Proc. Nati. Acad. Sci. USAVol. 86, pp. 5636-5640, July 1989Neurobiology

Adrenalectomy decreases nerve growth factor in young adultrat hippocampus

(hormonal regulation/corticosterone/chone acetyltransferase/nerve growth factor receptor)

LUIGI ALOEIstituto di Neurobiologia, Consiglio Nazionale della Ricerca, Viale Marx 15, 00156 Rome, Italy

Communicated by Rita Levi-Montalcini, April 3, 1989

ABSTRACT The effect of adrenalectomy on the level ofnerve growth factor (NGF) in the hippocampus and on thedistribution of choline acetyltransferase i ureactivity inforebrain cholinergic neurons of developing rats was studied.Biological and immunohistochemical determinations indicatedthat in 40-day-old rats, adrenalectomy reduced the NGF levelin the hippocampus and the choline acetyltrsferase immuno-reactivity in the septal lateral bands. Furthermore, autoradio-graphic studies showed that adrenalectomy causes changes inthe distribution and expression of NGF receptors in the hip-pocampus. These results suggest that adrenal hormones areinvolved in the regulation of the NGF level in the hippocampusand of NGF receptors in the septum.

Nerve growth factor (NGF) is a polypeptide found in both theperipheral (1) and the central (2) nervous system. In the ratcentral nervous system, the hippocampus contains one of thehighest levels of NGF and of mRNA encoding NGF (2-5).Several lines of evidence indicate that NGF is synthesizedwithin the hippocampus, taken up selectively by cholinergicneurons of the septal lateral band (SLB) terminals, andtransported retrogradely to the cell of origin (6-9). Intrahip-pocampal injection of NGF antibodies causes granule cellloss (10), further suggesting that NGF plays a physiologicalrole in the central nervous system. The hippocampus is amajor target site of glucocorticoid hormones, particularlycorticosterone, and administration ofglucocorticoids early indevelopment influences brain and somatic growth (11). Fur-thermore, recent in vitro and in vivo studies have shown thatglucocorticoids have a profound influence on the survival ofhippocampal neurons (12). Whether or not these hormonalactivities affect the level ofNGF in the hippocampus and thedistribution of NGF receptors in the basal forebrain cholin-ergic neurons has yet to be explored. The present studiesshow that in developing rats, adrenalectomy significantlyreduces the level of NGF in the hippocampus and of cholineacetyltransferase (ChAT) immunoreactivity in the choliner-gic neurons of the SLB while enhancing a concomitantincrease of NGF receptor expression in the septal neurons.A preliminary account of these studies has been presented.*

MATERIALS AND METHODSAnimals and Adrenalectomy. Forty-day-old Sprague-

Dawley rats (90-100 g) were purchased from Nossan (Calco,Italy). Rats were subjected either to bilateral adrenalectomyor to sham operation (laparotomy) under ether anesthesia.Corticosterone (Sigma) was dissolved in ethanol/saline, 1:9(vol/vol), and injected intradermally into the adrenalecto-mized (Adx) rats at doses of2 gg every other day for 2 weeks.

Adx rats were given 0.9% NaCl in place of drinking water tocompensate for the loss of salt.Mouse NGF (2.5S) was purified from male mouse sub-

maxillary glands (13). Polyclonal NGF antibodies were pre-pared (14, 15) and purified by affinity chromatography (16).

