BIG-2-binding partner might be upre-
gulated in the absence of BIG-2.
Intranasal irrigation of ZnSO4causes acute OE destruction
that accompanies deafferentation of olfactory axons projecting
we tested whether the binding partner of BIG-2 is derived from
OSNs or other types of cells in OB. A selective deafferentation
of olfactory axons was confirmed by the loss of OMP immunore-
cells projecting into glomeruli were unaffected by ZnSO4treat-
ment as revealed by immunostaining of protocadherin-21, a mi-
tral cell marker (Figures 7D–7F; Nakajima et al., 2001; Nagai
et al., 2005). ZnSO4-induced deafferentation of olfactory axons
led to almost complete loss of the BIG-2/AP binding to glomeruli
(Figure7C).Thisresult indicatesthattheBIG-2-binding molecule
is located on axon terminals of OSNs, but not on dendrites of OB
neurons, in glomeruli.
BIG-2 Mediates Olfactory Axon Targeting
840 Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc.
The mosaic pattern of BIG-2/AP-binding to OB glomeruli is
reminiscent of different levels of endogenous BIG-2 expression
among individual glomeruli. To correlate them, adjacent OB
sections from BIG-2-deficient mice were subjected to OR
immunohistochemistry and BIG-2/AP overlay assay. As men-
tioned above, MOR28 glomeruli are strongly positive for BIG-
2, while OR-I7 glomeruli are very weakly positive (Figure 4). In
BIG-2-deficient mice, an intense binding of BIG-2/AP was
observed in MOR28 glomeruli on both the medial and lateral
sides (Figures 7G, 7H, 7K, and 7L), whereas OR-I7 glomeruli
showed little binding of BIG-2/AP (Figures 7I, 7J, 7M, and
7N). Thus, there appears to be a correlation among the OR
gene choice, the BIG-2 expression level, and the BIG-2/AP
binding intensity reflecting the expression level of a putative
The key conclusions of this study are as follows: (1) BIG-2 is ex-
pressed by a subset of OSNs in correlation with the OR gene
choice. (2) BIG-2 expression in OB glomerular array shows a
mosaic pattern distinct from Kirrel2 and ephrin-A5. (3) BIG-2
expression is regulated by neural activity. (4) BIG-2-deficient
mice show aberrant projection of OSN axons to multiple glomer-
uli. (5) A heterophilic binding partner of BIG-2 is present on OSN
axon terminals. These results indicate that BIG-2 is an axon
Figure6. AbnormalGlomerularTargeting of
Olfactory Axons in BIG-2-Deficint Mice
(A) Gene targeting strategy to generate BIG-2-
deficient mice. A tandem cassette consisting of
a pgk-neo selection marker (neo) and a trans-
cription/translation stop signal (stop) was inserted
at a translation start codon (ATG) in the exon
2 of BIG-2 gene. DTA, diphtheria toxin A sub-
unit; pBS, pBluescriptII; E, EcoNI; H, HindIII;
(B) PCR genotyping of wild-type (+/+), heterozy-
gous (+/?), and homozygous (?/?) BIG-2 mutant
(C) Western blot analysis. BIG-2 protein is absent
in OB homogenates of the homozygous mutant
mice, whereas other cell recognition molecules
(NCAM, OCAM, NP-1, NP-2, and Kirrel2) are ex-
pressed at equivalent levels in all the genotypes.
(D–G) Nissl-stained OE sections of wild-type (D
and E) and BIG-2-deficient (F and G) mice. No
obvious difference in the gross anatomy of OE is
observed between the two genotypes at both
low (D and F) and high (E and G) magnifications.
(H–Q) Representative examples of ectopically tar-
geted MOR28-positive (H–L) and OR-I7-positive
(M–Q) olfactory axons in BIG-2-deficient mice at
postnatal 6 weeks. Distances from the rostral tip
of OB are shown in (H)–(K) and (M)–(P). (L) and
(Q) are higher-magnified views of the boxed re-
gions in (K) and (P), respectively. Arrows, glomeruli
at normal positions; arrowheads, glomeruli at
(R and S) Quantitative analysis of abnormal
glomerular targeting of olfactory axons in BIG-
2-deficient mice at postnatal 6–8 weeks. The
percentages of half-bulbs with multiple glomer-
uli (R) and the numbers of glomeruli per half-bulb
(S) are shown for three ORs (MOR28, mOR-EG,
deficient (?/?) mice. The numbers of half-bulbs
examined are shown in the columns. Error bars
indicate SEM *p < 0.05; **p < 0.01 (two-tailed
Scale bars: (D and F) 1 mm, (E and G) 25 mm, (H–K)
and (M–P) 400 mm, (L and Q) 100 mm.
BIG-2 Mediates Olfactory Axon Targeting
Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc. 841
guidance molecule that mediates proper neuronal wiring in the
mouse olfactory system.
