0306-4522/92 $5.00 + 0.00 Pergamon Press Ltd

Neuroscience Vol. 50, No. 2, pp. 467482, 1992

Printed in Great Britain

IBRO

DEVELOPMENTAL PATTERN A N D DISTRIBUTION OF NERVE GROWTH FACTOR LOW-AFFINITY RECEPTOR IMMUNOREACTIVITY IN H U M A N SPINAL CORD A N D DORSAL ROOT GANGLIA: COMPARISON WITH SYNAPTOPHYSIN, N E U R O F I L A M E N T A N D NEUROPEPTIDE IMMUNOREACTIVITIES A. M. SUBURO,* X.-H. Gu,* G. M o s c o s o , t A. ROSS,+ + G. TERENGHI*§ and J. M. POLAK* *Department of Histochemistry, Royal Postgraduate Medical School, Hammersmith Hospital, Du Cane Road, London WI20NN, U.K. ]'Department of Morbid Anatomy, Kings College Hospital, Denmark Hill, London SE5 8RX, U.K. ~:Worcester Foundation for Experimental Biology, 222 Maple Avenue, Shrewsbury, MA 01545, U.S.A. Abstract Immunocytochemical expression of the low-affinity nerve growth factor receptor was studied in human fetal and adult tissues using the monoclonal antibody ME20.4. In dorsal root ganglia, a few immunoreactive neurons were first detected in nine-week-old fetuses and many more were found in the following weeks of gestation. However, none was present in adult ganglia. The ME20.4-positive cells were larger than neurons immunostained by substance P, calcitonin gene-related peptide or galanin antibodies. In the spinal cord, fibres immunostained by ME20.4 appeared in a characteristic pattern that differed from the spatial and temporal distributions of synaptophysin- and neurofilament-immunoreactive fibres. Those expressing the low-affinity nerve growth factor receptor were only detected in regions containing collaterals of primary sensory axons: (i) in the dorsal funiculus between seven and 18 weeks of gestation; (ii) in a ventrodorsal bundle reaching the ventral horn from weeks 12 14: (iii) in the medial region of the dorsal horn between weeks 12 and 20; (iv) in the superficial layers and lateral portion of the dorsal horn after the 14th week of gestation and also in adult spinal cord. During the fetal period, ME20.4 immunoreactivity was also found in motoneurons and peripheral nerve fibres in the skin, myotomes and gut. Sheaths of peripheral nerves and the adventitia of blood vessels were stained both in fetal and adult tissues. Thus, the low-affinity nerve growth factor receptor is: (i) strongly expressed in the developing human nervous system; (ii) transiently associated with a subset of large primary sensory neurons and with motoneurons: (iii) transiently and sequentially expressed by various groups of sensory afferents to the spinal cord; (iv) permanently expressed by fibres in the superficial layers of the dorsal horn, Clarke's column, nerve sheaths and the adventitia of blood vessels.

Nerve growth factor ( N G F ) is a neurotrophin 23'3~ that interacts with membrane-linked cell surface receptors of high- or low-affinity. 42'55 High-affinity receptors have been associated with N G F biological activity, ~8,54 whereas the role of the low-affinity receptors is still under investigation. 4 High-affinity receptors could result from the association of the low-affinity receptor plus the trk proto-oncogene product 2° but there is also evidence suggesting that the trk product could constitute the high-affinity N G F receptor by itself. 26 The low-affinity receptors can be demonstrated by species-specific antibodies 6"2~32,44that have already been used to study the development of the spinal cord and ganglia in rats and primates. 46'61 The pattern of expression of the N G F low-affinity receptor during human development may provide further insight into the role of this molecule in neural and non-neural §To whom correspondence should be addressed. Abbrevh~tions : CGRP, calcitonin gene-related peptide; NGF, nerve growth factor; PBS, phosphate-buffered saline. 467

differentiation in man. Thus, an immunohistochemical study with monoclonal antibody (ME20.4) raised against the human N G F low-affinity receptor 44 has been made, trying to answer the following questions: (1) Is it possible to demonstrate immunoreactivity for the low-affinity N G F receptor in human neuronal and non-neuronal cells, either during a transient developmental period or as a persistent phenotypical trait in the adult? (2) Is there any correlation between the development of ME20.4immunoreactive neurons and that of other central or peripheral structures? For this the spinal cord, peripheral nerves, myotomal derivatives, skin and gut from human fetuses and adults were screened to establish the presence of ME20.4 immunoreactivity in different neuronal and non-neuronal cell types. The ME20.4 staining pattern was compared with the distribution of immunoreactivity for other neuronal markers including antibodies against neurofilament subunits, the well-known proteins of neuronal intermediate filaments, 7'29 antibodies against synaptophysin, an integral membrane protein of synaptic

468

A, M. SuBuRo et al.

Table 1. Specimens used for immunocytochemistry and regions and segmental levels studied _ Age

N

Regi_0_ns studied . . . . . . . . . . Spinal cord DRG SG C T L S T L S T

Gestational weeks 7 1 --9 2 -10 3 -11 8 2 12 12 3 13 1 14 1 -15 1 . . 16 2 -17 3 1 18 2 -1 1 20 Adults 8 1

.

1

1

2 3 5 7

3 6

1 1

2 2 5 8

l

1

1 1

...... 1

-2 2

1

2 3 3

-

1

1

--

2

.

1 2 2

.

.

.

1

4

--

--

1 ...... 2 1 5 1 1 . . . .

2 2 -1

.

l

..... 2 2 . . . . 2

1

N, number of specimens; DRG, dorsal root ganglia; SG, sympathetic ganglia; C, cervical; T, thoracic; L, lumbar; S, sacral. The medulla oblongata was also available in a 10-week-old fetus.

vesicles9 appearing at an early stage o f neuronal differentiation in vivo and in vitro. 3'53 C o m p a r i s o n s with substance P, calcitonin gene-related peptide ( C G R P ) and galanin were also made, since these neuropeptides characterize different subsets o f dorsal root ganglion neurons. 15.24,34

EXPERIMENTAL

PROCEDURES

Tissues

Samples of human tissues were obtained from material fixed for routine pathological studies and are detailed in Table 1. Estimation of gestational age was done as previously described)6 Adult spinal cords and dorsal root ganglia were obtained at post mortem examination done less than 12h after death in subjects dying from nonneurological causes. Prenatal tissues were fixed in Bouin's fluid for 6 h. Some specimens were also fixed in 4.0% paraformaldehyde in pH 9.4 borate buffer for 4 h, but no qualitative differences in immunostaining were detected between the fixatives. Adult tissues were fixed in the same paraformaldehyde solution. After fixation, all tissues were washed and stored at 4°C in 15% (w/v) sucrose in 0.01 M phosphate-buffered 0.15 M saline pH 7.4 (PBS) containing 0.1 percent (w/v) sodium azide (BDH). Immunohistoehemistry

