- Email: [email protected]
FLRT3, a cell surface molecule containing LRR repeats and a FNIII domain, promotes neurite outgrowth
BBRC Biochemical and Biophysical Research Communications 313 (2004) 1086–1091 www.elsevier.com/locate/ybbrc
FLRT3, a cell surface molecule containing LRR repeats and a FNIII domain, promotes neurite outgrowthq Lyuji Tsuji,a,b,c Toshihide Yamashita,d,* Tateki Kubo,b Tomas Madura,a,b,c Hiroyuki Tanaka,a,c Ko Hosokawa,b and Masaya Tohyamaa,c a
c
Department of Anatomy and Neuroscience, Osaka University Graduate School of Medicine, Osaka, Japan b Department of Plastic Surgery, Osaka University Graduate School of Medicine, Osaka, Japan Core Research for Evolutional Science and Technology (CREST) of Japan Science and Technology Corporation (JST), Kawaguchi, Japan d Department of Neurobiology, Graduate School of Medicine, Chiba University, Chiba, Japan Received 7 December 2003
Abstract The mature peripheral nervous system has the ability to survive and to regenerate its axons following axonal injury. After nerve injury, the distal axonal and myelin segment undergoes dissolution and absorption by the surrounding cellular environment, a process called Wallerian degeneration. Using cDNA microarrays, we isolated FLRT3 as one of the up-regulated genes expressed in the distal segment of the sciatic nerve 7 days after transection relative to those of the intact sciatic nerve. FLRT3 is a putative type I transmembrane protein containing 10 leucine-rich repeats, a fibronectin type III domain, and an intracellular tail. The neurons plated on CHO cells expressing FLRT3 extended significantly longer neurites than those plated on wild-type CHO cells, demonstrating that FLRT3 promotes neurite outgrowth. FLRT3 mRNA was especially abundant in the basal ganglia, the granular layer of cerebellum, and the hippocampus, except the CA1 region in the adult rat brain. Thus, FLRT3 may contribute to regeneration following axonal injury. Ó 2003 Elsevier Inc. All rights reserved. Keywords: cDNA microarrays; Injury-induced molecules; Nerve regeneration; Schwann cell; Adhesion molecules
In contrast to the central glia, following axonal injury, Wallerian degeneration occurs in the distal segment of the injured site, where Schwann cells proliferate and form the band of B€ ungner (Schwann cell column) to support and guide axoplasmic sprouts from the proximal portion of severed axons within the bands [1,2]. Schwann cells in the distal segment of the injured peripheral nerve express various neurotrophic factors and adhesion molecules, such as L1, NCAM, and N-cadherin, that promote neurite outgrowth by regulating contact between axons and Schwann cells [3–6]. Identification of the molecular basis of this process is expected to lead to opportunities to rescue injured neurons that q Abbreviations: EST, expressed sequence tag; N-CAM, neural cell adhesion molecule; FNIII, fibronectin type III. * Corresponding author. Fax: +81-6-6879-3229. E-mail address: [email protected] (T. Yamashita).
0006-291X/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2003.12.047
do not survive or regenerate after axotomy. In an attempt to explore this molecular mechanism, we used a cDNA library and microarrays derived from the distal stumps of the post-injury sciatic nerve which was rich in non-myelinating Schwann cells [7]. Here we report that FLRT3, one of the up-regulated genes, is expressed in the distal segment of sciatic nerve 7 days after transection relative to those of the intact sciatic nerve [7]. FLRT, a family of cell surface molecules containing LRR repeats and a FNIII domain, consists of an N-terminal putative signal peptide, 10 leucine-rich repeats (LRR) flanked by cysteine-rich regions, fibronectin type III domains, the putative transmembrane domains, and short intracellular tails. Potential N-linked glycosylation sites appear four times (226N, 282N, 296N, and 508N) within the extracellular region of FLRT3. FLRT3 mRNA is broadly expressed in various tissues such as kidney, skeletal muscle, lung, and brain [8].
L. Tsuji et al. / Biochemical and Biophysical Research Communications 313 (2004) 1086–1091
Materials and methods All procedures involving animals were reviewed and approved by the local Institutional Animal Care and Use Committee. Construction of the FLRT3 expression vectors. All expression vectors were constructed using the GATEWAY system (Invitrogen). We constructed two destination vectors, pcDNA5FRT/Igjleader/HA/ ccdB and pcDNA/ccdB/HA. An oligonucleotide coding for the HA tag, and a GATEWAY Reading Frame Cassette B, fragment were sequentially ligated into the KpnI–EcoRV sites of the pSecTag2/C vector (Invitrogen) to generate the pSecTag2/HA/ccdB vector. An NheI–PmeI fragment, containing a murine Igj chain V-J2-C signal peptide from the pSecTag2 vector, a HA tag and a GATEWAY cassette B was generated by PCR amplification. This fragment was ligated into NheI–EcoRV sites of pcDNA5FRT (Invitrogen) to generate pcDNA5FRT/Igjleader/HA/ccdB. pcDNA/ccdB/HA was generated by sequential ligation of a GATEWAY Reading Frame Cassette B fragment and an oligonucleotide coding for the HA tag into pcDNA3.1/Zeo (Invitrogen). All destination vectors were linealized by a cut at the BamHI site. DNA fragments for 29-FLRT3 (amino acids 29–649) and FLRT3-C (full-length FLRT3 with the kozak consensus skipping stop codon) were amplified by PCR with the human FLRT3 clone, KIAA1469, from Kazusa DNA Research Institute (Chiba, Japan), as a template. These fragments were integrated into pDONR221 using BP clonase to generate pDONR 29-FLRT3 and pDONR FLRT3-C entry vectors, respectively. Expression vectors for HAFLRT3 were generated by LR reaction with pDONR 29-FLRT3 and linealized pcDNA5FRT/Igjleader/HA/ccdB, and FLRT3-HA with pDONR FLRT3-C and linealized pcDNA/ccdB/HA. All PCR amplified fragments were verified by DNA sequencing. Transient transfections and immunofluorescence. CHO cells were routinely maintained in DMEM/F12 medium supplemented with 10% FCS. For transient transfections, the cells were grown on 35 mm dishes and transfected