A quantitative study of the developmental expression of nerve growth factor (NGF) receptor in rats

DEVELOPMENTAL

BIOLOGY

121,139-148

(198’7)

A Quantitative Study of the Developmental Expression of Nerve Growth Factor (NGF) Receptor in Rats QIAO YAN AND EUGENE M. JOHNSON, JR.~ Department

of Pharmacology,

Washington

University

School of Medicine,

Received July 31, 1986; accepted in revised

660 South Euclid Avenue, St. Louis, Missouri

63110

form December 9, 1986

The developmental expression of nerve growth factor (NGF) receptor was quantitated in either homogenates or plasma membrane-enriched preparations from whole rat embryos or from isolated tissues. The assay involved crosslinking ‘%I-NGF to receptors followed by immunoprecipitation with a monoclonal antibidy to rat NGF receptor. In some cases, the pellet was resuspended and subjected to a sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) autoradiographic analysis. The NGF receptor was found in whole embryo homogenates as early as embryonic Day 10 (ElO) (earliest age examined). The NGF receptor content in whole embryos per milligram protein increased about 3fold from El1 to El8 and decreased slightly at E20. SDS-PAGE autoradiography showed that the molecular weights of I25I-NGF-bound receptors did not vary with age. The NGF receptor content in sciatic nerve homogenates decreased 23-fold from newborn to adulthood. The change of NFG receptor level in hindleg muscle had a profile similar to that seen in sciatic nerve. The NGF receptor content in superior cervical ganglion (SCG) or dorsal root ganglion (DRG) homogenate preparations was expressed in two ways. (1) On a per milligram protein basis, in SCG, the receptor density was decreased slightly from E20 to adulthood; in DRG, it was relatively constant from El5 through postnatal Day 0 (PND-0) and then dropped 6.7-fold in adults. (2) On a per ganglion basis, in SCG, it increased 4.4-fold from E20 to adult; in DRG, it increased g-fold from El5 to PND-0 and then stayed constant through adulthood. In brain membrane preparations, the NGF receptor level decreased ll-fold from El5 to adulthood. In spinal cord membrane preparations, it decreased ‘7-fold from El8 to adulthood. Levels of receptor in cord were always greater than in brain. These data suggest that alterations in the NGF receptor density may have a role in changes in tissue responsiveness to NGF during development. 0 1987 Academic Press, Inc. INTRODUCTION

Nerve growth factor (NGF) is a protein which is essential for the development and maintenance of the function of peripheral sympathetic neurons and neural crest-derived sensory neurons (Levi-Montalcini and Angeletti, 1968; Thoenen and Barde, 1980; Pearson et ah, 1983). Administration of exogenous NGF causes hypertrophy of NGF target cells (Levi-Montalcini, 1966) and prevents naturally occurring cell death in those tissues (Hendry and Campbell, 1976; Hamburger et aZ., 1981). Systemic exposure of animals to antibodies against NGF results in the destruction of sympathetic neurons (LeviMontalcini and Booker, 1960) and sensory neurons (Johnson et al, 1980). There is strong evidence indicating that neurons obtain NGF from their target tissues which they innervate. Interrupting the connections of the neurons with their peripheral targets produces the same effects as treatment with antibodies to NGF (Hendry, 1975; Hendry and Campbell, 1976; Thoenen et al, 1978). NGF receptors present on the surface of the responsive neurons mediate the binding, internalization, and transport of NGF from their terminals to their cell bodies (Hendry et al., 1974; C,ampenot, 1977; Johnson et al., i To whom reprint

requests should be addressed. 139

1978). Pharmacological destruction of sympathetic nerve terminals by 6-hydroxydopamine or by blockade of axonal transport with colchicine results in a rapid decrease of the NGF level in the sympathetic ganglia and increases the NGF levels in the sympathetic target tissues (Korsching and Thoenen, 1985). Furthermore, endogenous NGF has been detected in sympathetic effector organs by two-site enzyme immunoassay (Korsching and Thoenen, 1983). NGF appears to be synthesized locally; sufficient amounts of NGF messenger RNA have been detected and correlate well with the amounts of NGF found and with the density of sympathetic innervation (Shelton and Reichardt, 1984; Heumann et ah, 1984). In contrast to the large number of studies about the presence and biological effects of NGF, few studies address the issue of changes in responsiveness of NGFdependent neurons to NGF during development. Sympathetic neurons become dependent upon NGF at certain stages of development, after embryonic date 14 (E14) for mice (Coughlin et aZ.,1977; Kessler and Black, 1980) and El5 for rats (Lahtinen et al., 1986). Responsiveness of chick dorsal root ganglia neurons to NGF decreases several-fold from E8 to about El6 (Herrup and Shooter, 1975; Greene, 1977; Rohrer and Barde, 1982). Likewise, in the central nervous system (CNS), exogenous intra0012-1606/87 $3.00 Copyright All rights

0 1987 by Academic Press, Inc. of reproduction in any form reserved.