Cryostat sections (14#m) were cut with orientation through the longest axis of the ganglia. Each ganglion was cut throughout and all sections mounted on slides. They

were processed as previously described52 and incubated overnight at 4"C alternatively with one of the primary antibodies described in Table 2. The slides were then rinsed in PBS and incubated for l h with biotinylated horse anti-mouse or goat anti-rabbit IgG (Vector Laboratories. U.K.) and, after further rinsing, with peroxidase-linked avidin-biotin complex (Vector Laboratories, UK.). Peroxidase activity was demonstrated by a nickel-enhanced procedure. 4~ Nonspecific staining was assessed by omission of the primary antibodies in duplicate sets of slides. No immunolabelling was observed in these preparations. The specificity of antisera raised against neuropeptides was checked by absorption of optimally diluted antisera with 1 nmol/ml of the corresponding peptide (Peninsula Laboratories). No immunostaining of neural structures was obtained when the preabsorbed galanin, CGRP and substance P antisera were used. For general neuronal staining and precise morphological identification of spinal cord areas, sections from each tissue-block were stained with 0.1% (w/v) Cresyl Fast Violet. Nomarski optics were used for the visual inspection of sections and for microphotographs. Co-localization studies

One of the sections immediately before or after those incubated with ME20.4 antibody was immunostained with one of the neuropeptide antibodies (substance P, CGRP or galanin). In order to evaluate the co-localization of immunocytochemical markers, the second section of each pair was reversed on the cryostat knife blade before picking it up so that they were mounted with adjacent mirror-image faces upwards. ~ Morphometric analysis

Because the image analysis system used for the quantification of the results is not equipped with a microscope with Nomarski optics, the procedure of staining consecutive sections with antisera and Cresyl Violet had to be adopted. For each gestational period, measurements were made in all stained cells of consecutive sections cut through the middle portion of two or more different thoracic ganglia. Although sections were cut throughout each ganglia and all were stained, morphometric analysis was only done in those passing through the middle of the ganglia, i.e. in the largest sections of each tissue-block. All the stained cells with a nucleus containing a clearly identifiable nucleolus were scored, to obtain profiles that were approximately through the middle of the cell body. The perimeters of stained ceils were interactively defined by measuring frames and their areas were measured with a VIDAS image analyser? 7 ANOVA was used for statistical comparisons. RESULTS Since several anatomical regions at different stages o f development were studied with various i m m u n o histochemical markers, observations in the peripheral and the central nervous systems are reported separately. F o r each anatomical localization, the staining

Table 2. Antibody characteristics Antibody to (reference) CGRP, rat synthetic Galanin~ natural pig Low-affinity NGF receptor (ME20.4) Neurofilament subunits chicken (R39) 200,000 mol. wt neurofilament subunit, phosphorylated (RT97) Substance P, synthetic Synaptophysin

From Rabbit Rabbit Mouse Rabbit Mouse

polyclonal polyclonal monoclonal polyclonal monoclonal

Rabbit polyclonal Rabbit polyclonal

Code

Optimals dilution and source Hammersmith Hospital, London 1s'33 Hammersmith Hospital, London3~ A. Ross, Worcester~ D. Dahl, Boston7.33 B. H. Anderton, London 2~

1204 1152 1338 1472 1420

1/2000 1/2000 1/100 1/2000 I/25

1651 1624

1/3000 Hammersmith Hospital, London 33'34 1/4000 Biometra, U.K. 53

NGF receptor in human spinal cord and ganglia pattern obtained with the ME20.4 antibody in specimens of different ages is described and then compared with the staining pattern of other antibodies. Several segmental levels were studied but, unless specified in the text, results refer to the thoracic level.

Dorsal root ganglia and peripheral nerves ME20.4 antibody. Positive neurons were first detected in nine-week-old thoracic and in ll-week-old lumbar ganglia. Their number and size increased during the gestational period included in this study (Fig. IA, B, D, E, G, H, Table 3). Thus, in 12-week-old fetuses the areas of immunostained neurons were between 0.73 and 3.20x 102/~m: whereas in 20-week-old fetuses they were between 2.84 and 13.56 x 102/~m2. No immunostained neurons could be detected in adult dorsal root ganglia or in fetal sympathetic ganglia. Comparison with serial sections stained with Cresyl Violet showed that ME20.4 immunoreactivity was present in a fraction of the total neuronal population (Fig. 2) including the largest neurons in the ganglia. Although the histograms show more large cells in the ME20.4 immunostained sections than in the neighbouring Cresyl Violet stained one, visual controls were carried out to confirm that there were no neurons that stained for ME20.4 but not for Cresyl Violet. The difference is most probably due to the selection of stained cells (with nucleolus, cf. Morphometric analysis method) and to a possible slight underestimation of the area of Cresyl Violet stained cells. Indeed, the Nissl substance stained by Cresyl Violet tends to compact slightly around the cells nucleus, whereas ME20.4 staining is more diffused and it occupies the whole cell profile. This does not detract from the results, as even with adjustment for underestimation the distribution of Cresyl Violet cells will occupy the whole range of cell areas under consideration, while ME20.4 stained cells appear only in a narrower range. Staining of neuronal satellite cells was observed only in adult dorsal root ganglia (Fig. 3A). However, the sheaths surrounding spinal ganglia (Figs 1A, B) and peripheral nerves showed ME20.4-immunoreactivity during the fetal period. The latter were weakly stained between the ninth and l lth weeks (Fig. 4A), but intense immunoreactivity was observed in older specimens. The capsule surrounding sympathetic ganglia was also strongly stained. In 10-12-week-old fetuses, ME20.4-immunoreactive fibres were found additionally in the muscular layers of the gut, and within the myotomes, where they appeared as small nerve bundles or surrounding individual myotubes (Fig. 4C). Isolated nerve fibres and nerve bundles could also be observed in the skin dermis (Fig. 4D). The morphology and distribution of ME20.4-stained neuronal terminals in fetal skin was consistent with that of nerves immunoreactive for protein gene product 9.5, a pan neuronal marker (Terenghi, personal communication). Indeed it was

469

observed that at 10 weeks gestational age the developing nerves appear thick and clubqike in shape, and only at later gestational age (16-18 weeks) do they achieve a morphological appearance similar to that of adult nerve fibres. Synaptophysin. Synaptophysin-immunoreactive neurons were first detected in 10-week-old fetuses. Their number and size increased with age (Fig. IC, I: Table 3), and almost every neuron was positive in 20-week-old fetuses. However, synaptophysin immunoreactivity was not detected in the cell body of adult primary sensory neurons. In 12-week-old fetuses, profiles of synaptophysinimmunoreactive neurons measured between 0.52 and 2.92 x 102#m 2 whereas in 17-18 week-old fetuses they ranged between 0.83 and 10.2 x 102~tmz. No significant differences could be demonstrated between the average areas of synaptophysin- and ME20.4-immunoreactive neurons of these developmental stages (Table 3). In 20-week-old fetuses, the range of synaptophysin-positive cell bodies (1.10-13.58 x 1021tm2) was broader than that of ME20.4-positive cells since synaptophysin stained smaller cells than those revealed by ME20.4. No synaptophysin immunoreactivity was found in neuronal somata of sympathetic ganglia but it was present in nerve fibres within the ganglia. NeuroI~laments. The antiserum R39, raised against all neurofilament subunits, immunostained some of the neurons and fibres within the fetal spinal ganglia from the 10th week onwards. In 17 18-week-old fetuses, the range and average area of immunostained neurons did not differ from those stained by Cresyl Violet (Table 4). Every neuron was stained in adult ganglia. Using antibody RT97, that detects a phosphorylated epitope of the 200,000 mol. wt subunit, weak immunostaining of a few cells was observed in IO-11week-old fetuses. Stronger staining was found from the 12th week onwards (Fig. IF). Although a large number of fibres was always found within the fetal ganglia, immunostained neurons were few. and always less than neurons stained by R39 antiserum in consecutive sections. In 17 18-week-old fetuses, the average area of RT97-immunostained cells could not be distinguished from that of neurons stained with Cresyl Violet or antiserum R39 (Table 4). All adult sensory neurons were RT97-immunoreactive (Fig. 3). No immunostained neurons were detected in sympathetic ganglia, but fibres were strongly immunoreactive. Neuropeptides. Substance P-immunoreactive neurons were already present in thoracic ganglia of I1week-old fetuses. In 17-18-week-old fetuses, their areas, ranging between 0.84 and 3.71 x 102 llm :, were on average smaller than those of ME20.4-immunoreactive neurons (Table 3). Thus, substance Pimmunoreactivity was not found in cells as large as those stained by antibody ME20.4. Broader differences between these two classes of neurons were