using LipofectAMINE (Invitrogen) according to the manufacturer’s recommendations. After 5 h, equal volumes of DMEM/F12 medium with 10% FCS were added to each well and the cells were grown for an additional 19 h. The cells were fixed with 4% PFA for 20 min at room temperature. Incubation with primary antibodies in PBS containing 0.5% BSA and 0.1% Triton-X was performed overnight at 4 °C and with secondary antibodies in PBS for 60 min at room temperature. The cells were washed in PBS after each incubation. Triton-X was omitted where indicated as the non-permeabilized condition. We confirmed that cells transfected with tagged Rho kinase showed no staining under this condition. The following antibodies were used at the dilution of 1:1000: monoclonal anti-HA antibody, HA.11 (Covance), and Alexa 488 goat anti-mouse IgG conjugate (Molecular Probes). Immunofluorescence images were acquired with a Zeiss LSM 510 laser scanning confocal microscope. Generation of FLRT3-CHO cells. The Flp-In System (Invitrogen) was used to generate FLRT3 expressing cells according to the manufacturer’s directions. pcDNA5FRT/Igjleader/HA/29-FLRT3 and pOG44 were co-transfected into Flp-in CHO cells and stable expressing cells were generated after growth in medium containing Hygromycin B (600 lg/ml; Invitrogen) for 2 weeks. Expression of HAFLRT3 was confirmed by Western blot (Fig. 3A) and immunocytology (data not shown). Western blot analysis. HA-FLRT3 was extracted by adding SDS sample buffer containing 12% b-mercaptoethanol directly to FLRT3CHO cells and then the sample was loaded on a 10% SDS–PAGE gel. This gel was then transferred to a PVDF membrane. The protein was detected by an anti-HA antibody (HA.11), an HRP-linked secondary antibody (Cell Signaling), and ECL detection reagents (Amersham). Neurite outgrowth assays. Cerebellar granule cells from postnatal day 8 (P8) rat pups were dissociated by trypsinization (0.25% trypsin in PBS for 10 min at 37 °C) followed by resuspension in serum-containing medium to inhibit trypsin and trituration. After three washes with
1087
PBS, cells were plated on confluent monolayers of either control CHO cells or CHO cells expressing FLRT3 in 35-mm plates at the density of 160 cells/mm2 . Cultures were grown in a serum-free DMEM/F12 medium. Twenty-four hours after plating, the cells were fixed with 4% PFA in PBS and immunocytology was performed using anti-neuronal class III b-tubulin (TUJ1) monoclonal antibody (Covance). Over 150 neurites were chosen for measurement. Each selected neurite could be observed emerging from a distinguishable cell body, was longer than the diameter of the neuronal cell body, and was identifiable over its entire length. Fluorescence pictures were acquired with a Zeiss Axioplan2 microscope equipped with a digital CCD camera (PVCAM; Photometrics). The images were reversed to enhance contrast and the length of each neurite was measured after tracing them using Scion Image software (Scion). Neurite lengths were compared between groups using a one-way ANOVA test. In situ hybridization. In situ hybridization histochemistry was performed as previously described [9]. Two RNA probes, ranging from bases 1041–1443, and 890–1231, were designed. Selected regions were amplified by PCR and subcloned into pGEM-T vector (Promega), and the sequence was confirmed. These vectors were then cut by NcoI and NdeI separately and used as a DNA template to generate 35 S-labeled RNA probes. Briefly, hybridization was performed overnight at 55 °C. After dehydration, X-ray film (Kodak Biomax) was placed on the uncoated sections for 3 days. The slides were coated with Ilford Nuclear Research Emulsions K5 diluted in distilled water (1:1). After exposure for 4 weeks in a dark box at 4 °C, the slides were developed in Kodak D-19 and fixed with photographic fixer. Selected slides were counterstained with 0.1% thionine and examined under bright- and dark-field illumination.
Results FLRT3 is localized in the plasma membrane Previously, we performed a comprehensive analysis of gene expression following injury in the distal stump of the axotomized sciatic nerve using cDNA microarrays [7]. An EST clone, AI227034, was found to be induced 3.9-fold 7 days after axotomy, and was selected for further study. Induction of the gene expression was confirmed by RT-PCR in the previous study. AI227034 had significant overlap with the RIKEN cDNA, AK017456, which has a putative open reading frame. This protein coded by the clone is a homolog of human fibronectin leucine-rich transmembrane protein 3 (FLRT3) (AF169677). We obtained the full-length coding region of human FLRT3 cDNA (KIAA 1469). Next, to determine the subcellular localization of the FLRT3 protein, we constructed an expression vector carrying the HA-tag. Fig. 2 shows the immunofluorescence analysis of CHO cells transfected with HA-tagged FLRT3. We failed to detect the tag fused to the amino terminus of a putative signal peptide of FLRT3 in transfected CHO cells with both immunocytostaining and immunoblotting. Therefore, we truncated the first 28 amino acids of FLRT3 and generated an HA-FLRT3 construct containing a murine Ig kappa-chain V-J2-C signal peptide followed by an HA tag fused to amino acids 29–649 of FLRT3 (Fig. 1). Immunostaining for HA-FLRT3 in the transfected CHO cells with an
1088
L. Tsuji et al. / Biochemical and Biophysical Research Communications 313 (2004) 1086–1091
Fig. 1. FLRT3 protein structure showing the relative positions of a signal peptide, the leucine rich repeats, the fibronectin III like domain, and the transmembrane domain.
anti-HA antibody under a non-permeabilized condition showed that the amino-terminus of FLRT3 accumulated on the cell surface (Fig. 2B). However, we observed no immunoreactivity for the HA-tagged protein under the same condition (Fig. 2D), when CHO cells were transfected with the vector coding FLRT3-HA, a full-length FLRT3 fused with a carboxy-terminal HA tag (Fig. 1). These findings suggest that FLRT3 is a type 1 plasma membrane protein, with its amino terminus being present extracellularly.