BIG-2: A Cell Adhesion Molecule of the Contactin Family
BIG-2 belongs to the contactin subgroup of the Ig superfamily,
which includes other fiveGPI-anchoring glyocoproteins: contac-
tin, TAG-1, BIG-1, NB-2, and NB-3. They are closely related in
structure harboring the same domain organization but show
differential patterns of expression in the developing and adult
brains. Based on phenotypes of knockout mice, crucial roles of
several members of this subgroup have been reported in various
aspects of neural development and functions. For example, con-
tactin and TAG-1 are essential for the molecular organization of
paranode and juxtaparanode of myelinated fibers, respectively
(Boyle et al., 2001; Traka et al., 2003). Contactin-deficient mice
display abnormal cerebellar architecture accompanying severe
ataxia (Berglund et al., 1999) and impaired long-term depression
in the hippocampus (Murai et al., 2002). TAG-1 is required for
proper neuronal migration of the precerebellar nuclei (Denaxa
et al., 2005). NB-2-deficient mice show aberrant responses to
acoustic stimuli (Li et al., 2003). NB-3 deficiency leads to im-
paired motor coordination (Takeda et al., 2003). In terms of the
olfactory system, however, there was no report on functional
analysis of the contactin subgroup members. In this study, we
identified BIG-2 as an axon guidance molecule required for
establishment of the olfactory neural circuitry.
binding activities to various types of molecules. Contactin inter-
acts with the L1 subgroup members of the Ig superfamily (L1,
NrCAM, neurofascin), the extracellular matrix molecules (tenas-
cin-R and -C), receptor tyrosine phosphatase b/phosphacan,
voltage-gated sodium channels, Notch, and Caspr/paranodin
(Falk et al., 2002). TAG-1 binds heterophilically to L1, NrCAM,
tenascin-C, and NCAM, and also homophilically to TAG-1 itself
(Rutishauser, 2000). In the BIG-2/AP-overlay experiment on OB
sections from BIG-2-deficient mice, we have demonstrated the
presence of heterophilic binding partner of BIG-2 in glomeruli.
Interestingly, BIG-2/AP bound strongly to MOR28 glomeruli
(high BIG-2) and weakly to OR-I7 glomeruli (low BIG-2). Thus,
the putative counter-receptor for BIG-2 shows a mosaic pattern
of expression correlated with that of BIG-2 itself. In addition,
ZnSO4-induced deafferentation experiment revealed that the
putative counter-receptor for BIG-2 is present on OSN axons.
Taken together, these results suggest that the heterophilic
adhesion between BIG-2 and its counter-receptor may mediate
olfactory axon fasciculation and glomerular targeting in close
relation to the OR gene choice in OSNs. Identification and char-
a deeper mechanistic insight into the axon-axon interaction
mediated by BIG-2.
As well as the olfactory system, BIG-2 is present in different
subsets of neuronal types in various brain regions (Yoshihara
et al., 1995). For example, BIG-2 is highly expressed in granule
cells in the anterior folia (lobules 1–6) and Purkinje cells in the
lobules 9 and 10 in the cerebellum. In the hippocampus, BIG-2
is expressed strongly in the CA1 pyramidal cells but only
weakly in the CA3 pyramidal cells and the dentate gyrus granule
cells. Thus, it is likely that an axon guidance molecule BIG-2
plays functional roles in the formation and maintenance of
neural circuits also in these regions. Recently, human BIG-2
(CNTN4) locus (3p26.2–3p26.3) was identified as a candidate
gene responsible for 3p deletion syndrome characterized by
developmental delay, postnatal growth retardation, and dys-
morphic features (Fernandez et al., 2004) and spinocerebellar
ataxia type 16 (SCA16) characterized by cerebellar degenera-
tion (Miura et al., 2006). With reference to the phenotypes of
these human disorders that are possibly caused by BIG-2
gene mutations, detailed histological and behavioral analyses
of BIG-2-deficient mice are now in progress to elucidate the
Figure 7. Presence of Heterophilic BIG-2-Binding Partner on
Olfactory Axon Terminals
(A–C) Representative images of BIG-2/AP binding to the OB sections from
wild-type (A), BIG-2-defieicnt (B), and ZnSO4-treated BIG-2-deficient (C)
mice. BIG-2/AP strongly binds onto OB glomeruli from BIG-2-deficient mice
in a mosaic fashion (B), whereas only a weak binding to glomeruli is observed
in wild-type mice (A). BIG-2-AP also binds to cell bodies of mitral cells. Deaf-
ferentation of olfactory axons by intranasal irrigation of ZnSO4results in the
loss of BIG-2/AP binding to OB glomeruli (C). Similar results were reproducibly
obtained from three wild-type mice, three BIG-2-deficient mice, and four
ZnSO4-treated BIG-2-deficient mice.
(D–F) Double labeling with antibodies against OMP (red) and protocadherin-21
(green) of the main OB sections from wild-type (D), BIG-2-defieicnt (E), and
ZnSO4-treated BIG-2-deficient (F) mice. Deafferentation of OSN axons is
confirmed by the loss of OMP immunoreactivity in glomeruli.