140

DEVELOPMENTALBIOLOGY

ventricular injection of NGF increases choline acetyltransferase activity in neonatal rat forebrain (Gnahn et ah, 1983) and striatum (Mobley et aL, 1985) but produces little or no increase in adult brains, suggesting that the sensitivity of these cholinergic neurons to NGF decreases during development. Recently, an autoradiographic study with lz51-NGF shows that NGF binding sites are developmentally regulated in chicken (Raivich et aZ.,1985). This study, although providing an anatomical description, was not quantitative. In addition, no systematic study of NGF receptor expression in mammalian development has been done. Providing such a quantitative description of the developmental regulation of NGF receptor in mammals would be important in understanding the role of NGF in neural development. A recently developed mouse monoclonal antibody 192 (192-IgG) against the rat NGF receptor (Chandler et ah, 1984; Taniuchi and Johnson, 1985) provides a tool to study the NGF receptor. An immunoprecipitation assay using 192-IgG (Taniuchi et al., 1986a) allows sensitive quantitation of the NGF receptor levels in tissues. In the present study, a modification of the assay was used to study quantitatively the developmental expression of NGF receptor in rat. Some of the data have been submitted in abstract form (Yan and Johnson, 1986). MATERIALS

AND

METHODS

(1) 1251-NGFPreparation Mouse NGF (2.5 S) was purified from male mouse submaxillary glands by the method of Bocchini and Angeletti (1969) and iodinated with Na? and lactoperoxidase (Marchalonis, 1969). lz51-NGF was separated from unincorporated “‘1 and NGF aggregates by chromatography on a Bio-Gel P-100 column (Bio-Rad). lz51-NGF with the specific activity of 1800-2800 cpm/fmole was used in the present experiments. (2) Preparation of Mono&ma1 and Polyckmal Antibodies Mouse anti-rat NGF receptor monoclonal antibody 192 (192-IgG) was purified from mouse ascites by precipitation with 50% saturated ammonium sulfate followed by affinity chromatography on a monoclonal antibody 187 (mAb-187)/Sepharose-4B column. mAb-187 is a rat anti-mouse immunoglobulin K-light chain monoclonal antibody (Yelton et ah, 1981). New Zealand White rabbits (Boswell, Pacific, MO) were immunized with purified mouse IgG in complete Freund’s adjuvant. The polyclonal antibodies were affinity purified on a mouse IgG-Sepharose-4B column.

(3)

VOLUMEEl,1987

Tissue Dissection

Sprague-Dawley rats (Chappel Breeder, St. Louis, MO) were used in all studies. Vaginal plug positive was recorded as embryonic Day 0 (EO). Parturition was between E21 to E22; postnatal Day 0 (PND-0) was the day of birth. Adult animals were 3- to 4-month-old females. Timed pregnant rats were killed by decapitation. Embryos were quickly removed from the uterus and placed on ice. Brains, spinal cords, superior cervical ganglia (SCG), dorsal root ganglia (DRG), sciatic nerves, and hindleg muscles were carefully dissected out under a dissecting microscope and frozen on dry ice. The numbers of SCG and DRG were counted prior to freezing. Postnatal tissues were collected in a similar manner. For whole embryo experiments, intact embryos were immediately frozen on dry ice after removal from the uterus. Tissues were stored in a -70°C freezer until use. (4) Homogenate and Membrane Preparations Whole embryos, sciatic nerves, SCG, or DRG were homogenized in a glass tissue grinder in 1 mM phenylmethylsulfonylfluoride (PMSF)/phosphate-buffered saline (PBS), pH 6.5, on ice followed by 1OOOg centrifugation for 10 min. The pellets were resuspended, rehomogenized, and centrifuged once more and the supernatants were collected and pooled. This is referred to as homogenate preparation. Plasma membrane-enriched preparations of brains, spinal cords, and muscles were prepared according to the method of Costrini and Bradshaw (1979). The final plasma membrane-enriched pellets were resuspended in 1 mM PMSF/PBS, pH 6.5, and referred to as membrane preparation. Protein concentrations of homogenate and membrane preparations were determined (Lowry et al., 1951). Bovine serum albumin (BSA) was used as the standard. Homogenate and membrane preparations were stored at -70°C. Preparations were subsequently thawed and assayed. Samples were never refrozen since experimentation showed that repeated freezing and thawing resulted in loss of receptor binding. (5)

Immunoprecipitation

Assay

A modification of the previously reported immunoprecipitation method (Taniuchi et al., 1986a) was used: Duplicates of 0.5 ml of homogenate or membrane preparation suspended in 1 mM PMSF/PBS, pH 6.5, were incubated with 2 nM lz51-NGF in the presence or absence of 2 PM unlabeled NGF at 35°C for 60 min. Tubes with a lOOO-fold excess of cold NGF provided blank or nonspecific binding for every tissue of each age tested. After incubation, 20 yl of ethyldimethylisopropylaminocarbodiimide (EDAC, Pierce) solution was added to bring