FLRT3 promotes neurite outgrowth from cerebellar granule neurons The cell surface localization of FLRT3 and its upregulation in the distal stump of axotomized nerves suggest that FLRT3 might promote nerve regeneration by adhesive interactions as has been observed with other adhesion molecules. We used a neurite outgrowth assay to measure the ability of FLRT3 to promote neurite outgrowth. Cerebellar granule cells from P8 rat pups were dissected, dissociated, and plated at low density onto confluent monolayers of either CHO cells expressing FLRT3 or control CHO cells. Twenty-four hours after plating, cells were fixed and stained with antibodies to class III b-tubulin to visualize the neurites. Representative photomicrographs of these cultures demonstrate an increase in neurite outgrowth from neurons cultured on CHO cells expressing FLRT3 (Fig. 3B). To quantitate this effect, the length of the longest neurite per neuron, which did not have contacts with nearby neurons, was measured. At least 150 neurons per each condition were measured in three independent experiments (Fig. 3C). Neurons grown for 24 h on CHO cells expressing FLRT3 had neurites with an average length of 72 2.6 lm, whereas neurites from neurons grown on control CHO cells were 59 1.9 lm in length (p < 0:001; ANOVA). These results demonstrate that FLRT3 promotes neurite outgrowth. Uneven distribution of FLRT3 in the adult rat brain
Fig. 2. FLRT3 was localized on the plasma membrane. Confocal microscopic analysis of CHO cells transfected with HA-FLRT3 and FLRT3-HA. Staining of permeabilized cells with an anti-HA antibody (A,C). Staining of non-permeabilized cells with an anti-HA antibody detecting the N-terminal tag of HA-FLRT3 (B) and the C-terminal tag of FLRT3-HA (D).
To gain insight into FLRT3 function, we examined the localization of its mRNA in the brain. Fig. 4 shows the in situ hybridization analysis of the distribution of FLRT3 mRNA in sagittal sections of the adult rat brain. The signal for FLRT3 mRNA obtained using the 35 S-labeled RNA antisense probe was heteroge-
L. Tsuji et al. / Biochemical and Biophysical Research Communications 313 (2004) 1086–1091
1089
Fig. 3. FLRT3 promoted neurite outgrowth from primary cultured cerebellar granule neurons. (A) Expression of HA-FLRT3 in Flp-in CHO cells. (B) Representative image of cerebellar granule neurons of P8 rat pups plated on control CHO and FLRT3-CHO cells. The images are reversed to enhance contrast. (C) Neurite length of at least 150 neurons per each condition in three independent experiments is presented as the mean neurite length. ***p < 0:001; one-way ANOVA. Scale bar, 50 lm.
Fig. 4. In situ hybridization analysis of the distribution of FLRT3 mRNA in sagittal sections of adult rat brain. Film autoradiogram of whole brain (A). Dark-field photomicrographs at low magnification (B,D) and bright-field photomicrographs at high magnification (C,E) of emulsion-dipped sections counterstained with 0.1% thionine. (B,C) hippocampus, (D,E) basal ganglia. Cellular localization of the silver grains in CA3 pyramidal cells (C, asterisk) and globus pallidus (E, arrowheads). DG, dentate gyrus; Cpu, caudate-putamen; GP, globus pallidus; ac, anterior commissure; and ic, internal capsule.
1090
L. Tsuji et al. / Biochemical and Biophysical Research Communications 313 (2004) 1086–1091
neously distributed throughout the rat brain. No signal was observed with the 35 S-labeled sense probe (data not shown), showing the specificity of the probe. We used two probes and found an identical distribution of the mRNA, strengthening the validity of the data. Examination of sections counterstained with thionine suggested that FLRT3 mRNA was present preferentially in neurons (Figs. 4C and E). FLRT3 mRNA was widely distributed in the rat brain, especially in basal ganglia, hippocampus, and cerebellum (Fig. 4A). In the hippocampus, the highest levels of signals were seen in CA3 and CA4 pyramidal cells and dentate granular cells (Figs. 4B and C), whereas signals in CA1 pyramidal cells were lower than those seen in CA3, CA4 (Fig. 4B). In the olfactory bulb, signals were high in the internal granular layer and moderate in mitral cells, while the glomerular, external granular, and plexiform layers showed a very low expression (data not shown). In the cerebellum, the highest density was in the granular layer and moderate signals were detected in Purkinje’s cells (data not shown). There was intense labeling throughout the basal ganglia, including the caudate-putamen, globus pallidus, and amygdaloid complex (Figs. 4A and D).
Discussion Upon axotomy, Schwann cells in the distal nerve segment undergo extensive changes along with axonal degeneration. Myelin sheaths are dissoluted and subsequently absorbed by the surrounding cellular environment, a process called Wallerian degeneration. During dissolution and absorption of myelin sheaths by macrophages, the remaining Schwann cells divide and align longitudinally within basal lamina tubes while extending slender cytoplasmic processes, which form the cellular strands called the band of B€ ungner (the Schwann cell column). The Schwann cells in the band provide effective pathways for the growth of regenerating axons, expressing adhesion molecules on the surface of neighboring cells as scaffolds and produce various trophic factors [1,2,10]. Thus, some cues for regenerating axons involve adhesive cell–cell contacts between the growth cone and adhesion molecules [11,12]. Many of the glycoproteins involved in the adhesion of regenerating neural cells belong to several major structural families. One is the immunoglobulin superfamily, e.g., neural cell adhesion molecule (NCAM) and L1. Another comprises a group of structurally related glycoproteins called cadherins such as N-cadherin. Immunoglobulin superfamily molecules, such as NCAM and L1, are characterized by structural motifs in their extracellular portions that are homologous to immunoglobulin constant-region (Ig) domains and fi-
bronectin type III (FNIII) domains. Ig domains are reported to mediate their homophilic adhesion [13]. There is little evidence for a role of FNIII domains in the homophilic interactions that occur between cells; however, there is evidence showing homophilic interaction in a cis manner between FNIII domains expressed in a single cell [13,14]. Although N-CAM and L1 are developmentally expressed on the plasma membranes of axons and Schwann cells, they are down-regulated in mature myelinated nerves, but not unmyelinated fibers [11]. Once denervation occurs, NCAM and L1 are up-regulated and re-expressed on the surfaces of Schwann cells [4]. When regenerating axons grow into Schwann cell columns, N-CAM and L1 are expressed on the plasma membrane at the axonSchwann cell attachment [2]. As Schwann cells begin to form myelin lamellae on the axon, N-CAM and L1 are rapidly down-regulated and become undetectable [15]. N-cadherin is also involved in the promotion of outgrowth of regenerating axons in the Schwann cell column in vivo [16]. Moreover, ninjurin-1 and -2, classified as type 3b transmembrane proteins, are upregulated in both neurons and Schwann cells after injury and promote neurite extension by homophilic adhesion [17,18]. In this report, we described the characterization of FLRT3. FLRT3 was identified using cDNA microarrays and was demonstrated to be up-regulated in the distal segment of the transected sciatic nerve [7]. FLRT3 has a putative signal peptide, leucine rich repeats flanked by a cysteine rich domain, fibronectin type III domains, and a single transmembrane domain [8]. Therefore, FLRT3 might promote axon outgrowth by contact with neurons following axonal injury, working as an adhesion molecule that is expressed in glial cells. Other adhesion molecules, such as N-CAM, L1, N-cadherin, and ninjurin-1,2, which promote axon outgrowth, are reported to function by homophilic binding [2,17,18]. However, our preliminary experiments using immunoprecipitation show no evidence for a homophilic interaction of FLRT3 (data not shown). As the fibronectin type III domain was reported to be involved in neurite outgrowth [19,20], neurite elongation effects of FLRT3 might be attributable, at least partly, to this domain. In summary, using cDNA microarrays we identified FLRT3 as one of the up-regulated genes expressed in the distal segment of the sciatic nerve 7 days after transection relative to the intact sciatic nerve. The neurons plated on CHO cells expressing FLRT3 extended significantly longer neurites than those plated on wild-type CHO cells, suggesting that FLRT3 contributes to axonal outgrowth. Distribution of FLRT3 mRNA in adult rat brain suggests that FLRT3 may play a role in neuronal development as well as regeneration following axonal injury.