(G–N) Representative images showing different signal intensities of BIG-2/AP
binding onto distinct OR-defined glomeruli in BIG-2-deficient mice. BIG-2/AP
binds strongly onto the MOR28 glomeruli (G, H, K, and L), but not onto the
OR-I7 glomeruli (I, J, M, and N) on both the medial (G–J) and lateral sides
ibly obtained from three BIG-2-deficient mice.
Scale bars: 200 mm.
BIG-2 Mediates Olfactory Axon Targeting
842 Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc.
physiological role of BIG-2 in brain development, morphogene-
sis, and functions.
Molecules in the Formation of Olfactory Neural Circuitry
Two basic cellular principles underlie the establishment of func-
tional odor map in OB: the single OR gene choice by individual
OSNs and the precise wiring of OSN axons to target glomeruli.
Various axon guidance molecules appear to be involved in the
process of olfactory axon pathfinding from OE to OB. We pro-
pose a four-step hierarchical model for the formation of neural
circuitry in the primary olfactory system. (1) Initial outgrowth of
olfactory axons from OE is regulated by the permissive effect
of chemokine Sdf-1 (Cxcl12) which acts on its receptor Cxcr4
transiently expressed in the developing OSNs of zebrafish
(Miyasaka et al., 2007). (2) Navigation of olfactory axons toward
OB is partly directed by the chemorepellent Slit and its receptor
Robo in zebrafish (Miyasaka et al., 2005). (3) Coarse targeting of
olfactory axons to appropriate domains of OB appears to be
mediated by the repulsive interactions between neuropilins
and their ligands semaphorins (Schwarting et al., 2000; Walz
et al., 2002; Taniguchi et al., 2003). In particular, spontaneous
OR/Gs-derived cAMP signals control the expression level of
NP-1 that directs olfactory axon projection along the anteropos-
terior axis of OB (Imai et al., 2006). (4) Convergence of olfactory
axons onto target glomeruli is controlled by combinatorial
actions of guidance molecules including BIG-2, Kirrel2, Kirrel3,
ephrin-A5, and EphA5 (Cutforth et al., 2003; Serizawa et al.,
most complex among the four steps. This process involves a
combination of three different modes of axon-axon interactions:
heterophilic adhesion, homophilic adhesion, and repulsion.
From this study, the Ig superfamily molecule BIG-2 is likely to
mediate the glomerular convergence of olfactory axons through
the heterophilic binding to its unknown counter-receptor. Kirrel2
and Kirrel3, the two closely related members of the Ig super-
family, are expressed in complementary subsets of OSNs and
respectively mediate the glomerular targeting through their
homophilic-binding properties (Serizawa et al., 2006). In con-
trast, ephrin-A5 and EphA5 that are expressed also in a com-
plementary fashion exert their functions through the repulsive
activity (Cutforth et al., 2003; Serizawa et al., 2006).
The expression levels of these axon guidance molecules are
closely correlated with the OR gene choice in individual OSNs
and are regulated by neural activity. Interestingly, the CNGA2-
deficient mouse and unilateral naris occlusion experiments re-
vealed that the properties of neural activity dependence are
different among the molecules in terms of the direction and the
extent. Kirrel2 and EphA5 are highly expressed in high-activity
OSNs, while BIG-2, ephrin-A5, and Kirrel3 show higher levels
inlow-activityOSNs.Theexpression ofephrin-A5 isdramatically
influenced by the neural activity, while BIG-2 is regulated to
a lesser extent. Such a difference may serve to generate the
variability in expression levels of axon guidance molecules in in-
dividual OSNs with different types of OR genes. Taken together,
the combinatorial codes of axon guidance molecules defined
by neural activity may be a crucial mechanism underlying the
selective convergence of olfactory axons to target glomeruli
In BIG-2-deficient mice, many olfactory axons aberrantly pro-
axon targeting varied among three OR types. Unexpectedly, the
abnormal targeting was most obvious for OR-I7-positive axons
that normally express a relatively low level of BIG-2, whereas
the abnormality was relatively mild for MOR28-positive axons
that normally display the highest BIG-2 expression. Two notions
may account for this apparently contradictory result as follows.
First, the stability of glomeruli appears different among individual
ORs. There are several reports demonstrating that some OR-de-
fined glomeruli are ‘‘labile’’ with variability of their positions and
numbers across different OBs (Strotmann et al., 2000; Schaefer
et al., 2001; Oka et al., 2006). We assume that MOR28 glomeruli
may be relatively ‘‘stable,’’ while OR-I7 glomeruli may be very
Figure 8. A Proposed Model for the Combinatorial Codes of
Guidance Molecules in Olfactory Axon Targeting
(A) Three types of OSNs expressing distinct ORs are schematically shown
(white, gray, and black). Different combinations of guidance molecules
(BIG-2: red, molecule A: green, molecule B: blue, and molecule C: purple) are
expressed on axons of individual OR-defined OSNs. In the wild-type mice,
OSNs expressing a given OR converge their axons precisely to the target
glomerulus through the combinatorial codes of axon guidance molecules.
