Vol. 2, 1123.1134,
0 1989 by Cell Pre+
Structure and Developmental Expression of the Nerve Growth Factor Receptor in the Chicken Central Nervous System Thomas H. Large: Cisela Weskamp: Judith C. HelderFMonte J. Radeke,+ Thomas P Misko,+ Eric M. Shooter,+ and Louis F. Reichardt* * Department of Physiology and the Howard Hughes Medical Institute University of California, San Francisco San Francisco, California 94143-0724 +Department of Neurobiology Stanford University School of Medicine Stanford, California 94305
Summary Chicken nerve growth factor (NGF) receptor cDNAs have been isolated and sequenced in an effort to identify functionally important receptor domains and as an initial step in determining the functions of the NCF receptor in early embryogenesis. Comparisons of the primary amino acid sequences of the avian and mammalian NCF receptors have identified several discrete domains that differ in their degree of conservation. The highly conserved regions include an extracellular domain, likely to be involved in ligand binding, in which the positions of 24 cysteine residues and virtually all negatively charged residues are conserved; a transmembrane region, including flanking stretches of extracellular and cytoplasmic amino acids, which has properties suggesting it interacts with other proteins; and a cytoplasmic PEST sequence, which may regulate receptor turnover. Transient expression of NGF receptor mRNA has been seen in many regions of the developing CNS. Experiments suggest that both NGF and its receptor help regulate development of the retina.
(Levi-Montalcini and Aloe, 1985; Raivich et al,, 1987; Yan and Johnson, 1988b). Understanding the possible diverse actions of NGF requires, in part, an understanding of the cellular events initiated by the binding of NGF to its receptor. NGF binding studies have identified high- (Kd = 10 PM) and low- (Kd = 1 nM) affinity classes of NGF binding sites (Sutter et al., 1979; Landreth and Shooter, 1980). The low-affinity NGF receptor, cloned from rat (Radeke et al., 1987) and human (Johnson et al., 1986), has no significant homology to other peptide receptors and contains no regions that are homologous to known enzymes, such as protein kinases. High-affinity NGF receptors appear to be complexes of the low-affinity receptor and additional proteins (Landreth and Shooter, 1980; Block and Bothwell, 1983; Hosang and Shooter, 1985) and seem to mediate most, if not all, biological responses to NGF (see Greene, 1984). While the additional protein constituents in high-affinity receptor complexes have not been identified, ras p21 and other G proteins appear to be required for signal transduction (see Hagag et al., 1986; Cremins et al., 1986). Because the chicken is an excellent system for embryological studies, we have cloned the chicken homolog of the low-affinity NGF receptor as a first step in understanding its function in the developing nervous system. The expression patterns of NGF and its receptor in the embryonic chick CNS suggest that these proteins are required for normal development of many CNS regions. In addition, comparison of the predicted amino acid sequence of the chicken NGF receptor with those of rat and human has identified several evolutionarily conserved receptor domains that appear likely to be important for receptor function.
Results tntroduction Isolation The critical role of nerve growth factor (NGF) in the survival and development of peripheral sensory and sympathetic neurons has been well characterized over the past 35 years (Levi-Montalcini, 1987). Recent studies have indicated that NGF also may be important for the development and maintenance of intrinsic (Mobley et al., 1985) and projection cholinergic neurons in the basal forebrain of mammals (Gnahn et al., 1983; Williams et al., 1986; Hefti, 1986). The tissues innervated by these peripheral and central neurons have been shown to synthesize NGF, supporting the classic model that NGF, in large part, is a target-derived trophic factor (Shelton and Reichardt, 1984, 1986; Heumann et al., 1984; Korsching et al., 1985; Whittemore et al., 1986; Large et al., 1986; Davies et al., 1987; Ayer-LeLievre et al., 1988). NGF also may have effects on a variety of additional cell types within the developing nervous system since NGF binding sites have been detected early in the development of many avian and mammalian CNS nuclei
An embryonic day 13 (E13) screened at low stringency 221-1247) from the coding rat NGF receptor cDNA (for
chick brain hgtl0 library was with a Stul fragment (bases sequence of the low-affinity structure, see Radeke et al.,
1987). A screen of 250,000 plaques yielded three clones containing 3.1 kb inserts with identical restriction maps (Figure 1A). Preliminary sequencing of one of these clones, named 3.1, indicated that the 5’ end was homologous to the fourth cysteine repeat and transmembrane domain of the rat NGF receptor and the 3’ end contained a poly(A) tail. Clones containing additional 5’sequence were obtained by screening an El3 chick brain hgtll library with two probes: an EcoRI-Sac1 fragment (bases 540-1471 in Figure 3) from the Send of clone 3.1 at high stringency and a Pstl fragment (bases 135-396) from the Send of the rat NGF receptor coding sequence at low stringency. A screen of 250,000 plaques yielded three clones hybridizing to both probes, one of which contained a 3.6 kb insert and was named clone 3.6 (Fig-
IO kb I
3.0 kb I
GCC~CffiCU;CCCCCM~ffi~~C~G~TG~~G~C~~CC -,g*rrrrr -10 -1 t ATGCCCGCCTTCGTACCCCTGCTG CTGCTCCTGCTGC(x. GCGGGACCCACCTGGR;C TCCAAGGAGA4G Mot Ala Gly Phr Vd Pro Leu Lsu Lsu Lsu Pro Thr Trp Gly SW Lyr Glu Lyr TECT AAGATG:A”c ACGACGAGCGGTGAG LcuThr Ly, lie, TyrThrThrSer Gly Glu
AACCTGGGCGAGGGCGTGGTGCAGCCC Am Leu Gly Glu Gly Vd Vd Gln Pro ACCTACTCGGACACAG% AGTGCCACA Thr Tyr SW Asp Thr VII SW Ala Thr
639 711 783 655 927 999 1071 I143 1215 1207 1368
**n ‘an I .a” KC CTGCCCGGTTCT GACAGCCCCGAACCAATCACCAGGGACCCCTTC PACACCGAGGGGATGCCCACCACC Ssr Lcu Al9 Gly SW Atp kr Pro Glu Pro Ile Thr Ar9 p*rp Pro Ph. Aan Thr Glu Gly Met Ala ThrThr 210 220 200 CTGGCAGACATTGTEACCACCGTCATGGGCAGCTCGCAGCCCGTCGTGAGCCGTGGCACCGCCGATAACCTC Lau Al, Asp 11s VII Thr Thr VI1 Met Gly SW SW Gin Pro VII Vd SW Ar9 GlyThr Al, Asp Am E I””
230 240 ATCCCTGTGT( GTt GTGGTGGGGCTGGTG GCTTAE ATT GCCTTC AAAAGGTGG Ile Pro VII Tyr Cyt SW 110 Lou Ala Ala VII Vd YeI Gly Lsu Vd Alp Tyr Iln Ale PhslLys Ar9 Trp I.?^ ^... LJ” LB” AACAGCTGCAAGCAGMC AAGCAGGGACCCAACAAC‘XC CCTGTGAAC CAGACGCCTTCCCCAGAGGGGGAG Am Ser Cus Lus Gln Am L&n Cln Glu Ala Asn Am Am Pro VII Am Gln Thr Pro Sar Pm Glu Glu Glu 270 . 280 . 290 . AAGCTGCACAGCGACAGCGGCATC Tffi GTGGACAGCCAGAGCTTGCACGACCAGCAGCCGCcc AACCAGAGC Lyr Leu fflr SW Asp kr Gly Ilo kr VII Asp hr Gin kr Leu HI, Asp Gln Gln Pro Pro Am Gln Ser 300 310 ACACAGCGACCAGCACCAAAGCGAGAT GGGAGCTTG TACCCCAGCCTG CCCCCCAGCAAGCAGGAGGAGGTG Thr Gln Gly Pro Ala Pro Lyr Gly Asp Gly Yr LWJTyr Ala SW Lou Pro Pro SW Lyr Gin Glu Glu Val 320 330 GAGAM CTGCTGAGCAGCTCGGCGGA&GAGACGTQ. CGGCAGCTGCCCGGGGAG,TG GGCTACAAG“2 GAC Glu Lyr Lou Lw SW SW SW Ala Flu Glu Thr Trp At.9 Gin Lou Al, Gly Glu Lou Gly Tyr Lys Cl” Asp 350 360 CTCATAGACTGCTTCACACGGGAGGAATCGCCCGCCCGACCCCTGCTGGCCGACTGGTCGGCCMC GAGACG Lou II8 Asp Cc Phs Thr Ar9 Glu Glu Se,’ Pro Ale At.9 Ala Lau Lou Air Pap Trp Ssr A,, Lyr G,” Thr 370 380 CCCACTCTCGACGCCCTC CTGGTGGCCCTGCCCk+G ATCCAACGTCCCGACATCGCTGAGAGCCTGTACAGC Ale Thr Lsu Asp Air Lsu Lou Val Ale Leu Ar9 Lys 110 Gln A19 Gly bp 11s Al8 Glu SW Lsu Tyr SW 390 GAGTCCACCGCCACCTCACCCGTGTGA [email protected]