L. Tsuji et al. / Biochemical and Biophysical Research Communications 313 (2004) 1086–1091
References [1] Y.J. Son, W.J. Thompson, Schwann cell processes guide regeneration of peripheral axons, Neuron 14 (1995) 125–132. [2] J.W. Fawcett, R.J. Keynes, Peripheral nerve regeneration, Annu. Rev. Neurosci. 13 (1990) 43–60. [3] J.L. Bixby, J. Lilien, L.F. Reichardt, Identification of the major proteins that promote neuronal process outgrowth on Schwann cells in vitro, J. Cell Biol. 107 (1988) 353–361. [4] R. Martini, M. Schachner, Immunoelectron microscopic localization of neural cell adhesion molecules (L1, N-CAM, and myelinassociated glycoprotein) in regenerating adult mouse sciatic nerve, J. Cell Biol. 106 (1988) 1735–1746. [5] F. Rieger, M. Nicolet, M. Pincon-Raymond, M. Murawsky, G. Levi, G.M. Edelman, Distribution and role in regeneration of NCAM in the basal laminae of muscle and Schwann cells, J. Cell Biol. 107 (1988) 707–719. [6] G.A. Brook, D.A. Houweling, R.G. Gieling, T. Hermanns, E.A. Joosten, D.P. Bar, W.H. Gispen, A.B. Schmitt, P. Leprince, J. Noth, W. Nacimiento, Attempted endogenous tissue repair following experimental spinal cord injury in the rat: involvement of cell adhesion molecules L1 and NCAM? Eur. J. Neurosci. 12 (2000) 3224–3238. [7] T. Kubo, T. Yamashita, A. Yamaguchi, K. Hosokawa, M. Tohyama, Analysis of genes induced in peripheral nerve after axotomy using cDNA microarrays, J. Neurochem. 82 (2002) 1129–1136. [8] S.E. Lacy, C.G. Bonnemann, E.A. Buzney, L.M. Kunkel, Identification of FLRT1, FLRT2, and FLRT3: a novel family of transmembrane leucine-rich repeat proteins, Genomics 62 (1999) 417–426. [9] S. Miyake, T. Yamashita, M. Taniguchi, M. Tamatani, K. Sato, Y. Kawai, E. Senba, N. Mitsuda, O. Hori, A. Yamaguchi, M. Tohyama, Expression of mitochondrial tricarboxylate carrier TCC mRNA and protein in the rat brain, Brain Res. Mol. Brain Res. 100 (2002) 67–73.
1091
[10] R.P. Bunge, Expanding roles for the Schwann cell: ensheathment, myelination, trophism and regeneration, Curr. Opin. Neurobiol. 3 (1993) 805–809. [11] R. Martini, Expression and functional roles of neural cell surface molecules and extracellular matrix components during development and regeneration of peripheral nerves, J. Neurocytol. 23 (1994) 1–28. [12] J.A. Brodkey, M.A. Gates, E.D. Laywell, D.A. Steindler, The complex nature of interactive neuroregeneration-related molecules, Exp. Neurol. 123 (1993) 251–270. [13] F.S. Walsh, P. Doherty, Neural cell adhesion molecules of the immunoglobulin superfamily: role in axon growth and guidance, Annu. Rev. Cell Dev. Biol. 13 (1997) 425–456. [14] D. Peck, F.S. Walsh, Differential effects of over-expressed neural cell adhesion molecule isoforms on myoblast fusion, J. Cell Biol. 123 (1993) 1587–1595. [15] B. Seilheimer, M. Schachner, Studies of adhesion molecules mediating interactions between cells of peripheral nervous system indicate a major role for L1 in mediating sensory neuron growth on Schwann cells in culture, J. Cell Biol. 107 (1988) 341–351. [16] C. Ide, Peripheral nerve regeneration, Neurosci. Res. 25 (1996) 101–121. [17] T. Araki, J. Milbrandt, Ninjurin, a novel adhesion molecule, is induced by nerve injury and promotes axonal growth, Neuron 17 (1996) 353–361. [18] T. Araki, J. Milbrandt, Ninjurin2, a novel homophilic adhesion molecule, is expressed in mature sensory and enteric neurons and promotes neurite outgrowth, J. Neurosci. 20 (2000) 187–195. [19] K. Husmann, A. Faissner, M. Schachner, Tenascin promotes cerebellar granule cell migration and neurite outgrowth by different domains in the fibronectin type III repeats, J. Cell Biol. 116 (1992) 1475–1486. [20] U. Dorries, J. Taylor, Z. Xiao, A. Lochter, D. Montag, M. Schachner, Distinct effects of recombinant tenascin-C domains on neuronal cell adhesion, growth cone guidance, and neuronal polarity, J. Neurosci. Res. 43 (1996) 420–438.