(B) In BIG-2-deficient mice, the combinatorial code of guidance molecules in
the white OSNs becomes identical to that in the black OSNs. This abnormal
situation results in confused wiring of some of OSN axons and mistargeting
to ectopic locations.
BIG-2 Mediates Olfactory Axon Targeting
Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc. 843
‘‘labile,’’ leading to such a phenotypic variability of axon mistar-
geting in BIG-2-deficient mice. Second, the disruption of BIG-2
gene results in a decrease of the combinatorial code repertoire
of guidance molecules on olfactory axons. Because of the loss
of BIG-2, several distinct OR-defined OSNs that normally ex-
press different levels of BIG-2 in wild-type mice should present
a similar combinatorial code of other axon guidance molecules.
Thus, it will become difficult for individual axons to recognize
their fellow travelers reaching the same target glomeruli in BIG-
2-deficient mice (Figure 8B). In summary, BIG-2 is required for
sound projection of olfactory axons to target glomeruli in combi-
nation with other axon guidance molecules.
In Situ Hybridization
In situ hybridization was performed as described previously (Yoshihara et al.,
Anti-BIG-2 antibody was produced by immunizing guinea pigs with recombi-
nant BIG-2/Fc fusion protein consisting of rat BIG-2 extracellular region and
human IgG1Fc region. Antibodies against mouse MOR-28, mOR-EG, and
OR-I7 were produced by immunizing guinea pigs with keyhole limpet hemocy-
anin-conjugated synthetic peptides corresponding to the carboxyl-terminal 20
RNKDVKDTVKKIIGTKVYSS; OR-I7, RTLHLAQGQDANTKKSSRDG). Individ-
ual OR-directed antibodies intensely labeled cilia, cell bodies, and axon termi-
nals of OSNs, as reported previously (Barnea et al., 2004; Strotmann et al.,
2004). Anti-protocadherin-21 antibody was produced by immunizing rabbits
with a synthetic peptide corresponding to the carboxyl-terminal 20 amino-
acid residue of mouse protocadherin-21 (LVSELKQKFEKKSLDNKAYI). This
antibody specifically labeled soma and dendrites of mitral cells.
Immunohistochemistry was performed as described previously (Yoshihara
et al., 2005). Primary antibodies used are as follows: guinea pig anti-BIG-
2/Fc (1:1000); goat anti-contactin-4/BIG-2 (1:500, R&D Systems); rat anti-
NCAM (1:500, Chemicon); rabbit anti-OCAM (1:1000) (Yoshihara et al.,
1997); goat anti-NP-1 (1:400, R&D Systems); goat anti-NP-2 (1:400, R&D
Systems); rabbit anti-VGluT2 (1:700, a gift from Dr. Takeshi Kaneko at Kyoto
University) (Kaneko et al., 2002); rabbit anti-Kirrel2 (1:2000, a gift from
Dr. Hitoshi Sakano at The University of Tokyo) (Serizawa et al., 2006); goat
anti-ephrin-A5 (1:75, R&D Systems); guinea pig anti-MOR28 (1:1000); guinea
pig anti-mOR-EG (1:1000); guinea pig anti-OR-I7 (1:5000); rabbit anti-
mOR256-17 (1:1000, a gift from Dr. Heinz Breer at Hohenheim University)
(Strotmann et al., 2004); rabbit anti-CNGA2 (1:200, Alomone Labs); rabbit
anti-TH (1:300, Chemicon); goat anti-OMP (1:10000, a gift from Dr. Frank L.
Margolis at University of Maryland); rat anti-GFP (1:1000, Nacalai Tesque).
Cy3- and Cy5-conjugated secondary antibodies were purchased from Jack-
son ImmunoResearch. Alexa 488-conjugated secondary antibodies were
purchased from Molecular Probes.
For triple-immunofluorescence labeling of BIG-2, Kirrel2, and ephrin-A5
proteins, OB sections were treated as follows: (1) a mixture of guinea pig
anti-BIG-2 and goat anti-ephrin-A5 antibodies, (2) a mixture of anti-guinea
pig and anti-goat secondary antibodies, (3) antigen-retrieval medium (HistoVT
One, Nacalai Tesque), (4) a mixture of guinea pig anti-BIG-2, goat anti-ephrin-
A5, and rabbit anti-Kirrel2 antibodies, (5) a mixture of anti-guinea pig, anti-
goat, and anti-rabbit secondary antibodies.
Fluorescent digital images were captured using Axioplan epifluorescence
microscopy (Carl Zeiss) equipped with DP70 CCD camera (Olympus) and
Fluoview FV1000 confocal laser scanning microscopy (Olympus). To quantify
the expression levels of BIG-2, Kirrel2, and ephrin-A5 in each glomerulus, the
mean pixel intensity within the region surrounded by the periglomerular cell
nuclei was measured using the public domain ImageJ program (NIH). A
three-dimensional glomerular map of BIG-2 expression was reconstructed
from 40 sections using DeltaViewer 2.1 software.