~GGCTGTCT~A Glu Ser Thr Al8 Thr kr Pro Val END
TTGCCTCCCCCCTCCCTC CCCCGGAGAGGAGGGTCAGGAAGAGGW CXECKXCTCCTU;tTTTATW CCTGTATCCACTGAGCTCCTGTllTCGGTATCCCCCCCTCCCACD%GcCT~CCTGG~TAACCTG~CGGGGTcc TGCTCGGTGCCTGCCCCGUGCAGccTTGAGGGcAGGGGCGGGUc CACTCTCACCTCTCTCCTCCTCTCTCTATTTDZAGCCCC TTATCCTGGTTCCAGTTCTC~T~TACTCTTTTTCCCCA
TTAAAGACTGGCCTTAAAGACTCAATGGAGAATGCAQXAA ATGCACACATGWXCTC#&~AT~GT~G~GAAA~T~AAAAG~ TTGAAGGATTTGUCATCCCAGGTTGTTGC~ TGlXTTTGCTGGCTGmtTCHTGACTGTcAQ3AEcAcTTcA GGTTGDXACATKGTTGTGGATTCTCCTTCCTGGAGATcTcc#&WCc CACCTGGAGGTGGTCATCAT~C~TTGG~XG CACCAttT~akrGTaCTCCIGCTTUGTCMTCTCG TTTAAAAGCCTTTGGAATGT~GCTGCCTGTGTTTUGAAAATMCAGAG~A~AGG~TT~T~TGGAT~A~GTTGG~~ATGTT~ ACGCCTCACGCAAAGCATCTGTGCAGcAAGAGcAAAGACCcCTTCTcATGc&A~cCcTTcAT GAGAAU%TTGTGfXATKTT JOB.8 TG~~CUiTCTaK*C~C~TTGGI\TGMiMMCT AAUG%CTAcGTTTTTCCGTcTGcATcccGAAT CTCTCC~TCPCTGAAGTCA~A~T~~~TG~~~~~A~~T~T~ TTGGGACGAGCGTAAATGCTGGTTGTTGCGTTGTT~GUTCffiT~TTTTTG,TT GGAAGMGTGACTCAAKC,WCGCTG GGTACTGCCAGCTCCGCATCC&UACTGccTG~cAGTCTGGccAAGCcCAk,3TGTGcAGT~~TTTTCTTC AGGACAATATTGCC~GTc,?GGTTTCTATTGcCAT AGUAAUXTCGTTTCCTCACCCTT [email protected]
35 18 AACACAGACMSCTCAGCATCCAAAGAGAGG~CA~GGGATG~~T~GAAT~~~A~C~-TGC~AAATAATAA~AA~ATTAA~AATAA $WTACAAAAAAA
1. Chick NGF Acid Sequences
(A) Schematic representation of chicken NCF receptor cDNAs and sequencing strategy. The top map represents clone 3.6, and the bottom map represents clone 3.1. The coding region is represented by a box, and untranslated sequence is represented by a line. The locations of several common restriction sites are indicated. The 1.5 kb Sac1 fragment from the 5’ end of clone 3.6 was used for sequencing in both directions the coding region, and clone 3.1 was used for sequencing the 3’ untranslated region. The open and closed arrows below the maps indicate the direction and length of the sequences derived from subclones of clones 3.1 and 3.6, respectively. The poly(A) tail on clone 3.1 is indicated by A,. (B) Chicken NCF receptor cDNA sequence and predicted amino acid sequence. The nucleotides are numbered in the left margin, and the amino acid residues are numbered above the sequence. The figure includes sequence information obtained from both clones. The predicted signal sequence is numbered from -19 to -1. The out-of-frame ATG at bases 40-42 and the TGA stop codon at bases 136-138 are overlined. The nucleotides flanking ATG(40) and ATG(63) that are predicted to form the stems of hairpin loops are indicated by asterisks. In the extracellular region, the only signal for asparagine-linked glycosylation, residues 33-35, is overlined and the cysteine residues are boxed. The membrane-spanning sequence, residues 22l242, is enclosed. The polyadenylation signal (AATAAA) near the 3’ end is underlined.
NCF Receptor 112s
ure 1A). Because of the instability of large inserts in M13, a 1.5 kb EcoRI-Sac1 fragment from clone 3.6 was subcloned for determination of the 5’ noncoding and coding sequences. Clone 3.1 was used for determination of the 3’ noncoding sequence.
Chicken NGF Receptor and Sequence
The structure (Figure 1A) and sequence of the chicken NGF receptor cDNA (Figure 1B) are homologous to those of rat (Radeke et al., 1987) and human NCF receptor cDNAs (Johnson et al., 1986). An ATG at base 63 occurs in the optimal context (ACCATGG) for initiation by eukaryotic ribosomes (Kozak, 1986). An upstream ATG, at base 40, is out of frame with respect to ATG(63) and produces an overlapping reading frame that terminates at base 136. Computer predictions of RNA secondary structure indicate that sequences flanking both ATG(40) and ATG(63) may form hairpin loops (Tinoco et al., 1973). These are potentially important for translational regulation and are described more completely in the discussion. The single long open reading frame of 1191 nucleotides is followed by a 3’ untranslated sequence of 2.3 kb, slightly longer than is found in either rat (1.9 kb) or human (2.0 kb). A single consensus polyadenylation signal (AATAAA) is found 6 bases upsteam of a poly(A) tail (Proudfoot and Brownlee, 1974), slightly closer to the tail than is typically found (Birnstiel et al., 1985), but similar to the distance in rat and human NGF receptor
8) Fz;,_o= c w=I I
-6 m kb
- 9.4 -65 -43
-I3 -10 Figure 2. DNA Blot Analysis of Chicken Cenomic Chicken and Rat NGF Receptor cDNA Probes
Genomic DNA (5 pg) from chicken muscle was resolved on a 0.7% agarose gel, transferred to GeneScreen Plus, and hybridized with 3ZP-labeled, random primer probes from either (A) chicken or \d) rat NCF receptor cDNAs. Hybridization and washing conditions for the chicken probe were 17°C and 12°C below Tm, respectively. Hybridization and washing conditions for the rat probe, assuming approximately 30% nucleotide mismatch, were 14°C and 7OC below Tm, respectively. Molecular weight markers, expressed in kilobases, are indicated in the right margin.
cDNAs. The predicted amino acid sequence of the long open reading frame, as is the case for rat and human NGF receptor cDNAs, contains a signal peptide followed by a cysteine-rich extracellular domain, a single membrane-spanning cytoplasmic domain. quencesfor chicken, the identification of presented later in the
sequence, and a relatively short Alignment of the amino acid serat, and human NGF receptors and conserved receptor domains are results section (see Figure 3).
DNA Blot Analysis To test the cross-hybridization of chicken and rat NGF receptor cDNAs, chicken genomic DNA was hybridized under high stringency with the chicken NGF receptor cDNA and under low stringency with the rat NGF receptor cDNA (Figure 2). The chicken probe spanned the coding sequence except for the first cysteine repeat and the signal sequence, while the rat probe spanned the coding sequence except for the C-terminal domain. Both probes strongly hybridized to 14.5 kb fragments of genomic DNA digested with EcoRl and Bglll and to a 4.4 kb fragment of genomic DNA digested with Hindlll. Taken together, the labeling of the same genomic DNA fragments and the strong conservation of the predicted amino acid sequence indicated that the chicken gene was the homolog of the rat low-affinity NGF receptor gene.
