Age-related increase in soluble and cell surface-associated neurite-outgrowth factors from rat muscle

Brain Research, 509 (1990) 309-320

309

Elsevier BRES 15210

Age-related increase in soluble and cell surface-associated neurite-outgrowth factors from rat muscle Julie L. Rosenheimer and Dean O. Smith Department of Physiology, University of Wisconsin, Madison, WI 53706 (U.S.A.) (Accepted 18 July 1989)

Key words: Motoneuron; Denervation; Aging; Muscle; Neuromuscular junction

While the number of nerve terminals per endplate decreases with age in the rat extensor digitorum longus (EDL) muscle, the number of endplates exhibiting ultraterminal sprouting, characteristic of denervation, increases. To determine if these changes associated with aging are accompanied by alterations in the production of muscle-derived neurite-outgrowth factors, we examined the effects of soluble and cell surface-associated components from innervated and denervated 10- and 25-month rat EDL muscles on a motoneuron-enriched fraction of embryonic chick spinal cord cells in vitro. Cells were cultured for 72 h with muscle extract or on muscle cross-sections. While soluble components of the extract affected initiation of neurite outgrowth, muscle cell surface-associated molecules influenced neurite elongation. Both muscle extract and muscle cross-sections from 10-month denervated animals were more effective in promoting neurite outgrowth than 10-month innervated muscle. There was no difference between 25-month innervated and 25-month denervated muscle. However, 25-month innervated and denervated muscles were significantly more effective in promoting neurite outgrowth than t0-month innervated muscle, but not different from 10-month denervated muscle. These results suggest that an age-related increase in muscle-derived soluble and cell surface-associated neurite-outgrowth factors may contribute to denervation-like morphological changes associated with aging at the neuromuscular junction.

INTRODUCTION The n u m b e r of m o t o r nerve terminals p e r e n d p l a t e decreases with age in the rat extensor digitorum longus ( E D L ) muscle 43. Likewise, the n u m b e r of terminal sprouts within E D L endplates declines. This age-related reduction in terminal arborization is a c c o m p a n i e d by c o r r e s p o n d i n g functional changes such as decreases in n e u r o t r a n s m i t t e r synthesis and release ~3"s4. In contrast, there is a significant increase in the n u m b e r of endplates exhibiting ultraterminal sprouting b e t w e e n 10 and 25 month of age in this muscle a4. These are sprouts which e m a n a t e from the endplate and usually t e r m i n a t e on neighboring muscle fibers. A l t h o u g h ultraterminal sprouting is occasionally seen in normal muscles, these sprouts are characteristically observed following partial denervation 3'6'23'41 or paralysis 7'2~j. W h y the n u m b e r of p r e s u m a b l y functional nerve terminals located within the endplate region declines with age and the n u m b e r of ultraterminal sprouts increases is currently unknown. These age-related changes which are similar to changes o b s e r v e d following denervation may result from an imbalance in the trophic interactions b e t w e e n m o t o n e u r o n s and the muscles these neurons innervate.

Neurons require both soluble growth factors and a permissive substrate to e l a b o r a t e outgrowth. In this respect, the influence of skeletal muscle on m o t o n e u r o n s has been widely studied using in vivo and in vitro techniques (for reviews, see refs. 2, 8, 48, 53, 64). During d e v e l o p m e n t and following d e n e r v a t i o n , the muscle contains elevated levels of soluble protein factors ~6' ~9,s6.sv whose compositions have not yet been identified, and cell surface-associated molecules 9,15.17,29,38,47,59, such as laminin, a h e p a r a n sulfate proteoglycan, fibronectin and N - C A M , a neuronal cell adhesion molecule, which support survival and growth of neurons. Most likely, the production of basal levels of these c o m p o n e n t s is necessary for maintenance of e n d p l a t e architecture at times besides these most critical periods. To d e t e r m i n e w h e t h e r age-related morphological changes o b s e r v e d at the n e u r o m u s c u l a r junction are associated with alterations in the production of musclederived soluble or cell surface neurite-outgrowth factors, enriched fractions of e m b r y o n i c m o t o n e u r o n s were grown in vitro in the presence of muscle extracts or on muscle cross-sections p r e p a r e d from E D L muscles of adult and aged rats. T h e results show that aging is a c c o m p a n i e d by a significant increase in both soluble and cell surface-associated neurite-outgrowth factors.