Biological Assays. At intervals, control and treated ratswere killed with an overdose of ether vapor and the hippo-campi were dissected out and stored at -80TC until use. Eachhippocampus was homogenized in distilled water (1 ml/100mg), the homogenate was centrifuged at 10,000 x g for 20min, and the supernatant was used for biological assay. Thesupernatant of Adx rat hippocampus was concentrated 10-fold (lyophilized and then dissolved in water to 1/10th of itsoriginal volume) to better evaluate the amount of NGF. ForNGF determination, a very sensitive cell culture bioassaywas used (17, 18). Since the hippocampus contains neuro-trophic factors that promote fiber outgrowth from NGF-unresponsive neurons of ciliar and sensory ganglia (19), ahighly NGF-responsive cell population was utilized. Nervecells from superior cervical ganglia (SCG) of 17- to 18-day-oldrat fetuses were mechanically dissociated and plated incollagen-coated 96-well Falcon tissue culture dishes contain-ing Dulbecco's modified Eagle's medium alone or with var-ious amounts of NGF or tissue samples to be tested. Aspreviously reported (17, 18), in the absence of NGF only asmall number of neurite-bearing cells appeared in the culturedish, while the addition of NGF (starting from 0.02 ng/ml)yielded a dose-related increase in the number of neurites.After 24-48 hr in culture, neurons bearing neurites ofa lengthmore than twice the diameter of the cell were scored. Thus,a standard NGF dose-response curve was obtained, permit-ting the evaluation of the NGF level in the tissue samplestested. To determine whether extracts of other brain regionscaused neurite outgrowth, cerebellum and spinal cord ex-tracts from the same rats were also tested.

Immunohistochemistry. For NGF immunohistochemistry,animals were killed and the brains were quickly removed andimmediately fixed by immersion in 2% paraformaldehyde/0.4% p-benzoquinone/50 mM phosphate buffer, pH 7.4, for6 hr at room temperature. After fixation, the brains werewashed and soaked in buffered 30o sucrose for 24 hr at 4TC.Thirty-micrometer coronal sections containing the hippo-campus were cut on a cryostat, and alternate sections wereprocessed for immunohistochemistry. Sections were incu-bated overnight with either affinity-purified polyclonal NGFantibodies, diluted 1:600-1:3000 (initial concentration, 20mg/ml), or monoclonal NGF antibodies (kindly provided byW. C. Mobley, University of California, San Francisco),diluted 1:30 (initial concentration 1.0 mg/ml), and then wereprocessed for immunoperoxidase staining (4, 20). Staining

Abbreviations: NGF, nerve growth factor; SLB, septal lateral band;Adx, adrenalectomized; ChAT, choline acetyltransferase; SCG,superior cervical ganglion (ganglia).*Aloe, L. & Levi-Montalcini, R., International Symposium onPerinatal Biology, April 14-16, 1988, Rome, p. 61 (abstr.).

5636

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

July

25,

202

1

Proc. Natl. Acad. Sci. USA 86 (1989) 5637

specificity was assessed by (i) preadsorption of the firstantibody with an excess ofNGF for 24 hr, (ii) omission of theprimary antibody, and (iii) incubation of sections with pre-immune rabbit serum or with commercially available anti-bodies to epidermal growth factor.For ChAT immunohistochemistry, animals were anesthe-

tized with Nembutal and perfused with 100 ml of salinefollowed by 300 ml of 4% paraformaldehyde in 0.1 M phos-phate buffer (pH 7.4). Brains were quickly removed andimmersed in the same fixative for 4 hr and then in phosphatebuffer containing 30%o sucrose for 24 hr at 40C. Serial coronalsections (40 ,um thick) were cut on a Bright cryostat. Thesections were incubated with monoclonal ChAT antibodies(Boehringer Mannheim), at a dilution of 1:25 in phosphate-buffered saline, for 24 hr at 40C and then processed with theperoxidase-antiperoxidase method (20, 21). A quantitativeanalysis of lightly and darkly stained immunoreactive nervecells in the SLB of Adx and sham-operated rats was carriedout on seven consecutive sections following a rostrocaudaldirection.NGF-Receptor Antibody lodination and Autoradiography.