Conservation of the Structural of the NGF Receptor
Alignment of the predicted amino acid sequence of the chicken NGF receptor with those of the rat and human NGF receptors is shown in Figure 3A. Although 71% of the residues are identical among the mature NGF receptors from the three species, discrete domains of the receptor can be identified by differences in degree of conservation (Figure 3C). Within the signal peptide of the pro-receptor, 37% of the residues are identical, in particular, the string of hydrophobic leucine residues that are predicted to penetrate the membrane (von Heijne, 1985). However, the chicken pro-NGF receptor differs from that of rat and human in the predicted site of cleavage of the signal peptide (von Heijne, 1986). The predicted N-terminal amino acid of the mature chicken receptor is serine, while the N-terminal amino acid of purified rat and human receptors has been reported to be the adjacent lysine (+2), consistent with the predicted sites of signal peptide cleavage in each of these receptor precursors (Marano et al., 1987; Radeke et al., 1987). As is the case for rat and human NGF receptors, the extracellular domain of the chicken receptor is rich in cysteine residues and can be divided into four repeating elements of 35-42 residues each (Figures 3A and 3B). Although the positions of all 24 cysteine residues are conserved among the three species (Figure 3A), the amino acid sequences of the first and second repeats are somewhat better conserved, 83% of the residues being identical, than the third and fourth repeats, 63%-64% of the residues being identical (Figure 3C). However, the third
A) MAGFVPLLLLLLPAGPTWG-SKEKCLTKM MRRAGAACSAMDRLRLLLLLILGVSS-GGA-KETCSTGL MG TGR 1 GP nL[ L[‘IAbh
TVSATEPCKPCTOCVGLHSMSAPCVESDDAV:RC:YtYF VVSATEPCKPCTECLGLOSMSAPCVEADDAVCRCAYGYV I M-l r-----l l----l
IUU’U’I UTkobd~~~~u~~1.~ 0DELSGSCKECSICEVGFGLMFPCRDSODTVCiEC;EGTF ODEETGHCEACSVCEVGSGLVFSCODKON
umno OR// SDEANFVDPCLPCT ICEENEVMVKECTATSDA&LHP SDEANHVDPCLPCTVCEDTERQLRECTPWADAECEEIPG
UT 0 [[I [IA
RWTTHTPSLAGSDSPEP ITRDP---FNTEGMATTLAD RWIPRSTPPEGSDSTAPSTOEPEVPPEODLVPSTVADW
DWSAKETATLDALLVALRKIQRGDIAESLVSESTATSPV SWGAQDSATLDALLAALRRlC!RADlVESLCSESTATSPV IAT r? tlouL uL_i
wgatwe Net --Flendue, Ckgc
and fourth repeats are strongly negatively charged, and 17 of 19 aspartate and glutamate residues are conserved within these repeats (Figure 38). The only potential site for N-linked glycosylation (Asn-X-Ser/Thr) in the extracellular domain of the chicken receptor is at the end of the first cysteine repeat, residue 33, and this site is conserved in rat and human receptors (Figure 3A). The rat NGF receptor contains a second site, on residue 42 at the beginning of the second repeat, that is absent in chicken as well as human receptors. The highly conserved cysteine repeats are followed by a poorly conserved domain, residues 164-201, named the variable domain (Figure 3C). Although the amino acid sequence is poorly conserved, a peak of hydrophobicity just before the transmembrane domain is preserved in each species (data not shown). This variable domain has been proposed to be the site of O-linked glycosylation in the human receptor because of its proximity to the transmembrane domain and the relative abundance of serine and threonine residues (Johnson et al., 1986; Marano et al., 1987). Although there is as yet no consensus sequence for sites of O-linked glycosylation, the conserved residues from position 177-188 also are found at known sites of O-linked glycosylation in other proteins (Marshall, 1972). The conserved 19 amino acid sequence between the variable domain and the membrane-spanning residues also contains several serine and threonine residues and, thus, may be an additional site of O-linked glycosylation The most strongly conserved part of the NGF receptor is the membrane-associated domain (Figures 3A and 3C). This region consists of the transmembrane sequence and the sequences immediately flanking it, the last 19 residues of the extracellular region and the first 46 residues of the cytoplasmic region. The amino acid sequence of this region is 95% identical among receptors in all three species. The 22 hydrophobic residues that are predicted to span the membrane are identical, and computer-assisted protein structure analysis indicates that this sequence has a relatively high helical hydrophobic moment (Eisenberg et al., 1984). Analysis of the cytoplasmic flanking sequence indicates it has a flexible structure (Karplus and Schulz, 1985) and contains a
Figure 3. Sequence Rat, and Human
(A) The amino acid sequences for chicken (top line), rat (middle line), and human (bottom line) NGF receptors have been aligned, and the residues conserved in all three species have been boxed. The sequence for the human NCF receptor uohnson et al., 1986) has been omitted except where it differs from the rat NCF receptor (Radeke et al., 1987). The predicted cleavage site for the signal sequence of the chicken pro-NGF receptor is indicated by a vertical arrow. The boundaries of the four cysteine repeats are marked with horizontal arrows. The conserved N-linked glycosylation site in the first cysteine repeat is indicated with a closed circle. The membrane-spanning sequence is bracketed with heavy lines. The PESTsequence in the cytoplasmic region is indicated with asterisks. The sequence of the synthetic peptide, residues 248-274, used to generate antibodies to the cytoplasmic region of the chicken receptor, is indicated by horizontal arrows. Dashes indicate gaps introduced to preserve alignment. (6) Cysteine repeats of the chicken NCF receptor. The four cysteine
repeats of the chicken NGF receptor have been aligned, and the negatively charged residues conserved in chick, rat, and human have been boxed. For negative residue conservation, the numerator is the number of conserved negatively charged residues and the denominator is the total number present in the chicken sequence. The net negative charge of each repeat is Itsted. (C) A schematic model for the chicken NGF receptor showing the degree of amino acid conservation (percent identity) for each domain. Cleavage of the signal peptide is indicated by an arrow. The four cysteine repeats are shown as hatched boxes, and the N-linked glycosylation site in the first repeat is indicated. The membranespanning sequence within the membrane-associated domaln is shown as a black box. In the deduced chicken NCF receptor amino acid sequence, the signal peptide domain is residues -19 to -1, the cysteine repeat domain is residues 1-163, the variable domain IS residues 164-201, the membrane-associated domain is residues 202-288, the cytoplasmic domain is residues 289-365, and the C-terminal domain is residues 366-397.
PEST sequence, a region rich in proline, aspartate, glutamate, serine, and threonine residues (Rogers et al., 1986; Rechsteiner et al., 1987). Although sequence identity of the next 77 amino acids of the intracellular region falls to 55%, the C-terminal 32 residues appear to be relatively highly conserved, as 81% of the residues are identical and the remainder are substitutions of similar amino acids.
RNA Blot Analysis The size of the chicken NGF receptor mRNA was determined by RNA blot analysis (Figure 4). A single-stranded chicken cDNA probe detected at high stringency a single band on blots of total RNA from El0 dorsal root ganglia, but not liver, and an mRNA of the same size in a poly(A)+ RNA sample from El3 chicken retina. The size of the chicken NGF receptor mRNA, using 28s (4.6 kb) and 185 (1.8 kb) ribosomal RNA (rRNA) as markers, was calculated to be 4.5 kb. The estimated size of the chicken NGF receptor mRNA is 0.9 kb longer than the composite cDNA sequence (Figure lB), and primer extension studies are required to determine the amount of additional 5’ sequence present in the mRNA. Comparison with poly(A)+ RNA from PC12 ceils and total RNA from postnatal day 35 (P35) rat basal forebrain indicated that the chicken NGF receptor mRNA is approximately 600 bases longer than the rat NGF receptor mRNA. Much of this difference is due to the fact that the 3’ untranslated sequence of the NGF receptor cDNA is 400 bases longer in chicken than in rat. RNA blot analysis of various regions of the chicken CNS indicated that the NGF receptor is transiently expressed throughout the CNS during embryonic development (Figure 5, upper panel). The expression pattern of NGF receptor mRNA during development is listed in Table 1, with values arbitrarily normalized to the receptor mRNA content in El3 retina. NGF receptor mRNA levels throughout the chicken CNS at El3 were relatively uniform, differing only 3-fold between spinal cord, the highest, and brain stem, the lowest. In all areas examined, receptor expression appeared to be highest at E6, the earliest age examined, decreased up to IO-fold by the time of hatching, and remained low in adults. To compare the distribution of NGF receptor expression with sites of NGF production in the embryonic CNS, the developmental expression of NGF mRNA in various regions also was determined (Figure 5, lower panel). The chicken NGF genomic probe detected a single mRNA of 1.3 kb in each sample at high stringency. NGF mRNA levels, also normalized to El3 retina, are shown in Table 1 below the values for NGF receptor mRNA. In the CNS at E13, NGF expression varied more than NGF receptor expression, e.g., the NGF mRNA content was at least TO-fold higher in the retina than in the optic lobe and forebrain-thalamus. In addition, NGF expression in the CNS did not exhibit the same developmental decrease as did NGF receptor expression. NGF mRNA content was relatively low both at El3 and in the adult in the optic lobe, cortex-striatum, and spinal cord.
