Résumé
Nogo recombinant proteins To express Amino-Nogo, the human Nogo-A cDNA for residues 1±1,040 was ligated into pcDNA3.1MycHis (Invitrogen, Burlingame, California) with an in-frame Myc-His tag. We transfected this plasmid into HEK293T cells and Amino-Nogo was puri®ed with a Ni 2+ resin 10. The human Nogo-66 sequence was ligated into pcAP-5 (ref. 10) in frame with the signal sequence, 6´His tag and placental AP coding region. This plasmid was transfected into HEK293 cells, and secreted AP±Nogo was puri®ed by Ni 2+ af®nity chromatography. GST±Nogo-66 has been described 1. Nogo-66 receptor binding assays and expression cloning To detect AP±Nogo binding, cultures were washed with Hanks balanced salt solution containing 20 mM sodium HEPES, pH 7.05, and 1 mg ml-1 bovine serum albumin (BSA) (HBH). The plates were then incubated with AP±Nogo in HBH for 2 h at 23 8C. We detected and quanti®ed bound AP±Nogo as for AP±Sema3A 11. For saturation analysis of AP±Nogo bound to COS-7 cells, bound AP±Nogo protein was eluted with 1% Triton X-100. After heat inactivation of endogenous AP, we measured AP±Nogo using p-nitrophenyl phosphate as substrate. For expression cloning of a Nogo-66 receptor, pools of 5,000 arrayed clones from a mouse adult-brain cDNA library (Origene Technologies, Rockville, Maryland) were transfected into COS-7 cells, and AP±Nogo binding was assessed. We isolated single NgR cDNA clones by sib selection and sequenced them. A Myc±NgR vector was created in pSecTag2-Hygro (Invitrogen) using the signal peptide of pSecTag2 fused to Myc and residues 27±473 of NgR. Human NgR cDNA was predicted from a human genomic cosmid sequence (AC007663). Oligonucleotide primers based on the predicted human cDNA ampli®ed the cDNA from a human adult-brain cDNA library (Origene Technologies). To assess binding of Myc±NgR to the cell membrane, particulate fractions were treated with or without 5 U PI-PLC (Sigma, St. Louis, MO) per mg of HEK293T cell protein for 1 h at 30 8C in HBH. After centrifugation at 100,000g for 1 h, we analysed equal proportions of the soluble and particulate fractions. To assess the physical interaction of NgR with Nogo-66, we incubated the PI-PLC extract (50 mg total protein) with 10 mg GST±Nogo-66 or buffer, or 10 mg GST for 1 h at 23 8C. We added glutathione-coupled agarose to bind GST and associated proteins. We analysed bound proteins by anti-Myc immunoblot. RNA analysis For northern blots, 1 mg poly(A) + RNA from each adult mouse tissue on a nylon membrane (Origene Technologies) was hybridized with a full-length 32 P-labelled probe 1,12. We used digoxegenin-labelled riboprobes (nucleotides 1±1,222) and adult mouse brain sections 1,12 for in situ hybridization. The sense probe produced no signal. Nogo-66 receptor antibodies A GST±NgR (residues 27±447) fusion protein was puri®ed from Escherichia coli and used to immunize rabbits. We diluted immune serum 3,000-fold for immunoblots and 1,000fold for immunohistology on tissue-culture samples that had been ®xed by formalin. Staining of tissue was totally abolished by addition of 5 mg ml-1 GST±NgR. Cell spreading, neurite outgrowth and viral infection To measure spreading rates, subcon¯uent NIH 3T3 ®broblasts or COS-7 cells were plated for 1 h in serum-containing medium before ®xation and staining with rhodaminephalloidin. Glass coverslips were precoated with 100 mg ml-1 poly-L-lysine, washed, and then 3 ml drops of PBS containing 15 pmol Amino-Nogo, 15 pmol GST±Nogo-66, 15 pmol poly-Asp (M r 35 K, Sigma), or no protein were spotted and dried. We added soluble Nogo protein preparations (100 nM) at the time of plating. Amino-Nogo was added alone or after a pre-incubation with a twofold molar excess of anti-Myc 9E10 antibody, or with a twofold excess of anti-Myc plus a twofold excess of puri®ed goat anti-mouse IgG. Chick E5 spinal cord, chick E7±E13 DRG, chick E7 retina and mouse P4 cerebellar neuron culture, growth-cone-collapse assays and neurite-outgrowth assays have been described 1,10±13. Here, outgrowth from dissociated neurons was assessed after 12±24 h. For the substrate-bound experiments, glass chamber slides were coated with 100 mg ml-1 poly-L-lysine, washed, and then 3 ml drops of PBS containing 15 pmol Amino-Nogo, 15 pmol GST±Nogo-66, 15 pmol poly-Asp, or no protein were spotted and dried. After three PBS washes, we coated slides with 10 mg ml-1 laminin. After aspiration of laminin, dissociated neurons were added. For the soluble Nogo experiments, slides were coated with poly-Llysine and laminin in the same fashion, and then 100 nM Amino-Nogo, 100 nM clustered Amino-Nogo, or 100 nM GST±Nogo-66 was added to the culture medium at the time of plating. After 1 d in vitro, some DRG explants were treated for 30 min with 1 unit ml-1 PI-PLC (Sigma) before the growth-cone-collapse assay. An HSV-Myc±NgR stock was prepared as described 10. We infected E7 retinal explants for 24 h. We stained some cultures infected with HSV-Myc±NgR with anti-Myc antibody to verify protein expression. Error bars re¯ect the s.e.m. from 4±8 determinations.