- Email: [email protected]
BMP5 expression in the adult rat brain
Neuroscience 284 (2015) 972–987
BMP5 EXPRESSION IN THE ADULT RAT BRAIN Y. KUSAKAWA, a,b S. MIKAWA a AND K. SATO a*
eral biological events (Bragdon et al., 2011). BMPs act on a cell via the induction of a heterodimeric complex of type I and type II BMP serine/threonine kinase receptors including bone morphogenetic protein receptor type I (BMPRIA, BMPRIB) and type II (BMPRII) (Bragdon et al., 2011). By binding to the type I BMPRs, BMPs activate the receptor-activated Smads (R-Smads; Smad1/5/ 8) which oligomerize with common-mediator Smad (Co-Smad; Smad4) in the cytoplasm. Then, the Smad complex enters the nucleus and initiates transcription (Moustakas and Heldin, 2009). Although most of the biological effects of BMPs have been related to the Smad-dependent signaling pathways, Smad-independent signaling pathways have been also shown (Massague, 2003). Functions of BMPs are also controlled extracellularly by secreted antagonistic regulators such as noggin, chordin, follistatin, and neurogenesin-1, which are reported to bind BMPs and inhibit their interaction with their receptors (Cho and Blitz, 1998; Ueki et al., 2003). Bone morphogenetic protein-5 (BMP5) is also a member of the TGF-b superfamily. BMP5 is reported to be expressed in the thymus, bone marrow, spleen, skeletal muscle, heart, kidney, lung, pancreas, and prostate (Bragdon et al., 2011), and function in bone and cartilage morphogenesis, limb development and connective soft tissue morphogenesis (Bragdon et al., 2011). It is also known that BMP5 mutations cause shortened, slightly ruffled external ears due to a defective cartilage framework (King et al., 1994). BMP5 has also been reported to be involved in many critical developmental phenomena in the central nervous system (CNS). For example, BMP5 is known to be abundantly expressed in the dorsal neuroepithelium and be involved in the genesis of noradrenergic locus coeruleus neurons (Vogel-Hopker and Rohrer, 2002; Tilleman et al., 2010). In addition, Brederlau et al. (2002) have shown that BMP5 also acts on precursors of the dopaminergic and astroglial lineage and induces their differentiation. Furthermore, in the developing peripheral nervous system, BMP5 has been reported to enhance dendritic growth in cultured sympathetic neurons (Beck et al., 2001). BMP5 expression has been investigated in the early development of the CNS (Furuta et al., 1997; VogelHopker and Rohrer, 2002). However, there is little information about BMP5 expression in the adult CNS. Furthermore, BMP receptors (Miyagi et al., 2011, 2012), and BMP antagonists (Mikawa and Sato, 2011, 2014), have been also known to be abundantly expressed in the adult rat CNS. Thus, It is needed to investigate BMP5 expression more widely and more in detail in the adult rat brain. In the present study, we show that BMP5 protein is widely
a
Department of Anatomy & Neuroscience, Hamamatsu University School of Medicine, 1-20-1 Handayama, Higashiku, Hamamatsu, Shizuoka 431-3192, Japan b Department of Rehabilitation, Seirei Hamamatsu General Hospital, 2-12-12 Sumiyoshi, Nakaku, Hamamatsu, Shizuoka 430-8558, Japan
Abstract—Bone morphogenetic protein-5 (BMP5), a member of the transforming growth factor-b (TGF-b) superfamily, has many effects in several biological events. Although BMP5 expression has been well reported in the early development of the central nervous system (CNS), there is little information about its expression in the adult CNS. Thus, we analyzed BMP5 expression in the adult rat CNS by immunohistochemistry. Abundant BMP5 expression was observed in most neurons, and their dendrites and axons. Furthermore, strong BMP5 expression was also detected in the neuropil of the gray matters with high plasticity, such as the molecular layer of the cerebellum, locus coeruleus, and nucleus of the solitary tract. In addition, we showed BMP5 expression also in astrocytes, ependymal cells and meninges. Our data suggest that BMP5 is widely expressed throughout the adult CNS, and this abundant expression in the adult brain strongly supports the idea that BMP5 plays important roles not only in the developing brain but also in the adult brain. Ó 2014 IBRO. Published by Elsevier Ltd. All rights reserved.
Key words: axon, neuropil, immunohistochemistry.
INTRODUCTION Bone morphogenetic proteins (BMPs) belong to the transforming growth factor b (TGF-b) superfamily (Bragdon et al., 2011). BMPs were at the beginning detected by their ability to promote ectopic bone formation, and are now reported to have various effects in sevGrant sponsor: The Ministry of Education, Science and Culture of Japan. *Corresponding author. Tel/fax: +81-53-435-2582. E-mail address: [email protected] (K. Sato). Abbreviations: BMP, bone morphogenetic protein; BMP5, bone morphogenetic protein-5; BMPR, bone morphogenetic protein receptor; BSA, bovine serum albumin; CNS, central nervous system; ELISA, enzyme-linked immunosorbent assay; GFAP, glial fibrillary acidic protein; IgG, immunoglobulin G; IR, like immunoreactivity; LTD, long-term depression; LTP, long term potentiation; NeuN, neuronal nuclei; PB, phosphate buffer; PBS, phosphate-buffered saline; SDS– PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; SVZ, subventricular zone; TGF-b, transforming growth factor b; TTBS, Tris-buffered saline. http://dx.doi.org/10.1016/j.neuroscience.2014.07.057 0306-4522/Ó 2014 IBRO. Published by Elsevier Ltd. All rights reserved. 972
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
expressed throughout the adult CNS, and that BMP5 protein is expressed in neurons, astrocytes, ependymal cells and meninges.
EXPERIMENTAL PROCEDURES Animals and section preparation Male Wister rats (n = 15, 7 weeks old; Japan SLC Inc., Shizuoka, Japan) were deeply anesthetized with diethyl ether, and then perfused with saline followed by 0.1 M phosphate buffer (PB, pH 7.4) containing 4% paraformaldehyde and 0.2% picric acid. The brains were quickly removed, and postfixed in the same fixative for 2 h at 4 °C. All brains were immersed in 10%, 20%, 25% buffered sucrose each, overnight at 4 °C, respectively. Frozen slices (20 lm for immunoperoxidase staining or 10 lm for immunofluorescence in thickness) were serially sectioned on a cryostat. All experiments conformed to the Guidelines for Animal Experimentation at Hamamatsu University School of Medicine on the ethical use of animals. Immunohistochemistry For immunoperoxidase staining, the sections were treated with 10% normal rabbit serum, 2% bovine serum albumin (BSA) and 0.2% Triton X-100 in 0.1 M PB for 2 h at room temperature (RT), and incubated further in goat anti-BMP5 (diluted 1:100, the final concentration, 1 lg/ml; R&D Systems, Inc., Minneapolis, MN, USA) overnight at 4 °C. After being washed with 0.1 M PB, the sections were incubated in rabbit anti-goat immunoglobulin G (IgG) with peroxidase complex (no dilution, ready-to-use; N-HistofineÒ Simple Stainä Mouse MAX PO (G); Nichirei, Tokyo, Japan) for 2 h at RT. After being washed with 0.1 M PB, immunoreaction was visualized with 3,30 -diaminobenzidine (Wako, Osaka, Japan). For double immunofluorescence with goat anti-BMP5 antibody (diluted 1:50; the final concentration, 2 lg/ml; R&D Systems, Inc.) and mouse anti-Tbx21 antibody (diluted 1:30; the final concentration, 17 lg/ml; Novus Biologicals, LLC, Littleton, CO, USA), or mouse antineuronal nuclei (NeuN) antibody (diluted 1:100; the final concentration, 10 lg/ml; Millipore, Temecula, CA, USA), or mouse anti-glial fibrillary acidic protein (GFAP) antibody (diluted 1:1000; the initial concentration is not available; Millipore), the sections were treated with 10% normal donkey serum, 2% BSA and 0.2% Triton X-100 in 0.1 M PB for 2 h at RT, and incubated further in goat anti-BMP5 antibody, mouse anti-NeuN antibody and mouse anti-GFAP antibody overnight at 4 °C and in mouse anti-Tbx21 antibody for three nights at 4 °C. After being washed with 0.1 M PB, sections were incubated in both Alexa Fluor 594 donkey anti-goat IgG (diluted 1:250; the final concentration, 8 lg/ml; Molecular Probes, Inc., Oregon, USA) and Alexa Fluor 488 donkey anti-mouse IgG (diluted 1:500; the final concentration, 4 lg/ml; Molecular Probes, Inc.) for 1.5 h at RT. Furthermore, for double immunofluorescence
973
with goat anti-BMP5 antibody (diluted 1:50) and mouse Fluoro anti-pan neuronal marker antibody cocktail (diluted 1:100; the initial concentration is not available; Millipore), the sections were treated with 10% normal donkey serum, 2% BSA and 0.2% Triton X-100 in 0.1 M PB for 2 h at RT, and incubated further in goat anti-BMP5 antibody overnight at 4 °C. After being washed with 0.1 M PB, the sections were incubated in both Alexa Fluor 594 donkey anti-goat IgG (diluted 1:250) and mouse Fluoro anti-pan neuronal marker antibody cocktail (diluted 1:100) for 1.5 h at RT. Brightfield and fluorescence images were recorded with an Eclipse 80i equipped with a DS-Ri CCD camera (Nikon, Tokyo, Japan) and were further processed by Adobe Photoshop (Tokyo, Japan). Western blotting A rat whole brain, or dissected rat cerebral cortex, hippocampus, thalamus, brain stem and cerebellum were homogenized in a 50 mM Tris–HCl (pH 7.4) buffer containing protease inhibitor cocktail (Nacalai Tesque, Kyoto, Japan) to avoid degradation of proteins, and solubilized by adding 2 sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE) sample buffer. The sample was separated by SDS– PAGE (14% acrylamide; PAGEL, ATTO Corporation, Tokyo, Japan) and then transferred to PVDF membrane (Immobilon-P; Merck Millipore, Tokyo, Japan) by electroblotting (40 V overnight at 4 °C) using a transfer buffer (25 mM Tris, 192 mM glycine) containing 20% methanol. The membranes were blocked with 3% BSA in 50 mM Tris-buffered saline containing 0.05% Tween20 (TTBS) for 2 h at RT and then incubated with goat anti-BMP5 antibody (diluted 1:200; the final concentration, 0.5 lg/ml; R&D Systems, Inc.) in TTBS containing 1% BSA for 40 min at RT. After being washed with TTBS, the blots were incubated with horseradish peroxidase-linked rabbit anti-goat Ig (diluted 1:4000; the final concentration, 0.125 lg/ml; KPL, Inc., Gaithersburg, MD, USA) in TTBS containing 1% BSA for 40 min at RT. After being washed with TTBS several times, signals were detected by SuperSignal West Pico chemiluminescent Substrate (Thermo Fisher Scientific, Inc., Waltham, MA, USA). The membranes were incubated with 15% H2O2 in phosphate-buffered saline (PBS) for 30 min to inactivate horseradish peroxidase. They were blocked with TTBS containing 3% BSA overnight at 4 °C and then incubated with rabbit antitubulin-b antibody (diluted 1:1000; the final concentration, 0.2 lg/ml; Thermo Fisher Scientific Inc.) in TTBS containing 1% BSA for 40 min at RT. After being washed with TTBS, the blots were incubated with horseradish peroxidase-linked goat anti-rabbit IgG (diluted 1:5000; the final concentration, 0.08 lg/ml; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) in TTBS containing 1% BSA for 40 min at RT. After being washed with TTBS several times, signals were detected by SuperSignal West Pico chemiluminescent Substrate (Thermo Fisher Scientific, Inc.).
974
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
Enzyme-linked immunosorbent assay (ELISA) Solutions of 100 ng/ml of the antigen (BMP5: PeproTech Inc., Rocky Hill, NJ, USA; BMP6: PeproTech Inc.; and BMP7: HumanZyme Inc., Chicago, IL, USA) in 0.1 M carbonate buffer (pH 9.6) were prepared, 50 ll of the antigen solutions was added to each well (96-well EIA/ RIA Plate; Corning Incorporated, Corning, NY, USA) and then were incubated overnight at 4 °C. After washing the plate with 0.01 M PBS, the wells were filled with 3% BSA in 0.1 M carbonate buffer and then were incubated overnight at RT. After the plate was washed three times with 0.01 M PBS, 50 ll of anti-BMP5 antibody (diluted 1:100; the final concentration, 1.0 lg/ ml; R&D Systems, Inc.) was added to each well and then incubated for 2 h at RT. After the plate was washed three times with 0.01 M PBS, 50 ll of horseradish peroxidase-conjugated rabbit anti-goat Ig (diluted 1:1000; the final concentration, 0.5 lg/ml; KPL, Inc.) was added to each well and then incubated for 2 h at RT. After the plate was again washed three times with 0.01 M PBS, 200 ll of ABTS substrate solution (Roche Applied Science, Basel, Switzerland) was added to each well and then incubated for 30 min at RT. To quantitate the binding, the results were read at 405 nm with a microplate reader.