kb 4.5 3.9 -
Rat PC12 cells (lane l), chicken El3 retina (lane 3), and chicken liver (lane 5) are poly(A)+ RNA samples (5 pg); rat P35 basal forebrain (lane 2) and chicken El0 dorsal root ganglion (DRC; lane 4) are total RNA samples (25 bg). Chicken and rat RNA samples were resolved on a 1.5% agarose-formaldehyde gel, transferred to GeneScreen, and hybridized with single-stranded, jZP-labeled cDNA probes from either chicken or rat NCF receptor cDNA. Hybridization and washing conditions were 28OC and 14’C below Tm, respectively, for the chicken probe and 16OC and 9’C below Tm, respectively, for the rat probe. The first four lanes were exposed for 1 day; (ane 5 was exposed for 5 days. The positions of chicken 18s and 28s rRNA are indicated in the right margin.
slightly during development, while expression in the cerebellum appeared to increase approximately 3-fold between El3 and adult. NGF expression appeared to be highest in the retina, where NGF mRNA content reached a peak at El3 and remained elevated in adults, even though NGF receptor mRNA levels had decreased lo-fold.
Discussion Two lines of evidence indicate that the cDNAs isolated from the El3 chick brain libraries encode the lowaffinity, or fast-dissociating, form of the NGF receptor. First, the predicted amino acid sequence of the chicken cDNAs was homologous to those of the low-affinity NGF receptor from rat and human, with several domains being very highly conserved. Second, the chicken clones and the low-affinity rat NGF receptor cDNA hybridized to the same size fragments on blots of chicken genomic DNA.
in the Chtck
168 k 76 (2) 46 + 5 (21
122 + 29 13) 79 t 14 (3)
100 k 16 (4) 100 & 11 (4)
12 + 2 (5) 98 k 2 (5)
17 * 7 (3) 91 k 8 (5)
74 * 20 (2) 4 k 1 (2)
56 + 9 (4) 10 * 2 (4)
10 5 2 (5)
6 k 2 (2)
44 * 13 (4) 18 k 3 (3)
12 f 1 (2) 10 + 1 (2)
8 k 3 (4) 14 f 4 (3)
43 (1) 3 (1)
12 If 1 (2) 17 * 1 (2)
11 f 2 (4) 19 f 8 (3)
55 * 14 (4) 12 + 4 (4)
23 (1) 7 (1)
22 & 4 (4) 33 + 5 (3)
36 f 13 (4) 26 + 1 (4)
10 * 4 (2) 17 + 6 (2)
11 i 2 (5) 11 + 4 (3)
809 f 2 f
185 k 17 (3) 7 f 1 (3)
190 (2) 1 (2)
109 * 15 (3) 11 f 3 (3)
8 + 2 (2) 16 f 5 (2)
6 (1) 5 (1) 13 k 6 (2) 16 f. 2 (21
RNA blots were quantified by scanning laser densitometry, and the amount of NGF and NCF receptor mRNA was normalized to ‘*P-labeled poly(T) binding (Large et al., 1986). El3 chicken retina was chosen arbitrarily as the 100% value, and a sample was run on every blot. The values represent the mean + SEM; the number of samples is indicated in parentheses.
Developmental Expression Receptor in the CNS
of the Chicken
Analysis of the sites and timing of NGF receptor mRNA expression in the embryonic chicken CNS indicated that the NGF receptor is transiently expressed throughout the developing chicken brain. The high level of NGF receptor mRNA between E6 and El3 agrees with the finding of NGF receptors in chicken brain homogenates from these ages (Frazier et al., 1974) and, in general, with recent RNA blot and in situ data of Ernfors et al. (1988). The developmental decrease in NGF receptor mRNA expression also agrees, in general, with the developmental disappearance of 1251-labeled NGF binding sites in sections of the embryonic chicken brain (Raivich et al., 1985, 1987). Two regions of the chicken CNS that were of particular interest because of their strong, transient expression of NGF receptor mRNA were the spinal cord and retina.
Spinal Cord Within the E6 spinal cord, we found the highest levels of NGF receptor mRNA measured in the chicken CNS. As axons do not contain significant levels of RNA, NGF receptor mRNA must be associated with cell bodies in the spinal cord rather than receptor-bearing processes, either from sensory neurons or ascending and descending fiber tracts. The high level of NGF receptor mRNA at E6 is entirely consistent with recent findings that motoneurons of the chicken lateral motor column contain both NGF receptor mRNA (Ernfors et al., 1988) and r251-labeled NGF binding sites between E6 and E8 (Raivich et al., 1985). The transient expression of NGF receptors by motoneurons also explains the finding that exogenous 12sl-labeled NGF injected into limb buds at E5-E7 is retrogradely transported to cell bodies in the lateral motor column (Wayne and Heaton, 1988). The
source of NGF for motoneurons would appear to be outside of the spinal cord, since NGF mRNA was barely detectable in the cord at E6 and remains low at later times (Table 1). Muscle masses (Hulst and Bennett, 1986; Goedert, 1986) and cells in peripheral nerve (Heumann et al., 1987; Bandtlow et al., 1987) synthesize NGF in early development and, therefore, may supply NGF to developing motoneurons. The role of NGF in the development of motoneurons remains unclear. Exogenous NGF does not diminish the naturally occurring death of either mammalian or avian motoneurons that occurs during development (Oppenheim et al., 1982; Yan et al., 1988a) and does not increase choline acetyltransferase activity in mammalian motoneurons (Van et al., 1988a). NGF has been shown to affect process outgrowth and transmitter synthesis in subpopulations of neurons that do not require it for survival (see Collins and Dawson, 1983; Heaton, 1987). Thus, NGF receptors may modulate facets of motoneuron differentiation, such as process outgrowth, even though it does not regulate their survival.
Retina The retina had the second highest level of NGF receptor mRNA measured in the embryonic chicken CNS. The cell types expressing NGF receptors in the embryonic retina have not been identified. In the developing avian and mammalian retina, r2514abeled NGF and anti-receptor antibody binding appears first in the ganglion cell layer, indicating that ganglion cells may express NGF receptors (Raivich et al., 1987; Yan and Johnson, 1988b; Schatteman et al., 1988). In the adult mammalian retina, Miiller glia appear to express NGF receptors (Schatteman et al., 1988), as do embryonic chicken Mijller glia grown in vitro (Large and Weskamp, unpublished data). In contrast to the down-regulation of NGF receptor
NCF Receptor 1129
s -6 5 &
2 5 I Y
,M$M w WQW
5. NGF and NCF Receptor mRNA in the Developing Chicken CNS
Poly(A)+ RNA (5 pg) from various regions of the chicken CNS was blotted and hybridized with single-stranded, 32P-labeled probes from chicken NCF receptor cDNA (upper panel) and chicken NCF genomic clone flower panel). Hybridization and washing conditions were 28OC and 14°C below Tm, respectively, for the NGF receptor probe and 22°C and 11°C below Tm, respectively, for the NGF probe. The amount of poly(A)+ RNA per lane was determined by hybridization with a 3*P-labeled poly(T) probe (Large et al., 1986). The positions of 28s and 18S;RNA are indicated in the left margin.