YAN

AND JOHNSON

NGF

Receptor

in Develophg

Rats

141

the final EDAC concentration to 10 mM. The crosslinking RESULTS reaction was allowed to proceed for 20 min at 22°C (the Because of marked differences in receptor density same temperature for the following steps) and was among tissues and the mass of tissue available from the stopped by adding 30 ~1 o’f 1 M Tris-HCl, pH 7.6. Memvarious structures, two general preparations were used brane proteins were solubilized by 2% octylglucoside in the present experiments. In small structures with high (OG)/O.5% BSA for 60 min. Twenty microliters of 10% receptor density (SCG, DRG, sciatic nerve), homogenate Formalin-fixed Staphylococcus aureus (Pansorbin, Calpreparations (post-1000g supernatants) were analyzed biochem) was added to preabsorb radioactive species, conveniently. In structures with lower density (brain, which bind to Pansorbin, in the following steps. The spinal cord, and muscle) but large mass, membrane mixtures were clarified by SOOOgcentrifugation for 10 preparations made as described (Costrini and Bradshaw, min. Supernatants were incubated for 60 min with 5 pg 1979) produced receptor enrichment and enhanced senof 192-IgG to allow binding to lz51-NGF:receptor complexes and then 40 ~1 of 10% Pansorbin which had been sitivity of the assay. By direct comparison of E20 spinal prebound with rabbit ami-mouse IgG antibodies were cord homogenate preparations (post-1000g supernaadded for 45 min. ‘251-NG1F:receptor:antibody:Pansorbin tants) and plasma membrane-enriched preparations, the complexes were pelleted and resuspended with 150 ~1 of NGF receptor content was enriched about 3.5-fold in the 1.3% OG/l mM PMSF/0.5% BSA/PBS, pH 7.4, on ice. latter (Fig. 1A). In a similar experiment with PND-0 In duplicates, 60 ~1 of that resuspension was layered on muscle, the enrichment was about 5-fold (data not the top of a 250-~1 cushion of 0.15 M sucrose/l.3% OG/ shown). Figure 1 shows the range of linearity of the immu0.5% BSA/PBS, pH 7.4, in 400~~1polyethylene tubes noprecipitation assay on E20 spinal cord membrane (Curtin Matheson Scientific) and followed by an SOOOg preparation (Fig. lA, upper curve) and homogenate centrifugation for 1 min. Tubes were immediately frozen preparation (lower curve). The assay was relatively linin a methanol/dry ice bath and tips with pellet were cut ear to about 1 mg protein per milliliter (0.5 mg protein and their radioactivities were determined in a gamma per tube). With increasing amounts of protein, the assay counter (Beckman Gamma 4000). In control experiments, became nonlinear. When the protein concentrations were several nonsense mouse monoclonal antibodies were used kept constant by systematically altering the amount of in place of 192-IgG. Thes#eantibodies consistently gave adult brain membrane preparation (low NGF receptor the same blank (nonspecific) values obtained by using content) and the amount of E20 spinal cord membrane excess unlabeled NGF in the initial binding step. All the preparation (high NGF receptor content), the values for data reported here were duplicate determinations from receptor quantity obtained were additive and generated a single tissue preparation pooled from several animals. The actual number of animals in each pooled tissue a straight line (Fig. 1B). The additive and linear result preparation varied according to the mass of tissues and of this mixing experiment with tissues of high and low the age of animals. Very similar experiments were re- NGF receptor density indicates that when constant peated at least once for e,ach tissue and the patterns of amounts of protein from different tissues are analyzed, the NGF receptor expression were basically the same the differences in the NGF binding sites observed accurately reflect the relative amounts of receptor in the as in the experiments reported here. samples. Thus, as long as the protein concentrations were kept constant, the NGF receptor contents of dif(6) SDS-PAGE Autoradiography ferent tissues measured by this assay could be directly To examine the apparent molecular weights of the compared. Similar linear results were obtained in control NGF receptor, the preciplitated pellets were subjected experiments with homogenate preparations of sciatic to a sodium dodecyl sulfate-polyacrylamide gel electro- nerve containing different receptor densities (not shown). In the experiments below, the protein concenphoresis (SDS-PAGE) autoradiographic analysis. After trations used were 1 mg/ml (0.5 mg protein per tube) their radioactivities had been determined (see above), for all tissues except for SCG, where the protein conthe pellets were resuspended in 100 ~1of double-strength centration used was 0.5 mg/ml (0.25 mg per tube). reducing SDS-PAGE sample buffer (4% NaDodS0/5% 2-mercaptoethanol/20% glycero1/0.008% bromphenol blue/0.125% M Tris-HCl, pH 6.8), heated in a boiling (1) NGF Receptor in Whole Embryo water bath for 5 min, and1electrophoresed on 7% polyacrylamide gels (Laemmli, 1970). Gels were stained with To determine when the NGF receptor first appeared Coomassie brilliant blue R, destained, and dried. Au- and whether the total amount of NGF receptor molecule toradiograms were made with Kodak X-Omat AR film changed during development, whole embryo homogenate with DuPont lightening-plus intensifying screen. preparations as early as El0 were examined. Figure 2A

142

DEVELOPMENTAL

A IOC

VOLUME 121, 1987

BIOLOGY

6 . SPECIFIC (1 SPECIFIC

f MI (HI

l5C

SPECIFIC 0 NON-SPECIFIC

l

60 z E

IOC I \ z E c

40

5a 20

0 l0

I 0.25

1 0.5 PROTEIN

I 0.75 (mgltube)

I 1.0

0 L20 SC Adu Bf.

I

I

0 (1.5

0.1 0.4

I

0.2 0.3 PROTEIN

I

I

I

0.3 0.2

0.4 0.1

0.5 0

(mgltube)

FIG. 1. (A) Different amounts of protein from homogenate preparations (open circles) and membrane preparations (filled circles) of E20 spinal cord were assayed for the NGF receptor (see Materials and Methods). The values represent the means of two independent determinations of specific NGF receptor contents. With the same amount of protein, there was about 3.5 times more NGF receptor in membrane preparation than in homogenate preparation. (B) Protein concentrations were kept constant (1 mg/ml or 0.5 mg per tube) by mixing the membrane preparation of E20 spinal cord (high NGF receptor content) with the membrane preparation of adult brain (low NGF receptor content). The assay was linear and additive. The vertical lines indicate the variation range of the duplicates; it is within the dimension of the symbols in those cases without the vertical lines. The abscissas indicate the amount of protein per tube. The ordinates indicate the amount of NGF receptor (fmole) per mg protein.