RESULTS Because the amino acids sequences of BMP5, BMP6, and BMP7 are very similar, we first investigated whether the antibody (goat anti-BMP5) can discriminate among them by using the ELISA method. As shown in Fig. 1A, B, the antibody could specifically recognize only BMP5
protein. In addition, Western blotting with the antibody for the whole rat brain exhibited a single band of about 17 kDa (Fig. 1C), the size coinciding with the theoretically calculated value. Furthermore, as controls, we performed immunohistochemistry without the first antibody, the control sections did not show any staining throughout the CNS (Fig. 2A2–H2). Taken together, these data indicate that the antibody can specifically recognize BMP5 protein. General expression patterns We show the overview of BMP5 expression in the adult rat brain in Fig. 2. BMP5-like immunoreactivity (IR) was seen throughout the adult rat brain. We observed abundant BMP5-IR in the olfactory bulb (Fig. 2A1), basal ganglia (Fig. 2B1, C1), cerebral cortex (Fig. 2C1–E1), hippocampus (Fig. 2D1, E1), thalamus (Fig. 2D1), hypothalamus (Fig. 2D1, E1), midbrain (Fig. 2E1), cerebellum (Fig. 2F1), brainstem (Fig. 2F1, G1), and spinal cord (Fig. 2H1). Moreover, abundant BMP5-IR is also seen located the border zones, such as the pia mater and ependyma throughout the CNS (Fig. 2). In addition, using the same antibody, we also performed Western blotting for dissected brain regions. Fig. 3 shows that a single band corresponding to BMP5 protein was detected in the cortex, hippocampus, thalamus, brain stem, and cerebellum, respectively, further supporting the data obtained by immunohistochemistry. We summarized the relative intensity of BMP5-IR in the rat CNS in Table 1. Telencephalon Olfactory bulb. BMP5-IR was abundantly observed throughout the olfactory bulb, where very strong expression was seen in the olfactory nerve layer (Fig. 4A, D). In the glomerular layer, we detected moderately stained periglomerular neurons (arrowheads in Fig. 4B) and glomeruli also showed weak neuropil staining (asterisks in Fig. 4B). In the external plexiform layer, strongly stained tufted cells and moderately stained neuropil were seen (Fig. 4A). In the mitral cell layer, the cell bodies of mitral cells were strongly labeled (arrowheads in Fig. 4C). In the granular layer, we observed moderately stained granular neurons (Fig. 4A). Interestingly, very strong BMP5-IR was also observed on the pia mater (arrows in Fig. 4D) and perivascular space (arrowheads in Fig. 4D).
Fig. 1. Specificity of the anti-BMP5 antibody. Wells with BMP5 antibody show intense ELISA reaction only when coated by BMP5 protein and not coated by BMP6 and BMP7 protein. Wells without the first antibody (control) show no ELISA reaction (A). Western blot analysis using anti-BMP5 antibody for the whole rat brain shows a single band of about 17 kDa (B).
Septum and nuclei of the diagonal band of Broca. In the medial septal nuclei and diagonal band of Broca, many neurons were strongly stained, in addition, weak neuropil staining was also seen (Fig. 4E). The lateral septal nucleus showed strong BMP5-IR (Fig. 4F). Interestingly, very strong BMP5-IR was also observed on the ependyma of the lateral ventricle (arrowheads in Fig. 4F). Ventral pallidum and islands of Calleja. In the ventral pallidum, large-sized neurons showed strong BMP5-IR
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
975
Fig. 2. BMP5 expression in the whole rat brain (A1, B1, C1, D1, E1, F1, G1, H1). Control sections without the primary antibody (A2, B2, C2, D2, E2, F2, G2, H2). 12, hypoglossal nucleus; Amy, amygdala; CN, cerebellar nuclei; Cor, cerebral cortex; CPu, caudate putamen; DH, dorsal horn; Hi, hippocampus; IO, inferior olive; MG, medial geniculate; MV, medial vestibular nucleus; OB, olfactory bulb; Sol, nucleus of the solitary tract; Sp5, spinal trigeminal nucleus; Th, thalamus, VH, ventral horn. Scale bar = 0.5 mm for A, G, H; 1 mm for B–F.
976
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
in the somata (arrowheads in Fig. 4G) and dendrites (arrow in Fig. 4G). Many moderately stained neurons were scattered in the islands of Calleja (Fig. 4H).
Piriform cortex. In layers I and III, moderate BMP5-IR was observed in neuropil (Fig. 4I). In layer II, we observed many moderately stained pyramidal neurons (arrowheads in Fig. 4I). Cerebral cortex. Strong BMP5-IR-positive cells and moderate neuropil staining were seen throughout the cerebral cortex (Fig. 5A). Interestingly, very strong BMP5-IR was also observed on the pia mater of the cerebral cortex (arrows in Fig. 5A). In the layer I, where short-axon cells make synapses with the apical dendrites of pyramidal neurons, strong BMP5-IR was detected in the terminals of apical dendrites (arrows in Fig. 5B) and neuropil (asterisks in Fig. 5B). In the layer V, the cell bodies of pyramidal neurons were strongly stained (arrowheads in Fig. 5C). Interestingly, we observed strong BMP5-IR also in the apical dendrites of pyramidal neurons (arrows in Fig. 5C).
Fig. 3. Western blot analysis for dissected brain regions. BS, brain stem; Cor, cerebral cortex; Cer, cerebellum; Hip, hippocampus; Th, thalamus.
Hippocampus. BMP5-IR was seen throughout the hippocampus (Fig. 5D). We observed moderate BMP5IR in pyramidal cells of Ammon’s horn (arrowheads in
Table 1. Distribution and intensity of BMP5-IR in the rat CNS Area I. Telencephalon Olfactory bulb Olfactory nerve layer Glomerular layer External plexiform layer Mitral cell layer Granular layer Subependymal layer Cerebral cortex Hippocampal formation Amygdala Basal ganglia Caudate putamen Globus pallidus Islands of Calleja Corpus callosum II. Diencephalon Thalamus Reticular nucleus Medial habenular nucleus Lateral habenular nucleus Ventroposterior nucleus Lateral geniculate nucleus Other nuclei Hypothalamus Supraoptic nucleus Paraventricular hypothalamic nucleus Ventromedial hypothalamic nucleus Other nuclei III. Midbrain Red nucleus Substantia nigra Interpeduncular nucleus IV. Pons and medulla Motor system Oculomotor nucleus Trigeminal motor nucleus Facial nucleus
Intensity
++++ ++ ++ +++ ++ + +++ ++ + ++ ++ ++ +
++ ++ + ++ ++ + ++ + + + +++ ++ +++
+++ +++ +++
Area
Intensity
Hypoglossal nucleus General somatosensory system Trigeminal mesencephalic nucleus Trigeminal spinal nucleus Cuneate nucleus Gracile nucleus General visceromotor system Dorsal nucleus of the vagus General viscerosensory system The nucleus of the solitary tract Special somatosensory system Auditory system Dorsal cochlear nucleus Ventral cochlear nucleus Inferior colliculus Vestibular system Medial vestibular nucleus Lateral vestibular nucleus Visual system Superior colliculus Pontine dorsal tegmental nucleus Pontine nucleus Locus coeruleus Inferior olive V. Cerebellum Cortex Molecular layer Purkinje cell layer Granule cell layer Cerebellar nuclei VI. Spinal cord Dorsal horm Ventral horn VII. Other areas Meninges Ependyma Choroidal plexus
+++
Relative intensities were estimated by visual comparison of immunostained slide: +, low; ++, moderate; +++, strong; ++++, very strong.
++ +++ +++ +++ ++++ ++++
+++ +++ ++ +++ ++ +++ ++++ +++ ++++ +++
+++ ++++ ++ +++ ++++ +++ ++++ ++++ ++++
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
977
Fig. 4. BMP5 expression in the olfactory bulb (A–D), medial septal nucleus (E), lateral septal nucleus (F), ventral pallidum (G), islands of Calleja (H) and piriform cortex (I). Note that the pia mater (arrows in D), perivascular space (arrowheads in D), and ependyma of the lateral ventricle (arrowheads in F) were very strongly stained. In the olfactory bulb, the soma of mitral cells (arrowheads in C) was strongly stained, and periglomerular neurons (arrowheads in B) were moderately stained, and glomeruli (asterisks in B) were weakly stained. In the ventral pallidum, the somas (arrowheads in G) and the dendrites (arrow in G) of neurons were strongly stained. In the piriform cortex, pyramidal neurons (arrowheads in I) were moderately stained. I–III, layers I–III of the piriform cortex; DB, diagonal band; EPI, external plexiform layer; Gl, glomerular layer; Gr, granular layer; ICj, islands of Calleja; LS, lateral septal nucleus; Mi, mitral cell layer; MS, medial septal nucleus; ON, olfactory nerve layer; Pir, piriform cortex; SE, subependymal layer; VP, ventral pallidum. Scale bar = 400 lm for E; 160 lm for A, I; 80 lm for F; 40 lm for B–D, G, H.
Fig. 5E). Furthermore, the apical dendrites of pyramidal cells also showed strong BMP5-IR (arrows in Fig. 5E). In the dentate gyrus, moderate BMP5-IR was seen in granule cells, and we also observed moderate neuropil staining in the polymorphic layer and stratum molecular (Fig. 5F). Corpus callosum. BMP5-IR-positive cells were scattered in the corpus callosum, (arrowheads in
Fig. 5G). In addition, we also observed weak neuropil staining (Fig. 5G). The basal ganglia. In the caudate putamen, we observed moderately positive neurons and moderate neuropil staining in the gray matter (Fig. 6A). In the globus pallidus, moderately positive neurons with stained dendrites and weak neuropil staining were observed in the gray matter (Fig. 6B).
978
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
Fig. 5. BMP5 expression in the cerebral cortex (A–C), hippocampus (D–F) and corpus callosum (G). In the cerebral cortex, the pia mater (arrows in A) was very strongly stained, and the apical dendrites (arrows in B and C) and the somas (arrows in C) of pyramidal cells in the layer V, and neuropil in the layer I (asterisks in B) were strongly stained. In the hippocampus, the apical dendrites of pyramidal cells (arrows in E) were strongly stained, and pyramidal cells of the Ammon’s horn (arrowheads in E) were moderately stained, and weak BMP5-IR-positive cells were observed in the corpus callosum (arrowheads in G). I–VI, layers I–VI; cc, corpus callosum; CA1, 3, field CA1, 3 of Ammon’s horn; DG, dentate gyrus; Gr, granular layer; Mo, stratum molecular; Or, stratum oriens; Po, polymorphological layer; Py, stratum pyramidale; Ra, stratum radiatum. Scale bar = 400 lm for D; 160 lm for A; 80 lm for F; 40 lm for B, C, E, G.
The amygdala. In all nuclei of the amygdala, we detected weak BMP5-IR (Fig. 2D1). Diencephalon In all nuclei in the thalamus, we observed BMP5-IRpositive neurons and neuropil (Fig. 2D1, Fig. 6C).
Interestingly, very strong BMP5-IR was also seen in the pia mater (arrows in Fig. 6C). In the medial habenular nucleus, moderate BMP5-IR-positive neurons and moderate neuropil staining were seen (Fig. 6D), while, in the lateral habenular nucleus, BMP5-IR was weak (Fig. 6D). Moderately positive neurons and weak neuropil staining were observed in the reticular thalamic
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
979
Fig. 6. BMP5 expression in the basal ganglia (A, B), thalamus (C–G) and hypothalamus (H–K). Note that the pia mater (arrows in C), perivascular space (arrowheads in E and F), and the ependyma of the third ventricle (arrowheads in K) were very strongly stained. CPu, caudate putamen; GP, globus pallidus; LG, lateral geniculate nucleus; LHb, lateral habenular nucleus; MHb, medial habenular nucleus; Po, posterior thalamic nucleus; PV, paraventricular hypothalamic nucleus; Rt, reticular thalamic nucleus; SO, supraoptic nucleus; VMH, ventromedial hypothalamic nucleus; VP, ventroposterior thalamic nucleus. Scale bar = 400 lm for C; 160 lm for G, I, K; 80 lm for A, B, D, H, J; 40 lm for E, F.