Correspondence: J.L. Rosenheimer, Department of Physiology, University of Wisconsin, 130(I University Avenue, Madison, WI 53706, U.S.A. 0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

310 T’ABLE 1 EDL muscle wet weights und totalsoluble

muscleprotein

Values represent the mean (a S.E.M.) of 6 sets of measurements. Cuntraluteral refers to the muscles contralateral to the denervated muscles. Innervated refers to the muscles obtained from unoperated animals. Superscripts a and b indicate that the corresponding values are different at the 0.05 levels. Muscle wet weights (g)

Denervated Contralateral Innervated

Rad protein assay with bovmc: y-globulm srandarcla. Average vnlues shown in Table I. The extract was then dialyzed (Spectrapore I dialysis membrane, mol. wt. cutoff of h.llOO-X.000) for ll-16 h against 1000 vols. of extraction buffer minus the proteasc inhibitor cocktail, concentrated by lyophilization to 20 mg protein!ml, filter-sterilized, quick-frozen, stored at -HO “C and assayed within I week. Embryonic chick and neonatal rat muscle extracts were prepared in a similar manner, however muscles i’rom each animal were pooled. are

Totulsoluble protein (mg)

IO months

2.5 months

IO months

25 months

0.129+0.006 0.135+0.004 0.145+0.004

0.123f0.002”.h 0.137+0.001” 0.135f0.004h

6.0t0.5 6.5kO.7 7.7f0.H

6. I kO.5 6.3kO.5 6.4kO.S

Preparation

Of

cryostat-cut muscle section.\

Frozen EDL muscle pieces were mounted in Tissue-Tek (Miles Laboratories, Inc., Elkart, IN) which was frozen quickly in a dry ice powder and sections were cut in a cryostat. Five to 7 h-pm-thick muscle cross-sections were collected on UV-sterilized 15-mm glass coverslips. Muscle sections from either lo- or 25-month innervated or denervated muscles were collected on duplicate coverslips for each experiment. Cells were plated onto the coverslips within 1 h following sectioning. Fractionation of spinal cord cells

MATERIALS

AND METHODS

Tissue culture medium and trypsin were purchased from GIBCO Inc. (Grand Island, NY). Vitrogen 100 Collagen was obtained from Collagen Corporation, Palo Alto, CA. Tetramethylrhodamine isothiocyanate isomer R was purchased from BBL Microbiology Systems. All other chemicals were obtained from Sigma Chemical Company (St. Louis, MO) unless noted otherwise. White Leghorn chicks (Sunnyside Hatchery, Oregon, WI) were incubated at 37.5 “C in a humidified forced-draft incubator with an automatic rotator. Fischer 344 male rats were obtained from the National Institute on Aging contract colonies (Harlan SpragueDawley, Indianapolis, IN). Muscle denervation The sciatic nerve was exposed at the thigh and cut distal to the

sciafic notch, and a 2- to 3-mm section of nerve was removed from the right limbs of 6 lo- (adult) and 25-month (aged) rats while under chloral hydrate anesthesia (0.5 mgig b. wt.). Care was taken to avoid damaging nearby blood vessels, Animal behavior was observed periodically to ensure that dorsi- and plantar-flexion of the ankle was eliminated. Seven days following unilateral denervation, the animals were sacrificed along with age-matched unoperated rats. For experiments using muscle extract, EDL muscles from the denervated and the contralateral limbs of denervated animals and the limbs of unoperated (innervated) animals were dissected. freeze-clamped immediately in liquid nitrogen, weighed and stored at -80 “C until extracts were prepared. Average muscle wet weights are shown in Table I. In other experiments using muscle extract, 3to 4-day neonatal rat hind-limb or 12-day embryonic chick leg muscles were removed (without prior denervation) and treated in a similar manner. For experiments involving cryostat-cut muscle cross-sections, muscles were removed from the denervated limbs of 6 lo- and 25-month animals and from limbs of age-matched unoperated animals, cut into thirds, frozen in liquid nitrogen and stored at -80 “C until the muscles were sectioned. Preparation of skeletal muscle extract