expression late in embryonic development, NGF expression in the chicken retina reached a maximum by El3 and remained elevated throughout development and in the adult. The avian retina is innervated by the isthmooptic nucleus, and the survival of neurons in this structure appears to be dependent on competition for a factor produced in limiting amounts within the retina (O’Leary and Cowan, 1984). The period of naturally occurring cell death in the isthmo-optic nucleus begins at E13, the time when the level of NGF mRNA in the retina has reached a maximum, raising the possibility that NGF produced in the retina may act as a target-derived trophic factor for neurons in this region. However, the elevated levels of NGF receptor mRNA in the embryonic retina between E6 and El3 suggest that NGF produced in the retina also may have local effects. To date, experiments testing the effects of NGF on retinal cell development have yielded negative results. Thus, NGF has no effect, in vitro, on the survival of either avian (Nurcombe and Bennett, 1981) or mammalian ganglion cells Uohnson et al., 1986) and does not appear to regulate the development of cholinergic markers in retinal amacrine cells (Hofmann, 7988). In recent years, NGF has been shown to regulate the differentiation of several discrete populations of neu-
rons, including expression of cholinergic enzymes by mammalian basal forebrain neurons (see Gnahn et al., 1983; Mobley et al., 1985). Most strikingly, in the premetamorphic Xenopus brain, NGF induces an increase in size and cell number of diverse nuclei, particularly those containing aminergic and peptidergic neurons (Levi-Montalcini and Aloe, 1985). The effects of NGF no longer appear limited to regulation of neuronal survival and transmitter synthesis. During development of neural crest derivatives, NGF is capable of stimulating cell proliferation and regulating cell determination both in vitro and in vivo (see Aloe and Levi-Montalcini, 1979; Stemple et al., 1988). It has been suggested that NGF also may regulate cell migration from proliferative zones in the CNS (Schatteman et al., 1988). It can clearly promote axonal process growth by neurons not dependent upon it for survival (see Collins and Dawson, 1983). NGF also functions outside of the nervous system, e.g., as a local growth factor that indirectly stimulates proliferation and differentiation of hemopoietic cells (Matsuda et al., 1988). While the NGF receptor must function as signal transducer in each of the above examples, it may have other functions. A truncated form of the NGF receptor has been found in serum, where it may function to scavenge NGF (DiStefano and Johnson, 1988). NGF receptors have
been detected on glia, where tard diffusion of NGF from sites crease local concentrations of rons (Taniuchi et al., 1986). It these questions in the chicken, tigating embryonic development.
Conservation of Structural NGF Receptor
they may function to reof synthesis and thus inNGF available for neuis now possible to study an ideal system for inves-
Understanding the actions of NGF requires elucidation of the mechanisms of NGF receptor-mediated signal transduction. Alignment of the amino acid sequences for the NGF receptor from chicken, rat, and human reveals discrete domains that differ in their degree of evolutionary conservation (Figure 3). The hydrophobic signal sequence is present in all three species, as would be expected for a cell surface receptor. However, the signal sequence of the chick NGF pro-receptor differs in that translation appears to initiate at ATG(63), corresponding to the second ATG codon in rat and human and resulting in a signal peptide of 19 residues for avian versus 28-29 residues for mammalian pro-receptors. The presence of additional upstream start sites has not been ruled out directly because no upstream, in-frame stop codon was detected in the cDNAs sequenced in this project. It appears unlikely, though, since this would result in a highly charged, arginine-rich signal peptide of over 40 amino acids that would not resemble typical signal sequences (von Heijne, 1985). The chick NGF receptor mRNA contains an upstream start codon, ATG(40), that is out of frame with respect to ATG(63) and the long open reading frame. Although the context of ATG(40) is less than optimal (GGGATGC), ribosomes have been shown to initiate translation at such sites (Kozak, 1986). Because the ATG(40) reading frame overlaps ATG(63) and terminates at base 136, ribosomes initiating at ATG(40) would result in synthesis of a 32 amino acid peptide rather than the pro-NGF receptor. RNA sequences surrounding ATG(40) and ATG(63) are predicted to form stable hairpin loops. The hairpin loop containing ATG(40) is formed by 5’ stem bases 33-38 (CCGGGG) and 3’ stem bases 51-56 (CCCCGG); the hairpin loop containing ATG(63) is formed by 5’ stem bases 52-57 (CCCGGC) and 3’stem bases 66-71 (GCCGGG). The hairpin loop containing ATG(40) is slightly more stable, AG (25°C) = -17.2 kcal, than the hairpin loop containing ATG(63), AC (25°C) = -16.4 kcal. Because the stem bases 52-56 are shared by both hairpin loops, it follows that only one hairpin loop can form at any time. The availability of either ATG(40) or ATG(63) for the initiation of translation and the synthesis of either the short peptide or the pro-NGF receptor may be regulated by these RNA secondary structures, as has been proposed for the translational regulation of anionic trypsinogen and amylase synthesis (Steinhilber et al., 1988). The finding that the cysteine-rich part of the extracellular domain is highly conserved between avian and mammalian NGF receptors, while most of the remaining extracellular sequence is highly variable, supports the proposal that the NGF binding site is formed by the cys-
teine repeats (Johnson et al., 1986). Computer-based analysis of the cysteine repeats from each species indicates that flexible, hydrophobic regions are conserved near the boundaries of each repeat (Karplus and Schulz, 1985), while hydrophobic regions tend to be conserved in the middle of the repeats, particularly the second, third, and fourth repeats (Kyte and Doolittle, 1982). This indicates that the ends of each repeat are likely to be exposed to the aqueous environment, while the middles are likely to be shielded. The conservation of negative charges within the third and fourth repeats indicates that these regions may participate in binding the highly positively charged NGF dimer (Angeletti and Bradshaw, 1971; Bradshaw et al., 1977). While the first and second repeats do not contain as many negatively charged residues, these residues are also conserved, and the primary amino acid sequence of each repeat is more highly conserved than in the third and fourth repeats. Thus, these repeats also are likely to be important for NGF binding. The most highly conserved part of the NGF receptor is the membrane-associated domain, consisting of the 22 membrane-spanning residues and flanking sequences. The membrane-spanning sequence, because of its relatively high helical hydrophobic moment, falls into a class of integral membrane proteins that are thought to be stabilized by interaction with other transmembrane helices (Eisenberg et al., 1984). Although the NGF receptor sequence appears unrelated to any other receptors, the membrane-spanning sequence does bear some resemblance to those of the immunoglobulin binding a subunits of high-affinity IgG and IgE F, receptors. These F, receptor subunits appear to require interaction with membrane-bound B and y subunits for their stable expression and function in cell membranes (Metzger and Kinet, 1988). Similarly, the formation of the high-affinity NGF receptor complex may occur by coupling of the low-affinity NGF receptor to other receptor subunits or signal transduction components within the plane of the membrane. This model also would explain the selective cross-linking of high-affinity, slow-dissociating, NGF receptors to other proteins by hydrophobic, but not hydrophilic, cross-linking agents (see Massague et al., 1981; Hosang and Shooter, 1985; Green and Greene, 1986). The 46 cytoplasmic residues flanking the transmembrane sequence and the 32 residues at the C-terminus are highly conserved parts of the cytoplasmic region and also may be involved in receptor function, cellular localization, or receptor turnover. These conserved cytoplasmic sequences may play a role in the preferential association of high-affinity NGF receptors with cytoskeletal elements (Schechter and Bothwell, 1981; Vale et al., 1985). An interesting conserved feature of the cytoplasmic flanking sequence is the presence of a PEST sequence, residues 258-270. Proteins containing sequences rich in proline, glutamate, aspartate, serine, and threonine residues bound by positively charged residues have been correlated with rapid turnover rates of less than 5 hr (Rechsteiner et al., 1987). Although the mechanism by