980
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
nucleus (Fig. 6E). We also observed moderately positive neurons and weak neuropil staining in the ventroposterior thalamic nucleus (Fig. 6F). Interestingly, BMP5-IR was also seen in the perivascular space of the nuclei (arrowheads in Fig. 6E, F). In the lateral geniculate nucleus, we observed moderately stained neurons and weakly stained neuropil (Fig. 6G). In the hypothalamus, we detected BMP5-IR-positive neurons and neuropil staining in all nuclei (Fig. 2D1, E1). Moderate BMP5-IR was observed in the supraoptic nucleus (Fig. 6H). In addition, weak BMP5-IR was seen in the paraventricular hypothalamic nucleus (Fig. 6I) and ventromedial hypothalamic nucleus (Fig. 6J). Interestingly, we also observed very strong BMP5-IR in the ependyma of the hypothalamus (arrowheads in Fig. 6K). Midbrain The interpeduncular nucleus exhibited strong BMP5-IR (Fig. 7A). We observed moderate BMP5-IR in the central gray (Fig. 7A). Interestingly, BMP5-IR was also seen in the pia mater (arrows in Fig. 7A) and ependyma in this region (arrowheads in Fig. 7A). In the red nucleus, large neurons having many neurites exhibited strong BMP5-IR (arrowheads in Fig. 7B), while neuropil was stained weakly (Fig. 7B). Moderate BMP5-IR was detected in the substantia nigra (Fig. 7A). Moderately stained neurons (arrowheads in Fig. 7C) and weak neuropil staining (Fig. 7C) were seen in the pars compacta of the substantia nigra. Pons and medulla Motor system. The nuclei of the general somatomotor and branchiomotor system, such as the oculomotor nucleus, motor trigeminal nucleus (Fig. 7D), facial nucleus (Fig. 7E) and hypoglossal nucleus (Fig. 7G), exhibited abundant BMP5-IR in neuronal cell bodies and their dendrites. Interestingly, in the facial nerve BMP5IR was observed in axon fibers (arrows in Fig. 7F). General somatosensory system. Large primary afferent neurons were stained moderately in the trigeminal mesencephalic nucleus (arrowheads in Fig. 7H). In the trigeminal spinal nucleus, strongly stained neurons were observed (Fig. 7I). Strongly positive neurons were also seen in the cuneate and gracile nuclei (Fig. 7J).
Special somatosensory system. Auditory system: The dorsal and ventral cochlear nuclei contained strongly positive neurons (Fig. 8C). In the inferior colliculus, moderate BMP5-IR was detected. Vestibular system: we observed strongly stained small neurons and moderate neuropil staining in the medial vestibular nucleus (Fig. 8D). Interestingly, BMP5-IR was also seen in the ependyma of the fourth ventricle (arrowheads in Fig. 8D). In the lateral vestibular nucleus, large neurons exhibited moderate BMP5-IR, while neuropil staining was weak (Fig. 8E). Interestingly, we found strongly stained axons in the vestibulocochlear nerve (arrows in Fig. 8F). Visual system: In the superior colliculus, we observed strongly stained neurons and moderate neuropil staining (Fig. 8G). The superficial part showed relatively strong BMP5-IR than the other parts (asterisks in Fig. 8G). Interestingly, very strong BMP5-IR was also seen in the pia mater above the superior colliculus (arrows in Fig. 8G). Other lower brain stem areas. We observed very strong BMP5-IR was observed in the pontine dorsal tegmental nucleus (Fig. 9A, B). Strongly stained neurons and moderate neuropil staining was also seen in the pontine nucleus (Fig. 9C). Interestingly, we observed very strongly stained neurons and very strong neuropil staining in the locus coeruleus (Fig. 9D, E). Interestingly, BMP5-IR was also seen in the ependyma of the fourth ventricle (arrowheads in Fig. 9A, D). In the gigantocellular reticular nucleus, the cell bodies (arrowheads in Fig. 9F) and dendrites (arrows in Fig. 9F) of large neurons were very strongly stained. In the inferior olive, we observed strongly stained neurons and moderate neuropil staining (Fig. 9G). Cerebellum In the cerebellum, we observed very strong staining in the Purkinje cell layer. Very strong BMP5-IR was seen in the cell bodies of Purkinje neurons (Fig. 10A). In addition, small Bergmann glias around the cell bodies of Purkinje cells also seemed to show very strong BMP5-IR. The molecular cell layer contained many strongly positive neurons, and strong neuropil staining was also observed (Fig. 10A). Moderate BMP5-IR was also seen in granular cells (Fig. 10A). Strong neuronal cell body staining and moderate neuropil staining were observed in the cerebellar nuclei (Fig. 10B). Spinal cord
General visceromotor system. We observed many very strong BMP5-IR-positive neurons and strong neuropil staining in the dorsal nucleus of the vagus (Fig. 8A, B). General viscerosensory system. In the nucleus of the solitary tract, very strong BMP5-IR-positive neurons and strong neuropil staining were seen (Fig. 8A, B). Interestingly, BMP5-IR was also seen in the ependyma of the fourth ventricle (arrowheads in Fig. 8A).
We observed abundant BMP5-IR in the gray matter (Fig. 10C). Furthermore, also in the white matter, we found that all fiber tracts contain BMP5-IR (Fig. 10C). Further investigation showed that in the white matter, besides strongly stained astrocyte-like cells (arrowheads in Fig. 10D), a lot of axons exhibited abundant BMP5-IR (arrows in Fig. 10D). We also observed very strong neuropil and neuronal cell body staining in the layers I and II of the dorsal horn (Fig. 10E, F). In addition, large motor neurons in the ventral horn also showed strong
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
981
Fig. 7. BMP5 expression in the midbrain (A–C), motor system (D–G) and general somatosensory system (H–J). Note that the pia mater (arrows in A) and the ependyma (arrowheads in A) were very strongly stained, that neurons in the red nucleus (arrowheads in B) and the axons in the facial nerve (arrows in F) were strongly stained, and that neurons in the pars compacta of the substantia nigra (arrowheads in C) and the trigeminal mesencephalic nucleus (arrowheads in H) were moderately stained. 5, motor trigeminal nucleus; 7, facial nucleus; 7n, facial nerve; 12, hypoglossal nucleus; CG, central gray; Cu, Cuneate nucleus; Gr, Gracile nucleus; IP, interpeduncular nucleus; Me5, trigeminal mesencephalic nucleus; MG, medial gray; R, red nucleus; SN, substantia nigra; Sp5, trigeminal spinal nucleus. Scale bar = 400 lm for A; 80 lm for H, J; 40 lm for B–G, I.
982
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
Fig. 8. BMP5 expression in the general viscerosensory system (A, B) and special somatosensory system (C–G). Note very strong BMP5-IR in the nucleus of the solitary tract and dorsal nucleus of the vagus (A, B). Note that the ependyma of the fourth ventricle (arrowheads in A and D) and the pia mater (arrows in G) were very strongly stained, and that axons in the vestibulocochlear nerve (arrows in F) were strongly stained. 8n, vestibulocochlear nerve; 10, vagus nucleus; 12, hypoglossal nucleus; DC, dorsal cochlear nucleus; LVe, lateral vestibular nucleus; MVe, medial vestibular nucleus; SC, superior colliculus; Sol, nucleus of the solitary tract; VC, ventral cochlear nucleus. Scale bar = 160 lm for A, C–E, G; 80 lm for B; 40 lm for F.
BMP5-IR in the cell bodies (arrowheads in Fig. 10G) and dendrites (arrows in Fig. 10G). Other areas In the subventricular zone (SVZ), we observed very strong BMP5-IR in ependymal cells, cell bodies and neuropils (Fig. 10H). In addition, choroidal plexus also expressed abundant BMP5 proteins (Fig. 10I). Double fluorescence immunohistochemistry Using double fluorescence immunohistochemistry with anti-Tbx21 (a specific marker for mitral neurons)
antibody, we tried to confirm that mitral neurons express BMP5 in the olfactory bulb. As shown in Fig. 11A1-3, co-expression of BMP5 and Tbx21 was seen in the mitral cell layer, indicating that mitral neurons express BMP5 proteins. In addition, using anti-NeuN (a pan neuronal marker) antibody, we tried to validate that granular neurons in the olfactory bulb express BMP5. As shown in Fig. 11B1-3, co-expression of BMP5 and NeuN was seen in the granular cell layer, indicating that granular neurons express BMP5 proteins. Furthermore, to investigate that neurons in the caudate putamen, reticular thalamus, cerebral cortex, and hippocampus express BMP5 proteins, we performed double fluorescence immunohistochemistry with anti-NeuN
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
983
Fig. 9. BMP5 expression in the other lower brain stem areas (A–G). Note that the locus coeruleus (D, E) and the ependyma of the fourth ventricle (arrowheads in A and D) show very strong BMP5-IR. In the gigantocellular reticular nucleus, the somas (arrowheads in F) and the dendrites (arrows in F) of neurons were very strongly stained. IO, inferior olive; LC, locus coeruleus; Me5, mesencephalic trigeminal nucleus; PDTg, pontine dorsal tegmental nucleus; Pn, pontine nucleus; Rt, reticular nucleus. Scale bar = 160 lm for A, D; 80 lm for G; 40 lm for B, C, E, F.
antibody. Fig. 11C–F clearly shows that neurons in these regions express BMP5 proteins. To confirm that BMP5-IR-positive cells in the white matter and hippocampus are astrocytes. We performed double fluorescence immunohistochemistry with antiGFAP antibody. As shown in Fig. 11G, H, co-expression of BMP5 and GFAP was seen in the corpus callosum and hippocampus, indicating that astrocytes express BMP5 proteins. Furthermore, we performed double fluorescence immunohistochemistry using anti-panneuronal marker antibody cocktail in the facial nerve and vestibulocochlear nerve to confirm that axons exactly express BMP5 proteins. As shown in Fig. 11I, J, pan-neuronal marker-IR-positive structures co-expressed
BMP5, indicating that axons express BMP5 proteins in both nerves.
DISCUSSION Up to the present, little information is available for BMP5 expression in the adult brain. In this study, we first exhibit that BMP5 is widely expressed throughout the adult CNS. Moreover, besides abundant BMP5 expression in neurons, we observed BMP5 expression in astrocytes, ependymal cells and meninges. These data suggest that BMP5 is produced by neurons, astrocytes, ependymal cells and meninges throughout the adult CNS. So far BMP receptors (Miyagi et al., 2011, 2012)
984
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
Fig. 10. BMP5 expression in the cerebellum (A, B), spinal cord (C–G), subventricular zone (H) and choroidal plexus (I). Note that axons (arrows in D) and astrocyte-like cells (arrowheads in D) in the white matter and the somas (arrowheads in G) and the dendrites (arrows in G) of motor neurons in the ventral horn were strongly stained. Note very strong BMP-IR in the Purkinje layer (A) and layers I and II of the dorsal horn of the spinal cord (F). I–III, layers I–III of the spinal cord; Ch, choroidal plexus; CN, cerebellar nuclei; DH, dorsal horn; Gr, granular layer; Mol, molecular layer; Pur, Purkinje cell layer; SVZ, subventricular zone; VH, ventral horn; wm, white matter. Scale bar = 400 lm for C; 160 lm for B, E; 80 lm for F, I; 40 lm for A, G, H; 16 lm for D.
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
985
Fig. 11. Double staining study showing that BMP5-IR-positive cells in the mitral cells (A1) are also positive for Tbx21 (A2), BMP5-IR-positive cells in the granular cells (B1), caudate putamen (C1), reticular thalamic nucleus (D1), cerebral cortex (E1) and hippocampus (F1) are also positive for NeuN (B2–F2), BMP5-IR-positive cells in the corpus callosum (G1) and hippocampus (H1) are also positive for GFAP (G2, H2), BMP5-IR-positive axons in the facial nerve (I1) and vestibulocochlear nerve (J1) are also positive for pan-neuronal marker (I2, J2) in a merged photomicrograph (A3–J3). cc, corpus callosum; Cor, cerebral cortex; CPu, caudate putamen; Gr, granular cell; Hip, hippocampus; Mi, mitral cell; Rt, reticular thalamic nucleus; 7n, facial nerve; 8n, vestibulocochlear nerve. Scale bar = 20 lm.