Monoclonal antibodies (192-IgG; a generous gift from E. M.Johnson, Jr., Washington University, Saint Louis) wereradioiodinated with Na1251 (Amersham) by the chloramine-Tprocedure (22-24), and 1SI-labeled antibodies (125I-192-IgG)were purified by Sephadex G-25 column chromatography.Labeled ligands with a specific activity of 1.0-1.5 Ci/,umol (1Ci = 37 GBq) were obtained and used for retrograde transportexperiments. Control and Adx rats were injected with 4 jkl of125I-192-IgG (107 cpm) in the anterior hippocampus (coordi-nates: A, 2.6 mm; L, 2.2 mm; V, 3.0 mm). Twenty-four hourslater, animals were anesthetized and perfused through theheart with 4% paraformaldehyde in 0.1 M phosphate buffer(pH 7.4), brains were removed, and coronal sections were cut40 ,um thick. Sections containing the septum, correspondingto plates 13-17 of the rat brain atlas (25), were washed twiceand the 125I in each was measured with a y counter. Theywere then mounted on glass slides, air-dried, and coated withnuclear tracking emulsion (Ilford K-2). Slices were developed20 days later in Kodak developer D-19 and sections werecounterstained with toluidine blue. The total number ofradiolabeled neurons in the septum of control and Adx ratswere counted in seven consecutive sections following arostrocaudal direction. Statistical evaluations were per-formed by analysis of variance. Groups were compared bythe two-tailed Student t test.

RESULTSExtracts of rat hippocampus caused neurite outgrowth fromrat SCG neurons in the NGF bioassay. Adrenalectomy,performed on 40-day-old rats, caused a decrease of the NGFlevel in the hippocampus. This decrease was already appar-ent 5 days after the operation and reached its lowest levelafter about 12 days. By this time, the mean level of NGFdetected in the hippocampus of Adx rats was reduced to lessthan half the amount found in the hippocampus of sham-operated rats. The biological assay showed that the amountofNGF in hippocampus of sham-operated rats was about 1.7ng/g of hippocampal wet weight, whereas in Adx rats hip-pocampal NGF decreased to 0.8 ng/g (Table 1). The neurite-promoting activity of the hippocampal extract was dose-dependent and saturable, as has been shown with chicksensory ganglia (26, 27) and dissociated nerve cells (17, 18).Further, the biological activity of the extract was completelyinhibited by NGF antibodies. SCG neurons cultured in thepresence of cerebellum and spinal cord extracts did notproduce any neurite outgrowth.Adrenalectomy resulted in a pronounced reduction ofNGF

immunoreactivity in the hippocampal formation, particularly

Table 1. NGF in the hippocampus of control (sham-operated)and Adx rats

BodyGroup weight, gControl 143 + 10Adx 136 ± 12

Brainweight, g1.8 + 0.31.7 + 0.4

Hippocampusweight, mg

88 ± 789 ± 9

NGF,ng/g

1.7 ± 0.30.8 ± 0.2*

Hippocampi were removed 12 days after adrenalectomy or shamoperation, and NGF levels were measured by bioassay using SCGneurons. Values (ng/g of wet weight) are mean + SEM of four rats.*P < 0.001, significantly different from control.

in the CA2 and CA3 regions and in the dentate gyrus (Fig. 1).This reduction of immunostaining was evident after 1 weekand reached a maximum 3 weeks after adrenalectomy. Noimmunoreactive cells were observed when hippocampal sec-tions were incubated with preimmune serum, NGF antibod-

CA1

.# Dp-..

s..

Ws~~~~~~~~~~~~

CA2 '^*

CA3'

..v..*'...-

A

B

CFIG. 1. (A and B) Distribution of NGF-immunoreactive cells in

the hippocampus of control (A) and Adx (B) rats stained with theperoxidase-antiperoxidase technique. In A, a gradient of NGF im-munostaining can be observed within the field CA, with the highestlevel present in CA2, intermediate amounts in the CA3 areas, andlowest amounts in CAL. Note also the high level of NGF immuno-reactivity in the dentate gyrus (DG). (C) A section from the samespecimen shown in A, but exposed to purified antibodies pread-sorbed for 24 hr with an excess of NGF. (x 18.)

Neurobiology: Aloe

Dow

nloa

ded

by g

uest

on

July

25,

202

1

Proc. Natl. Acad. Sci. USA 86 (1989)

6 44 . v

. , s t0O*ie ,0,

t b

4 10.