which these regions accelerate degradation is not known, degradation of PEST-containing proteins appears to be regulated by interactions with other molecules (Rogers et al., 1986). Conceivably, exposure of the PEST sequence through interactions with other proteins may accelerate degradation of the NGF receptor. in summary, we have identified conserved domains of the NGF receptor that are likely to be important for receptor function. Although little is known about the precise mechanisms of NGF receptor signaling, the deletion or replacement of domains can be tested for disruption of receptor-mediated signal transduction. Alternatively, antibodies directed to conserved extracellular and cytoplasmic domains can be tested for their ability to block NGF binding or to disrupt receptor coupling to accessory molecules of the signal transduction system, respectively. Recently, a synthetic peptide corresponding to amino acids 248-274 of the chick NGF receptor, a highly conserved cytoplasmic domain (Figure 3), has been used to generate antibodies that recognize both chick and mammalian NGF receptors in immunoprecipitations and immunocytochemistry (Large and Weskamp, unpublished data). These antibodies should be useful for future studies on receptor expression. They may also be useful for identifying NGF receptors in more evolutionarily distant species. Experimental
Reagents and Solutions White Leghorn chickens were purchased from Feather Hill Farm (Pet&ma, CA), and Sprague-Dawley rats were from Simonson Laboratories (Gilroy, CA). Nitrocellulose filters were from Schleicher and Schuell (Keene, NH). GeneScreen and GeneScreen Plus nylon membranes were from New England Nuclear (Boston, MA). Restriction enzymes, Klenow fragment of DNA polymerase. and T4 polynucleotide kinase were from New England Biolabs (Beverly, MA) and Boehringer Mannheim Dragnostics (Houston, TX). Exonuclease III and other enzymes used to generate unidirectional deletions of inserts in Ml3 were from Promega (Madison, WI). Sequenase enzyme, sequencing reagents, and random hexanucleotide primers were from United States Biochemical (Cleveland, OH). OligofdT)-cellulose type 7 was from Pharmacia (Piscataway, NJ). [a%]dATP, [y-a2P]ATP, and [aJ*P]dCTP were from Amersham (Arlington Heights, IL). Cronex intensifying screens were from DuPont (Wilmington, DE). All other chemicals were purchased from Sigma (St. Louis, MO). Denhardt’s solution (lx) is 0.02% BSA, 0.02% polyvinyl pyrrolidone, and 0.02% Ficoll type 400. Hybridization buffer (HYB) is 900 mM sodium chloride, 5 mM EDTA, 50 mM sodium phosphate (pH 7.4), 5x Denhardt’s solution, and 200 gglml salmon sperm DNA. lx SSC is 150 mM sodium chloride and 15 mM sodium citrate fpH 7.0). Screening of cDNA library El3 chicken brain cDNA libraries in hgtl0 and lcgtll wereobtained from Dr. B Ranscht (1988). The libraries were plated on E. coli Y1090, and replica filters were prepared according to standard procedures (Maniatis et al., 1982). The filters screened at low stringency with rat NGF receptor cDNA probes were prehybridized for 18 hr at 42OC in HYB containing 10% formamide and 1% SDS and then hybridized for 24 hr in the same buffer containing lob cpmlml probe. The filters screened with chicken NCF receptor cDNA probes were prehybridized for 18 hr at 50°C in HYB containing 40% formamide and 1% SDS and then hybridized for 24 hr in the &me buffer containing lob cpmlml probe. Primer extension of random hexanucleotides annealed to denatured fragments of the
YCF receptor cDNAs was used to generate ?‘P-labeled probes (Feinberg and Vogelstein, 1983). The filters were washed for 12 hr rn their respective prehybridization buffers and autoradiographed. The hybridization and washing conditions for the rat and chicken cDNA probes, assuming a 30% mismatch between the nucleotide sequences of rat and chicken NGF receptors, were 27°C and 28°C below the melting temperature of duplex DNA (Tm), respectively (Thomas and Dancis, 1973; Bonner et al.. 1973). DNA Sequencing The coding region of the chicken NGF receptor was sequenced using a 1.5 kb EcoRI-Sac1 fragment from the 5’ end of clone 3.6. The noncoding sequence was determined from clone 3.1. Unidirectional deletions of the inserts in M13mp18 were obtained by digestion with exonuclease Ill (Henikoff, 19841, and the subclones were sequenced by the dideoxy chain termination method (Sanger et al., 1977). The sequence was merged and edited using the GEL sequence management program on Bionet (Intelligenetics Corp.. Mountain View, CA). Nucleic Acid and Amino Acid Sequence Analysis The nucleotide and predicted amino acid sequence of the chicken NGF receptor was analyzed by the package of programs contained within PCGENE (Intelligenetics Corp., Mountain View, CA). Formation of hairpin loops in RNA sequences was predicted by the program HAIRPIN, based on the methods of Tinoco et al. (1973). Signal sequence cleavage analysis was performed by the program PSICNAL, based on the methods of von Heijne (1986). Hydrophobrcity plots were generated by the program SOAP based on the methods of Kyte and Doolittle (1982). Antigenic sites and exposed regions of the protein were predicted by the programs ANTIGEN and FLEXPRO, based on the methods of Hopp and Woods (1981) and Karplus and Schulz (19851, respectively. The membrane-spanning sequence was predicted by the programs SOAP and RAOARGOS, based on the methods of Klein et al. (1985) and Rao and Argos (1986), respectively. In addition, the membrane-spanning sequence was classified as multimeric by the program HELIXMEM, based on the methods of Eisenberg et al. (1982, 1984). Identification of the cytoplasmic PEST sequence was performed by the program PESTFIND, based on the methods of Rogers et al. (1986). DNA Blot Analysis Genomic DNA from chicken muscle was digested with several restriction enzymes, and 5 pg was resolved on a 0.7% agarose gel, blotted ontoGeneScreen Plus membranes bycapillarytransfer, and hybridized with 32P-labeled, random primer probes of either rat or chick NGF receptor cDNA (Southern, 1975). The gel-purified inserts used for probe synthesis were a Maelll-Sac1 fragment (bases 249-1470) of chicken clone 3.6 and a Stul fragment (bases 221-1247) of the rat NGF receptor cDNA (for structure, see Radeke et al., 1987). The chicken probe was hybridized for 18 hr at 65°C rn HYB containing 40% formamide and 1% SDS (17OC below Tm) dnd washed at 65°C in 0.1x SSC, 1% SDS (12°C below Tm). The rat probe was hybridized for 18 hr at 55°C in HYB containing 10% formamide and 1% SDS (14OC below Tm) and washed at 55OC in lx SSC, 1% SDS (7°C below Tm). RNA Blot Analysis RNA blots were performed as described previously by Large et al. (1986) with the modifications described by Clegg et al. (1988). Briefly, total cellular RNA was prepared by lysis of tissues in 7.6 M guanidine HCI, 0.1 M potassium acetate, precipitation with 35% ethanol (Cheley and Anderson, 1984), and extraction with phenol and chloroform. Some samples were enriched for poly(A)+ RNA by a single pass over oligofdT)-cellulose columns (Aviv and Leder, 1972). Total (25 ug) or poly(A)+ (5 ug) RNA samples were resolved on 1.5% agarose-formaldehyde gels, transferred to GeneScreen membranes by electroblotting, and cross-linked by UV Irradiation (Church and Gilbert, 1984). RNA blots were prehybridized at 50°C for 18 hr in HYB containing 50% formamide and 5% SDS, hybridized at 50°C for 36 hr in the same buffer containing lo6 cpm/ml 3LP-labeled, single-stranded cDNA probes, and washed at various temperatures in 0.1 x SSC, 0.1% SDS. Chicken NGF receptor mRNA
was detected by hybridization with a probe from clone 3.6 (28°C below Tm) and washing at 67.5”C (14’C below Tm). The chicken NCF receptor probe was stripped from the blot, and NGF mRNA was detected by hybridization with a probe from a chicken NGF clone (16’C below Tm) and washing at 60°C (9°C below Tm). The chicken NGF clone was generated by subcloning a Pstl-Pvull fragment (bases 240-1056) of a chick genomic clone into M13mplY (for structure, see Ebendal et al., 1986). The chicken genomic clone encoding NCF was a generous gift of Dr. D. Shelton. Blots containing rat RNA were hybridized (22°C below Tm) with a probe from a rat NGF receptor clone, generated by subcloning an EcoRI-Bglll fragment (bases 1-1683) of the rat NGF receptor cDNA into M13mp18, and washed at 65’C (11’C below Tm). Finally, poly(A)+ RNA was detected by hybridization with oligo(dT) (18-mer) end-labeled with [“PIATP and T4 polynucleotide kinase. The blots were exposed to X-ray film with intensifying screens at -70°C. Autoradiograms were scanned with a laser densitometer, and the amount of chicken NGF receptor or chicken NGF mRNA was normalized to [?2P]oligo(dT) binding. Acknowledgments The authors would like to acknowledge the generous contribution of the chicken b-NGF genomic clone from Dr. D. Shelton and the El3 chicken brain libraries from Dr. B. Rantsch. We thank Drs. J. LaVail, C. Guthrie, and M. lgnatius for their helpful comments on the manuscript and Marion Meyerson for expert typing of the manuscript. C. W. is supported by Deutscher Akademischer Auslandsdienst, Gentechnologie Programm. This work was supported by NIH grants NS21824 and NS04270, the American Cancer Society, the lsabelle M. Niemela Trust, and the Howard Hughes Medical Institute. L. F. R. is an investigator of the Howard Hughes Medical Institute. ReceIveA
References Aloe, L., and Levi-Montalcini, R. (1979). Nerve growth factorinduced transformation of immature chromaffin cells in vivo into sympathetic neurons: effect of antiserum to nerve growth factor. Proc. Natl. Acad. Sci. USA 76, 1246-1250. Angeletti, R. H., and Bradshaw, from mouse submaxillary gland: Acad. Sci. USA 68, 2417-2420.