shows the changes of NGF receptor content in the whole embryo homogenate preparation. The NGF receptor content was 12.2 fmole/mg protein in El1 embryos and increased about threefold to 34.9 fmole/mg protein in El8 and then dropped slightly in E20 embryos. In a separate experiment the NGF receptor content in El0 embryos was found to be about half of that found in El1 embryos (not shown). After the amount of radioactivity in the samples was determined, the same pellets were subjected to SDS-PAGE autoradiographic analysis (Fig. 2B). The same major bands previously observed by this methodology in adult animals, a major band of 92 kDa and a minor band of 220 kDa (Taniuchi et aZ., 1986a,b) were observed. In addition to these two bands, there is an additional 123-kDa band in all the embryonic homogenates which is not seen as clearly in the adult (Fig. 2B). Some of the radioactivity shown at the dye front of the SDS-PAGE autoradiogram (Fig. 2B) is ‘251-NGF nonspecifically bound to the immunoprecipitated pellet. A major component of the radioactivity at the dye front comes from either the dissociation of NGF dimer which bound the receptor but was not crosslinked or the dissociation of one monomer of the NGF dimer which was

crosslinked to the NGF receptor but was not crosslinked to the other monomer by EDAC. In either of these two cases the radioactivities were specific and counting the radioactivity in the pellets should quantitatively reflect the relative NGF receptor content with the assumption that one receptor binds one NGF dimer.

(2) NGF Receptor in

Sciatic

Nerve and Hindleg Muscle

The sciatic nerve, the largest peripheral mixed nerve, was chosen for studing the developmental change of NGF receptor in peripheral nerve. The NGF receptor contents of PND-0, PND-10, and adult sciatic nerve homogenate preparations were measured. As shown in Fig. 3A, the NGF receptor content decreased 23-fold from 276.1 fmole/mg protein in PND-0 sciatic nerve to 12.8 fmole/mg protein in adult. With muscle membrane preparations, a pattern very similar to that in sciatic nerve (Fig. 3B) was observed. Because of the enrichment produced by the membrane preparation procedure (about &fold), the receptor density in the hindleg muscle was considerably lower than that in the nerve. The similarity of the time course and magnitude of the change

YAN AND JOHNSON

NGF

Receptor

in

Developins

143

Rats

B A * SPECIFIC s\ NON-SPECIFIC

El1 El3 ababababab

El5

El8

E20

Adu ab

116.2592.566.2 -

45-

o-l-l-r,

I

Eli

El3

I

II

El5

I

El8

I

I

I

E20

AGE (day)

FIG. 2. (A) The same amount of protein (0.5 mg per tube) from whole embryo homogenates of Ell, E13, E15, E18, and E20 was assayed for the NGF receptor content (see Materials and Methods). The values represent the means of two independent determinations. The vertical lines indicate the variation range of the duplicates; it is within the dimension of the symbol in those cases without the vertical lines. (B) After the radioactivities were determined,, the immunoprecipitates, the same as shown in (A) were subjected to a 7% SDS-PAGE electrophoresis and autoradiography. Molecular weight marks are shown on the left. The assays were carried out with (lane a) or without (lane b) the presence of lOOO-fold excess unlabeled NGF (2 FM) (see Materials and Methods). Besides the 220- and 92-kDa bands, a distinctive band at 123 kDa can be seen. Adult spinal cord membrane preparation was subjected to the same procedure and is shown in the right. The bands at 66 are artifacts due to large amounts of BSA in the gel.

in NGF receptor content of the tissues is consistent with the idea that the NGF receptor present in the muscle tissue could reflect the NGF receptor within the nerve in the muscles. The exact cellular location of NGF receptor molecule in the muscle tissue will require anatomical localization by immunohistochemical techniques.

fmole/ganglion). Postnatally, it stayed very constant into adulthood (Fig. 4B). These data might indicate that the decrease of the NGF receptor density in the postnatal DRG was due to the increase in the size of the ganglion instead of a decline of absolute numbers of NGF receptors in the whole ganglion. (.4) NGF Receptor in CNS

(3)

NGF Receptor in SCG and DRG

NGF receptor content in SCG was measured starting at E20. The density of NG.F receptor decreased slightly from 80.0 fmole/mg protein in E20 SCG homogenate preparation to 50.4 fmole/mg protein in adult (Fig. 4A, closed circles). Because the SCG is a discrete structure, the same data could be expressed on a per ganglion basis as shown in Fig. 4A (open circles). The NGF receptor content per ganglion continuously increased from 1.82 fmole/SCG in E20 to 7.87 fmole/SCG in adult. On a per milligram protein basis, the NGF receptor content in DRG stayed relatively constant prenatally (El5 to PND-0). It reached a peak at E20 (113.6 fmole/ mg protein) and decreased greatly from PND-0 (109.5 fmole/mg protein) to adulthood (14.7 fmole/mg protein) (Fig. 4B). When the same data were expressed on per ganglion basis, the NGF receptor content increased 9.4fold from El5 (0.184 fmole/ganglion) to PND-0 (1.73

Recent evidence shows that NGF not only has a critical role in the development of the peripheral nervous system but also has effects on CNS neurons (Gnahn et al, 1983; Mobley et al, 1985; Shelton and Reichardt, 1986). The NGF receptor species in brain appears to be the same as in the periphery (Taniuchi et ah, 1986a). Here we studied the NGF receptor content in the CNS during development. In whole brain membrane preparations, the NGF receptor content decreased 11-fold from El5 (68.9 fmole/mg protein) to adulthood (6.2 fmole/mg protein) (Fig. 5A). In the spinal cord, the NGF receptor content varied little from El5 to PND-0. After PND-0, it decreased from 156.0 fmole/mg protein to 25.4 fmole/ mg protein in the adult animal (Fig. 5B). At any developmental stage, the NGF receptor content in spinal cord was considerably higher than in the brain. The temporal change of the NGF receptor in spinal cord was different from that in brain but very similar to that in DRG.