A.M. Su~uRo et al.

470

detected at 20 weeks of gestation (Table 3), when substance P-immunostained neurons ranged between 1.20 and 4.45 x 102/~m 2, indicating that this immunoreactivity was now present in a group of very small neurons not detected by ME20.4 antibody. Substance P-immunoreactive cells, however, were within the size range of those stained by synaptophysin antiserum. Galanin- and C G R P - i m m u n o r e a c t i v e cells were first detected after the 16th week o f gestation. Both neuronal subsets showed the same average size, which fell within the lower half of the range of cytoplasmic areas of ME20.4-immunoreactive cells (Fig. 2, Table 4). A few C G R P - i m m u n o r e a c t i v e neurons were found in adult dorsal root ganglia (Fig. 3). The differences in the size of substance P-, C G R P - and galanin-immunoreactive neurons were consistent with the lack of evidence for co-localization of these neuropeptides and the low-affinity N G F receptor in the same neurons.

Spinal cord ME20.4 antibody. Immunocytochemical expression of the low-affinity N G F receptor followed characteristic temporal and regional sequences that are summarized in Table 5. Positive fibres were only present in regions containing primary sensory afterents: the dorsal root, the dorsal and dorsolateral funiculi and restricted regions of the dorsal horn. They were not found in the ventrolateral funiculus, Immunostained fibres were first observed in the periphery of the dorsal funiculus (Fig. 5A), and became more widely distributed in older fetuses (Fig. 5B, C). In the upper spinal cord they could be identified

b c df dlf dh e fg fp lc

Table 3. Cytoplasmic areas ofimmunostained cells in dorsal root ganglia of human fetuses Immunocytochemical marker Weeks 12 17-18 20

ME20.4 0.68-5.57 2.12-t-0.34 0.92-9.39 3.61 +0.15 2.84-13.56 6.88 _+0.40

Synaptophysin 0.52--2.92 1.52 + 0.43 0.83 __+1 0 . 2 3.20_+0.26 1.10- 1 3 . 5 8 5.67 __+0.71

Substance P 0.72 -3.44 IA4 + 0.34 0.843.71 1.99_+0.1"7 1.20~,.45 2.80 + 0.32

Cell areas of primary sensory neurons are expressed in 102#m 2. For each age, the first line indicates the range and the second the mean + S.D. of cell areas. Statistical comparisons were made using ANOVA. Differences between time-points are highly significant for each marker (P > 0.0001). No significant differences between markers are found at 12 weeks. At 17-18 weeks, there is no difference between the average areas of ME20.4- and synaptophysin-immunoreactive neurons but both are significantly larger than the average area of substance P-immunoreactive neurons (P > 0.0001). In 20-week-old fetuses, the average area of ME20.4-immunoreactive neurons is larger than that of synaptophysin-immunoreactive neurons and the latter is larger than that of substance P-immunoreactive neurons (P > 0.0001). within the cuneatus and gracilis tracts, and were also observed in the medulla oblongata of an 1 l-week-old fetus, up to the level of the cuneatus and gracilis nuclei (dorsal column nuclei). N o immunostaining of the dorsal funiculus was observed after the 18th week of gestation. Immunoreactivity in the gray matter and the dorsolateral funiculus appeared after the 12th week of gestation, depending on the cephalocaudal level (Table 5). Fibres with a dorsoventral orientation

Abbreviations used in the figures blood vessel m myotube capillary mc medial collaterals dorsal funiculus r dorsal root dorsolateral funiculus s posterior septum dorsal horn vc long ventral collaterals epidermis ve ventricular lumen fasciculus gracilis vh ventral horn fasciculi proprii vlf ventrolateral funiculus lateral collaterals

Fig. t. Micrographs of human fetal sensory ganglia from the thoracic level stained with antibodies ME20A (A, B, D, E, G, H), antisynaptophysin (C and I) and RT97 (F). Micrographs C-I were taken under Nomarski optics. (A, B) These are low-power micrographs of ganglia from 11- (A) and 17-week-old (B) fetuses, to compare the distribution of ME20.4 immunoreactivity at two different developmental stages. Few small positive neurons (thin arrows) can be observed in the 1l-week-old ganglion (A), that contained many more unstained neurons, as demonstrated by the study of consecutive serial sections using Cresyl Violet (not shown). Staining in one of the ganglion roots (r) indicates immunoreactivity of nerve sheaths. A larger number of immunostained neurons (thin arrows) is observed in the 17-week-old fetus (B). Immunostaining of the ganglion capsule and septa is also observed (thick arrows). Scale bar = 250#m. (C-I) These micrographs, taken at a higher magnification than A and B, illustrate immunostaining of representative neurons at different stages of development. Scale bar = 50/tm. (C, D) Micrographs from the 11-week-old ganglion showing in C a small neuron stained with antisynaptophysin and in D a larger cell stained with ME20.4. (E, F) Sections from a 12-week-old ganglion. In E, neurons immunostained with ME20.4 are larger and have a stronger immunoreactivity than those in younger specimens. In F, immunostaining with RT97 shows numerous nerve fibres and some cell bodies. (G, H) ME20.4-positive neurons are observed in a 17-week-old ((3) and a 20-week-old ganglion (H). Compare the increase in the size ofimmunostained neurons in D, E, G and H. (I) A section from a 20-week-old ganglion demonstrating the presence of synaptophysin immunoreactivity in large neurons.

8

9.

A. M. SunuRo et al.

472 4~

2

1

1

2

3

4

5

6

7

Areas

40t 30

MF-20A

20

10.

1

2

3

4

6

5

7 Areas

40-

30 ¸ .

¢n 20-

CGRP

E Z

lO

o

,, r - ~ r , I , - , " 1

2

3

.

4

.

.

.

.

5

.

.

.

.

.

.

6

.

.