4 4X c, ;

-Wi,1 I.8-4 "

*X

_w X

A

.4 %

*8i

G

ies preadsorbed with an excess ofNGF, or epidermal growthfactor antibodies.ChAT immunoreactivity and, to a lesser extent, the num-

ber of darkly immunostained cells in the SLB, decreasedfollowing adrenalectomy (Fig. 2). A time-course studyshowed that 14 days after adrenalectomy, the immunoreac-tivity in the SLB cholinergic neurons was significantly re-duced. Quantitative determination of immunostained neurondistribution showed a drastic decrease in ChAT immuno-reactivity rather than a reduction in the total number ofcholinergic neurons (Table 2). This decrease in ChATimmunoreactivity became even more pronounced after 3-4weeks. Thereafter, the rats began to lose weight and furtherimmunohistochemical differences were not considered. Al-ternate brain sections stained with toluidine blue did notshow any evident cytological alterations, such as chromatol-ysis, reduction of cell size, or cell damage in the septal regionof the Adx rats.

It has been shown (3, 6, 7, 9) that NGF synthesized in thehippocampus is internalized and transported retrogradely bycholinergic neurons located in the SLB. To determinewhether septohippocampal retrograde transport was affectedby adrenalectomy, radiolabeled monoclonal antibodies

FIG. 2. ChAT-immunoreactive neu-ronal perykaria in the septum of sham-operated (A) and Adx (B) rats. Note amarked decrease in immunoreactive cell

f̂D bodies in the SLB of Adx rats. (x190.)

against NGF receptor were injected into the hippocampusand, at various time intervals, the topographic distribution oflabeled cells in the SLB was studied. Consistent with previ-ous reports (8, 28, 29), monoclonal NGF-receptor antibodiesinjected into the neuronal terminal field reached the cell oforigin, located in the septum, within 20-24 hr. The accumu-lation of radiolabeled receptor antibodies in the septum ofAdx rats was significantly enhanced (Fig. 3). Histologicalexamination ofthe brains ofAdx and control rats showed that24 hr after injection of antibodies against NGF receptors, thenumber of labeled neurons in the septum of Adx rats wasgreater than the number of labeled neurons in the septum ofcontrol rats. Table 2 shows the morphometric determinationof retrogradely labeled cells in the septum of sham-operatedand Adx rats 2 weeks after operation. A significant increasein the number of radiolabeled cells occurred following adre-nalectomy (Fig. 4). The distribution of NGF receptors wasalso examined with the immunoperoxidase technique. Thesestudies showed that adrenalectomy enhanced the distributionof immunostained NGF receptors in the SLB (data notshown). Subcutaneous injection of corticosterone for 2weeks into Adx rats reversed the decrease of the NGF levelin the hippocampus and restored ChAT immunoreactivity

Table 2. ChAT immunoreactivity and NGF receptors in septum of control (sham-operated) andAdx rats

No. of ChAT-immunoreactive cells counted*

Group Total Light Dark Total no. of cells labeledtControl 1265 ± 213 36 ± 6 1228 ± 204 233 ± 38Adx 1133 ± 65t 913 ± 122§ 220 ± 39§ 469 ± 27§Data are mean ± SEM of five rats.

*For quantitative analysis of ChAT-immunoreactive cells, rats were killed 2 weeks after operation, andbrain tissue was processed for peroxidase-antiperoxidase immunohistochemistry. Immunostainedcells of the SLB were analyzed at x400 magnification in seven consecutive sections in a rostrocaudaldirection. Cells were classified as darkly stained (cell body and neurites stained dark brown) and lightlystained (cell body weakly stained and neurite profile undefined). See Fig. 2.

tFor detection of receptor-bearing cells, iodinated antibodies against NGF receptor were injected intothe hippocampus 2 weeks after operation. After 24 hr, rats were killed and brain tissues wereprocessed. All labeled SLB cells (showing autoradiographic grain density at least 4 times as high asbackground) in seven consecutive sections were counted.tP < 0.05 vs. control.§P < 0.001 vs. control.

..

.4

I

A.

4.