Frozen EDL muscles were pulverized individually with a mortar and pestle cooled in liquid nitrogen and homogenized in 9 ~01s. (wt./voI.) of extraction buffer (137 mM NaCI, 5.3 mM KCI, 0.67 mM Na,HPO,, 0.22 mM KHzPO,, 15 mM HEPES and 0.03 mM Phenol red, pH 7.4) supplemented with a protease inhibitor cocktail (IOOpM phenylmethylsulfonyl fluoride, 100 KIU Aprotinin, 42 PM leupeptin, 100 ,uM benzamidine, 1 mM EDTA and 5 mM EGTA). and centrifuged for 1 h at 32,000 g. The supernatant was recentrifuged for 2 h at 100,000 g, and the total protein concentration was determined for the resulting soluble extract using the Bio

Lumbar spinal cords from 6.5-day (ref. 25; stage 28 in our hands) embryonic chicks were dissected sterilely, freed of their meninges and dorsal root ganglia and collected in ice-cold extraction buffer (see above) supplemented with 5.6 mM glucose, 100 U/ml penicillin and 100 yg/ml streptomycin and adjusted to 340 mOsm with NaCI. Cells were dissociated by incubation at 37 “C with 0.05% trypsin and 0.005% DNase I. Enzymatic dissociation was terminated by addition of soybean trypsin inhibitor (final concentration 0.04%) followed by gentle trituration. Motoneuron-enriched cell fractions were generated on the basis of their buoyant density using the technique of Schnaar and SchaffnerJ’. Briefly, dissociated cells were collected by centrifugation at 4 “C, resuspended in Dulbecco’s modified Eagle medium (DME; No. 430-2100 supplemented with 5.6 mM glucose, 100 U/ml penicillin, 100 pg/ml streptomycin, 1.5 mM HEPES, 0.185 g/l NaHCO,, pH 7.4, osmolarity adjusted to 340 mOsm), pipetted onto a sterile metrizamide solution with a density of 1.035 g/ml and centrifuged. This resulted in a distinct band of cells which could be collected. resuspended in DME, counted with a Coulter Counter or hemacytometer, and plated. Cells were examined visually using phase-contrast optics prior to plating. These observations indicated that motoneurons represented 12.5 ? 0.4% of the dissociated spinal cord cells. This yield is consistent with the yields of motoneurons obtained by others using this enrichment procedure “-” Following fractionation, they were enriched to 92.1 + 0.6% purity (cf. ref. 49). Exclusion of the vital dye, Trypan blue indicated that 95.5 + 0.4% of the cells were viable when plated. Retrograde lubeling of motoneurons

The effectiveness of this procedure for enrichment of motoneurons was verified by labeling motoneurons specifically prior to dissociation. In 3 experiments 6.0-day embryonic chicks were placed in oxygenated buffer, decapitated and eviscerated, and a ventral laminectomy was performed, Lumbosacral spinal nerves LS3 through LS6 were then cut bilaterally”“, and the fluorescent dye rhodamine (3.6 mg in 20~1 dimethylsuifoxide further diluted in I ml water) was carefully pressure-injected onto each cut stump. Any remaining dye was quickly rinsed away with fresh buffer. Embryos were then incubated in oxygenated buffer for 14-19 h at 26 “C, after which the labeled region of the spinal cord was removed and either fixed in 4% formaldehyde, frozen, sectioned in a cryostat and photographed, or dissociated, fractionated and plated onto collagen gels. Two to 3 h after plating, cells were fixed in 4% formaldehyde, photographed and analyzed. Cell culture Cells incubated

hydrated

with muscle extract. Cells were plated onto collagen gels in 16-mm tissue culture wells (Corning