7

Areas

Fig. 2. Histograms showing the distribution of labelled neurons according to cell area. in serial sections of a thoracic dorsal root ganglion from the 17th week of gestation. Each histogram shows the sizes of all the cells in a section of the same ganglion for three consecutive sections stained with Cresyl Violet, the ME20.4 antibody and CGRP antiserum. respectively. Neuronal areas are given in 10~#m2. Since Cresyl Violet stained all the neurons present in the section (n = 297), the first histogram reflects the size distribution of the whole neuronal population. Antibodies for ME20.4 and CGRP stained different subsets of the whole neuronal population. The ME20.4-immunoreactive subset (n = 190) is formed by relatively large cells whereas the CGRP-immunoreaetive subset (n = 78) is formed by smaller neurons.

through the middle of the dorsal horn (Fig. 5B) were observed between the 1lth and 13th weeks of gestation. Their localization suggested that they were ventral collaterals forming a connection between the dorsal funiculus and the ventral horn. In older fetuses, the abundant positive fibres present along the medial aspect of the dorsal horn (Figs 5C, 6A) probably corresponded to medial collaterals, from axons in the dorsal funiculus to the deep layers of the dorsal horn, where an immunostained plexus was observed. From the 14th to the 18th weeks of gestation strong immunoreactivity was always found in the most medial portions of laminae V and VI (medial basilar region) (Fig. 6A, B). Some of the medial fibres extended to lamina VII (nucleus dorsalis

or Clarke's column) (Fig. 6A). Staining of the medial fibres decreased after the 17th week. Another group of positive fibres surrounded the lateral aspect of the dorsal horn and appeared to form a connection between the dorsal root, the dorsolateral funiculus and deep laminae of the dorsal horn. A third group crossed through the middle of the dorsal horn. The strongest expression of the low-affinity NGF receptor in dorsal horn fibres was observed in 17-18-week-old fetuses. Immunostaining in the dorsotateral funiculus was first observed in 12-week-old fetuses and became stronger after the 17th week of gestation (Figs 5C, 6A, B). This region remained positive in the adult spinal cord, where immunoreactivity was also found in the superficial laminae of the dorsal horn (Fig. 7), in the same laminae stained by substance P, galanin and CGRP antisera (not shown). A few immunostained fibres were occasionally observed in Clarke's column. The low-affinity N G F receptor was also expressed by fetal motoneurons (Table 5) and by cells in the nucleus of the hypoglossus and the dorsal motor nucleus of the vagus of an l l-week-old fetus. Immunostaining of spinal motoneurons was stronger between the 12th and 14th weeks, when they were surrounded by a plexus of immunostained fibres. Their staining decreased thereafter and was not observed in the adult. Synaptophysin. Synaptophysin immunostaining of fibres in the spinal cord followed a characteristic pattern easily distinguished from that of ME20,4immunoreactive fibres. Thus, in seven-week-old embryos they were only found in the ventrotateral funiculi and the ventral gray. They appeared in the dorsal funiculi and the dorsal horn at a later stage, around the 10th week of gestation (Fig. 8A). In 12-week-old fetuses, they were observed in the ventral portion of the dorsal funiculus, in the region corresponding to the fasciculi proprii of the adult spinal cord (Fig. 8B). In contrast, ME20.4-immunostained fibres were concentrated in the most medial and dorsal portions of the dorsal funiculus (cf. Figs 5C, 8B). Little immunoreactivity remained in the white matter after the 14th week of gestation. In older fetuses, synaptophysin-immunoreactive fibres surrounded the dorsal and lateral aspects of the dorsal horn. and were also present in the deep layers of the dorsal horn and the medial basilar region (not shown), partially resembling the distribution of ME20.4-positive fibres. In fetal stages, staining of the neuropile was stronger in the ventral than in the dorsal horn (Fig. 8B), whereas in the adult, all regions of the gray matter showed a similar staining intensity. Neurofilaments and neuropeptides. The strong staining of numerous groups of fibres within the fetal spinal cord with antiserum R39 has already been described? 3'52 A more selective immunostaining was obtained with antibody RT97. This revealed all the

N G F receptor in human spinal cord and ganglia

Fig. 3. Sections of adult human dorsal root ganglia photographed under Nomarski optics. Scale bar = 50 #m. (A) Staining with ME20.4 antibody shows immunoreactivity at the level of perineuronal satellite cells. Neuronal cell bodies and nerve fibres are negative. (B) In contrast, antibody RT97 reveals a large number of neurons and fibres in the ganglia. Some axons form a glomerulus around some cell bodies. Notice that the unstained perineuronal satellite sheath separates the cell body from this glomerulus. (C) A sensory neuron showing C G R P immunoreactivity together with several unstained neurons. No immunoreaction is found in satellite cells. NSC 50/2-- H

473

474

A.M. SUBUROet

al.

Fig. 4. Different regions of a 10-week-old fetus, photographed under Nomarski optics. (A, B) Adjacent sections through the sciatic nerve. In A, staining with antibody ME20.4 demonstrates that axons within the nerve trunk are negative whereas nerve sheaths (arrows) are weakly immunoreactive. By contrast, the connective sheath around blood vessels is strongly immunoreactive. In B, staining with RT97 shows the strong immunoreactivity of axons (arrows). Scale b a r = 100pro. (C) A section through the myotome, showing a ME20.4-immunoreaetive fibre (arrow) adjacent to a myotube. Scale bar = 50 pro. (D) In the skin overlying the lumbosacral spine, ME20.4-immunoreactive nerves (arrows) are found immediately beneath the epidermis. The erythrocytes in the small blood vessels appear larger than 10 pro, because fetal red blood cells are larger than those of an adult (> 10.5/~m diameter before 12 weeks' gestation). Scale bar = 100 pro.

white matter, but the strongest immunoreactivity corresponded to the dorsal funiculus (Fig. 5D). In the dorsal horn, RT97 only stained some of the fibre groups shown by ME20.4, including those in the region o f the medial collaterals (Fig. 5D). Few or no immunostained fibres appeared in the lateral portion

of the dorsal horn. Immunoreactive neurons were never detected in the spinal gray matter; however, positive fibres were present in the ventral commissure at all ages of gestation. In the adult spinal cord, numerous RT97-immunoreactive fibres were found in all regions of the gray matter.

NGF receptor in human spinal cord and ganglia Table 4. Cytoplasmic areas of neurons in dorsal root ganglia of 17-18-week-old human fetuses Staining

Range

Cresyl Violet 0.38-9.68 Neurofilaments (R39) 0.66-9.73 200,000 mol. wt neurofilament subunit (RT97) 0.95-5.51 Galanin 0.86-5.17 CGRP 0.49-4.42 ME20.4 0.92-9.39

Mean + S.D. 2.37 _+0.17 2.83 + 0.22 2.52 + 0.51 2.21 _.+0.33 2.01 + 0.17 3.61 +0.15

The range and average area of stained cells is given in 1021zm2 + S.D. Statistical comparisons were made using ANOVA. Average areas of cells stained by all markers were significantly lower than the average area of ME20.4-immunoreactivecells (P > 0.01), but no significant differenceswere found between the average areas of neurons stained by the remaining markers.

Observations in specimens stained for substance P, galanin and CGRP were consistent with results already described in developing and adult human spinal cord. 17'33'52 DISCUSSION The molecule detected by antibody ME20.4, generated against the human low-affinity N G F receptor, was expressed by neurons in human fetal dorsal root ganglia and their fibres, whereas in the spinal cord it was found in regions occupied by primary sensory collaterals and also in motoneurons. Immunostaining of these structures decreased in later stages of development, suggesting that the trophic factor/s interacting with the low-affinity N G F receptor might play a role during the establishment of connections between the periphery and the spinal cord.