J~9

44

* 4"~4

'~I4ft )*

5638 Neurobiology: Aloe

rA.

lb

'

Dow

nloa

ded

by g

uest

on

July

25,

202

1

Proc. Natl. Acad. Sci. USA 86 (1989)

3000 --

2500-

CL

0-u4)(D

0)

N

2000-

1500-

1000-

500

0- I--40p

[F--

1 .1121-1(X) II! ,161-'0 III

;I7Septum thickness (rostro-Caudal direction)

FIG. 3. Histogram illustrating distribution of NGF receptor (as detected by 125I-labeled antibodies, 125I-192-IgG) in the rat septal area ina rostrocaudal direction. 125I-192-IgG (2 Al, 2.6 x 106 cpm) was injected into the hippocampus (coordinates: A, 2.6 mm; L, 2.2 mm; V, 3 mm)of control (sham-operated; dark bars) and Adx (light bars) rats. Twenty-four hours later, animals were perfused with 4% paraformaldehyde andbrains were removed. Coronal brain sections (40 1Lm thick) containing the septum were cut on a cryostat and coded, and radioactivity wasmeasured in a y counter. Each bar represents the mean ± SEM of four rats. Groups were compared by analysis of variance (P < 0.001). A,Atm.and the distribution of NGF receptors in the septum to thelevels observed in sham-operated rats.

DISCUSSIONThe hippocampus is a target tissue for a variety of steroidhormones (30). Glucocorticoids, secreted during stress, have

been shown to influence behavioral response and the struc-tural organization of the central nervous system in bothyoung and adult animals (11, 12, 31). Further, adrenocorticalhormones influence the rate of hippocampal neuronal deaththat occurs following neuropathological insults and in agingprocesses (32). Such studies, while providing experimentalevidence that the adrenocortical hormone receptor system in

.I

.*- t'

A. *T 4

w-i^ # a.

BS as ',,^s.

*.R A;.X,>s.sv.a*#W

mu:.* _;

t * W

,. *:'..tE

A.

as

A. *

A.* +

A

*01

FIG. 4. Photomicrographs illustratingthe topographical accumulation of NGF-receptor antibodies (125I-192-IgG) in SLBcells of sham-operated (A) and Adx (B)rats. Rats were anesthetized with nem-butal, placed in a stereotaxic apparatus,and injected in the hippocampus with 4 ulof iodinated antibodies (coordinates as inFig. 3). Eighteen hours later, rats wereperfused with 4% paraformaldehyde, andbrains were removed, sectioned (40 Am),and processed for autoradiographic stud-ies. Note the increase of labeled neuronsin the septum ofAdx rats as compared to

B the septum of sham-operated rats.t(x280.)

Neurobiology: Aloe 5639

T

i181-120 11

tI-.

I

Dow

nloa

ded

by g

uest

on

July

25,

202

1

Proc. Natl. Acad. Sci. USA 86 (1989)

the brain displays a considerable plasticity, have offered aconceptual and methodological approach for investigatingthe role of corticosteroids in regulating the level of NGF andother brain neurotrophic factors (2, 19) in the hippocampusand in different brain regions.

Several lines of published evidence indicate that NGF ispresent in the hippocampus, that it is retrogradely trans-ported to the SLB neurons (2, 6, 7, 29), and that injection ofexogenous NGF into the hippocampus enhances the activityof ChAT, a key enzyme for the synthesis of acetylcholine inthe forebrain neurons (6-8, 28, 29, 33). Impairment of theseptohippocampal pathway is involved in the pathophysiol-ogy of dementia of the Alzheimer type and perhaps ofnormalaging as well (34-38). The observations that the NGF level inthe hippocampus decreases with age (39) and that exogenousNGF can significantly reduce the degenerative events ofbasal cholinergic neurons (37, 44) support the notion thatNGF may be involved in some neurodegenerative diseasesinvolving central cholinergic neuron damage (2, 4, 8, 38, 40).However, how synthesis, storage, and release ofNGF in thecentral nervous system are modulated is largely unknown.Similarly, the correlation between the brain NGF level andcirculating adrenocortical hormones during brain develop-ment and in aging is not known. The present study shows thatadrenalectomy in young rats causes a drastic decrease of theNGF level in the hippocampus, a decrease in ChAT immuno-reactivity in septal neurons, and a concomitant change in thedistribution and immunological reactivity of NGF receptorsin the SLB neurons, as detected with the monoclonal anti-bodies used. Injection of corticosterone into Adx rats re-stores NGF levels, ChAT immunoreactivity, and NGF re-ceptor expression, suggesting that lack of adrenal hormonesis the primary cause of this deficit. These studies, showingthat Adx causes a decrease in the NGF level in the hippo-campus, indicate that the hippocampal NGF level and NGFreceptor expression are hormonally regulated.Recent reports that thyroid hormones modulate the distri-