986
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
and BMP antagonists (Mikawa and Sato, 2011, 2014) have been reported to be abundantly expressed in the adult rat CNS. BMP5 might take on important roles as an agonist regulating BMP signaling in the adult brain. BMP5 expressions in neurons We found that almost all neuronal cell bodies express BMP5 proteins with different intensities. What are the roles of BMP5 secreted from neuronal cell bodies? BMP5 has been reported to control the differentiation of various neurons. For example, BMP5 is known to be abundantly expressed in the dorsal neuroepithelium and be involved in the genesis of noradrenergic locus coeruleus neurons (Vogel-Hopker and Rohrer, 2002; Tilleman et al., 2010). Interestingly, even in the adult CNS, we also found very strong BMP5 expression in the locus coeruleus. In the locus coeruleus, both cell bodies of noradrenergic neurons and neuropil were strongly stained (Fig. 9D, E), suggesting that even in the adult noradrenergic neurons need abundant BMP5 signaling to keep their phenotype. In addition, Brederlau et al. (2002) have reported that BMP5 also acts on precursors of the dopaminergic and induces their differentiation. Interestingly, we also found that BMP5 is expressed in dopaminergic neurons (Fig. 7C), further supporting the idea that BMP5 signaling may be necessary to keep the identities of various neurons even in the adult CNS. BMP5 expressions in dendrites Interestingly, we found intense BMP5 protein expressions in apical dendrites of pyramidal neurons in the cerebral cortex, and hippocampus. In addition, we found many BMP5-IR-positive dendrites throughout the CNS. What are the functions of BMP5 in dendrites? BMP signalings have been previously reported to be involved in modulating dendritic morphology (Lein et al., 1995; Le Roux et al., 1999; Withers et al., 2000). Withers et al. showed that addition of BMP7 to cultured hippocampal neurons promotes dendritic growth, and also enhances synaptogenesis (Withers et al., 2000). Moreover, Althini et al. (2004) have recently reported that BMP4 potentiates neurotrophin 3 and neurturin-induced neurite outgrowth of peripheral neurons from the E9 chicken embryo. Furthermore, BMP2 is also reported to promote mitral/ tufted cell dendritic outgrowth (Tran et al., 2008). Interestingly, BMP5 has been also reported to promote dendritic growth in cultured sympathetic neurons (Beck et al., 2001). Taken together, BMP5 released from dendrites may regulate dendrite morphology and synaptic homeostasis. BMP5 expressions in axons Surprisingly, we observed that BMP5 protein is expressed in many kinds of axons, such as in the facial nerve, and vestibulocochlear nerve. Our data revealed, almost all axons are positive for BMP5 expression. What is the physiological significance of BMP5 protein expressed in axons? Retrograde BMP signaling is known to modulate
neuronal terminal differentiation and synaptic efficacy (da Silva and Wang, 2011). For example, in Drosophila, the BMP homolog Glass bottom boat (Gbb) acts as a muscle-derived retrograde signal that enhances synaptic growth and neurotransmitter release (McCabe et al., 2003). In mammals, target-derived retrograde BMP signaling has been also shown to determine the number of neurons in the trigeminal ganglion (Guha et al., 2004), and to regulate trigeminal sensory neuron identities (Hodge et al., 2007). Furthermore, Xiao et al. (2013) have reported that BMP5 plays a role in the development of a large and fast CNS synapse, and also in the elimination of competing synaptic inputs. These data strongly suggest that, also in the mammalian brain, many neurons are involved in this kind of target-derived retrograde BMP signal by using BMP5. If so, there are two possibilities for the origin of BMP5 protein observed in the axon. One is that BMP5 protein is anterogradely transported from the cell body to the axon terminal. Another possibility is that BMP5 protein released from the postsynaptic neurons or surrounding astrocytes is up-taken by the presynapse and retrogradely transported to the cell body. BMP5 expressions in the neuropil In this study, we found intense BMP5-IR in the neuropil of the molecular layer of the cerebellum (Fig. 10A), the nucleus of the solitary tact (Fig. 8B), and the locus coeruleus (Fig. 9D). The molecular layer of the cerebellum is known to have high plasticity, such as long-term depression (LTD) and long term potentiation (LTP) (Ito, 2001). In the molecular layer of the cerebellum, parallel fiber axons of granule cells and climbing fiber axons from the inferior olive neurons make synapses with the dendrites of Purkinje cells. Interestingly, we detected abundant BMP5-IR in the inferior olive neurons, granule cells, and Purkinje cells, suggesting that BMP5 in the neuropil might play a role to keep plasticity by regulating important phenomena, such as LTD and LTP. BMP5 expressions in ependymal cells and astrocytes In this study, we showed that ependymal cells express BMP5 protein. It has been reported that, in the adult SVZ, a BMP antagonist noggin released from the ependymal cells makes a niche for adult neurogenesis (Lim et al., 2000). In the present study, we also found BMP5 is also expressed in the SVZ, suggesting that BMP5 plays a role in the regulation of neurogenesis. Furthermore, we showed that BMP5 protein is expressed in astrocytes. What does BMP5 released from astrocytes do? One possibility may be that BMP5 secreted from astrocytes affects astrocytes themselves via autocrine or paracrine manners. Interestingly, BMP5 is also known to control astroglial lineage and induce their differentiation. These findings raise the possibility that BMP5 might also play a role in regulating the functions of astrocyte. Another possibility may be that BMP5 secreted from astrocytes control local environment, such as synaptic conditions.
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
BMP5 expressions in the meninges and choroidal plexi In this study, we found that the meninges express BMP5 protein. What is the role of BMP5 in the meninges? Interestingly, other members of BMP proteins, such as BMP4 and BMP7 have been reported to be expressed in the meninges (Machida et al., 2014). In addition, Choe et al. (2012) have also reported that meninges produce BMP7 in the embryonic brain. And the overexpression of BMP7 inhibits callosal axon outgrowth, resulting in callosal agenesis. Recently, it has been proposed that the brain and circulating immune cells engage in a continuous dialog that takes place within the brain’s territory, though outside the parenchyma, occurring within the brain’s borders – the choroidal plexi, the brain meninges and the cerebrospinal fluid (Ron-Harel et al., 2011). Therefore, one possibility might be that BMP5 released from the meninges and choroidal plexi also plays a role in this kind of dialog by giving BMP signals to the neighbor structures.
REFERENCES Althini S, Usoskin D, Kylberg A, Kaplan PL, Ebendal T (2004) Blocked MAP kinase activity selectively enhances neurotrophic growth responses. Mol Cell Neurosci 25:345–354. Beck HN, Drahushuk K, Jacoby DB, Higgins D, Lein PJ (2001) Bone morphogenetic protein-5 (BMP-5) promotes dendritic growth in cultured sympathetic neurons. BMC Neurosci 2:12. Bragdon B, Moseychuk O, Saldanha S, King D, Julian J, Nohe A (2011) Bone morphogenetic proteins: a critical review. Cell Signal 23:609–620. Brederlau A, Faigle R, Kaplan P, Odin P, Funa K (2002) Bone morphogenetic proteins but not growth differentiation factors induce dopaminergic differentiation in mesencephalic precursors. Mol Cell Neurosci 21:367–378. Cho KW, Blitz IL (1998) BMPs, Smads and metalloproteases: extracellular and intracellular modes of negative regulation. Curr Opin Genet Dev 8:443–449. Choe Y, Siegenthaler JA, Pleasure SJ (2012) A cascade of morphogenetic signaling initiated by the meninges control corpus callosum formation. Neuron 73:698–712. da Silva S, Wang F (2011) Retrograde neural circuit specification by target-derived neurotrophins and growth factors. Curr Opin Neurobiol 21:61–67. Furuta Y, Piston DW, Hogan BL (1997) Bone morphogenetic proteins (BMPs) as regulators of dorsal forebrain development. Development 124:2203–2212. Guha U, Gomes WA, Samanta J, Gupta M, Rice FL, Kessler JA (2004) Target-derived BMP signaling limits sensory neuron number and the extent of peripheral innervation in vivo. Development 131:1175–1186. Hodge LK, Klassen MP, Han BX, Yiu G, Hurrell J, Howell A, Rousseau G, Lemaigre F, Tessier-Lavigne M, Wang F (2007) Retrograde BMP signaling regulates trigeminal sensory neuron identities and the formation of precise face maps. Neuron 55:572–586. Ito M (2001) Cerebellar long-term depression: characterization, signal transduction and functional roles. Physiol Rev 81:1143–1195.
987
King JA, Marker PC, Seung KJ, Kingsley DM (1994) BMP5 and the molecular, skeletal, and soft-tissue alterations in short ear mice. Dev Biol 166:112–122. Le Roux P, Behar S, Higgins D, Charette M (1999) OP-1 enhances dendritic growth from cerebral cortical neurons in vitro. Exp Neurol 160:151–163. Lein P, Johnson M, Guo X, Rueger D, Higgins D (1995) Osteogenic protein-1 induces dendritic growth in rat sympathetic neurons. Neuron 15:597–605. Lim DA, Tramontin AD, Trevejo JM, Herrera DG, Gracia-Verdugo JM, Alvarez-Buylla A (2000) Noggin antagonizes BMP signaling to create a niche for adult neurogenesis. Neuron 28:713–726. Machida A, Okuhara S, Harada K, Iseki S (2014) Difference in apical and basal growth of the frontal bone primordium in Foxc1ch/ch mice. Congenit Anom (Kyoto) 54:172–177. Massague J (2003) Integration of Smad and MAPK pathways: a link and a linker revisited. Genes Dev 17:2993–2997. McCabe BD, Marques G, Haghighi AP, Fetter RD, Crotty ML, Haerry TE, Goodman CS, O’Connor MB (2003) The BMP homolog Gbb provides a retrograde signal that regulates synaptic growth at the Drosophila neuromuscular junction. Neuron 39:241–254. Mikawa S, Sato K (2011) Noggin expression in the adult rat brain. Neuroscience 184:38–53. Mikawa S, Sato K (2014) Chordin expression in the adult rat brain. Neuroscience 258:16–33. Miyagi M, Mikawa S, Hasegawa T, Kobayashi S, Matsuyama Y, Sato K (2011) Bone morphogenetic protein receptor expressions in the adult rat brain. Neuroscience 176:93–109. Miyagi M, Mikawa S, Sato T, Hasegawa T, Kobayashi S, Matsuyama Y, Sato K (2012) BMP2, BMP4, noggin, BMPRIA, BMPRIB, and BMPRII are differentially expressed in the adult rat spinal cord. Neuroscience 203:12–26. Moustakas A, Heldin CH (2009) The regulation of TGFbeta signal transduction. Development 136:3699–3714. Ron-Harel N, Cardon M, Schwartz M (2011) Brain homeostasis is maintained by danger signals stimulating a supportive immune response within the brain borders. Brain Behav Immun 25:1036–1043. Tilleman H, Hakim V, Novikov O, Liser K, Nashelsky L, Di Salvio M, Krauthammer M, Scheffner O, Maor I, Mayseless O, Meir I, Kayam G, Sela-Donenfeld D, Simeone A, Brodski C (2010) Bmp5/7 in concert with the mid-hindbrain organizer control development of noradrenergic locus coeruleus neurons. Mol Cell Neurosci 45:1–11. Tran H, Chen H, Walz A, Posthumus JC, Gong Q (2008) Influence of olfactory epithelium on mitral/tufted cell dendritic outgrowth. PLoS One 3:e3816. Ueki T, Tanaka M, Yamashita K, Mikawa S, Qiu ZF, Maragakis NJ, Hevner RF, Miura N, Sugimura H, Sato K (2003) A novel secretory factor, neurogenesin-1, provides neurogenic environmental cues for neural stem cells in the adult hippocampus. J Neurosci 23:11732–11740. Vogel-Hopker A, Rohrer H (2002) The specification of noradrenergic locus coeruleus (LC) neurons depends on bone morphogenetic proteins (BMPs). Development 129:983–991. Withers GS, Higgins D, Charette M, Banker G (2000) Bone morphogenetic protein-7 enhances dendritic growth and receptivity to innervation in cultured hippocampal neurons. Eur J Neurosci 12:106–116. Xiao L, Michalski N, Kronander E, Gjoni E, Genoud C, Knott G, Schneggenburger R (2013) BMP signaling specifies the development of a large and fast CNS synapse. Nat Neurosci 16:856–864.