Primary sensory neurons expressing the low-affinity nerve growth factor receptor In the human fetal ganglia, the low-affinity N G F receptor appeared in a cell subpopulation that excluded the smallest neurons. During development, a small cell could either be a growing large neuron or a differentiated small neuron. Then, ME20.4

475

immunoreactivity in cells larger than those expressing other markers could imply either that the low-affinity N G F receptor was not expressed by growing neurons, or that the low-affinity N G F receptor was not expressed by the neuronal subsets of smaller size. Comparison of ME20.4-immunoreactive neurons and those expressing substance P, CGRP or galanin indicated that the low-affinity N G F receptor and those neuropeptides were expressed by neurons of different sizes. This is consistent with the negative result of co-localization studies. Therefore, the small ME20.4-positive neurons in 12-week-old fetuses probably represent the expression of the low-affinity N G F receptor in growing neurons. On the other hand, synaptophysin immunoreactivity was present in neurons of all sizes. Synaptophysin is a protein of synaptic vesicles appearing in all neurons.9 It appears in neuronal cell bodies at an early stage of differentiation but its accumulation in nerve terminals requires a long maturation process. 53Thus, its presence in neurons smaller than those showing ME20.4 immunoreactivity further confirms that some neurons differentiate without expressing the low-affinity NGF. A correlation between the localization of the lowaffinity N G F receptor and developmental actions of NGF cannot be made, since the biological effects of this neurotrophin are usually attributed to the highaffinity receptorJ TM The lack of ME20.4 immunoreactivity in sympathetic neurons is consistent with this hypothesis. Development of substance P-containing neurons in rodents requires NGF. 25'38 Thus, the absence of ME20.4 immunoreactivity in the neuronal subset expressing substance P, further suggests that the N G F receptors involved in this effect lack the ME20.4 antigen. In contrast, the presence of ME20.4-immunoreactive fibres in the adult dorsal horn, in the same localization as substance P- and CGRP-immunoreactive fibres, ~7implies that the lowaffinity N G F receptor could be related to some of the postnatal effects of N G F on neurons containing those peptides. 3°'43'59'63However, species differences cannot be excluded, since in the rat, the ME20.4 antigen

Table 5. Localization of ME20.4 immunoreactivity in the thoracic and lumbar levels of the spinal cord Dorsal funiculus Weeks

Thor.

7-9 10 11 12 13-16 17-18 20 adult

2/2 2/2 5/5 6/6 2/2 3/3 0/1 0/4

Lumb.

2/4 6/6 4/4 0/1

Dorsolateral funiculus Thor. 0/2 0/2 0/5 5/6 2/2 3/3 1/1 4/4

Lumb.

0/4 4/6 4/4 1/1

Dorsal horn Thor. 0/2 0/2 0/5 5/6 2/2 3/3 1/1 4/4

Lumb.

1/4 4/6 4/4 1/1

Motoneurons Thor. 0/2 2/2 3/5 5/6 1/2 3/3 1/1 0/4

Lumb.

2/4 4/6

0/1

Thor., thoracic level; Lumb., lumbar. For each localization, the first number indicates the number of specimens where immunoreactivity was observed, whereas the second one indicates the number of specimens studied.

476

A.M. SUaUROet al.

Fig. 5. Sections through the spinal cord at different gestational ages stained with ME20.4 ( A ~ ) and RT97 (D) antibodies. (A) A 10-week-old thoracic spinal cord, photographed under Nomarski optics, showing immunostaining in the periphery of the dorsal funiculus. Scale bar = 250/~m. (B) A 12-week-oldthoracic spinal cord showing immunostaining of the dorsal funiculus and ventral collaterals. Scale bar = 500 l*m. (C, D) Comparable sections through the dorsal root level of a sacral spinal segment of a 13-week-oldfetus showing the localization of ME20.4 (C) and RT97-immunoreactive (D) fibres. Both antibodies label a group of fibres in the medial portion of the dorsal horn. Scale bar = 250 am.

is expressed by all developing primary sensory neurons. 6J Differences between human and rat spinal ganglia were also demonstrated by antibody RT97 which stained all adult human primary neurons, whereas in the rat it reveals all the large neurons and some small/medium size ones..,29

Fibres of the spinal cord expressing the low-affinity nerve growth factor receptor In the adult spinal cord, the dorsal root separates into a medial bundle of coarse fibres, concerned particularly, but not exclusively, with the formation of the dorsal funiculus and a smaller lateral group of

fine fibres which branch in the dorsolateral funiculus or marginal zone of Lissauer? This anatomical segregation is probably related to the sequential growth of different types of sensory afferents. As it has been experimentally demonstrated in rat and chick embryos, projections to the deep laminae of the dorsal horn develop before projections to superficial laminae.S, 12,49 The earliest ME20.4-immunoreactive fibres probably were ascending collaterals, as indicated by their presence in the dorsal funiculus of different spinal segments up to the dorsal column nuclei. Some of the fibres stained by the synaptophysin antibody in

NGF receptor in human spinal cord and ganglia

477

Fig. 6. Sections through the dorsal portion of the spinal cord stained with the ME20.4 antibody. (A) In a 14-week-old fetus, immunoreactivity is present in the most dorsal portion of the dorsal funiculus and the fasciculus gracilis, No staining is found in the region of the fasciculi proprii. Medial collaterals surround the dorsal horn and turn into the gray matter where a plexus of immunostained fibres is found in laminae IV and V. A spot of strong immunoreactivity appears in the medial basilar region (long arrow) and a few fibres extend to lamina VII. Lateral collaterals are more weakly stained. Short arrows indicate another group of fibres through the middle of the dorsal horn. Scale bar = 500 pm. B. In a 17-week-old fetus, immunostaining is found in the dorsal funiculus. Medial collaterals have almost disappeared but strong immunostaining is still present in the medial basilar region (long arrow). Strong immunoreactivity is now present in the superficial (I) and deep laminae (IV and V), and also in the fibres occupying the middle region of the dorsal horn (short arrows). Scale bar = 100 #m,

the dorsal funiculus would also represent ascending collaterals. In the gray matter, however, early synapt o p h y s i n - i m m u n o r e a c t i v e fibres a p p e a r i n g before M E 2 0 . 4 - i m m u n o r e a c t i v e ones, would p r o b a b l y represent p r o p r i o s p i n a l connections. Since the accumulation o f s y n a p t o p h y s i n in nerve terminals is a late step in differentiation, 53 the presence o f a synaptophysin-positive plexus in the ventral h o r n of seven-

week-old fetuses supports the hypothesis t h a t m o t o n e u r o n s receive functional propriospinal innerv a t i o n before c o n n e c t i o n s between p r i m a r y sensory fibres a n d spinal i n t e r n e u r o n s are made. 37'5s A similar growth p a t t e r n has been described in the rat fetus. 3 In 10-week-old fetuses, s y n a p t o p h y s i n i m m u n o reactivity was also c o n c e n t r a t e d in the m o s t lateral a n d ventral p o r t i o n o f the dorsal h o r n , where a

478

A.M. SuauRo et

al.