bution of NGF receptors in the basal forebrain neurons (41,42), and an earlier report that administration of thyroxine toadult mice results in an increased amount of NGF in hypo-thalamic regions (43), support the hypothesis that brain NGFis hormonally regulated. The mechanism(s) of action ofNGFhormone regulation remains, however, a question open tofurther investigation. The observation that NGF receptorsand ChAT immunoreactivity in the septum are also affectedby corticosteroid suggests that this hormone exerts, eitherdirectly or indirectly, a well-defined modulatory action oncholinergic cells projecting to the hippocampus. These find-ings, therefore, suggest a functional interrelationship amongcorticosteroid secretion, brain NGF level, and the septohip-pocampal cholinergic pathways. Further studies are neces-sary, however, to determine the extent to which circulatingadrenocortical hormones and NGF synthesis and availabilityaffect other forebrain cholinergic pathways in adult and agedrodents. This specific experimental approach will be mostvaluable for studying the effects of circulating steroid hor-mone levels, particularly in animal models carrying hormone-related disorders, such as hypo- or hyperactivity of adreno-cortical hormones. It will likewise be of great value forstudying NGF distribution and NGF gene expression inspecific brain regions.

I thank Dr. Rita Levi-Montalcini for her suggestions and intellec-tual contributions during the course of this work. This project was

supported in part by a grant from the Consiglio Nazionale dellaRicerca to Dr. Levi-Montalcini.

1. Levi-Montalcini, R. (1987) Science 237, 1154-1162.2. Thoenen, H., Bandtlow, C. & Heumann, R. (1987) Physiol. Bio-

chem. Pharmacol. 109, 146-171.3. Shelton, D. L. & Reichardt, L. F. (1986) Proc. Nati. Acad. Sci.

USA 83, 2714-2718.4. Whittemore, S. R., Ebendal, T., Larkfors, L., Olson, L., Seiger,

A., Stromberg, I. & Persson, H. (1986) Proc. Nati. Acad. Sci. USA83, 817-821.

5. Ayer-LeLievre, C., Olson, L., Ebendal, T., Seiger, A. & Persson,H. (1988) Science 240, 1339-1341.

6. Seiler, M. & Schwab, M. E. (1984) Brain Res. 300, 33-49.7. Gnahn, H., Hefti, F., Heumann, R., Schwab, M. E. & Thoenen, H.

(1983) Dev. Brain Res. 3, 229-238.8. Hefti, F. (1986) J. Neurosci. 6, 2155-2162.9. Dawbarn, D., Allen, S. J. & Semenenko, F. M. (1988) Brain Res.

440, 185-189.10. Springer, J. E. & Loy, R. (1985) Brain Res. Bull. 261, 127-131.11. McEwen, B. S., De Kloet, E. R. & Rostene, W. (1986) Physiol.

Rev. 66, 1121-1188.12. Meaney; M. J., Aitken, D. H., Van Berkel, C., Bhatnagar, S. &

Sapolsky, R. M. (1988) Science 239, 766-768.13. Bocchini, G. & Angeletti, P. U. (1969) Proc. Natl. Acad. Sci. USA

64, 787-794.14. Aloe, L., Cozzari, C., Calissano, P. & Levi-Montalcini, R. (1981)

Nature (London) 291, 358-366.15. Aloe, L. & Levi-Montalcini, R. (1980) Exp. Cell Res. 125, 15-22.16. St6ckel, K., Gagnon, C., Guroff, G. & Thoenen, H. (1976) J.