BIG-2-deficient mice were generated with a standard gene-targeting method.
An 8.7 kb genomic DNA fragment containing the exon 2 of mouse BIG-2 gene
was isolated and subcloned into a plasmid with diphtheria toxin A gene as
a negative selection marker. The initiation methionine codon (ATG) of BIG-2
gene was mutated into a stop codon (TAG) using GeneEditor in vitro Site-
Directed Mutagenesis System (Promega). A tandem cassette consisting of
the pgk-neo positive-selection marker and the transcription/translation stop
signal was inserted into a HindIII site in the exon 2. Electroporation into a
129/SvEv mouse ES cell line followed by G418 selection resulted in one pos-
itive homologous recombinant out of 576 clones as determined by PCR
screening and Southernblotanalysis. Chimeric C57B/6males thattransmitted
the mutant allele were obtained, backcrossed with C57B/6 mice nine times,
and used for the experiments. The following primers were used for genotyping
BIG-2 progeny: 50-aaaggcagatatgcagagaccataaggag-30and 50-ttcatcactcct
gaatcacacatgtcagg-30for the wild-type allele and 50-tggtagatggatcgatggcaaa
catgtccc-30and 50-ggttcacagaaggctcatggaaacagtc-50for the mutant allele.
Thirty-five cycles of PCR (96?C for 45 s, 61?C for 30 s, 72?C for 105 s) were
performed with ExTaq HS DNA polymerase (Takara).
The H-MOR28-ires-tauECFP transgenic mice (Serizawa et al., 2003, 2006)
were generous gifts from Dr. Hitoshi Sakano (The University of Tokyo). The
CNGA2 mutant mice were purchased from The Jackson Laboratory.
Unilateral Naris Occlusion
previously (Philpot et al., 1997) and analyzed after 6 months for immunohisto-
Western Blot Analysis
OB homogenates were prepared from adult BIG-2 +/+, +/?, and ?/? mice.
Western blot analysis was performed as described previously (Yoshihara
et al., 1991). BIG-2 protein on blots could be detected with guinea pig anti-
BIG-2 antibody only under a nonreducing condition. Primary antibodies
used are as follows: guinea pig anti-BIG-2/Fc (1:5000); rat anti-NCAM
(1:1000, Chemicon); rabbit anti-OCAM (1:2000); goat anti-NP-1 (1:1000,
R&D Systems), goat anti-NP-2 (1:1000, R&D Systems); rabbit anti-Kirrel2
(1:10000). Horseradish peroxidase-conjugated secondary antibodies were
purchased from Jackson ImmunoResearch. Protein bands were visualized
with a chemiluminescence reaction kit (ECL Plus Western Blotting Detection
System, Amersham) and an image analysis system (LAS-1000, Fujifilm).
BIG-2/AP Overlay Assay
BIG-2/AP fusion plasmid, pEF-BIG-2/AP, was constructed by inserting cDNA
encoding the extracellular region of mouse BIG-2 into pEF-AP vector which
harbors human elongation factor-1a promoter (Mizushima and Nagata,
1990) and human placental alkaline phosphatase (Cheng and Flanagan,
1994). pEF-BIG-2/AP was transfected into HEK293-F cells (FreeStyle 293-F
cells, Invitrogen) and BIG-2/AP protein in the culture supernatant was concen-
trated using Centriprep (Millipore). AP overlay assay was performed on para-
et al., 2000).
Zinc Sulfate Irrigation
Intranasal irrigation of ZnSO4was performed on adult mice as described
previously (McBride et al., 2003) and analyzed after 14 days for immunohisto-
chemistry and BIG-2/AP overlay assay.
The Supplemental Data for this article can be found online at http://www.
BIG-2 Mediates Olfactory Axon Targeting
844 Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc.
We thank K. Mori for continuous support and encouragement; H. Sakano for
gifts of anti-Kirrel2 antibody and H-MOR28-ires-tauECFP transgenic mice;
H. Breer for a gift of anti-mOR256-17 antibody; F.L. Margolis for a gift of
anti-OMP antibody; T. Kaneko for a gift of anti-VGluT2 antibody; Y.F. Sasaki
for help in production of anti-BIG-2 antibody; S. Mitsui for help in production
of anti-protocadherin-21 antibody; K. Kubota and M. Kawasaki for help in
construction of BIG-2 targeting vector; Y. Furutani for help in production of
recombinant proteins; BSI Research Resource Center (RRC) for help in gener-
ation of BIG-2-deficient mice; N. Miyasaka for comments on the manuscript;
and members of the Yoshihara lab for valuable discussions. This work was
supported in part by a Grant-in-Aid for Scientific Research (B) and a Grant-
in-Aid for Scientific Research on Priority Area (Cellular Sensor) to Y.Y. from
the Ministry of Education, Culture, Sports, Science, and Technology of Japan.