Fig. 7. A micrograph showing the dorsal portion of an adult human spinal cord at the level of the cervical enlargeraent. Fibres showing ME20.4 immunoreactivity are found in the dorsolaterat funieulus and the superficial layers of the dorsal horn. No immunoreactivity is observed in the surrounding white matter corresponding to the dorsal and the ventrolateral funiculi. Scale bar = I mm.

group of spinal motoneurons expressing neuropeptide Y and its C-flanking peptide has been recently identified)2 In the dorsal horn, many of the ME20.4-immunoreactive fibres observed in the region of the sensory afferents medial collaterals followed the typical course of large-diameter collaterals, turning dorsally

at the level of the deep laminae:°'4~,47Some others followed the pathway of arcuate and short coUatorats to the column of Clarke. 22 In fetuses of 1~18 weoks' gestation, the medial basilar region was strongly labelled, both by ME20.4 and synaptophysin antibodies. In the adult, this region receives a great number of propriospinal and primary sensory

NGF receptor in human spinal cord and ganglia

479

Fig. 8. Synaptophysin immunostaining in the thoracic spinal cord of 10- and 12-week-old fetuses (A, B). In the younger fetus there are numerous positive fibres in the ventrolateral and ventral horn. A moderate number of fibres is present in the dorsal funiculus and the medial portion of the dorsal and ventral horns (short arrows). Notice a region of reduced immunoreactivity corresponding to the developing dorsolateral funiculus. Synaptophysin immunoreactivity is also found in the lateral aspect of the dorsal horn (arrow). Scale bar = 250 #m. (B) In the 12-week-old fetus, fibres are mainly found in the ventrolateral funiculus and ventral horn. However, they are also present in the dorsal funiculus, close to the dorsal root and in the region corresponding to the fasciculi proprii. No immunostaining is observed in the dorsolateral funiculus, but fibres can be seen in the deep laminae of the dorsal horn. Scale bar = 500 ~m.

afferents from other spinal segments 4~ and visceral afferents entering the spinal cord through the dorsolateral funiculus. 35 The ME20.4-immunoreactive fibres of the dorsolateral funiculus and the lateral aspect of the dorsal horn were the last to appear. Thus, the expression of the low-affinity N G F receptor in areas of the spinal cord containing primary sensory afferents followed the same developmental sequence described for these axons in experimental animals. 8'12'49 Although substance P-immunoreactive fibres surround the lateral aspect of the dorsal horn in the 1 lth week of gestation, 33"52ME20.4-immunostained fibres were not observed in this localization until the 14th week. This time-lag is consistent with the expression of the low-affinity N G F receptor and substance P in sensory neurons of different sizes. Since ME20.4-immunoreactive fibres were found in the superficial layers of the adult dorsal horn, in the same localization as substance P-, C G R P - and galanin-immunoreactive fibres, ~7 it must be concluded that primary sensory

neurons expressing low-affinity N G F receptor in adult man would be different from those expressing it in early fetal stages. Overlapping between fibres expressing this antigen and those expressing C G R P is also found in the spinal dorsal horn of monkey and cat. ~4 Fibres immunostained by RT97 were absent in the dorsolateral funiculus and lateral regions of the dorsal horn, suggesting a situation analogous to that observed in rats where RT97 antigen is preferentially expressed by large sensory neurons mainly projecting to the dorsal funiculus/~,29

The low-affinity nerve growth factor receptor and neuronal and non-neuronal differentiation In developing rat trigeminal neurons, the amount of low-affinity N G F receptor increases at approximately the same time as initial contact. ~° Our findings suggest a similar pattern in human development, since ME20.4-immunoreactive nerve fibres were detected in the skin and myotome at about the same

480

A.M. SU~UROeta/.

time as the earliest immunostained neurons appeared in the segmental ganglia. The myotome ME20.4immunoreactive fibres appeared before immunostained motoneurons could be observed, suggesting their sensory origin. This is supported by the early appearance of ME20.4-immunostained fibres in the dorsal funiculus, since ascending collaterals include a significant number of proprioceptive afferents. ~-~'4~ The role of the low-affinity N G F receptor in developing motoneurons is at present unknown. It has not been detected in the adult except in a novel group of neurons in the ventral horn of the rat lumbar spinal cord 27 or after rhizotomy, 28 Although it is present in motoneurons during their period of embryonic cell death, 39"62 neither N G F nor brainderived neurotropic factor, another neurotrophin, can keep motoneurons alive in vitro. ~ Immunoreactivity for ME20.4 in non-neuronal structures was detected around blood vessels, meningeal coats and nerve sheaths. A similar developmental pattern has been observed in non-neuronal structures of human femoral nerves, where strong immunostaining of Schwann cells is present in weeks 13-14 of gestation but decreases after the 18th week. 45 Expression of the low-affinity N G F receptor reappears in peripheral neuropathies. 5° Similarly, adult rat Schwann cells re-express the low-affinity receptor after transection of the sciatic nerve. ~6 Since ME20.4 immunoreactivity could be demonstrated in adult neuronal satellite cells in situ, a physiological role for the low-affinity N G F receptor cannot bc excluded. Expression of the ME20.4 antigen could be related to sensitivity to other neurotrophins, since all of them bind similarly to the low-affinity N G F receptor. 51

Besides, neurites in dorsal root ganglion explants show an almost equivalent response to NGF, brainderived neurotrophic factor or neurotrophin-3/~ A role for the low-affinity N G F receptor unrelated to N G F is also suggested by its appearance in developing non-neural tissues and by the high level of its mRNA in early differentiating tissues/~-~ CONCLUSIONS The strong expression of the low-atfinity NGF receptor molecule in fetal sensory neurons and its disappearance in later stages indicate that this receptor plays an important role during differentiation of the human sensory system. Moreover, the sequential appearance of fibres expressing this molecule in different regions of the spinal cord and the periphery further implies that the low-affinity N G F receptor would be involved in a developmental event, perhaps related to the formation or maturation of connections between primary sensory neurons and their central or peripheral targets. The localization of the low-affinity N G F receptor molecule in different neuronal and non-neuronal localizations is not strictly correlated with known biological effects of NGF, suggesting that the developmental rote of this receptor is not primarily related to N G F binding. Acknowledgements--Research described in this paper was supported by the BUPA Medical Foundation, U.K.: the Grand Charity of Freemasons, U.K: the Motoneurone Disease Association, U.K, the COLT Foundation, U.K; and the Arthritis and Rheumatism Council. AMS was a fellow from the Commission of the European Communities. The Royal Society, U.K. and CONICET, Argentina. and the recipient of an IBRO travel grant.