144

VOLUME 121, 1987

DEVELOPMENTAL BIOLOGY

A

6

3ot

l SPECIFIC 0 NON-SPECIFIC

l SPECIFIC 0 NON-SPECIFIC

2oc

F \ 'ii: E

100

100

0

0 PI0

PO

Adu

AGE

E20

PO

PI0

Adu

AGE

FIG. 3. (A) Sciatic nerve homogenate preparations and (B) muscle membrane preparations were assayed for the NGF receptor content during development. The immunoprecipitation assay (see Materials and Methods) were conducted with 0.5 mg protein per tube. The results are expressed in fmole NGF receptor per mg protein. The values represent the means of two independent determinations of specific NGF receptor content. The vertical lines indicate the variation range of the duplicates; it is within the dimension of the symbol in those cases without the vertical lines. DISCUSSION

The use of the crosslinking/immunoprecipitation assay permits the quantitation of NGF receptor content in various tissues in the process of development and the visualization of the receptor molecules by SDS-PAGE autoradiography. The validity of this assay lies in the ability of 192-IgG to bind at a different site on the receptor from the NGF binding site which allows the receptor to be immunoprecipitated after crosslinking to 1251-NGF (Chandler et aZ., 1984; Taniuchi and Johnson, 1985; Taniuchi et ak, 1986a). The assay procedure used here was modified from that previously reported (Taniuchi et al., 1986a) in that there was no ultracentrifugation step and the number of washes of the final immunoprecipitate was reduced. Although these modifications resulted in a slight lowering of the sample to blank ratios, the modifications made the assay less time consuming and also reduced the influence of the dissociation events during washes. Therefore, the receptor

densities determined with the modified procedure were higher and presumably closer to the real values. It must be appreciated that since subsaturating (2 nM) levels of ‘251-NGF are used and since pure NGF receptor is not available as an internal standard to correct for recovery, the values obtained are underestimates of receptor numbers. However, as shown in Fig. 1, relative receptor densities are reliably obtained. (1) NGF Receptor and NGF Content in Development Although the present experiment has not resolved the question of when the NGF receptor first appears in the embryogenesis, it is clear that the NGF receptor appears very early and its expression is regulated by developmental processes. In chick embryo, the NGF binding site has been shown to appear as early as E4 (Raivich et al., 1985). The NGF receptor molecules revealed by SDSPAGE autoradiography consist of a major 92-kDa band and a minor 220-kDa band, previously observed in adult

YAN AND JOHNSON

NGF

Receptor

in Developing

145

Rats

B

A

125 SPECIFIC (fmol/mg) 0 NON-SPECIFIC 0 SPECIFIC (fmol/DRG)

l

too

100 SPECIFIC (fmol/mg) Q NON-SPECIFIC 0 SPECIFIC (fmoVSCG1 l

i

,i . 75

: \ 5 c 50

2s

2.5

25

0 E20

PO

PI0

Adu

AGE FIG. 4. (A) SCG homogenate preparations assayed for the NGF receptor content. The two independent determinations of specific the dimension of the symbol in those cases

El5

El8

E20

PO

PI0

Adu

AGE (day) (0.25 mg protein per tube) and (B) DRG homogenate preparations (0.5 mg protein per tube) were results are expressed in fmole NGF receptor per mg protein. The values represent the means of NGF receptor content. The vertical lines indicate the variation range of the duplicate; it is within without the vertical lines.

tissues (Taniuchi et al., :1986a), and in addition, a frequently observed 123-kDa band which appears more prominently in tissues from early developmental stages. NGF-like immunoreactivity has been detected both in CNS and PNS of El5 to YE17rat embryos (Seiger et al., 1985). In the rat brain, an NGF-like protein is shown to double from El6 to PND-21 by an enzyme-linked immunoassay, whereas NGF messenger RNA in the brain is detected only after postnatal Day 1 and increased about 20-fold to the adlult level by 3 weeks of age (Whittemore et al., 1986). Although there is no quantitative data to compare the NGF content in other tissues during development, the CNS data available indicate that the NGF content increases during development and stays relatively high and1constant postnatally. In contrast, the NGF receptor content, as reported here, decreased dramatically during development and stayed relatively low postnatally. Raivich et ah (1985) also noted a decrease in the NGF b’inding in many tissues of developing chicken by an autoradiographic method. The biological implications of the reciprocal expression of these two molecules in the development remains to be elucidated.

(2) NGF Receptor in SCG and in DRG The NGF receptor content on a per milligram protein basis in SCG remained relatively constant from E20 to adulthood; corresponding to this, the dependence of sympathetic neurons on NGF for maintenance of function and survival persists into adulthood. The dependence on NGF for the normal function of sympathetic neurons is reflected in atrophy (reductions of the size of the ganglionic cell bodies, the levels of tyrosine hydroxylase, and dopamine /3-hydroxylase) or cell death after treatment of the animals with purified antibodies to NGF (Goedert et ah, 1978) or in autoimmune animals (Otten et al., 1979; Gorin and Johnson, 1980; Johnson et al., 1980; Rich et aZ., 1984). In DRG, the NGF receptor content on a per milligram protein basis decreased about 7.5-fold from PND-0 to adulthood. Correlated with this, mature DRG neurons seem not to need NGF for survival (Levi-Montalcini and Angeletti, 1966, 1968; Johnson et ah, 1980; Rich et ah, 1984) but DRG neurons in adult animals retain a dependence on NGF for at least some function (Schwartz et ah, 1982; Rich et aL, 1984). Our data show that the NGF receptor content per ganglion