Received: August 10, 2007
Revised: December 6, 2007
Accepted: January 18, 2008
Published: March 26, 2008
Araneda, R.C., Kini, A.D., and Firestein, S. (2000). The molecular receptive
range of an odorant receptor. Nat. Neurosci. 3, 1248–1255.
Barnea, G., O’Donnell, S., Mancia, F., Sun, X., Nemes, A., Mendelsohn, M.,
and Axel, R. (2004). Odorant receptors on axon termini in the brain. Science
Berglund, E.O., Murai, K.K., Fredette, B., Sekerkowa, G., Marturano, B.,
Weber, L., Mugnaini, E., and Ranscht, B. (1999). Ataxia and abnormal cerebel-
lar microorganization in mice with ablated contactin gene expression. Neuron
Boyle, M.E., Berglund, E.O., Murai, K.K., Weber, L., Peles, E., and Ranscht, B.
(2001). Contactin orchestrates assembly of the septate-like junctions at the
paranode in myelinated peripheral nerve. Neuron 30, 385–397.
Brunet, L.J., Gold, G.H., and Ngai, J. (1996). General anosmia caused by a
targeted disruption of the mouse olfactory cyclic nucleotide-gated cation
channel. Neuron 17, 681–693.
Cheng, H.J., and Flanagan, J.G. (1994). Identification and cloning of ELF-1,
a developmentally expressed ligand for the Mek4 and Sek receptor tyrosine
kinases. Cell 79, 157–168.
Chess, A., Simon, I., Cedar, H., and Axel, R. (1994). Allelic inactivation regu-
lates olfactory receptor gene expression. Cell 78, 823–834.
Cutforth, T., Moring, L., Mendelsohn, M., Nemes, A., Shar, N.M., Kim, M.M.,
Frisen, J., and Axel, R. (2003). Axonal ephrin-As and odorant receptors:
Coordinate determination of the olfactory sensory map. Cell 114, 311–322.
Denaxa, M., Kyriakopoulou, K., Theodorakis, K., Trichas, G., Vidaki, M.,
Takeda, Y., Watanabe, K., and Karagogeos, D. (2005). The adhesion molecule
medulla but not of cortical interneurons. Dev. Biol. 288, 87–99.
Faivre-Sarrailh, C., and Rougon, G. (1997). Axonal molecules of the immuno-
globulin superfamily bearing a GPI anchor: their role in controlling neurite
outgrowth. Mol. Cell. Neurosci. 9, 109–115.
Falk, J., Bonnon, C., Girault, J.A., and Faivre-Sarrailh, C. (2002). F3/contactin,
a neuronal cell adhesion molecule implicated in axogenesis and myelination.
Biol. Cell 94, 327–334.
Fernandez, T., Morgan, T., Davis, N., Klin, A., Morris, A., Farhi, A., Lifton, R.P.,
and State, M.W. (2004). Disruption of contactin 4 (CNTN4) results in develop-
mental delay and other features of 3p deletion syndrome. Am. J. Hum. Genet.
Flanagan, J.G., Cheng, H.J., Feldheim, D.A., Hattori, M., Lu, Q., and Vander-
haeghen, P. (2000). Alkaline phosphatase fusions of ligands or receptors as
in situ probes for staining of cells, tissues, and embryos. Methods Enzymol.
Hansford, L.M., Smith, S.A., Haber, M., Norris, M.D., Cheung, B., and
Marshall, G.M. (2003). Cloning and characterization of the human neural cell
adhesion molecule, CNTN4 (alias BIG-2). Cytogenet. Genome Res. 101,
Imai, T., Suzuki, M., and Sakano, H. (2006). Odorant receptor-derived cAMP
signals direct axonal targeting. Science 314, 657–661.
Kaneko, T., Fujiyama, F., and Hioki, H. (2002). Immunohistochemical localiza-
tion of candidates for vesicular glutamate transporters in the rat brain.
J. Comp. Neurol. 444, 39–62.
Lewcock, J.W., and Reed, R.R. (2004). A feedback mechanism regulates
monoallelic odorant receptor expression. Proc. Natl. Acad. Sci. USA 101,
Li, H., Takeda, Y., Niki, H., Ogawa, J., Kobayashi, S., Kai, N., Akasaka, K.,
Asano, M., Sudo, K., Iwakura, Y., and Watanabe, K. (2003). Aberrant re-
sponses to acoustic stimuli in mice deficient for neural recognition molecule
NB-2. Eur. J. Neurosci. 17, 929–936.
Malnic, B., Hirono, J., Sato, T., and Buck, L.B. (1999). Combinatorial receptor
codes for odors. Cell 96, 713–723.
Maness, P.F., and Schachner, M. (2007). Neural recognition molecules of the
immunoglobulin superfamily: signaling transducers of axon guidance and
neuronal migration. Nat. Neurosci. 10, 19–26.