(Accepted 24 July 2014) (Available online 7 August 2014)
BMP5 EXPRESSION IN THE ADULT RAT BRAIN Y. KUSAKAWA, a,b S. MIKAWA a AND K. SATO a*
eral biological events (Bragdon et al., 2011). BMPs act on a cell via the induction of a heterodimeric complex of type I and type II BMP serine/threonine kinase receptors including bone morphogenetic protein receptor type I (BMPRIA, BMPRIB) and type II (BMPRII) (Bragdon et al., 2011). By binding to the type I BMPRs, BMPs activate the receptor-activated Smads (R-Smads; Smad1/5/ 8) which oligomerize with common-mediator Smad (Co-Smad; Smad4) in the cytoplasm. Then, the Smad complex enters the nucleus and initiates transcription (Moustakas and Heldin, 2009). Although most of the biological effects of BMPs have been related to the Smad-dependent signaling pathways, Smad-independent signaling pathways have been also shown (Massague, 2003). Functions of BMPs are also controlled extracellularly by secreted antagonistic regulators such as noggin, chordin, follistatin, and neurogenesin-1, which are reported to bind BMPs and inhibit their interaction with their receptors (Cho and Blitz, 1998; Ueki et al., 2003). Bone morphogenetic protein-5 (BMP5) is also a member of the TGF-b superfamily. BMP5 is reported to be expressed in the thymus, bone marrow, spleen, skeletal muscle, heart, kidney, lung, pancreas, and prostate (Bragdon et al., 2011), and function in bone and cartilage morphogenesis, limb development and connective soft tissue morphogenesis (Bragdon et al., 2011). It is also known that BMP5 mutations cause shortened, slightly ruffled external ears due to a defective cartilage framework (King et al., 1994). BMP5 has also been reported to be involved in many critical developmental phenomena in the central nervous system (CNS). For example, BMP5 is known to be abundantly expressed in the dorsal neuroepithelium and be involved in the genesis of noradrenergic locus coeruleus neurons (Vogel-Hopker and Rohrer, 2002; Tilleman et al., 2010). In addition, Brederlau et al. (2002) have shown that BMP5 also acts on precursors of the dopaminergic and astroglial lineage and induces their differentiation. Furthermore, in the developing peripheral nervous system, BMP5 has been reported to enhance dendritic growth in cultured sympathetic neurons (Beck et al., 2001). BMP5 expression has been investigated in the early development of the CNS (Furuta et al., 1997; VogelHopker and Rohrer, 2002). However, there is little information about BMP5 expression in the adult CNS. Furthermore, BMP receptors (Miyagi et al., 2011, 2012), and BMP antagonists (Mikawa and Sato, 2011, 2014), have been also known to be abundantly expressed in the adult rat CNS. Thus, It is needed to investigate BMP5 expression more widely and more in detail in the adult rat brain. In the present study, we show that BMP5 protein is widely
a
Department of Anatomy & Neuroscience, Hamamatsu University School of Medicine, 1-20-1 Handayama, Higashiku, Hamamatsu, Shizuoka 431-3192, Japan b Department of Rehabilitation, Seirei Hamamatsu General Hospital, 2-12-12 Sumiyoshi, Nakaku, Hamamatsu, Shizuoka 430-8558, Japan
Abstract—Bone morphogenetic protein-5 (BMP5), a member of the transforming growth factor-b (TGF-b) superfamily, has many effects in several biological events. Although BMP5 expression has been well reported in the early development of the central nervous system (CNS), there is little information about its expression in the adult CNS. Thus, we analyzed BMP5 expression in the adult rat CNS by immunohistochemistry. Abundant BMP5 expression was observed in most neurons, and their dendrites and axons. Furthermore, strong BMP5 expression was also detected in the neuropil of the gray matters with high plasticity, such as the molecular layer of the cerebellum, locus coeruleus, and nucleus of the solitary tract. In addition, we showed BMP5 expression also in astrocytes, ependymal cells and meninges. Our data suggest that BMP5 is widely expressed throughout the adult CNS, and this abundant expression in the adult brain strongly supports the idea that BMP5 plays important roles not only in the developing brain but also in the adult brain. Ó 2014 IBRO. Published by Elsevier Ltd. All rights reserved.
Key words: axon, neuropil, immunohistochemistry.
INTRODUCTION Bone morphogenetic proteins (BMPs) belong to the transforming growth factor b (TGF-b) superfamily (Bragdon et al., 2011). BMPs were at the beginning detected by their ability to promote ectopic bone formation, and are now reported to have various effects in sevGrant sponsor: The Ministry of Education, Science and Culture of Japan. *Corresponding author. Tel/fax: +81-53-435-2582. E-mail address: [email protected] (K. Sato). Abbreviations: BMP, bone morphogenetic protein; BMP5, bone morphogenetic protein-5; BMPR, bone morphogenetic protein receptor; BSA, bovine serum albumin; CNS, central nervous system; ELISA, enzyme-linked immunosorbent assay; GFAP, glial fibrillary acidic protein; IgG, immunoglobulin G; IR, like immunoreactivity; LTD, long-term depression; LTP, long term potentiation; NeuN, neuronal nuclei; PB, phosphate buffer; PBS, phosphate-buffered saline; SDS– PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; SVZ, subventricular zone; TGF-b, transforming growth factor b; TTBS, Tris-buffered saline. http://dx.doi.org/10.1016/j.neuroscience.2014.07.057 0306-4522/Ó 2014 IBRO. Published by Elsevier Ltd. All rights reserved. 972
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
expressed throughout the adult CNS, and that BMP5 protein is expressed in neurons, astrocytes, ependymal cells and meninges.
EXPERIMENTAL PROCEDURES Animals and section preparation Male Wister rats (n = 15, 7 weeks old; Japan SLC Inc., Shizuoka, Japan) were deeply anesthetized with diethyl ether, and then perfused with saline followed by 0.1 M phosphate buffer (PB, pH 7.4) containing 4% paraformaldehyde and 0.2% picric acid. The brains were quickly removed, and postfixed in the same fixative for 2 h at 4 °C. All brains were immersed in 10%, 20%, 25% buffered sucrose each, overnight at 4 °C, respectively. Frozen slices (20 lm for immunoperoxidase staining or 10 lm for immunofluorescence in thickness) were serially sectioned on a cryostat. All experiments conformed to the Guidelines for Animal Experimentation at Hamamatsu University School of Medicine on the ethical use of animals. Immunohistochemistry For immunoperoxidase staining, the sections were treated with 10% normal rabbit serum, 2% bovine serum albumin (BSA) and 0.2% Triton X-100 in 0.1 M PB for 2 h at room temperature (RT), and incubated further in goat anti-BMP5 (diluted 1:100, the final concentration, 1 lg/ml; R&D Systems, Inc., Minneapolis, MN, USA) overnight at 4 °C. After being washed with 0.1 M PB, the sections were incubated in rabbit anti-goat immunoglobulin G (IgG) with peroxidase complex (no dilution, ready-to-use; N-HistofineÒ Simple Stainä Mouse MAX PO (G); Nichirei, Tokyo, Japan) for 2 h at RT. After being washed with 0.1 M PB, immunoreaction was visualized with 3,30 -diaminobenzidine (Wako, Osaka, Japan). For double immunofluorescence with goat anti-BMP5 antibody (diluted 1:50; the final concentration, 2 lg/ml; R&D Systems, Inc.) and mouse anti-Tbx21 antibody (diluted 1:30; the final concentration, 17 lg/ml; Novus Biologicals, LLC, Littleton, CO, USA), or mouse antineuronal nuclei (NeuN) antibody (diluted 1:100; the final concentration, 10 lg/ml; Millipore, Temecula, CA, USA), or mouse anti-glial fibrillary acidic protein (GFAP) antibody (diluted 1:1000; the initial concentration is not available; Millipore), the sections were treated with 10% normal donkey serum, 2% BSA and 0.2% Triton X-100 in 0.1 M PB for 2 h at RT, and incubated further in goat anti-BMP5 antibody, mouse anti-NeuN antibody and mouse anti-GFAP antibody overnight at 4 °C and in mouse anti-Tbx21 antibody for three nights at 4 °C. After being washed with 0.1 M PB, sections were incubated in both Alexa Fluor 594 donkey anti-goat IgG (diluted 1:250; the final concentration, 8 lg/ml; Molecular Probes, Inc., Oregon, USA) and Alexa Fluor 488 donkey anti-mouse IgG (diluted 1:500; the final concentration, 4 lg/ml; Molecular Probes, Inc.) for 1.5 h at RT. Furthermore, for double immunofluorescence
973
with goat anti-BMP5 antibody (diluted 1:50) and mouse Fluoro anti-pan neuronal marker antibody cocktail (diluted 1:100; the initial concentration is not available; Millipore), the sections were treated with 10% normal donkey serum, 2% BSA and 0.2% Triton X-100 in 0.1 M PB for 2 h at RT, and incubated further in goat anti-BMP5 antibody overnight at 4 °C. After being washed with 0.1 M PB, the sections were incubated in both Alexa Fluor 594 donkey anti-goat IgG (diluted 1:250) and mouse Fluoro anti-pan neuronal marker antibody cocktail (diluted 1:100) for 1.5 h at RT. Brightfield and fluorescence images were recorded with an Eclipse 80i equipped with a DS-Ri CCD camera (Nikon, Tokyo, Japan) and were further processed by Adobe Photoshop (Tokyo, Japan). Western blotting A rat whole brain, or dissected rat cerebral cortex, hippocampus, thalamus, brain stem and cerebellum were homogenized in a 50 mM Tris–HCl (pH 7.4) buffer containing protease inhibitor cocktail (Nacalai Tesque, Kyoto, Japan) to avoid degradation of proteins, and solubilized by adding 2 sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE) sample buffer. The sample was separated by SDS– PAGE (14% acrylamide; PAGEL, ATTO Corporation, Tokyo, Japan) and then transferred to PVDF membrane (Immobilon-P; Merck Millipore, Tokyo, Japan) by electroblotting (40 V overnight at 4 °C) using a transfer buffer (25 mM Tris, 192 mM glycine) containing 20% methanol. The membranes were blocked with 3% BSA in 50 mM Tris-buffered saline containing 0.05% Tween20 (TTBS) for 2 h at RT and then incubated with goat anti-BMP5 antibody (diluted 1:200; the final concentration, 0.5 lg/ml; R&D Systems, Inc.) in TTBS containing 1% BSA for 40 min at RT. After being washed with TTBS, the blots were incubated with horseradish peroxidase-linked rabbit anti-goat Ig (diluted 1:4000; the final concentration, 0.125 lg/ml; KPL, Inc., Gaithersburg, MD, USA) in TTBS containing 1% BSA for 40 min at RT. After being washed with TTBS several times, signals were detected by SuperSignal West Pico chemiluminescent Substrate (Thermo Fisher Scientific, Inc., Waltham, MA, USA). The membranes were incubated with 15% H2O2 in phosphate-buffered saline (PBS) for 30 min to inactivate horseradish peroxidase. They were blocked with TTBS containing 3% BSA overnight at 4 °C and then incubated with rabbit antitubulin-b antibody (diluted 1:1000; the final concentration, 0.2 lg/ml; Thermo Fisher Scientific Inc.) in TTBS containing 1% BSA for 40 min at RT. After being washed with TTBS, the blots were incubated with horseradish peroxidase-linked goat anti-rabbit IgG (diluted 1:5000; the final concentration, 0.08 lg/ml; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) in TTBS containing 1% BSA for 40 min at RT. After being washed with TTBS several times, signals were detected by SuperSignal West Pico chemiluminescent Substrate (Thermo Fisher Scientific, Inc.).
974
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
Enzyme-linked immunosorbent assay (ELISA) Solutions of 100 ng/ml of the antigen (BMP5: PeproTech Inc., Rocky Hill, NJ, USA; BMP6: PeproTech Inc.; and BMP7: HumanZyme Inc., Chicago, IL, USA) in 0.1 M carbonate buffer (pH 9.6) were prepared, 50 ll of the antigen solutions was added to each well (96-well EIA/ RIA Plate; Corning Incorporated, Corning, NY, USA) and then were incubated overnight at 4 °C. After washing the plate with 0.01 M PBS, the wells were filled with 3% BSA in 0.1 M carbonate buffer and then were incubated overnight at RT. After the plate was washed three times with 0.01 M PBS, 50 ll of anti-BMP5 antibody (diluted 1:100; the final concentration, 1.0 lg/ ml; R&D Systems, Inc.) was added to each well and then incubated for 2 h at RT. After the plate was washed three times with 0.01 M PBS, 50 ll of horseradish peroxidase-conjugated rabbit anti-goat Ig (diluted 1:1000; the final concentration, 0.5 lg/ml; KPL, Inc.) was added to each well and then incubated for 2 h at RT. After the plate was again washed three times with 0.01 M PBS, 200 ll of ABTS substrate solution (Roche Applied Science, Basel, Switzerland) was added to each well and then incubated for 30 min at RT. To quantitate the binding, the results were read at 405 nm with a microplate reader.