Neurochem. 26, 1207-1211.17. Greene, L. A. (1974) Neurobiology 4, 286-292.18. Stephani, U., Sutter, A. & Zimmermann, A. (1987) J. Neurosci.

Res. 17, 25-35.19. Crutcher, K. A. & Collins, F. (1982) Science 217, 67-68.20. Sternberger, L. A., Hardy, P. H., Jr., Cuculis, J. J. & Meyer, H. G.

(1970) J. Histochem. Cytochem. 18, 315-333.21. Cuello, A. C., Priestley, J. V. & Sofroniew, M. V. (1983) J. Exp.

Physiol. 68, 545-578.22. Hunter, W. M. & Greenwood, F. C. (1962) Nature (London) 194,

495.23. Levi-Montalcini, R. & Aloe, L. (1985) Proc. NatI. Acad. Sci. USA

82, 7111-7115.24. Stbckel, K., Paravicini, U. & Thoenen, H. (1974) Brain Res. 76,

413-421.25. Palkovits, M. & Brownstein, M. J. (1988) Maps and Guide to

Microdissection of the Rat Brain (Elsevier, New York).26. Levi-Montalcini, R. & Booker, B. (1960) Proc. Natl. Acad. Sci.

USA 46, 373-384.27. Varon, S., Nomura, J., Perez-Polo, J. R. & Shooter, E. M. (1972)

Methods Neurochem. 3, 203-229.28. Meibach, R. & Siegel, A. (1977) Brain Res. 119, 1-20.29. Taniuchi, M., Schweitzer, J. B. & Johnson, E. M., Jr. (1986) Proc.

Natl. Acad. Sci. USA 83, 1950-1954.30. Stumpf, W. E. & Sar, M. (1978)'Am. Zool. 18, 435-445.31. Kaufman, H., Vadasz, C. & Lajtha, A. (1988) Brain Res. 453,

389-392.32. Landfield, P. W., Waymire, J. C. & Lynch, G. (1978) Science 202,

1098-1101.33. Mobley, W. C., Rutkowski, J. L., Tennekoon, G. I., Buchanan, K.

& Johnston, M. V. (1985) Science 229, 284-287.34. Fibiger, H. C. (1982) Brain Res. Rev. 4, 327-388.35. Bartus, R. T., Dean, R. L., III, Beer, B. & Lippa, A. S. (1982)

Science 217, 408-417.36. Coyle, J. T., Price, D. L. & DeLong, M. R. (1983) Science 219,

1184-1190.37. Hefty, F. & Weiner, W. J. (1986) Ann. Neurol. 20, 275-281.38. Lamour, Y., Dutar, P. & Jobert, A. (1987) Brain Res. 416, 277-282.39. Larkfors, L., Ebendal, T., Whittemore, S. R., Persson, H., Hoffer,

B. & Olson, L. (1987) Mol. Brain Res. 3, 55-60.40. Koh, S. & Loy, R. (1988) Brain Res. 440, 396-401.41. Patel, A. J., Hayashi, M. & Hunt, A. (1988) J. Neurochem. 50,

803-811.42. Kiss, J., Patel, A. J. & McGovern, J. (1988) Int. J. Dev. Neurosci.,

Suppl. 1, 6, 92 (abstr.).43. Walker, P., Weichsel, M. E., Fisher, D. A., Jr., Guo, S. M. &

Fisher, D. A. (1979) Science 204, 427-429.44. Fischer, W., Wictorin, K., Bjorklund, A., Williams, L. R., Varon,

S. & Gage, F. H. (1987) Nature (London) 329, 65-68.

5640 Neurobiology: Aloe

Dow

nloa

ded

by g

uest

on

July

25,

202

1