146

VOLUME 121. 198’7

DEVELOPMENTAL BIOLOGY

6

A

200

. SPECIFIC ‘, NON-SPECIFIC

60 . SPECIFIC 0 NON-SPECIFIC

60

E \ 5 E w

E” \ z E 40

100

20

) 0

El5

6

0 -LEGll+-i

w-Ir El6

E20

PO

PI0

Adu

El5

AGE (day)

El6

E20

PO

PI0

Adu

AGE (day)

FIG. 5. (A) Brain membrane preparations (0.5 mg protein per tube) and (B) spinal cord membrane preparations (0.5 mg protein per tube) were assayed for the NGF receptor content. The results are expressed in fmole NGF receptor per mg protein. The values represent the means of two independent determination, of specific NGF receptor content. The vertical lines indicate the variation range of the duplicates; it is within the dimension of the symbol in those cases without the vertical lines.

remained quite constant postnatally. Evidence indicates that other trophic factors are also important for the survival and function maintenance of adult DRG neurons (Johnson et aZ.,1986). NGF receptors have been found on nonneuronal cells isolated from ganglia and cultured for various times (Sutter et ab, 1979; Carbonetto and Stach, 1982; Zimmerman and Sutter, 1983; Rohrer, 1985). Whether nonneuronal NGF receptors are present on Schwann cells within SCG and DRG in viva is not known. Similarly, whether the NGF receptor on these glia cells is developmentally regulated and how much it contributes to the results reported here are interesting questions but also unknown. Morphological studies with 192-IgG as an immunohistochemical probe are currently under way and hopefully will provide some information on this point. The time course of changes in the NGF receptor content of DRG was very similar to that of spinal cord. This may indicate that most NGF receptors in the spinal cord are associated with the central processes of DRG neurons. This suggestion is supported by the observation of high densities of NGF binding sites in the superficial layers (laminae I and II) of the dorsal gray matter of

adult spinal cord which disappears after dorsal root section (Yip and Johnson, 1986). (3)

NGF Receptor in Peripheral Nerve

The NGF receptor content decreases about 23-fold from PND-0 to adulthood in the sciatic nerve. A similar time course was seen in the hindleg muscle. This concordance is probably a reflection of a situation in which the NGF receptors in muscle reside on the nerve within the muscle rather than on the muscle cells (Taniuchi, 1986b). The reduction of the NGF receptor content in the sciatic nerve and muscle from PND-0 to adulthood was striking. This decrease could be due to several factors. (1) The general growth including the myelination of the nerve may dilute the NGF receptor density. However, during the same period of time myelination is also taking place in the CNS. Between PND-0 and adulthood the NGF receptor content only decreased 6-fold in brain and spinal cord instead of the 23-fold change seen in the sciatic nerve. (2) The NGF receptor might be expressed at an early developmental stage by some cell type which does not express the receptor in mature nerve. Evidence suggests that Schwann cells might behave in this man-