RESULTS Because the amino acids sequences of BMP5, BMP6, and BMP7 are very similar, we first investigated whether the antibody (goat anti-BMP5) can discriminate among them by using the ELISA method. As shown in Fig. 1A, B, the antibody could specifically recognize only BMP5
protein. In addition, Western blotting with the antibody for the whole rat brain exhibited a single band of about 17 kDa (Fig. 1C), the size coinciding with the theoretically calculated value. Furthermore, as controls, we performed immunohistochemistry without the first antibody, the control sections did not show any staining throughout the CNS (Fig. 2A2–H2). Taken together, these data indicate that the antibody can specifically recognize BMP5 protein. General expression patterns We show the overview of BMP5 expression in the adult rat brain in Fig. 2. BMP5-like immunoreactivity (IR) was seen throughout the adult rat brain. We observed abundant BMP5-IR in the olfactory bulb (Fig. 2A1), basal ganglia (Fig. 2B1, C1), cerebral cortex (Fig. 2C1–E1), hippocampus (Fig. 2D1, E1), thalamus (Fig. 2D1), hypothalamus (Fig. 2D1, E1), midbrain (Fig. 2E1), cerebellum (Fig. 2F1), brainstem (Fig. 2F1, G1), and spinal cord (Fig. 2H1). Moreover, abundant BMP5-IR is also seen located the border zones, such as the pia mater and ependyma throughout the CNS (Fig. 2). In addition, using the same antibody, we also performed Western blotting for dissected brain regions. Fig. 3 shows that a single band corresponding to BMP5 protein was detected in the cortex, hippocampus, thalamus, brain stem, and cerebellum, respectively, further supporting the data obtained by immunohistochemistry. We summarized the relative intensity of BMP5-IR in the rat CNS in Table 1. Telencephalon Olfactory bulb. BMP5-IR was abundantly observed throughout the olfactory bulb, where very strong expression was seen in the olfactory nerve layer (Fig. 4A, D). In the glomerular layer, we detected moderately stained periglomerular neurons (arrowheads in Fig. 4B) and glomeruli also showed weak neuropil staining (asterisks in Fig. 4B). In the external plexiform layer, strongly stained tufted cells and moderately stained neuropil were seen (Fig. 4A). In the mitral cell layer, the cell bodies of mitral cells were strongly labeled (arrowheads in Fig. 4C). In the granular layer, we observed moderately stained granular neurons (Fig. 4A). Interestingly, very strong BMP5-IR was also observed on the pia mater (arrows in Fig. 4D) and perivascular space (arrowheads in Fig. 4D).
Fig. 1. Specificity of the anti-BMP5 antibody. Wells with BMP5 antibody show intense ELISA reaction only when coated by BMP5 protein and not coated by BMP6 and BMP7 protein. Wells without the first antibody (control) show no ELISA reaction (A). Western blot analysis using anti-BMP5 antibody for the whole rat brain shows a single band of about 17 kDa (B).
Septum and nuclei of the diagonal band of Broca. In the medial septal nuclei and diagonal band of Broca, many neurons were strongly stained, in addition, weak neuropil staining was also seen (Fig. 4E). The lateral septal nucleus showed strong BMP5-IR (Fig. 4F). Interestingly, very strong BMP5-IR was also observed on the ependyma of the lateral ventricle (arrowheads in Fig. 4F). Ventral pallidum and islands of Calleja. In the ventral pallidum, large-sized neurons showed strong BMP5-IR
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
975
Fig. 2. BMP5 expression in the whole rat brain (A1, B1, C1, D1, E1, F1, G1, H1). Control sections without the primary antibody (A2, B2, C2, D2, E2, F2, G2, H2). 12, hypoglossal nucleus; Amy, amygdala; CN, cerebellar nuclei; Cor, cerebral cortex; CPu, caudate putamen; DH, dorsal horn; Hi, hippocampus; IO, inferior olive; MG, medial geniculate; MV, medial vestibular nucleus; OB, olfactory bulb; Sol, nucleus of the solitary tract; Sp5, spinal trigeminal nucleus; Th, thalamus, VH, ventral horn. Scale bar = 0.5 mm for A, G, H; 1 mm for B–F.
976
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
in the somata (arrowheads in Fig. 4G) and dendrites (arrow in Fig. 4G). Many moderately stained neurons were scattered in the islands of Calleja (Fig. 4H).
Piriform cortex. In layers I and III, moderate BMP5-IR was observed in neuropil (Fig. 4I). In layer II, we observed many moderately stained pyramidal neurons (arrowheads in Fig. 4I). Cerebral cortex. Strong BMP5-IR-positive cells and moderate neuropil staining were seen throughout the cerebral cortex (Fig. 5A). Interestingly, very strong BMP5-IR was also observed on the pia mater of the cerebral cortex (arrows in Fig. 5A). In the layer I, where short-axon cells make synapses with the apical dendrites of pyramidal neurons, strong BMP5-IR was detected in the terminals of apical dendrites (arrows in Fig. 5B) and neuropil (asterisks in Fig. 5B). In the layer V, the cell bodies of pyramidal neurons were strongly stained (arrowheads in Fig. 5C). Interestingly, we observed strong BMP5-IR also in the apical dendrites of pyramidal neurons (arrows in Fig. 5C).
Fig. 3. Western blot analysis for dissected brain regions. BS, brain stem; Cor, cerebral cortex; Cer, cerebellum; Hip, hippocampus; Th, thalamus.
Hippocampus. BMP5-IR was seen throughout the hippocampus (Fig. 5D). We observed moderate BMP5IR in pyramidal cells of Ammon’s horn (arrowheads in
Table 1. Distribution and intensity of BMP5-IR in the rat CNS Area I. Telencephalon Olfactory bulb Olfactory nerve layer Glomerular layer External plexiform layer Mitral cell layer Granular layer Subependymal layer Cerebral cortex Hippocampal formation Amygdala Basal ganglia Caudate putamen Globus pallidus Islands of Calleja Corpus callosum II. Diencephalon Thalamus Reticular nucleus Medial habenular nucleus Lateral habenular nucleus Ventroposterior nucleus Lateral geniculate nucleus Other nuclei Hypothalamus Supraoptic nucleus Paraventricular hypothalamic nucleus Ventromedial hypothalamic nucleus Other nuclei III. Midbrain Red nucleus Substantia nigra Interpeduncular nucleus IV. Pons and medulla Motor system Oculomotor nucleus Trigeminal motor nucleus Facial nucleus
Intensity
++++ ++ ++ +++ ++ + +++ ++ + ++ ++ ++ +
++ ++ + ++ ++ + ++ + + + +++ ++ +++
+++ +++ +++
Area
Intensity
Hypoglossal nucleus General somatosensory system Trigeminal mesencephalic nucleus Trigeminal spinal nucleus Cuneate nucleus Gracile nucleus General visceromotor system Dorsal nucleus of the vagus General viscerosensory system The nucleus of the solitary tract Special somatosensory system Auditory system Dorsal cochlear nucleus Ventral cochlear nucleus Inferior colliculus Vestibular system Medial vestibular nucleus Lateral vestibular nucleus Visual system Superior colliculus Pontine dorsal tegmental nucleus Pontine nucleus Locus coeruleus Inferior olive V. Cerebellum Cortex Molecular layer Purkinje cell layer Granule cell layer Cerebellar nuclei VI. Spinal cord Dorsal horm Ventral horn VII. Other areas Meninges Ependyma Choroidal plexus
+++
Relative intensities were estimated by visual comparison of immunostained slide: +, low; ++, moderate; +++, strong; ++++, very strong.
++ +++ +++ +++ ++++ ++++
+++ +++ ++ +++ ++ +++ ++++ +++ ++++ +++
+++ ++++ ++ +++ ++++ +++ ++++ ++++ ++++
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
977
Fig. 4. BMP5 expression in the olfactory bulb (A–D), medial septal nucleus (E), lateral septal nucleus (F), ventral pallidum (G), islands of Calleja (H) and piriform cortex (I). Note that the pia mater (arrows in D), perivascular space (arrowheads in D), and ependyma of the lateral ventricle (arrowheads in F) were very strongly stained. In the olfactory bulb, the soma of mitral cells (arrowheads in C) was strongly stained, and periglomerular neurons (arrowheads in B) were moderately stained, and glomeruli (asterisks in B) were weakly stained. In the ventral pallidum, the somas (arrowheads in G) and the dendrites (arrow in G) of neurons were strongly stained. In the piriform cortex, pyramidal neurons (arrowheads in I) were moderately stained. I–III, layers I–III of the piriform cortex; DB, diagonal band; EPI, external plexiform layer; Gl, glomerular layer; Gr, granular layer; ICj, islands of Calleja; LS, lateral septal nucleus; Mi, mitral cell layer; MS, medial septal nucleus; ON, olfactory nerve layer; Pir, piriform cortex; SE, subependymal layer; VP, ventral pallidum. Scale bar = 400 lm for E; 160 lm for A, I; 80 lm for F; 40 lm for B–D, G, H.
Fig. 5E). Furthermore, the apical dendrites of pyramidal cells also showed strong BMP5-IR (arrows in Fig. 5E). In the dentate gyrus, moderate BMP5-IR was seen in granule cells, and we also observed moderate neuropil staining in the polymorphic layer and stratum molecular (Fig. 5F). Corpus callosum. BMP5-IR-positive cells were scattered in the corpus callosum, (arrowheads in
Fig. 5G). In addition, we also observed weak neuropil staining (Fig. 5G). The basal ganglia. In the caudate putamen, we observed moderately positive neurons and moderate neuropil staining in the gray matter (Fig. 6A). In the globus pallidus, moderately positive neurons with stained dendrites and weak neuropil staining were observed in the gray matter (Fig. 6B).
978
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
Fig. 5. BMP5 expression in the cerebral cortex (A–C), hippocampus (D–F) and corpus callosum (G). In the cerebral cortex, the pia mater (arrows in A) was very strongly stained, and the apical dendrites (arrows in B and C) and the somas (arrows in C) of pyramidal cells in the layer V, and neuropil in the layer I (asterisks in B) were strongly stained. In the hippocampus, the apical dendrites of pyramidal cells (arrows in E) were strongly stained, and pyramidal cells of the Ammon’s horn (arrowheads in E) were moderately stained, and weak BMP5-IR-positive cells were observed in the corpus callosum (arrowheads in G). I–VI, layers I–VI; cc, corpus callosum; CA1, 3, field CA1, 3 of Ammon’s horn; DG, dentate gyrus; Gr, granular layer; Mo, stratum molecular; Or, stratum oriens; Po, polymorphological layer; Py, stratum pyramidale; Ra, stratum radiatum. Scale bar = 400 lm for D; 160 lm for A; 80 lm for F; 40 lm for B, C, E, G.
The amygdala. In all nuclei of the amygdala, we detected weak BMP5-IR (Fig. 2D1). Diencephalon In all nuclei in the thalamus, we observed BMP5-IRpositive neurons and neuropil (Fig. 2D1, Fig. 6C).
Interestingly, very strong BMP5-IR was also seen in the pia mater (arrows in Fig. 6C). In the medial habenular nucleus, moderate BMP5-IR-positive neurons and moderate neuropil staining were seen (Fig. 6D), while, in the lateral habenular nucleus, BMP5-IR was weak (Fig. 6D). Moderately positive neurons and weak neuropil staining were observed in the reticular thalamic
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
979
Fig. 6. BMP5 expression in the basal ganglia (A, B), thalamus (C–G) and hypothalamus (H–K). Note that the pia mater (arrows in C), perivascular space (arrowheads in E and F), and the ependyma of the third ventricle (arrowheads in K) were very strongly stained. CPu, caudate putamen; GP, globus pallidus; LG, lateral geniculate nucleus; LHb, lateral habenular nucleus; MHb, medial habenular nucleus; Po, posterior thalamic nucleus; PV, paraventricular hypothalamic nucleus; Rt, reticular thalamic nucleus; SO, supraoptic nucleus; VMH, ventromedial hypothalamic nucleus; VP, ventroposterior thalamic nucleus. Scale bar = 400 lm for C; 160 lm for G, I, K; 80 lm for A, B, D, H, J; 40 lm for E, F.