311 24-well plates) at a density of 200 cells/mm2 (40,000 cells/well). They were maintained for 72 h in 1 ml (final volume) DME containing 1.85 g/l NaHCO3, and supplemented with 100 ktg/ml human transferrin, 5 /~g/ml insulin, 30 nM sodium selenite, 20 nM progesterone and 100/~M putrescine 5. In addition, bovine serum albumin (BSA) or extract from embryonic chick, neonatal rat, or either denervated, contralateral or other innervated 10- and 25-month rat muscles was added to duplicate wells approx. 2 h prior to addition of the cells. BSA or extract diluted with extraction buffer was added in constant (50 /A) volumes. Control wells contained extraction buffer only. Cell culture was performed in a humidified atmosphere of 9.5% CO 2, 90.5% air at 37.0 °C. After 72 h in culture, the cells were fixed in 2% formaldehyde diluted in extraction buffer. Two to 4 microscopic fields, representing approx. 125 cells per field, were then photographed at predetermined locations from each well. Examples are shown in Fig. 1. To quantify the effect of muscle extract on neurite outgrowth from the cells which remained attached to the collagen gel-coated wells, blind counts of cell and neurite numbers were made from the photographs. Neurites at least 1 cell diameter in length were counted and measured using a digital planimeter. Only those cells with neurite outgrowth were judged healthy at the time of fixation. During preliminary experiments, a number of the wells from each experiment were incubated briefly in 0.08% Trypan blue prior to fixation. Exclusion of the dye indicated that approx. 90% of the cells were viable. However, the majority of the cells not eliciting neurite outgrowth by this time appeared crenated or shrunken. Cells incubated on muscle cross-sections. Cells were plated at a density of 400 cells/ram 2 onto coverslips containing muscle sections which had been placed in 16-mm tissue culture wells. Cells were initially added to the wells in 200 ~1 of DME containing 10% heat-inactivated horse serum followed by addition of 800/~1 DME 1 h after plating. Cultures were incubated for 72 h in a humidified atmosphere containing 5.0% CO2, 95.0% air at 37 °C. For staining, cultures were fixed in 4% formaldehyde in 0.1 M phosphate buffer. Muscle endplates were stained following a 30-rain incubation in a bromoindoxylacetate dye-stain for acetylcholinesterase 4°. This resulted in a dark-blue reaction product at endplate regions. Cultures were then rinsed with Dulbecco's phosphate-buffered saline (PBS). Neurons were visualized after further incubations in rabbit anti-chick neurofilament IgG's which stain all 3 forms of the neurofilament protein (characterized and kindly provided by Dr. D. Dahl) followed by fluorescein isothiocyanate-conjugated goat anti-rabbit IgG (Polysciences, Inc., Warrington, PA). Antibodies were diluted in PBS containing 1.0% goat serum. Coverslips were then mounted in glycerol containing the antibleaching agent, p-phenylene-diamine 31. To quantify neurite outgrowth on muscle sections, coverslips were scanned using fluorescence optics. Only single neurons with neurites that did not contact other cells were selected for further evaluation. Likewise, only those neurons with processes extending between muscle fibers, as opposed to along blood vessel walls or nerve fiber bundles were included. Neurons adhering to the glass coverslip never extended processes (cf. ref. 15). Neurite lengths were measured using a reticle. The optics were then switched to bright field and the underlying muscle section was examined for endplate staining. Thirty neurons were measured per muscle type.

Analysis of muscle extract In some experiments, wells were preincubated overnight at 4 °C with muscle extract diluted in extraction buffer or at 37 °C with extract diluted in DME. Wells were rinsed 2x with DME prior to the addition of cells plated in DME plus defined supplements. In addition, the preincubated extract was transferred to other wells and incubated with cells to test its effectiveness in initiating neurite outgrowth. Ammonium sulfate precipitation. Crude muscle extract from 10-month denervated rats was fraetionated by ammonium sulfate precipitation at 4 °C. Saturated ammonium sulfate (5.7 M, adjusted

to pH 7.4) was added in a dropwise fashion to extract in 3 steps: 0-25%, 25-50% and 50-75% saturation. After each stepwise addition, samples were incubated for 1 h, and precipitated proteins were collected by a 30-rain centrifugation at 32,000 g. Pellets were resuspended in extraction buffer to 1/10th their original volume, dialyzed against this buffer for 16 h, concentrated by lyophilization, filter-sterilized and stored at -80 °C until analyzed. Electrophoresis and immunoblotting. Extract from 10-month denervated rats was examined following sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions 22. Samples were heated for 2 rain at 100 °C. Stacking and separating gels contained 5.0% and 9.5% acrylamide, respectively. Gels were stained with Coomassie Brilliant Blue. For immunoblotting (Western blots), samples were electrophoretically transferred to Immobilon P (Millipore Corp., Bedford, MA) according to the method of Towbin et al. 62 and probed with rabbit anti-mouse laminin IgG (kindly provided by Dr. H. Kleinman) followed by alkaline phosphatase-conjugated goat anti-rabbit IgG (Bio-Rad Laboratories, Richmond, CA). Bound antibody was detected by exposure to nitroblue tetrazolium (NBT) and 5-bromo-4-chloro3-indolyl phosphate (BCIP). Immunoprecipitation. Muscle extract (400 ~g protein plus protease inhibitor cocktail, see above) was incubated with antilaminin IgG (800 /~g) for 16 h at 4 °C with periodic agitation. This concentration of antibody was approximately 25x the amount found to be in excess of that necessary to totally block 400 fLg of purificd laminin. Laminin was immunoprecipitated by circulating the extract-antibody mixture over a protein A-sepharose column. Extract was then collected, dialyzed against extraction buffer, and assayed for its effect on neurite outgrowth or immunoblotted.