REFERENCES

1. Arakawa Y., Sendtner M. and Thoenen H. (1990) Survival effect of ciliary neurotrophic factor (CNTF) on chick embryonic motoneurons in culture: comparison with other neurotrophic factors and cytokines. J. Neurosci. 10, 3507-3515. 2. AriEns Kappers C. U., Huber G. C. and Crosby E. C. (1967) The Comparative Anatomy of the Nertous System of Vertebrates, Including Man, Vol. 1. Hafner, New York. 3. Bergmann M., Lahr G., Mayerhofer A. and Gratzl M. (199t) Expression of synaptophysin during the prenatal development of the rat spinal cord: correlation with basic differentiation processes of neurons. Neuroscience 42, 569-582. 4. Bothwell M. (1991) Keeping track of neurotrophin receptors. Cell 65, 915-918. 5. Byers M. R., Schatteman G. C. and Bothwell M. (1990) Multiple functions for NGF receptor in developing, aging and injured rat teeth are suggested by epithelia, mesenehymal and neural immunoreactivity. Development 109, 461~471. 6. Chandler C. E., Parsons L. M., Hosang M. and Shooter E. M. (1984) A monoclonal antibody modulates the interaction of nerve growth factor with PC12 cells. J. biol. Chem. 259, 6682q5889. 7. Dahl D. and Bignami A. (1977) Preparation of antisera to neurofilament protein from chicken brain and human sciatic nerve. J. comp. Neurol. 176, 645~57. 8. Davis B. M., Frank E., Johnson F. A. and Scott S. A. (1989) Development of central projections of lumbosacral sensory neurons in the chick. J. comp. Neurol. 279, 556-566. 9. De Camilli P. and Jahn R. (1990) Pathways to regulated exocytosis in neurons. A. Rev. Physiol. 52, 625-645. 10. Ernfors P., Henschen A., Olson L. and Persson H. (1989) Expression of nerve growth factor receptor mRNA is developmentally regulated and increased after axotomy in rat spinal cord motoneurons. Neuron 2, 1605-1613. l 1. Ferri G.-L., Sabani A., Abetli L., Polak J. M., Dahl D. and Portier M. M. (1990) Neuronal intermediate filaments in rat dorsal root ganglia: differential distribution of peripherin and neurofilament protein immunoreactivity and effect of capsaicin. Brain Res, 515, 331-335. 12. Fitzgerald M. (1987) Prenatal growth of fine diameter afferents into the rat spinal cord: a transganglionic tracer study. J. comp. Neurol. 261, 98-104. 13. Fitzgerald M. (1989) The course and termination of primary afferent fibres. In Textbook (~f Pain, 2nd edn (eds Wall P. D. and Melzack R.), pp. 46-q52. Churchill Livingstone, Edinburgh.

NGF receptor in human spinal cord and ganglia

481

14. Fried K., Risling M., Arvidsson U. and Paulie S. (1990) Nerve growth factor receptor-like immunoreactivity in nerve fibers in the spinal and medullary dorsal horn of the adult monkey and cat: correlation with calcitonin-gene related peptide-like immunoreactivity. Brain Res. 536, 321-326. 15. Gibson S. and Polak J. M. (1986) Neurochemistry of the spinal cord. In lmmunocytochemistry. Modern Methods and Applications, 2nd edn (eds Polak J. M. and Van Norden S.), pp. 360-389. John Wright, Bristol. 16. Gibson S. J., McGregor G., Bloom S. R., Sabate I. M., Mulderry P. M., Ghatei M. A., McGregor G. P., Morrison J. F. B., Kelly J. S., Evans R. M., Rosenfeld M. G. (1984). Calcitonin gene-related peptide immunoreactivity in the spinal cord of man and eight other species. J. Neurosci. 4, 3101-3111. 17. Gibson S. J., Polak J. M., Katagiri T., Su H., Weller R. O., Brownell D. B., Holland S., Hughes J. T., Kikuyama S., Ball J., Bloom S. R., Steiner T. J., de Belleroche J. and Clifford Rose J. (1988) A comparison of the distributions of eight peptides in spinal cord from normal controls and cases of motor neurone disease with special reference to Onuf's nucleus. Brain Res. 474, 255-278. 18. Green S. M., Rydel R. M., Connolly J. L. and Greene L. A. (1986) PCI2 mutants that possess low- but not high-affinity nerve growth factor receptors neither respond to nor internalize nerve growth factor. J. Cell Biol. 102, 830-843. 19. Hallb66k F., Ayer-LeLirvre C., Ebendal T. and Person H. (1990) Expression of nerve growth factor receptor mRNA during early development of the chicken embryo: emphasis on cranial ganglia. Development 108, 693 704. 20. Hempstead B. L., Martin-Zanca D., Kaplan D. R., Parada L. F. and Chao M. V. (1990) High-affinity NGF binding requires coexpression of the trk proto-oncogene and the low-affinity NGF receptor. Nature 350, 678-683. 21. Herlyn M., Steplewski Z., Herlyn D., Clark W. H., Ross A. H., Blaszczyk P., Pak K. Y. and Koprowski H. (1983) Production and characterization of monoclonal antibodies against human malignant melanoma. Cancer Ira,est. 1, 215 224. 22. Hogg I. D. (1944) The development of the nucleus dorsalis (Clarke's column). J. comp. Neurol. 81, 69-95. 23. Hohn A., Leibrock J., Bailey K. and Barde Y.-A. (1990) Identification and characterization of a novel member of the nerve growth factor/brain-derived neurotrophic factor family. Nature 344, 339 341. 24. Kai-Kai M. A. (1989) Cytochemistry of the trigeminal and dorsal root ganglia and spinal cord of the rat. Comp. Biochem. Physiol. 93A, 183 193. 25. Kessler J. A. and Black I. B. (1980) Nerve growth factor stimulates the development of substance P in sensory ganglia. Proc. natn. Acad. Sci. U.S.A. 77, 649-652. 26. Klein R., Jing S., Nanduri V., O'Rourke E. and Barbacid M. (199l) The irk proto-oncogene encodes a receptor for nerve growth factor. Cell 65, 189 197. 27. Koliatsos V. E., Shelton D. L., Mobley W. C. and Price D. L. (1991) A novel group of nerve growth factor receptor-immunoreactive neurons in the ventral horn of the lumbar spinal cord. Brain Res. 541, 121 128. 28. Kullas C. D., Henry M. A., Westrum L. E., Johnson L. R. and Bothwell M. (1991) Retrogasserian rhizotomy causes expression of nerve growth factor receptor-immunoreactive protein in motoneurons within the adult feline trigeminal motor nucleus. Neurosci. Lett. 126, 145-148. 29. Lawson S. N., Harper A. A., Harper E. I., Garson J. A. and Anderton B. H. (1984) A monoclonal antibody against neurofilament protein specifically labels a subpopulation of rat sensory neurones. J. comp. Neurol. 228, 263 272. 30. Lindsay R. M. and Harmar A. J. (1989) Nerve growth factor regulates expression of neuropeptide genes in adult sensory neurons. Nature 337, 362-364. 31. Maisonpierre P. C., Belluscio L., Squinto S., Ip N. Y., Furth M. E., Lindsay R. M. and Yancopoulos G. D. (1990) Neurotrophin-3: a neurotrophic factor related to H and BDNF. Science 247, 1446-1451. 32. Marano N., Dietzchold B., Earley J. J., Schatteman G., 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. 33. Marti E., Gibson S. J., Polak J. M., Facer P., Springall D. R., Van Aswegen G., Aitchison M. and Koltzenburg M. (1987) Ontogeny of peptide- and amine-containing neurones in motor, sensory, and autonomic regions of rat and human spinal cord, dorsal root ganglia and rat skin. J. comp. Neurol. 266, 332-359. 34. Merighi A., Polak J. M., Gibson S. J., Gulbenkian S., Valentino K. L. and Peirone S. M. (1988) Ultrastructural studies on calcitonin gene-related peptide-, tachykinins-, and somatostatin-immunoreactive neurones in rat dorsal root ganglia. Cell Tiss. Res. 254, 101-109. 35. Morgan C., Nadelhaft I. and de Groat W. C. (1981) The distribution of visceral primary afferents from the pelvic nerve to Lissauer's tract and the spinal grey matter and its relationship to the sacral parasympathetic nucleus. J. comp. Neurol. 201,415-440. 36. Moscoso G. and Pexieder T. (1990) Variations in microscopic anatomy and ultrastructure of human embryonic hearts subjected to three different modes of fixation. Path. Res. Pract. 186, 768 774. 37. Okado N. and Kojima T. (1984) Ontogenesis of the central nervous system: neurogenesis, fibre connection. synaptogenesis and myelination in the spinal cord. In Continuity o f Neural Functions from Prenatal to Postnatal L(~" (ed. H. F. R. Prechtl), pp. 31-45. Spastics International Medical, London. 38. Otten U., Goedert M., Mayer N. and Lembeck F. (1980) Requirement of nerve growth factor for development of substance P-containing sensory neurons. Nature 287, 158 159. 39. Raivich G., Zimmermann A. and Sutter A. (1985) The spatial and temporal pattern of beta NGF receptor expression in the developing chick embryo. Eur. molec. Biol. Org. J. 4, 637-644. 40. Ram6n y Cajal S. (1911) Histologie du Systkme Nerveux de L'homme et des Vertkbrks. Maloine, Paris, 41. Rrthelyi M. and Szentagothai J. (1973) Distribution and connections of afferent fibres in the spinal cord. In Handbook o f Sensory Physiology, Vol. H. Somatosensory System (ed. A. Iggo), pp. 207-252. Springer, Berlin. 42. Richardson P. M., Verge Issa V. M. K. and Riopelle R. J. (1986) Distribution of neuronal receptors for nerve growth factor in the rat. J. Neurosci. 6, 2312 2321. 43. Ritter A. M., Lewis G. R., Kremer N. E. and Mendell L. M. (1991) Requirement for nerve growth factor in the development of myelinated nociceptors in vivo. Nature 350, 500 502. 44. Ross A. H., Grob P., Bothwell M., Elder D. E,, Ernst C. S., Marano N., Ghrist B. F. D., Slemp C. S., Herlyn M., Atkinson B. and Koprowski H. (1984) Characterization of nerve growth factor receptor in neural crest tumors using monoclonal antibodies. Proc. natn. Acad. Sci. U.S.A. 81, 6681-6685. 45. Scarpini E. (1988) Expression of nerve growth factor receptor during human peripheral nerve development. Devl Biol. 12A, 301 308.