FLRT3, a cell surface molecule containing LRR repeats and a FNIII domain, promotes neurite outgrowthq Lyuji Tsuji,a,b,c Toshihide Yamashita,d,* Tateki Kubo,b Tomas Madura,a,b,c Hiroyuki Tanaka,a,c Ko Hosokawa,b and Masaya Tohyamaa,c a
c
Department of Anatomy and Neuroscience, Osaka University Graduate School of Medicine, Osaka, Japan b Department of Plastic Surgery, Osaka University Graduate School of Medicine, Osaka, Japan Core Research for Evolutional Science and Technology (CREST) of Japan Science and Technology Corporation (JST), Kawaguchi, Japan d Department of Neurobiology, Graduate School of Medicine, Chiba University, Chiba, Japan Received 7 December 2003
Abstract The mature peripheral nervous system has the ability to survive and to regenerate its axons following axonal injury. After nerve injury, the distal axonal and myelin segment undergoes dissolution and absorption by the surrounding cellular environment, a process called Wallerian degeneration. Using cDNA microarrays, we isolated FLRT3 as one of the up-regulated genes expressed in the distal segment of the sciatic nerve 7 days after transection relative to those of the intact sciatic nerve. FLRT3 is a putative type I transmembrane protein containing 10 leucine-rich repeats, a fibronectin type III domain, and an intracellular tail. The neurons plated on CHO cells expressing FLRT3 extended significantly longer neurites than those plated on wild-type CHO cells, demonstrating that FLRT3 promotes neurite outgrowth. FLRT3 mRNA was especially abundant in the basal ganglia, the granular layer of cerebellum, and the hippocampus, except the CA1 region in the adult rat brain. Thus, FLRT3 may contribute to regeneration following axonal injury. Ó 2003 Elsevier Inc. All rights reserved. Keywords: cDNA microarrays; Injury-induced molecules; Nerve regeneration; Schwann cell; Adhesion molecules
In contrast to the central glia, following axonal injury, Wallerian degeneration occurs in the distal segment of the injured site, where Schwann cells proliferate and form the band of B€ ungner (Schwann cell column) to support and guide axoplasmic sprouts from the proximal portion of severed axons within the bands [1,2]. Schwann cells in the distal segment of the injured peripheral nerve express various neurotrophic factors and adhesion molecules, such as L1, NCAM, and N-cadherin, that promote neurite outgrowth by regulating contact between axons and Schwann cells [3–6]. Identification of the molecular basis of this process is expected to lead to opportunities to rescue injured neurons that q Abbreviations: EST, expressed sequence tag; N-CAM, neural cell adhesion molecule; FNIII, fibronectin type III. * Corresponding author. Fax: +81-6-6879-3229. E-mail address: [email protected] (T. Yamashita).
0006-291X/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2003.12.047
do not survive or regenerate after axotomy. In an attempt to explore this molecular mechanism, we used a cDNA library and microarrays derived from the distal stumps of the post-injury sciatic nerve which was rich in non-myelinating Schwann cells [7]. Here we report that FLRT3, one of the up-regulated genes, is expressed in the distal segment of sciatic nerve 7 days after transection relative to those of the intact sciatic nerve [7]. FLRT, a family of cell surface molecules containing LRR repeats and a FNIII domain, consists of an N-terminal putative signal peptide, 10 leucine-rich repeats (LRR) flanked by cysteine-rich regions, fibronectin type III domains, the putative transmembrane domains, and short intracellular tails. Potential N-linked glycosylation sites appear four times (226N, 282N, 296N, and 508N) within the extracellular region of FLRT3. FLRT3 mRNA is broadly expressed in various tissues such as kidney, skeletal muscle, lung, and brain [8].
L. Tsuji et al. / Biochemical and Biophysical Research Communications 313 (2004) 1086–1091
Materials and methods All procedures involving animals were reviewed and approved by the local Institutional Animal Care and Use Committee. Construction of the FLRT3 expression vectors. All expression vectors were constructed using the GATEWAY system (Invitrogen). We constructed two destination vectors, pcDNA5FRT/Igjleader/HA/ ccdB and pcDNA/ccdB/HA. An oligonucleotide coding for the HA tag, and a GATEWAY Reading Frame Cassette B, fragment were sequentially ligated into the KpnI–EcoRV sites of the pSecTag2/C vector (Invitrogen) to generate the pSecTag2/HA/ccdB vector. An NheI–PmeI fragment, containing a murine Igj chain V-J2-C signal peptide from the pSecTag2 vector, a HA tag and a GATEWAY cassette B was generated by PCR amplification. This fragment was ligated into NheI–EcoRV sites of pcDNA5FRT (Invitrogen) to generate pcDNA5FRT/Igjleader/HA/ccdB. pcDNA/ccdB/HA was generated by sequential ligation of a GATEWAY Reading Frame Cassette B fragment and an oligonucleotide coding for the HA tag into pcDNA3.1/Zeo (Invitrogen). All destination vectors were linealized by a cut at the BamHI site. DNA fragments for 29-FLRT3 (amino acids 29–649) and FLRT3-C (full-length FLRT3 with the kozak consensus skipping stop codon) were amplified by PCR with the human FLRT3 clone, KIAA1469, from Kazusa DNA Research Institute (Chiba, Japan), as a template. These fragments were integrated into pDONR221 using BP clonase to generate pDONR 29-FLRT3 and pDONR FLRT3-C entry vectors, respectively. Expression vectors for HAFLRT3 were generated by LR reaction with pDONR 29-FLRT3 and linealized pcDNA5FRT/Igjleader/HA/ccdB, and FLRT3-HA with pDONR FLRT3-C and linealized pcDNA/ccdB/HA. All PCR amplified fragments were verified by DNA sequencing. Transient transfections and immunofluorescence. CHO cells were routinely maintained in DMEM/F12 medium supplemented with 10% FCS. For transient transfections, the cells were grown on 35 mm dishes and transfected using LipofectAMINE (Invitrogen) according to the manufacturer’s recommendations. After 5 h, equal volumes of