R. A. (1971). Nerve growth factor amino acid sequence. Proc. Nat/.
Aviv, H., and Leder, F! (1972). Purification of biologically globin messenger RNA by chromatography on oligothymidylic cellulose. Proc. Natl. Acad. Sci. USA 69, 1408-1412. Ayer-LeLievre, C., Olson, L., Ebendal, (1988). Expression of p-nerve growth neurons. Science 240, 1339-1341.
T., Seiger, A., and Persson, H. factor gene in hippocampal
Clegg, D. O., Bodary, S. C., Large, T. H., Shelton, D. L., and Reichardt, L. F. (1988). Quantitative measurement of nerve growth factor. In Nerve Growth Factors, R. A. Rush, ed. (West Sussex, England: John Wiley & Sons), in press. Collins. F., and Dawson, A. (1983). An effect of nerve growth factor on parasympathetic neunte outgrowth. Proc. Natl. Acad. Sci. USA 80, 2091-2094. Cremins, J., Wagner, J. A., and Halegoua, 5. (1986). Nerve growth factor action is mediated by cyclic AMP- and Ca+z/phospholipiddependent protein kinases. J. Cell Biol. 103, 887-893. Davies, A. M., Bandtlow, C., H., and Thoenen, H. (1987). synthesis in developing skin slon of the receptor. Nature
Heumann, R., Korsching, S., Rohrer, Timing and site of nerve growth factor in relation to innervation and expres326, 353-358.
DiStefano, P 5.. and Johnson, E. M. (1988). Identification cated form of the nerve growth factor receptor. Proc. Sci. USA 85, 270-274. Ebendal, T., Larhammar, D., and Persson, expression of the chicken p-nerve growth 1483-1487.
of a trunNat/. Acad.
H. (19861. Structure factor gene. EMBO
and J. 5,
Eisenberg, D., Weiss, R. M., and Terwilliger, T. C. (1982). The helical hydrophobic moment: a measure of the amphilicity of a helix, Nature 299, 371-374. Eisenberg, D., Schwarz, E., Komaromy, M., and Wall, R. (1984). Analysis of membrane and surface protein sequences with the hydrophobic moment plot. J. Mol. Biol. 179, 125-142. Ernfors, I?, Hallbook, F., Ebendal, T., Shooter, E. M., Radeke, M. J., Misko, 7. P., and Persson, H. (1988). Developmental and regional expression of b-nerve growth factor receptor mRNA in the chick and rat. Neuron I, 983-996. Feinberg, A. P, and Vogelstein, B. (1983). A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132, 6-13. Frazier, W. A., Boyd, L. F., Srutowicz, shaw, R. A. (1974). Specific binding tor in peripheral tissues and brain. mun. 57, 1096-1103.
A., Pulliam, M. W., and Bradsites for lZ51-nerve growth facBiochem. Biophys. Res. Com-
Gnahn, t-l., Hefti, F., Heumann, R., Schwab, M. E., and Thoenen, H. (1983). NGF-mediated increase of choline acetyltransferase (ChATI In the neonatal rat forebrain: evidence for a physiological role of NGF in the brain? Dev. Brain Res. 9, 45-52. Goedert, M. (1986). Molecular factorgene: mRNAdistribution them. Biophys. Res. Commun.
cloning ofthe chicken nerve growth in developingandadulttissues. Bio141, 1116-1122.
Green, S. H., and Greene, L. A. (1986). A single M, = 103,000 ILSIp-nerve growth factor-affinity-labeled species represents both the low and high affinity forms of the nerve growth factor receptor. J. Biol. Chem. 261, 15316-15326.
Bandtlow, C. E., Heumann, R., Schwab, M. E., and Thoenen, H. (1987). Cellular localization of nerve growth factor synthesis by in situ hybridization. EMBO J. 6, 891-899.
Greene, sponses Neurosci.
Blrnstiel, M. L., Busslinger, M., and Strub, K. (1985). Transcription termination and 3’ processing: the end is in site. Cell 41, 349-359.
Hagag, N., Halegoua, S., and Viola, M. (1986). Inhibition of growth factor-induced differentiation of PC12 cells by microinjection of antibody to ras ~21. Nature 319, 680-682.
Block, T., and Bothwell, M. (1983). The nerve growth tor on PC12 cells: interconversion between two forms binding properties. J. Neurochem. 40, 1654-1663.
factor recepwith different
L. A. (1984). The importance of both early and delayed rein the biological actions of nerve growth factor. Trends 7, 91-94.
Heaton, M. B. (1987). Influence of nerve growth factor on chick trlgeminal motor nucleus explants. J. Comp. Neural. 265, 362-366.
Bonner, T. I., Brenner, D. J., Neufeld, B. R., and Britten, R. I. (1973). Reduction in the rate of DNA reassociation by sequence divergence. J. Mol. Biol. 81, 123-135.
Hefti, F. 11986). Nerve growth factor promotes cholinergic neurons after fimbrial transections. 2355-2162.
Bradshaw, R. A., Jeng, I., Andres, R., Pulliam, M. W., Silverman, R. E., Rubin, J., and Jacobs, J. W. (1977). The structure and function of nerve growth factor. Proc. Vth Int. Congr. Endocrinol. 2, 206-212.
Henikoff, S. (1984). Unidirectional creates targeted breakpoints for 351-359.
Cheley, S., and Anderson, R. (1984). A reproducible method for the detection of specific RNA sequences bridization. Anal. Biochem. 137, 15-19. Church, G. M., and Gilbert, W. (1984). Natl. Acad. Sci. USA 81, 1991-1995.
microanalytical by dot-blot hysequencing.
survival of septal J. Neurosci. 6,
digestion with exonuclease DNA sequencing. Gene
Heumann, R., Korsching, S., Scott, J., and Thoenen, H. (1984). Relationship between levels of nerve growth factor (NGF) and its messenger RNA In sympathetic ganglia and peripheral target tissues. EMBO J. 3, 3183-3189. Heumann, R., Llndholm, D., M. J., Misko, T. P., Shooter,
Bandtlow, C., Meyer, E. M.. and Thoenen,
M., Radeke, H. (1987).
NGF Receptor 1133
Differential regulation of mRNA encoding nerve growth factor and its receptor in rat sciatic nerve during development, degeneration, and regeneration: role of macrophages. Proc. Natl. Acad. Sci. USA 84, 87358739.
Mobley, W. C., Rutkowski, 1. L., Tennekoon, C. I., Buchanan, K., and Johnston, M. V. (1985). Choline acetyltransferase activity in striatum of neonatal rats increased by nerve growth factor. Science 229, 284-287.
Hofmann, H.-D. (1988). Development of cholinergic retinal neurons from embryonic chicken in monolayer cultures: stimulation by glial cell-derived factors. J. Neurosci. 8, 1361-1369.
Nurcombe, V., and Bennett, M. R. (1981). Embryonic chick ganglion cells identified in vitro. Their survival IS dependent factor from the optic tectum. Exp. Brain Res. 44, 249-258.
Hopp, T. f?, and Woods, genie determinants from Sci. USA 78, 3824-3828.
O’Leary, D. D. M., and Cowan, W. M. (1984). optic neurons after early removal of one eye. 293-310.
K. R. (1981). Prediction amino acid sequences.
of protein antiProc. Natl. Acad.
Hosang, M.. and Shooter, E. M. (1985). Molecularcharacteristicsof nerve growth factor receptors on PC12 cells. J. Biol. Chem. 655-662.
H&t, J. R., and Bennett, M. (1986). Motoneurone survival factor produced by muscle increases in parallel with messenger RNA sequences homologous to P-NGF-cDNA. Dev. Brain Res. 25, 153-156. Johnson, Mercer, structure
D., Lanahan, A., Buck, C. R., Sehgal, A., Morgan, E.. Bothwell, M., and Chao, M. (1986). Expression of the human NGF receptor. Cell 47, 545-554.
Johnson, J. E., Barde, Y.-A., Schwab, M., and Thoenen, Brain-derived neurotrophic factor support5 the survival rat retinal ganglion cells. J. Neurosci. 6, 3031-3038. Karplus, P A., and Schulz, G. E. (1985). Prediction ity in proteins. Naturwlssenschaften 72, 212-213. Klein, P., Kanehisa, M., and DeLisI, classification of membrane-spanning Acta 8 15, 468-476.
C. (1985). proteins.
H. (1986). of cultured
The detection and Biochim. Biophys.