YAN AND JOHNSON

NGF

Receptor

in Developing

Rats

147

brain. These data suggest that alterations in NGF rener. The NGF receptor has been found on cultured Schwann cells of embryonic chicken sensory and sym- ceptor density may play a role in changes of tissue repathetic ganglia (Zimmermann and Sutter, 1983;Rohrer, sponsiveness to NGF during development. 1985). Specific NGF binding sites have been detected on ACKNOWLEDGMENTS rat Schwann cells immediately after isolation from sciatic nerve of PND-1 rats (P. DiStefano, unpublished). The authors gratefully acknowledge the experimental and editorial NGF receptors have also been demonstrated on Schwann assistance of Patricia Osborne. This work was supported by NIH Grant cells in vivo. Seven days after the transection of adult NS 18071, by the Washington University Alzheimers’ Disease Research sciatic nerve, the NGF receptor content increases about Center (NIH Grant AG 05681), and by a grant from the Monsanto Co., 50-fold in the nerve distd, but not proximal, to the ax- St. Louis, MO. otomy. NGF receptor molecules have been immunohisREFERENCES tochemically localized on the surface of Schwann cells. Receptors are not seen o’n Schwann cells of the normal BOCCHINI, V., and ANGELETTI, P. U. (1969). The nerve growth factor: Purification as a 30,000 molecular weight protein. Proc. Natl. Acad adult nerve (Taniuchi et al., 1986b). Since regeneration Sci. USA 64,78’7-794. processes often recapitulate processes of development, the same mechanism may mediate the regulation of the CAMPENOT, R. B. (1977). Local control of neurite development by nerve growth factor. Proc. Natl. Acad Sci USA 74,4516-4519. NGF receptor in the developing as in the regenerating CARBONETTO, S., and STACH, R. (1982). Localization of nerve growth nerve. We suggest that Schwann cells in developing factor bound to neurons growing nerve fibers in culture. Dev. Brain nerve, but not in mature adult nerve, may bear NGF Res. 3,463-473. receptors and that the (dramatic decrease in NGF re- CHANDLER, C. E., PARSONS, L. M., HOSANG, M., and SHOOTER, E. M. (1984). A monoclonal antibody modulates the interaction of nerve ceptor number in nerve reflects loss of Schwann cell regrowth factor with PC12 cells. J. Biol. Chem. 259, 6882-6889. ceptors. Obviously, electron microscopic examination of COSTRINI, N. V., and BRADSHAW, R. A. (1979). Binding characteristics the sciatic nerve at earl:y developmental stages will be and apparent molecular size of detergent-solubilized nerve growth required to resolve this issue. (3) Since the quantitation factor receptor of sympathetic ganglia. Proc. Natl. Acad. Sci. USA 76.3242-3245. of NGF receptors by this method is dependent on the COUGHLIN, M. D., BOYER, D. M., and BLACK, I. B. (1977). Embryologic recognition of receptors by 19%IgG, it is possible that development of a mouse sympathetic ganglion in viva and in vitro. the dramatic decrease in receptor density in nerve is Proc. Natl Acad. Sci. USA 74,3438-3442. partially because of the conversion of NGF receptors to GNAHN, H., HEFTI, F., HEUMANN, R. SCHWAB, M. E., and THOENEN, H. a form not recognized by 192-IgG. It appears that 192(1983). NGF-mediated increase of choline acetyltransferase (ChAT) in the neonatal rat forebrain: Evidence for a physiological role of IgG recognizes the “fast” form of the NGF receptor NGF in the brain? Dev. Brain Res. 9,45-52. (Chandler et ah, 1984; Taniuchi et ab, 1986a). Since GOEDERT, M., OTTEN, U., and THOENEN, H. (1978). Biochemical effects Schwann cells appear to bear only the fast receptor of antibodies against nerve growth factor on developing and differ(Zimmerman and Sutter, 1983; P. DiStefano, unpubentiated sympathetic ganglia. Brain Res. 148,264-268. lished), a decrease in receptors on Schwann cells GORIN, P. D., and JOHNSON, E. M., JR. (1980). Effects of long-term nerve growth factor deprivation on the nervous system of the adult rat: during development, as suggested above, would proAn experimental autoimmune approach. Brain Res. 198,27-42. duce a marked decrease in the NGF receptors detected GREENE, L. A., (1977). Quantitative in vitro studies on the nerve growth by 192-IgG. factor (NGF) requirement of neurons. II. Sensory neurons. Dev. BioL In summary, the NG:F receptor has been quantita58,106-113. tively studied in whole embryos, SCG, DRG, sciatic HAMBURGER, V., BRUNSO-BECHTOLD, J. K., and YIP, J. W. (1981). Neuronal death in the spinal ganglia of the chick embryo and its renerve, hindleg muscle, haain, and spinal cord of clevelduction by nerve growth factor. J. Neurosci 1,60-71. aping rats. The NGF receptor appeared very early in HENDRY, I. A. (1975). The response of adrenergic neurons to axotomy development. The gener,al picture of the change of the and nerve growth factor. Brain Res. 94,87-97. NGF receptor content is that it decreased with clevel- HENDRY, I. A., and CAMPBELL, J. (1976). Morphometric analysis of rat opment in all the tissues except SCG. Prenatally, the superior cervical ganglion after axotomy and nerve growth factor treatment. J. Neurocytol. 5.351-360. changes of NGF receptor content varied with different tissues: it increased slightly in the whole embryo, stayed HENDRY, I. A., STOOKEL, K., THOENEN, H., and IVERSEN, L. L. (1974). The retrograde axonal transport of nerve growth factor. Bruin Res. relatively constant in SCG, DRG, and spinal cord, and 68,103-121. decreased in brain and muscle. The NGF receptor density HERRUP, K., and SHOOTER, E. M. (1975). Properties of the B-nerve dramatically decreased postnatally in all the tissues exgrowth factor receptor in development. J. Cell Biol 67,118-125. aminecl except SCG. The relative receptor density varied HEUMANN, R. KORSCHING, S. SCOTT, J., and THOENEN, H. (1984). Relationship between levels of nerve growth factor (NGF) and its meswith age: in neonatal rats, the NGF receptor density senger RNA in sympathetic ganglia and peripheral target tissues. decreased in the order of sciatic nerve, DRG, SCG, musEMBO J. 3,3183-3189. de, spinal cord, and bra-in; and in adult rats, the order JOHNSON, E. M., JR., ANDRES, R. Y., and BRADSHAW, R. A. (1978). Charwas SCG, DRG, sciatic nerve, spinal cord, muscle and acterization of the retrograde transport of nerve growth factor