DMEM/F12 medium with 10% FCS were added to each well and the cells were grown for an additional 19 h. The cells were fixed with 4% PFA for 20 min at room temperature. Incubation with primary antibodies in PBS containing 0.5% BSA and 0.1% Triton-X was performed overnight at 4 °C and with secondary antibodies in PBS for 60 min at room temperature. The cells were washed in PBS after each incubation. Triton-X was omitted where indicated as the non-permeabilized condition. We confirmed that cells transfected with tagged Rho kinase showed no staining under this condition. The following antibodies were used at the dilution of 1:1000: monoclonal anti-HA antibody, HA.11 (Covance), and Alexa 488 goat anti-mouse IgG conjugate (Molecular Probes). Immunofluorescence images were acquired with a Zeiss LSM 510 laser scanning confocal microscope. Generation of FLRT3-CHO cells. The Flp-In System (Invitrogen) was used to generate FLRT3 expressing cells according to the manufacturer’s directions. pcDNA5FRT/Igjleader/HA/29-FLRT3 and pOG44 were co-transfected into Flp-in CHO cells and stable expressing cells were generated after growth in medium containing Hygromycin B (600 lg/ml; Invitrogen) for 2 weeks. Expression of HAFLRT3 was confirmed by Western blot (Fig. 3A) and immunocytology (data not shown). Western blot analysis. HA-FLRT3 was extracted by adding SDS sample buffer containing 12% b-mercaptoethanol directly to FLRT3CHO cells and then the sample was loaded on a 10% SDS–PAGE gel. This gel was then transferred to a PVDF membrane. The protein was detected by an anti-HA antibody (HA.11), an HRP-linked secondary antibody (Cell Signaling), and ECL detection reagents (Amersham). Neurite outgrowth assays. Cerebellar granule cells from postnatal day 8 (P8) rat pups were dissociated by trypsinization (0.25% trypsin in PBS for 10 min at 37 °C) followed by resuspension in serum-containing medium to inhibit trypsin and trituration. After three washes with
1087
PBS, cells were plated on confluent monolayers of either control CHO cells or CHO cells expressing FLRT3 in 35-mm plates at the density of 160 cells/mm2 . Cultures were grown in a serum-free DMEM/F12 medium. Twenty-four hours after plating, the cells were fixed with 4% PFA in PBS and immunocytology was performed using anti-neuronal class III b-tubulin (TUJ1) monoclonal antibody (Covance). Over 150 neurites were chosen for measurement. Each selected neurite could be observed emerging from a distinguishable cell body, was longer than the diameter of the neuronal cell body, and was identifiable over its entire length. Fluorescence pictures were acquired with a Zeiss Axioplan2 microscope equipped with a digital CCD camera (PVCAM; Photometrics). The images were reversed to enhance contrast and the length of each neurite was measured after tracing them using Scion Image software (Scion). Neurite lengths were compared between groups using a one-way ANOVA test. In situ hybridization. In situ hybridization histochemistry was performed as previously described [9]. Two RNA probes, ranging from bases 1041–1443, and 890–1231, were designed. Selected regions were amplified by PCR and subcloned into pGEM-T vector (Promega), and the sequence was confirmed. These vectors were then cut by NcoI and NdeI separately and used as a DNA template to generate 35 S-labeled RNA probes. Briefly, hybridization was performed overnight at 55 °C. After dehydration, X-ray film (Kodak Biomax) was placed on the uncoated sections for 3 days. The slides were coated with Ilford Nuclear Research Emulsions K5 diluted in distilled water (1:1). After exposure for 4 weeks in a dark box at 4 °C, the slides were developed in Kodak D-19 and fixed with photographic fixer. Selected slides were counterstained with 0.1% thionine and examined under bright- and dark-field illumination.
Results FLRT3 is localized in the plasma membrane Previously, we performed a comprehensive analysis of gene expression following injury in the distal stump of the axotomized sciatic nerve using cDNA microarrays [7]. An EST clone, AI227034, was found to be induced 3.9-fold 7 days after axotomy, and was selected for further study. Induction of the gene expression was confirmed by RT-PCR in the previous study. AI227034 had significant overlap with the RIKEN cDNA, AK017456, which has a putative open reading frame. This protein coded by the clone is a homolog of human fibronectin leucine-rich transmembrane protein 3 (FLRT3) (AF169677). We obtained the full-length coding region of human FLRT3 cDNA (KIAA 1469). Next, to determine the subcellular localization of the FLRT3 protein, we constructed an expression vector carrying the HA-tag. Fig. 2 shows the immunofluorescence analysis of CHO cells transfected with HA-tagged FLRT3. We failed to detect the tag fused to the amino terminus of a putative signal peptide of FLRT3 in transfected CHO cells with both immunocytostaining and immunoblotting. Therefore, we truncated the first 28 amino acids of FLRT3 and generated an HA-FLRT3 construct containing a murine Ig kappa-chain V-J2-C signal peptide followed by an HA tag fused to amino acids 29–649 of FLRT3 (Fig. 1). Immunostaining for HA-FLRT3 in the transfected CHO cells with an
1088
L. Tsuji et al. / Biochemical and Biophysical Research Communications 313 (2004) 1086–1091
Fig. 1. FLRT3 protein structure showing the relative positions of a signal peptide, the leucine rich repeats, the fibronectin III like domain, and the transmembrane domain.
anti-HA antibody under a non-permeabilized condition showed that the amino-terminus of FLRT3 accumulated on the cell surface (Fig. 2B). However, we observed no immunoreactivity for the HA-tagged protein under the same condition (Fig. 2D), when CHO cells were transfected with the vector coding FLRT3-HA, a full-length FLRT3 fused with a carboxy-terminal HA tag (Fig. 1). These findings suggest that FLRT3 is a type 1 plasma membrane protein, with its amino terminus being present extracellularly.