McBride, K., Slotnick, B., and Margolis, F.L. (2003). Does intranasal applica-
tion of zinc sulfate produce anosmia in the mouse? An olfactometric and
anatomical study. Chem. Senses 28, 659–670.
Miura, S., Shibata, H., Furuya, H., Ohyagi, Y., Osoegawa, M., Miyoshi, Y.,
Matsunaga, H., Shibata, A., Matsumoto, N., Iwaki, A., et al. (2006). The
contactin 4 gene locus at 3p26 is a candidate gene of SCA16. Neurology 67,
Miyasaka, N., Sato, Y., Yeo, S.Y., Hutson, L.D., Chien, C.B., Okamoto, H., and
Yoshihara, Y. (2005). Robo2 is required for establishment of a precise glomer-
ular map in the zebrafish olfactory system. Development 132, 1283–1293.
Miyasaka, N., Knaut, H., and Yoshihara, Y. (2007). Cxcl12/Cxcr4 chemokine
signaling is required for placode assembly and sensory axon pathfinding in
the zebrafish olfactory system. Development 134, 2459–2468.
Mizushima, S., and Nagata, S. (1990). pEF-BOS, a powerful mammalian
expression vector. Nucleic Acids Res. 18, 5322.
Mombaerts, P., Wang, F., Dulac, C., Chao, S.K., Nemes, A., Mendelsohn, M.,
Edmondson, J., and Axel, R. (1996). Visualizing an olfactory sensory map. Cell
Mori, K., Nagao, H., and Yoshihara, Y. (1999). The olfactory bulb: coding and
processing of odor molecule information. Science 286, 711–715.
Mori, K., Takahashi, Y.K., Igarashi, K., and Yamaguchi, M. (2006). Maps of
odorant molecular features in the mammalian olfactory bulb. Physiol. Rev.
Murai, K.K., Misner, D., and Ranscht, B. (2002). Contactin supports synaptic
plasticity associated with hippocampal long-term depression but not potenti-
ation. Curr. Biol. 12, 181–190.
Nadi, N.S., Head, R., Grillo, M., Hempstead, J., Granot-Reisfeld, N., and
Margolis, F.L. (1981). Chemical deafferentation of the olfactory bulb: plasticity
ofthelevelsoftyrosine hydroxylase, dopamine and norepinephrine.BrainRes.
Nagai, Y., Sano, H., and Yokoi, M. (2005). Transgenic expression of Cre
recombinase in mitral/tufted cells of the olfactory bulb. Genesis 43, 12–16.
Nagao, H., Yoshihara, Y., Mitsui, S., and Mori, K. (2000). Two mirror-image
sensory maps with domain organization in the mouse main olfactory bulb.
Neuroreport 11, 3023–3027.
O. (2001). Identification of three novel non-classical cadherin genes through
comprehensive analysis of large cDNAs. Brain Res. Mol. Brain Res. 94, 85–95.
Oka, Y., Katada, S., Omura, M., Suwa, M., Yoshihara, Y., and Touhara, K.
(2006). Odorant receptor map in the mouse olfactory bulb; in vivo sensitivity
and specificity of receptor-defined glomeruli. Neuron 52, 857–869.
BIG-2 Mediates Olfactory Axon Targeting
Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc. 845
Philpot, B.D., Lim, J.H., Halpain, S., and Brunjes, P.C. (1997). Experience-
dependent modifications in MAP2 phosphorylation in rat olfactory bulb.
J. Neurosci. 17, 9596–9604.
Ressler, K.J., Sullivan, S.L., and Buck, L.B. (1994). Information coding in the
olfactory system: evidence for a stereotyped and highly organized epitope
map in the olfactory bulb. Cell 79, 1245–1255.
Rutishauser, U. (2000). Defining a role and mechanism for IgCAM function in
vertebrate axon guidance. J. Cell Biol. 149, 757–759.
Schwarting, G.A., Kostek, C., Ahmad, N., Dibble, C., Pays, L., and Puschel,
A.W. (2000). Semaphorin 3A is required for guidance of olfactory axons in
mice. J. Neurosci. 20, 7691–7697.
Serizawa, S., Ishii, T., Nakatani, H., Tsuboi, A., Nagawa, F., Asano, M., Sudo,
K., Sakagami, J., Sakano, H., Ijiri, T., et al. (2000). Mutually exclusive expres-
sion of odorant receptor transgenes. Nat. Neurosci. 3, 687–693.
Serizawa, S., Miyamichi, K., Nakatani, H., Suzuki, M., Saito, M., Yoshihara, Y.,
and Sakano, H. (2003). Negative feedback regulation ensures the one recep-
tor-one olfactory neuron rule in mouse. Science 302, 2088–2094.
Serizawa, S., Miyamichi, K., Takeuchi, H., Yamagishi, Y., Suzuki, M., and
Sakano, H. (2006). A neuronal identity code for the odorant receptor-specific
and activity-dependent axon sorting. Cell 127, 1057–1069.