482

A, M. SUaURO et a/.

46. Schatteman G. C., Gibbs L., Lanahan A. A., Claude P. and Bothwell M. (1988) Expression of NGF receptor in the developing and adult primate nervous system. J. Neurosci. 8, 860-873. 47. Scheibel M, E. and Scheibel A. B. (1968) Terminal axon patterns in cat spinal cord. II The dorsal horn. Brain Re.~. 9, 32-58. 48. Shu S., Ju G. and Fan L. (1988) The glucose oxidase-DAB-nickel method in peroxidase histochemistry of the nervous system. Neurosci. Lett. 85, 169-171. 49. Smith C. L. (1983) The development and postnatal organization of primary afferent projections to the rat thoracic spinal cord. J. comp. Neurol. 220, 29-43. 50. Sobue G., Yasuda T., Mitsuma T., Ross A. H. and Pleasure D. (1988) Expression of nerve growth factor receptor in human peripheral neuropathies. Ann. Neurol. 24, 64-72. 51. Squinto S. P., Stitt T. N., Aldrich T. H., Davis S., Bianco S., Radziejewski C., Glass D. J., Masiakowski P., Furth M. E., Valenzuela D. M., DiStefano P. S. and Yancopoulos G. D. (1991) trkB encodes a functional receptor for brain-derived neurotrophic factor and neurotrophin-3 but not nerve growth factor. Cell 65, 885-893. 52. Suburo A. M., Gibson S. J., Moscoso G., Terenghi G. and Polak J. M. (1992) Transient expression of neuropeptide Y and its C-flanking peptide in the spinal cord and ganglia of human embryos and foetuses. Neuroscience 46, 571-584. 53. Suburo A. M., Wheatley S. C., Horn D. A., Gibson S. J., Jahn R., Fischer-Colbrie R., Wood J. N., Latchman D. S. and Polak J. M. (1992) Intracellular redistribution of neuropeptides and secretory proteins during differentiation of neuronal cell lines. Neuroscience 46, 881-889. 54. Sutter A., Riopelle R. J., Harris-Warrick R. M. and Shooter E. M. (1979) Nerve growth factor receptors: characterization of two distinct classes of binding sites on chick embryo sensory ganglia cells. J. biol. Chem. 254, 5972-5982. 55. Taniuchi M., Schweitzer J. B. and Johnson E. M., Jr. (1986) Nerve growth factor receptor molecules in rat brain. Proc. natn. Acad. Sci. U.S.A. 83, 1950-1954. 56. Taniuchi M., Clark H. B., Schweitzer J. B. and Johnson E. M. Jr (1988) Expression of nerve growth factor receptors by Schwann cells of axotomized peripheral nerves: ultrastructural location, suppression by axonal contact and binding properties. J. Neurosci. 8, 664-681~ 57. Terenghi G., Bunker C. B., Liu Y.-F., Springall D. R., Cowen T., Dowd P. M. and Potak J. M. (1991) Image analysis quantification of peptide-immunoreactive nerves in skin of patients with Raynaud's phenomenon and systemic sclerosis. J. Pathol. 164, 245-252. 58. Windle W. F. and Fitzgerald J. E. (1937) Development of the spinal reflex mechanism in human embryos. J. eomp. Neurol. 67, 493-509. 59. Wong J. and Oblinger M. M. (1991) NGF rescues substance P expression but not neurofilament or tubulin gene expression in axotomized sensory neurons. J. Neurosci. 11, 543-552. 60. Wyatt S., Shooter E. M. and Davis A. M. (1990) Expression of the NGF receptor gene in sensory neurons and their cutaneous targets prior to and during innervation. Neuron 2, 421-427. 61. Yan Q. and Johnson E. M. Jr (1988) An immunohistochemical study of the nerve growth factor receptor in developing rats. J. Neurosci. g, 3481-3498. 62. Yan Q., Snider W. D., Pinzone J. J. and Johnson E. M. Jr (1988) Retrograde transport of nerve growth factor (NGF) in motoneurons of developing rats: assessment of potential neurotrophic effects. Neuron 1, 335--343. 63. Yip H. K., Rich K. M., Lampe P. A. and Johnson E. M. (1984) The effects of nerve growth factor and its antiserum on the postnatal development and survival injury of sensory neurons in rat dorsal root ganglia. J. Neurosci. 4, 298(~2992. (Accepted 29 April 1992)

Developmental pattern and distribution of nerve growth factor low-affinity receptor immunoreactivity in human spinal cord and dorsal root ganglia: comparison with synaptophysin, neurofilament and neuropeptide immunoreactivities.

Immunocytochemical expression of the low-affinity nerve growth factor receptor was studied in human fetal and adult tissues using the monoclonal antib...
NAN Sizes 0 Downloads 0 Views