FLRT3 promotes neurite outgrowth from cerebellar granule neurons The cell surface localization of FLRT3 and its upregulation in the distal stump of axotomized nerves suggest that FLRT3 might promote nerve regeneration by adhesive interactions as has been observed with other adhesion molecules. We used a neurite outgrowth assay to measure the ability of FLRT3 to promote neurite outgrowth. Cerebellar granule cells from P8 rat pups were dissected, dissociated, and plated at low density onto confluent monolayers of either CHO cells expressing FLRT3 or control CHO cells. Twenty-four hours after plating, cells were fixed and stained with antibodies to class III b-tubulin to visualize the neurites. Representative photomicrographs of these cultures demonstrate an increase in neurite outgrowth from neurons cultured on CHO cells expressing FLRT3 (Fig. 3B). To quantitate this effect, the length of the longest neurite per neuron, which did not have contacts with nearby neurons, was measured. At least 150 neurons per each condition were measured in three independent experiments (Fig. 3C). Neurons grown for 24 h on CHO cells expressing FLRT3 had neurites with an average length of 72 2.6 lm, whereas neurites from neurons grown on control CHO cells were 59 1.9 lm in length (p < 0:001; ANOVA). These results demonstrate that FLRT3 promotes neurite outgrowth. Uneven distribution of FLRT3 in the adult rat brain
Fig. 2. FLRT3 was localized on the plasma membrane. Confocal microscopic analysis of CHO cells transfected with HA-FLRT3 and FLRT3-HA. Staining of permeabilized cells with an anti-HA antibody (A,C). Staining of non-permeabilized cells with an anti-HA antibody detecting the N-terminal tag of HA-FLRT3 (B) and the C-terminal tag of FLRT3-HA (D).
To gain insight into FLRT3 function, we examined the localization of its mRNA in the brain. Fig. 4 shows the in situ hybridization analysis of the distribution of FLRT3 mRNA in sagittal sections of the adult rat brain. The signal for FLRT3 mRNA obtained using the 35 S-labeled RNA antisense probe was heteroge-
L. Tsuji et al. / Biochemical and Biophysical Research Communications 313 (2004) 1086–1091
1089
Fig. 3. FLRT3 promoted neurite outgrowth from primary cultured cerebellar granule neurons. (A) Expression of HA-FLRT3 in Flp-in CHO cells. (B) Representative image of cerebellar granule neurons of P8 rat pups plated on control CHO and FLRT3-CHO cells. The images are reversed to enhance contrast. (C) Neurite length of at least 150 neurons per each condition in three independent experiments is presented as the mean neurite length. ***p < 0:001; one-way ANOVA. Scale bar, 50 lm.
Fig. 4. In situ hybridization analysis of the distribution of FLRT3 mRNA in sagittal sections of adult rat brain. Film autoradiogram of whole brain (A). Dark-field photomicrographs at low magnification (B,D) and bright-field photomicrographs at high magnification (C,E) of emulsion-dipped sections counterstained with 0.1% thionine. (B,C) hippocampus, (D,E) basal ganglia. Cellular localization of the silver grains in CA3 pyramidal cells (C, asterisk) and globus pallidus (E, arrowheads). DG, dentate gyrus; Cpu, caudate-putamen; GP, globus pallidus; ac, anterior commissure; and ic, internal capsule.
1090
L. Tsuji et al. / Biochemical and Biophysical Research Communications 313 (2004) 1086–1091
neously distributed throughout the rat brain. No signal was observed with the 35 S-labeled sense probe (data not shown), showing the specificity of the probe. We used two probes and found an identical distribution of the mRNA, strengthening the validity of the data. Examination of sections counterstained with thionine suggested that FLRT3 mRNA was present preferentially in neurons (Figs. 4C and E). FLRT3 mRNA was widely distributed in the rat brain, especially in basal ganglia, hippocampus, and cerebellum (Fig. 4A). In the hippocampus, the highest levels of signals were seen in CA3 and CA4 pyramidal cells and dentate granular cells (Figs. 4B and C), whereas signals in CA1 pyramidal cells were lower than those seen in CA3, CA4 (Fig. 4B). In the olfactory bulb, signals were high in the internal granular layer and moderate in mitral cells, while the glomerular, external granular, and plexiform layers showed a very low expression (data not shown). In the cerebellum, the highest density was in the granular layer and moderate signals were detected in Purkinje’s cells (data not shown). There was intense labeling throughout the basal ganglia, including the caudate-putamen, globus pallidus, and amygdaloid complex (Figs. 4A and D).
Discussion Upon axotomy, Schwann cells in the distal nerve segment undergo extensive changes along with axonal degeneration. Myelin sheaths are dissoluted and subsequently absorbed by the surrounding cellular environment, a process called Wallerian degeneration. During dissolution and absorption of myelin sheaths by macrophages, the remaining Schwann cells divide and align longitudinally within basal lamina tubes while extending slender cytoplasmic processes, which form the cellular strands called the band of B€ ungner (the Schwann cell column). The Schwann cells in the band provide effective pathways for the growth of regenerating axons, expressing adhesion molecules on the surface of neighboring cells as scaffolds and produce various trophic factors [1,2,10]. Thus, some cues for regenerating axons involve adhesive cell–cell contacts between the growth cone and adhesion molecules [11,12]. Many of the glycoproteins involved in the adhesion of regenerating neural cells belong to several major structural families. One is the immunoglobulin superfamily, e.g., neural cell adhesion molecule (NCAM) and L1. Another comprises a group of structurally related glycoproteins called cadherins such as N-cadherin. Immunoglobulin superfamily molecules, such as NCAM and L1, are characterized by structural motifs in their extracellular portions that are homologous to immunoglobulin constant-region (Ig) domains and fi-
bronectin type III (FNIII) domains. Ig domains are reported to mediate their homophilic adhesion [13]. There is little evidence for a role of FNIII domains in the homophilic interactions that occur between cells; however, there is evidence showing homophilic interaction in a cis manner between FNIII domains expressed in a single cell [13,14]. Although N-CAM and L1 are developmentally expressed on the plasma membranes of axons and Schwann cells, they are down-regulated in mature myelinated nerves, but not unmyelinated fibers [11]. Once denervation occurs, NCAM and L1 are up-regulated and re-expressed on the surfaces of Schwann cells [4]. When regenerating axons grow into Schwann cell columns, N-CAM and L1 are expressed on the plasma membrane at the axonSchwann cell attachment [2]. As Schwann cells begin to form myelin lamellae on the axon, N-CAM and L1 are rapidly down-regulated and become undetectable [15]. N-cadherin is also involved in the promotion of outgrowth of regenerating axons in the Schwann cell column in vivo [16]. Moreover, ninjurin-1 and -2, classified as type 3b transmembrane proteins, are upregulated in both neurons and Schwann cells after injury and promote neurite extension by homophilic adhesion [17,18]. In this report, we described the characterization of FLRT3. FLRT3 was identified using cDNA microarrays and was demonstrated to be up-regulated in the distal segment of the transected sciatic nerve [7]. FLRT3 has a putative signal peptide, leucine rich repeats flanked by a cysteine rich domain, fibronectin type III domains, and a single transmembrane domain [8]. Therefore, FLRT3 might promote axon outgrowth by contact with neurons following axonal injury, working as an adhesion molecule that is expressed in glial cells. Other adhesion molecules, such as N-CAM, L1, N-cadherin, and ninjurin-1,2, which promote axon outgrowth, are reported to function by homophilic binding [2,17,18]. However, our preliminary experiments using immunoprecipitation show no evidence for a homophilic interaction of FLRT3 (data not shown). As the fibronectin type III domain was reported to be involved in neurite outgrowth [19,20], neurite elongation effects of FLRT3 might be attributable, at least partly, to this domain. In summary, using cDNA microarrays we identified FLRT3 as one of the up-regulated genes expressed in the distal segment of the sciatic nerve 7 days after transection relative to the intact sciatic nerve. The neurons plated on CHO cells expressing FLRT3 extended significantly longer neurites than those plated on wild-type CHO cells, suggesting that FLRT3 contributes to axonal outgrowth. Distribution of FLRT3 mRNA in adult rat brain suggests that FLRT3 may play a role in neuronal development as well as regeneration following axonal injury.