Schaefer, M.L., Finger, T.E., and Restrepo, D. (2001). Variability of position of
the P2 glomerulus within a map of the mouse olfactory bulb. J. Comp. Neurol.
Shykind, B.M., Rohani, S.C., O’Donnell, S., Nemes, A., Mendelsohn, M., Sun,
Y., Axel, R., and Barnea, G. (2004). Gene switching and the stability of odorant
receptor gene choice. Cell 117, 801–815.
Stone, D.M., Wessel, T., Joh, T.H., and Baker, H. (1990). Decrease in tyrosine
hydroxylase, but not aromatic L-amino acid decarboxylase, messenger RNA
in rat olfactory bulb following neonatal, unilateral odor deprivation. Brain
Res. Mol. Brain Res. 8, 291–300.
Strotmann, J., Conzelmann, S., Beck, A., Feinstein, P., Breer, H., and
Mombaerts,P.(2000). Localpermutationsintheglomerular arrayofthemouse
olfactory bulb. J. Neurosci. 20, 6927–6938.
Strotmann, J., Levai, O., Fleischer, J., Schwarzenbacher, K., and Breer, H.
(2004). Olfactory receptor proteins in axonal processes of chemosensory
neurons. J. Neurosci. 24, 7754–7761.
Takeda, Y., Akasaka, K., Lee, S., Kobayashi, S., Kawano, H., Murayama, S.,
Takahashi, N., Hashimoto, K., Kano, M., Asano, M., et al. (2003). Imaired
motor coordination in mice lacking neural recognition molecule NB-3 of the
contactin/F3 subgroup. J. Neurobiol. 56, 252–265.
Taniguchi, M., Nagao, H., Takahashi, Y.K., Yamaguchi, M., Mitsui, S., Yagi, T.,
Mori, K., and Shimizu, T. (2003). Distorted odor maps in the olfactory bulb of
semaphorin 3A-deficient mice. J. Neurosci. 23, 1390–1397.
Traka, M., Goutebroze, L., Denisenko, N., Bessa, M., Nifli, A., Havaki, S.,
Iwakura, Y., Fukamauchi, F., Watanabe, K., Soliven, B., et al. (2003). Associa-
tion of TAG-1 and Caspr2 is essential for the molecular organization of juxta-
paranodal regions of myelinated fibers. J. Cell Biol. 162, 1161–1172.
Vassar, R., Chao, S.K., Sitcheran, R., Numez, J.M., Vosshall, L.B., and Axel, R.
(1994). Topographic organization of sensory projections to the olfactory bulb.
Cell 79, 981–991.
Walz, A., Rodriguez, I., and Mombaerts, P. (2002). Aberrant sensory innerva-
tion of the olfactory bulb in neuropilin-2 mutant mice. J. Neurosci. 22, 4025–
Yoshihara, Y., Oka, S., Watanabe, Y., and Mori, K. (1991). Developmentally
and spatially regulated expression of HNK-1 carbohydrate antigen on a novel
phosphatidylinositol-anchored glycoprotein in rat brain. J. Cell Biol. 115, 731–
Yoshihara, Y., Kawasaki, M., Tamada, A., Nagata, S., Kagamiyama, H., and
Mori, K. (1995). Overlapping and differential expression of BIG-2, BIG-1,
TAG-1, and F3: four members of an axon-associated cell adhesion molecules
subgroup of the immunoglobulin superfamily. J. Neurobiol. 28, 51–69.
Yoshihara,Y.,Kawasaki, M.,Tamada,A.,Fujita, H.,Hayashi, H.,Kagamiyama,
H., and Mori, K. (1997). OCAM: a new member of the neural cell adhesion
molecule family related to zone-to-zone projection of olfactory and vomero-
nasal axons. J. Neurosci. 17, 5830–5842.
Yoshihara, S., Omichi, K., Yanazawa, M., Kitamura, K., and Yoshihara, Y.
(2005). Arx homeobox gene is essential for development of mouse olfactory
system. Development 132, 751–762.
Zhao, H., and Reed, R.R. (2001). X inactivation of the OCNC1 channel gene
reveals a role for activity-dependent competition in the olfactory system.
Cell 104, 651–660.
Zeng, L., Zhang, C., Xu, J., Ye, X., Wu, Q., Dai, J., Ji, C., Gu, S., Xie, Y., and
Mao, Y. (2002). A novel splice variant of the cell adhesion molecule contactin
4 (CNTN4) is mainly expressed in human brain. J. Hum. Genet. 47, 497–499.
Zheng, C., Feinstein, P., Bozza, T., Rodriguez, I., and Mombaerts, P. (2000).
Peripheral olfactory projections are differentially affected in mice deficient in
a cyclic nucleotide-gated channel subunit. Neuron 26, 81–91.
BIG-2 Mediates Olfactory Axon Targeting
846 Neuron 57, 834–846, March 27, 2008 ª2008 Elsevier Inc.