Korsching, S., Auberger, G., Heumann, R., Scott, J., and Thoenen, H. (1985). Levels of nerve growth factor and its mRNA in the central nervous system of the rat correlate with cholinergic Innervation. EMBO J. 4, 1389-1393. Kozak, M. (1986). Point mutations AUC initiator codon that modulates somes. Cell 44, 283-292.
define a sequence flanking the translation by eukaryotic ribo-
Kyte. J., and Doolittle, R. F. (1982). A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105-132. Landreth, C. E., and Shooter, E. M. (1980). Nerve growth factor receptors on PC12 cells: ligand-induced conversion from low- to high-affinity states. Proc. Natl. Acad. Sci. USA 77, 4751-4755. Large, T. H., Bodary, S. C., Clegg, D. O., Weskamp, G., Otten, and Reichardt, L. F. (1986). Nerve growth factor gene expression the developing rat braln. Science 234, 352-355. Levi-Montalcini, Science 237,
R. (1987). 1154-1162.
U., in later.
Levi-Montalcini, R., and Aloe, L. (1985). Differentiating effects of murine nerve growth factor in the peripheral and central nervous systems of Xwopus /aev~s tadpoles. Proc. Natl. Acad. SCI. USA 82, 7111&7115. Maniatis, T., Fritsch, E. F., dnd Sambrook, ing: A Laboratory Manual (Cold Spring Spring Harbor Laboratory).
J. (1982). Harbor,
Molecular ClonNew York: Cold
Marano, N., Dietzschold, B., Early, J. J., Schatteman, C., Thompson, S., Grob, P, Ross, A. H., Bothwell, M., Atkinson, B. F., and Koprowski, H. (1987). Purification and amino terminal sequencing of human melanoma nerve growth factor receptor. J. Neurochem. 48, 225-232. Marshall, 673-702.
R D. (1972).
Oppenhetm. R. W., Maderdrut, J. L., and death of motoneurons in the chick embryo tion of naturally occurring cell death in the of Tern1 by nerve growth factor. J. Comp.
retinal on a
Survival of isthmoDev. Brain Res. J2,
Wells, D. J. (1982). Cell spinal cord. VI. Reducthoracolumbar column Neural. 210, 174-189.
Proudfoot, N. j., and Brownlee, G. C. (1974). Sequence at the 3’ end of globin mRNA shows homology with immunoglobulin light chain mRNA. Nature 252, 359-362. Radeke. M. J., Misko, T. P., Hsu, C., Herzenberg, L. A., and Shooter, E. M. (1987). Gene transfer and molecular cloning of the rat nerve growth factor receptor. Nature 325, 593-597. Raivlch, G., Zimmermann, A., and Sutter, A. (1985). The spatial and temporal pattern of P-NGF receptor expression in the developing chick embryo. EMBO J. 4, 637-644. Raivlch, C., Zimmermann, A., and Sutter, A. (1987). factor (NGF) receptor expression in chicken cranial J. Comp. Neurol. 256, 229-245.
Nerve growth development.
Ranscht. B. (1988). Sequence of contactin, a 130 kd glycoprotein concentrated in areas of interneuronal contact, defines a new member of the immunoglobulin supergene family in the nervous system. J. Cell Biol. 107, 1561-1573. Rao, J, K. M., and Argos, D. (1986). A conformational preference parameter to predict helices In integral membrane proteins. Biochem. Biophys. Acta 869, 197-214. Rechsteiner, M., and intracellular
Rogers, S., and Rote, K. (1987). Protein structure stability. Trends Biochem. Sci. 12, 390-394.
Rogers, S , Wells, quences common sis. Science 234,
R.. and Rechsteiner, to rapidly degraded 364-368.
M. (1986). Amino acid seproteins: the PEST hypothe-
Sanger, F., Nicklen. S., and Coulson, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463-5467. Schatteman, G. C., Gibbs, L., Lanahan, A. A., Claude, P., and Bothwell, M. (1988). Expression of NGF receptor in the developing and adult primate central nervous system. J. Neurosci. 8, 860-873. Schechter, A. L., and Bothwell, M. A. (1981). Nerve receptors on PC12 cells: evidence for two receptor differing cytoskeletal association. Cell 24, 867-874.
growth factor classes with
Shelton, D. L., and Reichardt, L. F. (1984). Expression of B-nerve growth factor gene correlates with the density of sympathetic innervation in effectororgans. Proc. Natl. Acad. Sci. USA81, 7951-7955. Shelton, D. L., and Reichardt, L. F. (1986). Studieson theexpression of the p-nerve growth factor (NGF) gene in the central nervous system: level and regional distribution of NGF mRNA suggest that NGF functions as a trophic factor for several distinct populations of neurons. Proc. Natl. Acad. Sci. USA 83, 2714-2718. Southern, E. M. (1975). Detection of specific DNA fragments separated by gel electrophoresis. 503-517.
sequences J. Mol.
among Biol. 98,
Massague, J., Guillette, B. J., Czech, M. t?, Morgan, C. J., and Bradshaw. R. A. (1981). Identification of a nerve growth factor receptor protein in sympathetic ganglia membranes by affinity labeling. J. Biol. Chem. 256, 9419-9424.
Steinhilber, W., Poensgen, J., Rausch, U., Kern, H. F. and Scheele, G. A. (1988). Translational control of anionic trypsinogen and amylase synthesis in rat pancreas in response to caerulein stimulation. Proc. Natl. Acad. Sci. USA 85, 6597-6601.
Matsuda, H , Coughlin, M. D., Bienenstock, J., and Denburg, 1. A. (1988). Nerve growth factor promotes human hemopoietic colony growth and differentiation. Proc. Natl. Acad. Sci. USA 85, 6508-6512.
Stemple, D. L., Mahanthappa, N. K., and Anderson, D. 1. (1988). Basic FGF induces neuronal differentiation, cell division, and NGF dependence in chromaffin cells: a sequence of events in sympathetic development. Neuron 1, 517-525.
Metzger, H., and Kinet, J.-P. (1988). Fc receptors. FASEB J. 2. 3-11.
Sutter, A., Riopelle, R. J., Harris-Warrick, (1979). Nerve growth factor receptors.
R. M., and Shooter, E. M. Characterization of two dis-
tinct classes of binding sites on chick J. Biol. Chem. 254, 5972-5982.
Taniuchi, M., Clark, H. B., and Johnson, nerve growth factor receptor in Schwann Natl. Acad. Sci. USA 83, 4090-4098.
E. M. (1986). Induction of cells after axotomy. Proc.
Thomas, C. A., and Dancis, 77, 44-55.
B. M. (1973).
Tinoco, I., Borer, I? N., Dengler, B., Levine, M. D., Uhlenbeck, 0. C., Crothers, D. N., and Gralla, J. (1973). Improved estimation of secondary structure in ribonucleic acids. Nature 246, 40-41. Vale, R. D., Ignatius, M. I., and Shooter, E. M. (1985). Association of nerve growth factor receptors with the Triton X-100 cytoskeleton of PC12 cells. J. Neurosci. 5, 2762-2770. von Heijne, C. (1985). Signal sequences: Mol. Biol. 184, 99-105.
the limits of variation.
von Heijne, G. (1986). A new method for predicting quence cleavage sites. Nucl. Acids Res. 14, 4683-4690. Wayne, D. B., and Heaton, NGF by early chick embryo 727, 220-223.
M. B. (1988). Retrograde spinal cord motoneurons.
transport of Dev. Biol.
Whittemore, S. R., Ebendal, T., Larkfors, L., Olson, L., Seiger, A., Stromberg, I., and Persson, H. (1986). Developmental and regional expression of B-nerve growth factor messenger RNA and protein in the rat central nervous system. Proc. Natl. Acad. Sci. USA 83, 817-821. Williams, L. R., Varon, S., Peterson, C. M., Wictorin, K., Fischer, W., Bjorklund, A., and Gage, F. H. (1986). Continuous infusion of nerve growth factor prevents basal forebrain neuronal death after fimbria fornix transection. Proc. Natl. Acad. Sci. USA 83, 9231-9235. Yan, Q., Snider, W. D., Pinzone, J. J., and Johnson, E. M., Jr. (1988a). Retrograde transport of nerve growth factor (NCF) in motoneurons of developing rats: assessment of potential neurotrophic effects. Neuron I, 335-343. Yan, Q., and Johnson, E. M., Jr. (1988b). study of the nerve growth factor receptor rosci. 8, 3481-3498.
An immunohistochemical in developing rats. J. Neu-