980
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
nucleus (Fig. 6E). We also observed moderately positive neurons and weak neuropil staining in the ventroposterior thalamic nucleus (Fig. 6F). Interestingly, BMP5-IR was also seen in the perivascular space of the nuclei (arrowheads in Fig. 6E, F). In the lateral geniculate nucleus, we observed moderately stained neurons and weakly stained neuropil (Fig. 6G). In the hypothalamus, we detected BMP5-IR-positive neurons and neuropil staining in all nuclei (Fig. 2D1, E1). Moderate BMP5-IR was observed in the supraoptic nucleus (Fig. 6H). In addition, weak BMP5-IR was seen in the paraventricular hypothalamic nucleus (Fig. 6I) and ventromedial hypothalamic nucleus (Fig. 6J). Interestingly, we also observed very strong BMP5-IR in the ependyma of the hypothalamus (arrowheads in Fig. 6K). Midbrain The interpeduncular nucleus exhibited strong BMP5-IR (Fig. 7A). We observed moderate BMP5-IR in the central gray (Fig. 7A). Interestingly, BMP5-IR was also seen in the pia mater (arrows in Fig. 7A) and ependyma in this region (arrowheads in Fig. 7A). In the red nucleus, large neurons having many neurites exhibited strong BMP5-IR (arrowheads in Fig. 7B), while neuropil was stained weakly (Fig. 7B). Moderate BMP5-IR was detected in the substantia nigra (Fig. 7A). Moderately stained neurons (arrowheads in Fig. 7C) and weak neuropil staining (Fig. 7C) were seen in the pars compacta of the substantia nigra. Pons and medulla Motor system. The nuclei of the general somatomotor and branchiomotor system, such as the oculomotor nucleus, motor trigeminal nucleus (Fig. 7D), facial nucleus (Fig. 7E) and hypoglossal nucleus (Fig. 7G), exhibited abundant BMP5-IR in neuronal cell bodies and their dendrites. Interestingly, in the facial nerve BMP5IR was observed in axon fibers (arrows in Fig. 7F). General somatosensory system. Large primary afferent neurons were stained moderately in the trigeminal mesencephalic nucleus (arrowheads in Fig. 7H). In the trigeminal spinal nucleus, strongly stained neurons were observed (Fig. 7I). Strongly positive neurons were also seen in the cuneate and gracile nuclei (Fig. 7J).
Special somatosensory system. Auditory system: The dorsal and ventral cochlear nuclei contained strongly positive neurons (Fig. 8C). In the inferior colliculus, moderate BMP5-IR was detected. Vestibular system: we observed strongly stained small neurons and moderate neuropil staining in the medial vestibular nucleus (Fig. 8D). Interestingly, BMP5-IR was also seen in the ependyma of the fourth ventricle (arrowheads in Fig. 8D). In the lateral vestibular nucleus, large neurons exhibited moderate BMP5-IR, while neuropil staining was weak (Fig. 8E). Interestingly, we found strongly stained axons in the vestibulocochlear nerve (arrows in Fig. 8F). Visual system: In the superior colliculus, we observed strongly stained neurons and moderate neuropil staining (Fig. 8G). The superficial part showed relatively strong BMP5-IR than the other parts (asterisks in Fig. 8G). Interestingly, very strong BMP5-IR was also seen in the pia mater above the superior colliculus (arrows in Fig. 8G). Other lower brain stem areas. We observed very strong BMP5-IR was observed in the pontine dorsal tegmental nucleus (Fig. 9A, B). Strongly stained neurons and moderate neuropil staining was also seen in the pontine nucleus (Fig. 9C). Interestingly, we observed very strongly stained neurons and very strong neuropil staining in the locus coeruleus (Fig. 9D, E). Interestingly, BMP5-IR was also seen in the ependyma of the fourth ventricle (arrowheads in Fig. 9A, D). In the gigantocellular reticular nucleus, the cell bodies (arrowheads in Fig. 9F) and dendrites (arrows in Fig. 9F) of large neurons were very strongly stained. In the inferior olive, we observed strongly stained neurons and moderate neuropil staining (Fig. 9G). Cerebellum In the cerebellum, we observed very strong staining in the Purkinje cell layer. Very strong BMP5-IR was seen in the cell bodies of Purkinje neurons (Fig. 10A). In addition, small Bergmann glias around the cell bodies of Purkinje cells also seemed to show very strong BMP5-IR. The molecular cell layer contained many strongly positive neurons, and strong neuropil staining was also observed (Fig. 10A). Moderate BMP5-IR was also seen in granular cells (Fig. 10A). Strong neuronal cell body staining and moderate neuropil staining were observed in the cerebellar nuclei (Fig. 10B). Spinal cord
General visceromotor system. We observed many very strong BMP5-IR-positive neurons and strong neuropil staining in the dorsal nucleus of the vagus (Fig. 8A, B). General viscerosensory system. In the nucleus of the solitary tract, very strong BMP5-IR-positive neurons and strong neuropil staining were seen (Fig. 8A, B). Interestingly, BMP5-IR was also seen in the ependyma of the fourth ventricle (arrowheads in Fig. 8A).
We observed abundant BMP5-IR in the gray matter (Fig. 10C). Furthermore, also in the white matter, we found that all fiber tracts contain BMP5-IR (Fig. 10C). Further investigation showed that in the white matter, besides strongly stained astrocyte-like cells (arrowheads in Fig. 10D), a lot of axons exhibited abundant BMP5-IR (arrows in Fig. 10D). We also observed very strong neuropil and neuronal cell body staining in the layers I and II of the dorsal horn (Fig. 10E, F). In addition, large motor neurons in the ventral horn also showed strong
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
981
Fig. 7. BMP5 expression in the midbrain (A–C), motor system (D–G) and general somatosensory system (H–J). Note that the pia mater (arrows in A) and the ependyma (arrowheads in A) were very strongly stained, that neurons in the red nucleus (arrowheads in B) and the axons in the facial nerve (arrows in F) were strongly stained, and that neurons in the pars compacta of the substantia nigra (arrowheads in C) and the trigeminal mesencephalic nucleus (arrowheads in H) were moderately stained. 5, motor trigeminal nucleus; 7, facial nucleus; 7n, facial nerve; 12, hypoglossal nucleus; CG, central gray; Cu, Cuneate nucleus; Gr, Gracile nucleus; IP, interpeduncular nucleus; Me5, trigeminal mesencephalic nucleus; MG, medial gray; R, red nucleus; SN, substantia nigra; Sp5, trigeminal spinal nucleus. Scale bar = 400 lm for A; 80 lm for H, J; 40 lm for B–G, I.
982
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
Fig. 8. BMP5 expression in the general viscerosensory system (A, B) and special somatosensory system (C–G). Note very strong BMP5-IR in the nucleus of the solitary tract and dorsal nucleus of the vagus (A, B). Note that the ependyma of the fourth ventricle (arrowheads in A and D) and the pia mater (arrows in G) were very strongly stained, and that axons in the vestibulocochlear nerve (arrows in F) were strongly stained. 8n, vestibulocochlear nerve; 10, vagus nucleus; 12, hypoglossal nucleus; DC, dorsal cochlear nucleus; LVe, lateral vestibular nucleus; MVe, medial vestibular nucleus; SC, superior colliculus; Sol, nucleus of the solitary tract; VC, ventral cochlear nucleus. Scale bar = 160 lm for A, C–E, G; 80 lm for B; 40 lm for F.
BMP5-IR in the cell bodies (arrowheads in Fig. 10G) and dendrites (arrows in Fig. 10G). Other areas In the subventricular zone (SVZ), we observed very strong BMP5-IR in ependymal cells, cell bodies and neuropils (Fig. 10H). In addition, choroidal plexus also expressed abundant BMP5 proteins (Fig. 10I). Double fluorescence immunohistochemistry Using double fluorescence immunohistochemistry with anti-Tbx21 (a specific marker for mitral neurons)
antibody, we tried to confirm that mitral neurons express BMP5 in the olfactory bulb. As shown in Fig. 11A1-3, co-expression of BMP5 and Tbx21 was seen in the mitral cell layer, indicating that mitral neurons express BMP5 proteins. In addition, using anti-NeuN (a pan neuronal marker) antibody, we tried to validate that granular neurons in the olfactory bulb express BMP5. As shown in Fig. 11B1-3, co-expression of BMP5 and NeuN was seen in the granular cell layer, indicating that granular neurons express BMP5 proteins. Furthermore, to investigate that neurons in the caudate putamen, reticular thalamus, cerebral cortex, and hippocampus express BMP5 proteins, we performed double fluorescence immunohistochemistry with anti-NeuN
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
983
Fig. 9. BMP5 expression in the other lower brain stem areas (A–G). Note that the locus coeruleus (D, E) and the ependyma of the fourth ventricle (arrowheads in A and D) show very strong BMP5-IR. In the gigantocellular reticular nucleus, the somas (arrowheads in F) and the dendrites (arrows in F) of neurons were very strongly stained. IO, inferior olive; LC, locus coeruleus; Me5, mesencephalic trigeminal nucleus; PDTg, pontine dorsal tegmental nucleus; Pn, pontine nucleus; Rt, reticular nucleus. Scale bar = 160 lm for A, D; 80 lm for G; 40 lm for B, C, E, F.
antibody. Fig. 11C–F clearly shows that neurons in these regions express BMP5 proteins. To confirm that BMP5-IR-positive cells in the white matter and hippocampus are astrocytes. We performed double fluorescence immunohistochemistry with antiGFAP antibody. As shown in Fig. 11G, H, co-expression of BMP5 and GFAP was seen in the corpus callosum and hippocampus, indicating that astrocytes express BMP5 proteins. Furthermore, we performed double fluorescence immunohistochemistry using anti-panneuronal marker antibody cocktail in the facial nerve and vestibulocochlear nerve to confirm that axons exactly express BMP5 proteins. As shown in Fig. 11I, J, pan-neuronal marker-IR-positive structures co-expressed
BMP5, indicating that axons express BMP5 proteins in both nerves.
DISCUSSION Up to the present, little information is available for BMP5 expression in the adult brain. In this study, we first exhibit that BMP5 is widely expressed throughout the adult CNS. Moreover, besides abundant BMP5 expression in neurons, we observed BMP5 expression in astrocytes, ependymal cells and meninges. These data suggest that BMP5 is produced by neurons, astrocytes, ependymal cells and meninges throughout the adult CNS. So far BMP receptors (Miyagi et al., 2011, 2012)
984
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
Fig. 10. BMP5 expression in the cerebellum (A, B), spinal cord (C–G), subventricular zone (H) and choroidal plexus (I). Note that axons (arrows in D) and astrocyte-like cells (arrowheads in D) in the white matter and the somas (arrowheads in G) and the dendrites (arrows in G) of motor neurons in the ventral horn were strongly stained. Note very strong BMP-IR in the Purkinje layer (A) and layers I and II of the dorsal horn of the spinal cord (F). I–III, layers I–III of the spinal cord; Ch, choroidal plexus; CN, cerebellar nuclei; DH, dorsal horn; Gr, granular layer; Mol, molecular layer; Pur, Purkinje cell layer; SVZ, subventricular zone; VH, ventral horn; wm, white matter. Scale bar = 400 lm for C; 160 lm for B, E; 80 lm for F, I; 40 lm for A, G, H; 16 lm for D.
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
985
Fig. 11. Double staining study showing that BMP5-IR-positive cells in the mitral cells (A1) are also positive for Tbx21 (A2), BMP5-IR-positive cells in the granular cells (B1), caudate putamen (C1), reticular thalamic nucleus (D1), cerebral cortex (E1) and hippocampus (F1) are also positive for NeuN (B2–F2), BMP5-IR-positive cells in the corpus callosum (G1) and hippocampus (H1) are also positive for GFAP (G2, H2), BMP5-IR-positive axons in the facial nerve (I1) and vestibulocochlear nerve (J1) are also positive for pan-neuronal marker (I2, J2) in a merged photomicrograph (A3–J3). cc, corpus callosum; Cor, cerebral cortex; CPu, caudate putamen; Gr, granular cell; Hip, hippocampus; Mi, mitral cell; Rt, reticular thalamic nucleus; 7n, facial nerve; 8n, vestibulocochlear nerve. Scale bar = 20 lm.