L. Tsuji et al. / Biochemical and Biophysical Research Communications 313 (2004) 1086–1091
References [1] Y.J. Son, W.J. Thompson, Schwann cell processes guide regeneration of peripheral axons, Neuron 14 (1995) 125–132. [2] J.W. Fawcett, R.J. Keynes, Peripheral nerve regeneration, Annu. Rev. Neurosci. 13 (1990) 43–60. [3] J.L. Bixby, J. Lilien, L.F. Reichardt, Identification of the major proteins that promote neuronal process outgrowth on Schwann cells in vitro, J. Cell Biol. 107 (1988) 353–361. [4] R. Martini, M. Schachner, Immunoelectron microscopic localization of neural cell adhesion molecules (L1, N-CAM, and myelinassociated glycoprotein) in regenerating adult mouse sciatic nerve, J. Cell Biol. 106 (1988) 1735–1746. [5] F. Rieger, M. Nicolet, M. Pincon-Raymond, M. Murawsky, G. Levi, G.M. Edelman, Distribution and role in regeneration of NCAM in the basal laminae of muscle and Schwann cells, J. Cell Biol. 107 (1988) 707–719. [6] G.A. Brook, D.A. Houweling, R.G. Gieling, T. Hermanns, E.A. Joosten, D.P. Bar, W.H. Gispen, A.B. Schmitt, P. Leprince, J. Noth, W. Nacimiento, Attempted endogenous tissue repair following experimental spinal cord injury in the rat: involvement of cell adhesion molecules L1 and NCAM? Eur. J. Neurosci. 12 (2000) 3224–3238. [7] T. Kubo, T. Yamashita, A. Yamaguchi, K. Hosokawa, M. Tohyama, Analysis of genes induced in peripheral nerve after axotomy using cDNA microarrays, J. Neurochem. 82 (2002) 1129–1136. [8] S.E. Lacy, C.G. Bonnemann, E.A. Buzney, L.M. Kunkel, Identification of FLRT1, FLRT2, and FLRT3: a novel family of transmembrane leucine-rich repeat proteins, Genomics 62 (1999) 417–426. [9] S. Miyake, T. Yamashita, M. Taniguchi, M. Tamatani, K. Sato, Y. Kawai, E. Senba, N. Mitsuda, O. Hori, A. Yamaguchi, M. Tohyama, Expression of mitochondrial tricarboxylate carrier TCC mRNA and protein in the rat brain, Brain Res. Mol. Brain Res. 100 (2002) 67–73.
1091
[10] R.P. Bunge, Expanding roles for the Schwann cell: ensheathment, myelination, trophism and regeneration, Curr. Opin. Neurobiol. 3 (1993) 805–809. [11] R. Martini, Expression and functional roles of neural cell surface molecules and extracellular matrix components during development and regeneration of peripheral nerves, J. Neurocytol. 23 (1994) 1–28. [12] J.A. Brodkey, M.A. Gates, E.D. Laywell, D.A. Steindler, The complex nature of interactive neuroregeneration-related molecules, Exp. Neurol. 123 (1993) 251–270. [13] F.S. Walsh, P. Doherty, Neural cell adhesion molecules of the immunoglobulin superfamily: role in axon growth and guidance, Annu. Rev. Cell Dev. Biol. 13 (1997) 425–456. [14] D. Peck, F.S. Walsh, Differential effects of over-expressed neural cell adhesion molecule isoforms on myoblast fusion, J. Cell Biol. 123 (1993) 1587–1595. [15] B. Seilheimer, M. Schachner, Studies of adhesion molecules mediating interactions between cells of peripheral nervous system indicate a major role for L1 in mediating sensory neuron growth on Schwann cells in culture, J. Cell Biol. 107 (1988) 341–351. [16] C. Ide, Peripheral nerve regeneration, Neurosci. Res. 25 (1996) 101–121. [17] T. Araki, J. Milbrandt, Ninjurin, a novel adhesion molecule, is induced by nerve injury and promotes axonal growth, Neuron 17 (1996) 353–361. [18] T. Araki, J. Milbrandt, Ninjurin2, a novel homophilic adhesion molecule, is expressed in mature sensory and enteric neurons and promotes neurite outgrowth, J. Neurosci. 20 (2000) 187–195. [19] K. Husmann, A. Faissner, M. Schachner, Tenascin promotes cerebellar granule cell migration and neurite outgrowth by different domains in the fibronectin type III repeats, J. Cell Biol. 116 (1992) 1475–1486. [20] U. Dorries, J. Taylor, Z. Xiao, A. Lochter, D. Montag, M. Schachner, Distinct effects of recombinant tenascin-C domains on neuronal cell adhesion, growth cone guidance, and neuronal polarity, J. Neurosci. Res. 43 (1996) 420–438.