148

DEVELOPMENTAL BIOLOCY

(NGF) using high specific activity (‘?)NGF. Bruin Res. 150, 319331. JOHNSON, E. M., JR., GORIN, P. D., BRANDEIS, L. D., and PEARSON, J. (1980). Dorsal root ganglion neurons are destroyed by exposure in utero to maternal antibody to nerve growth factor. Science210,916918. JOHNSON, E. M., JR., RICH, K. M., and YIP, H. K. (1986). The role of NGF in sensory neurons in viva. Trends Neurosci. 9,33-37. KESSLER, J. A., and BLACK, I. B. (1980). The effects of nerve growth factor (NGF) and antiserum to NGF on the development of embryonic sympathetic neurons in wivo. Bruin Res. 189,157168. KORSCHING, S., and THOENEN, H. (1983). Nerve growth factor in sympathetic ganglia and corresponding target organs of the rat: Correlation with density of sympathetic innervation. Proc. Natl. Acd Sci. USA 80,3513-3516. KORSCHING, S., and THOENEN, H. (1985). Treatment with 6-hydroxydopamine and colchicine decreases nerve growth factor level in sympathetic ganglia and increases them in the corresponding target tissues. J. Neurosci. 6,1058-1061. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227,680685. LAHTINEN, T., SOINILA, S., and ERANKO, 0. (1986). Age-dependent stimulation by atrium explants or nerve growth factor of nerve fibre outgrowth from cocultured embryonic rat sympathetic ganglia. Dev. Brain Res. 27,51-57. LEVI-M• NTALCINI, R. (1966). The nerve growth factor: Its mode of action on sensory and sympathetic nerve cells. Harvey Lect. 60,217-259. LEVI-M• NTALCINI, R., and ANGELETTI, P. U. (1966). Immunosympathectomy. PharmacoL Rev. 18,619-628. LEVI-M• NTALCINI, R., and ANGELETTI, P. U. (1968). The nerve growth factor. Physiol Rev. 48,534-569. LEVI-M• NTALCINI, R., and BOOKER, B. (1960). Destruction of the sympathetic ganglia in mammals by an antiserum to a nerve-growth protein. Proc NatL Acad. Sci. USA 46,38&391. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., and RANDALL, R. J. (1951). Protein measurement with the folin phenol reagent. J. BioL Chem. 193,265-275. MARCHALONIS, J. J. (1969). An enzyme method for the trace iodination of immunoglobulins and other proteins. B&hem. J. 113,229-305. MOBLEY, W. C., RUTKOWSKI, J. L., TENNEKOON, G. I., BUCHANAN, K., and JOHNSON, M. V. (1985). Choline acetyltransferase activity in striatum of neonatal rats increased by nerve growth factor. Science 229,284-287. OTTEN, U., GOEDERT, M., SCHWAB, M., and THIBAULT, J. (1979). Immunization of adult rats against 2.5s nerve growth factor: Effects on the peripheral sympathetic nervous system. Bruin Res. 176, 7990. PEARSON, J., JOHNSON, E. M., JR., and BRANDEIS, L. (1983). Effects of antibodies to nerve growth factor on intrauterine development of derivatives of cranial neural crest and placode in the guinea pig. Dew. BioL 96,32-36. RAIVICH, G., ZIMMERMANN, A., and SUTTER, A. (1985). The spatial and termporal pattern of BNGF receptor expression in the developing chick embryo. EMBO J. 4,637~644. RICH, K. M., YIP, H. K., OSBORNE, P. A., SCHMIDT, R. E., and JOHNSON, E. M., JR. (1984). Role of nerve growth factor in the adult dorsal root ganglia neuron and its response to injury. J. Camp. Neural. 230, 110-118.

VOLUME 121. 1987 ROHRER, H. (1985). Nonneuronal cells from chick sympathetic and dorsal root sensory ganglia express catecholamine uptake and receptors for nerve growth factor during development. Dev. BioL 111, 95-107. ROHRER, H., and BARDE, Y. A. (1982). Presence and disappearance of nerve growth factor receptors on sensory neurons in culture. Dev. Biol 89, 309-315. SCHWARTZ, J. P., PEARSON, J., and JOHNSON, E. M., JR. (1982). Effects of exposure to anti-NGF on sensory neurons of adult rats and guinea pigs. Brain Res. 244,378-381. SEIGER, A., AYER-LELIEVRE, C., EBENDAL, T., and OLSEN, L. (1985). NGF-like immunoreactivity in the central and peripheral nervous system of the fetal rat. Sot. Neurosci. Abs. 11, 933. SHELTON, D. L., and REICHARDT, L. F. (1984). Expression of the p-nerve growth factor gene correlates with the density of sympathetic innervation in effector organs. Proc. NatL Acad Sci. USA 81, 79517955. SHELTON, D. L., and REICHARDT, L. F. (1986). Studies on the expression 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. SUTTER, A. RIOPELLE, R. J., JARRIS-WARRICK, R. M., and SHOOTER, E. M. (1979). The heterogeneity of nerve growth factor receptors. In “Transmembrane Signalling” (M. Bilensky, R. J. Collier, D. F. Steiner, and C. F. Fox, Eds.), pp. 659-667. Alan R. Liss, New York. TANIUCHI, M., and JOHNSON, E. M., JR. (1985). Characterization of the binding properties and retrograde axonal transport of a monoclonal antibody directed against the rat nerve growth factor receptor. J. Cell BioL lOl, llOO-1106. TANIUCHI, M., SCHWEITZER, J. B., and JOHNSON, E. M., JR. (1986a). Demonstration of nerve growth factor receptor in the rat brain. Proc. NatL Acad Sci USA 83,1950-1954. TANIUCHI, M., CLARK, H. B., and JOHNSON, E. M., JR. (198613). Induction of nerve growth factor receptors in Schwann cells after axotomy. Proc. NatL Acad. Sci. USA 83,4094-4098. THOENEN, H., and BARDE, Y. A. (1980). Physiology of nerve growth factor. PhysioL Rev. 60,1284-1335. THOENEN, H., SCHWAB, M., and OTTEN, U. (1978). Nerve growth factor as a mediator of information between effector organs and innervating neurons. Symp. Sot. Dev. BioL 35,101-118. WHITTEMORE, S. R., EBENDAL, T., LARKSFORS, L., OLSON, L., SEIGER, A., STROMBERG, I., and PERSSON, H. (1986). Developmental and regional expression of p-nerve growth factor messenger RNA and protein in the rat central nervous system. Proc. NatL Acad Sci. USA 83,817-821. YAN, Q., and JOHNSON, E. M., JR. (1986). NGF receptor in developing rats. Sot. Neurosci. Abs. 12,1093. YELTON, D. E., DESAYMARD, C., and SCHARFF, M. D. (1981). Use of monoclonal anti-mouse immunoglobulin to detect mouse antibodies. Hybridoma 1,5-11. YIP, H. K., and JOHNSON,E. M., JR. (1986). Nerve growth factor receptors in rat spinal cord: An autoradiographic and immunohistochemical study. Neuroscience, in press. ZIMMERMAN, A., and SUTTER, A. (1983). P-Nerve growth factor (BNGF) receptors on glial cells. Cell-cell interaction between neurons and Schwann cells in cultures of chick sensory ganglia. EMBOJ. 2,879885.