986
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
and BMP antagonists (Mikawa and Sato, 2011, 2014) have been reported to be abundantly expressed in the adult rat CNS. BMP5 might take on important roles as an agonist regulating BMP signaling in the adult brain. BMP5 expressions in neurons We found that almost all neuronal cell bodies express BMP5 proteins with different intensities. What are the roles of BMP5 secreted from neuronal cell bodies? BMP5 has been reported to control the differentiation of various neurons. For example, BMP5 is known to be abundantly expressed in the dorsal neuroepithelium and be involved in the genesis of noradrenergic locus coeruleus neurons (Vogel-Hopker and Rohrer, 2002; Tilleman et al., 2010). Interestingly, even in the adult CNS, we also found very strong BMP5 expression in the locus coeruleus. In the locus coeruleus, both cell bodies of noradrenergic neurons and neuropil were strongly stained (Fig. 9D, E), suggesting that even in the adult noradrenergic neurons need abundant BMP5 signaling to keep their phenotype. In addition, Brederlau et al. (2002) have reported that BMP5 also acts on precursors of the dopaminergic and induces their differentiation. Interestingly, we also found that BMP5 is expressed in dopaminergic neurons (Fig. 7C), further supporting the idea that BMP5 signaling may be necessary to keep the identities of various neurons even in the adult CNS. BMP5 expressions in dendrites Interestingly, we found intense BMP5 protein expressions in apical dendrites of pyramidal neurons in the cerebral cortex, and hippocampus. In addition, we found many BMP5-IR-positive dendrites throughout the CNS. What are the functions of BMP5 in dendrites? BMP signalings have been previously reported to be involved in modulating dendritic morphology (Lein et al., 1995; Le Roux et al., 1999; Withers et al., 2000). Withers et al. showed that addition of BMP7 to cultured hippocampal neurons promotes dendritic growth, and also enhances synaptogenesis (Withers et al., 2000). Moreover, Althini et al. (2004) have recently reported that BMP4 potentiates neurotrophin 3 and neurturin-induced neurite outgrowth of peripheral neurons from the E9 chicken embryo. Furthermore, BMP2 is also reported to promote mitral/ tufted cell dendritic outgrowth (Tran et al., 2008). Interestingly, BMP5 has been also reported to promote dendritic growth in cultured sympathetic neurons (Beck et al., 2001). Taken together, BMP5 released from dendrites may regulate dendrite morphology and synaptic homeostasis. BMP5 expressions in axons Surprisingly, we observed that BMP5 protein is expressed in many kinds of axons, such as in the facial nerve, and vestibulocochlear nerve. Our data revealed, almost all axons are positive for BMP5 expression. What is the physiological significance of BMP5 protein expressed in axons? Retrograde BMP signaling is known to modulate
neuronal terminal differentiation and synaptic efficacy (da Silva and Wang, 2011). For example, in Drosophila, the BMP homolog Glass bottom boat (Gbb) acts as a muscle-derived retrograde signal that enhances synaptic growth and neurotransmitter release (McCabe et al., 2003). In mammals, target-derived retrograde BMP signaling has been also shown to determine the number of neurons in the trigeminal ganglion (Guha et al., 2004), and to regulate trigeminal sensory neuron identities (Hodge et al., 2007). Furthermore, Xiao et al. (2013) have reported that BMP5 plays a role in the development of a large and fast CNS synapse, and also in the elimination of competing synaptic inputs. These data strongly suggest that, also in the mammalian brain, many neurons are involved in this kind of target-derived retrograde BMP signal by using BMP5. If so, there are two possibilities for the origin of BMP5 protein observed in the axon. One is that BMP5 protein is anterogradely transported from the cell body to the axon terminal. Another possibility is that BMP5 protein released from the postsynaptic neurons or surrounding astrocytes is up-taken by the presynapse and retrogradely transported to the cell body. BMP5 expressions in the neuropil In this study, we found intense BMP5-IR in the neuropil of the molecular layer of the cerebellum (Fig. 10A), the nucleus of the solitary tact (Fig. 8B), and the locus coeruleus (Fig. 9D). The molecular layer of the cerebellum is known to have high plasticity, such as long-term depression (LTD) and long term potentiation (LTP) (Ito, 2001). In the molecular layer of the cerebellum, parallel fiber axons of granule cells and climbing fiber axons from the inferior olive neurons make synapses with the dendrites of Purkinje cells. Interestingly, we detected abundant BMP5-IR in the inferior olive neurons, granule cells, and Purkinje cells, suggesting that BMP5 in the neuropil might play a role to keep plasticity by regulating important phenomena, such as LTD and LTP. BMP5 expressions in ependymal cells and astrocytes In this study, we showed that ependymal cells express BMP5 protein. It has been reported that, in the adult SVZ, a BMP antagonist noggin released from the ependymal cells makes a niche for adult neurogenesis (Lim et al., 2000). In the present study, we also found BMP5 is also expressed in the SVZ, suggesting that BMP5 plays a role in the regulation of neurogenesis. Furthermore, we showed that BMP5 protein is expressed in astrocytes. What does BMP5 released from astrocytes do? One possibility may be that BMP5 secreted from astrocytes affects astrocytes themselves via autocrine or paracrine manners. Interestingly, BMP5 is also known to control astroglial lineage and induce their differentiation. These findings raise the possibility that BMP5 might also play a role in regulating the functions of astrocyte. Another possibility may be that BMP5 secreted from astrocytes control local environment, such as synaptic conditions.
Y. Kusakawa et al. / Neuroscience 284 (2015) 972–987
BMP5 expressions in the meninges and choroidal plexi In this study, we found that the meninges express BMP5 protein. What is the role of BMP5 in the meninges? Interestingly, other members of BMP proteins, such as BMP4 and BMP7 have been reported to be expressed in the meninges (Machida et al., 2014). In addition, Choe et al. (2012) have also reported that meninges produce BMP7 in the embryonic brain. And the overexpression of BMP7 inhibits callosal axon outgrowth, resulting in callosal agenesis. Recently, it has been proposed that the brain and circulating immune cells engage in a continuous dialog that takes place within the brain’s territory, though outside the parenchyma, occurring within the brain’s borders – the choroidal plexi, the brain meninges and the cerebrospinal fluid (Ron-Harel et al., 2011). Therefore, one possibility might be that BMP5 released from the meninges and choroidal plexi also plays a role in this kind of dialog by giving BMP signals to the neighbor structures.
REFERENCES Althini S, Usoskin D, Kylberg A, Kaplan PL, Ebendal T (2004) Blocked MAP kinase activity selectively enhances neurotrophic growth responses. Mol Cell Neurosci 25:345–354. Beck HN, Drahushuk K, Jacoby DB, Higgins D, Lein PJ (2001) Bone morphogenetic protein-5 (BMP-5) promotes dendritic growth in cultured sympathetic neurons. BMC Neurosci 2:12. Bragdon B, Moseychuk O, Saldanha S, King D, Julian J, Nohe A (2011) Bone morphogenetic proteins: a critical review. Cell Signal 23:609–620. Brederlau A, Faigle R, Kaplan P, Odin P, Funa K (2002) Bone morphogenetic proteins but not growth differentiation factors induce dopaminergic differentiation in mesencephalic precursors. Mol Cell Neurosci 21:367–378. Cho KW, Blitz IL (1998) BMPs, Smads and metalloproteases: extracellular and intracellular modes of negative regulation. Curr Opin Genet Dev 8:443–449. Choe Y, Siegenthaler JA, Pleasure SJ (2012) A cascade of morphogenetic signaling initiated by the meninges control corpus callosum formation. Neuron 73:698–712. da Silva S, Wang F (2011) Retrograde neural circuit specification by target-derived neurotrophins and growth factors. Curr Opin Neurobiol 21:61–67. Furuta Y, Piston DW, Hogan BL (1997) Bone morphogenetic proteins (BMPs) as regulators of dorsal forebrain development. Development 124:2203–2212. Guha U, Gomes WA, Samanta J, Gupta M, Rice FL, Kessler JA (2004) Target-derived BMP signaling limits sensory neuron number and the extent of peripheral innervation in vivo. Development 131:1175–1186. Hodge LK, Klassen MP, Han BX, Yiu G, Hurrell J, Howell A, Rousseau G, Lemaigre F, Tessier-Lavigne M, Wang F (2007) Retrograde BMP signaling regulates trigeminal sensory neuron identities and the formation of precise face maps. Neuron 55:572–586. Ito M (2001) Cerebellar long-term depression: characterization, signal transduction and functional roles. Physiol Rev 81:1143–1195.
987
King JA, Marker PC, Seung KJ, Kingsley DM (1994) BMP5 and the molecular, skeletal, and soft-tissue alterations in short ear mice. Dev Biol 166:112–122. Le Roux P, Behar S, Higgins D, Charette M (1999) OP-1 enhances dendritic growth from cerebral cortical neurons in vitro. Exp Neurol 160:151–163. Lein P, Johnson M, Guo X, Rueger D, Higgins D (1995) Osteogenic protein-1 induces dendritic growth in rat sympathetic neurons. Neuron 15:597–605. Lim DA, Tramontin AD, Trevejo JM, Herrera DG, Gracia-Verdugo JM, Alvarez-Buylla A (2000) Noggin antagonizes BMP signaling to create a niche for adult neurogenesis. Neuron 28:713–726. Machida A, Okuhara S, Harada K, Iseki S (2014) Difference in apical and basal growth of the frontal bone primordium in Foxc1ch/ch mice. Congenit Anom (Kyoto) 54:172–177. Massague J (2003) Integration of Smad and MAPK pathways: a link and a linker revisited. Genes Dev 17:2993–2997. McCabe BD, Marques G, Haghighi AP, Fetter RD, Crotty ML, Haerry TE, Goodman CS, O’Connor MB (2003) The BMP homolog Gbb provides a retrograde signal that regulates synaptic growth at the Drosophila neuromuscular junction. Neuron 39:241–254. Mikawa S, Sato K (2011) Noggin expression in the adult rat brain. Neuroscience 184:38–53. Mikawa S, Sato K (2014) Chordin expression in the adult rat brain. Neuroscience 258:16–33. Miyagi M, Mikawa S, Hasegawa T, Kobayashi S, Matsuyama Y, Sato K (2011) Bone morphogenetic protein receptor expressions in the adult rat brain. Neuroscience 176:93–109. Miyagi M, Mikawa S, Sato T, Hasegawa T, Kobayashi S, Matsuyama Y, Sato K (2012) BMP2, BMP4, noggin, BMPRIA, BMPRIB, and BMPRII are differentially expressed in the adult rat spinal cord. Neuroscience 203:12–26. Moustakas A, Heldin CH (2009) The regulation of TGFbeta signal transduction. Development 136:3699–3714. Ron-Harel N, Cardon M, Schwartz M (2011) Brain homeostasis is maintained by danger signals stimulating a supportive immune response within the brain borders. Brain Behav Immun 25:1036–1043. Tilleman H, Hakim V, Novikov O, Liser K, Nashelsky L, Di Salvio M, Krauthammer M, Scheffner O, Maor I, Mayseless O, Meir I, Kayam G, Sela-Donenfeld D, Simeone A, Brodski C (2010) Bmp5/7 in concert with the mid-hindbrain organizer control development of noradrenergic locus coeruleus neurons. Mol Cell Neurosci 45:1–11. Tran H, Chen H, Walz A, Posthumus JC, Gong Q (2008) Influence of olfactory epithelium on mitral/tufted cell dendritic outgrowth. PLoS One 3:e3816. Ueki T, Tanaka M, Yamashita K, Mikawa S, Qiu ZF, Maragakis NJ, Hevner RF, Miura N, Sugimura H, Sato K (2003) A novel secretory factor, neurogenesin-1, provides neurogenic environmental cues for neural stem cells in the adult hippocampus. J Neurosci 23:11732–11740. Vogel-Hopker A, Rohrer H (2002) The specification of noradrenergic locus coeruleus (LC) neurons depends on bone morphogenetic proteins (BMPs). Development 129:983–991. Withers GS, Higgins D, Charette M, Banker G (2000) Bone morphogenetic protein-7 enhances dendritic growth and receptivity to innervation in cultured hippocampal neurons. Eur J Neurosci 12:106–116. Xiao L, Michalski N, Kronander E, Gjoni E, Genoud C, Knott G, Schneggenburger R (2013) BMP signaling specifies the development of a large and fast CNS synapse. Nat Neurosci 16:856–864.
(Accepted 24 July 2014) (Available online 7 August 2014)