Statistics Data represent averages + S.E.M. Duplicate wells from 4 experiments using BSA, embryonic chick or neonatal rat muscle extract and 6 experiments using extract or muscle cross-sections derived from 10- and 25-month rats were analyzed. Student's 2-sidcd t-tests were performed to compare means. RESULTS

Enrichment of motoneurons Motoneurons

enriched on the basis of their buoyant

density could be identified by their comparatively large size r e l a t i v e to o t h e r s p i n a l c o r d cell p o p u l a t i o n s at this stage

o f t h e i r d e v e l o p m e n t ~7'49. T h e

motoneuron-en-

r i c h e d cell f r a c t i o n c o u l d b e f u r t h e r i d e n t i f i e d by its unique

growth

characteristics.

In p r e l i m i n a r y

experi-

m e n t s , s p i n a l c o r d cells w e r e s e p a r a t e d i n t o 3 f r a c t i o n s corresponding

to m e t r i z a m i d e

1.050 g/ml a n d

d e n s i t i e s o f 1.035 g/ml,

1.070 g/m149 a n d

c o l l a g e n gels in D M E

grown on hydrated

supplemented

with

10% horse

s e r u m , B y 72 h in v i t r o , f e w e r t h a n 1 0 % of t h e cells in the motoneuron-enriched

f r a c t i o n e x h i b i t e d n e u r i t e out-

growth unless the medium was supplemented with muscle e x t r a c t . In c o n t r a s t , n o n - m o t o n e u r o n

and non-neuronal

cell f r a c t i o n s c o u l d b e m a i n t a i n e d in v i t r o for u p to 3 w e e k s in t h e a b s e n c e o f m u s c l e e x t r a c t (cf. refs. 17, 49, 57). R e t r o g r a d e l a b e l i n g e x p e r i m e n t s s h o w e d t h a t in s p i n a l c o r d c r o s s - s e c t i o n s , t h e d y e a c c u m u l a t e d in t h e v e n t r a l born,

composed

of m o t o n e u r o n

cell b o d i e s ,

and the

dorsal funiculus, containing dorsal root ganglion fibers,

3t2

Fig. 1. Motoneuron-enriched cell fractions from embryonic chick spinal cords plated on hydrated collagen gels. Cells were cultured for 72 h in defined medium (A) without or (B) with muscle extract coincubation (25-month denervated muscle extract). Neurite outgrowth (B; arrows) is only observed following coincubation of the cells with muscle extract. Calibration bar indicates I(X)Itm. as seen in Fig. 2A and B (cf. ref. 60), Of the dissociated cells, only those cell bodies identified as m o t o n e u r o n s by their comparatively large size were stained. Furthermore, the e n r i c h m e n t procedure resulted in a 5- to 7-fold increase in strongly fluorescing cells, as shown in Fig.

2 C - F (cf. ref. 17). This was in agreement with the 7-fold enrichment calculated from phase-contrast observations.

Effects of muscle extract Comparison of embryonic chick and neonatal rat

313

muscle extract. Initial experiments using defined medium

were designed to compare the effects on embryonic chick motoneurons of extracts from embryonic chick leg

g

o

muscles with extracts from neonatal rat hind-limb muscles. The percentage of cells exhibiting neurite outgrowth

®

O

D

-----

F

Fig. 2. Phase-contrast and corresponding fluorescence photomicrographs of labeled motoneurons following retrograde transport of rhodamine injected into the spinal nerves. A and B: cross-sections of a lumbar spinal cord. The dye is localized in cell bodies of the ventral horn (*) and dorsal root ganglion fibers of the dorsal funiculus ('*). C-F: examples of dissociated spinal cord ceils before (C and D) and following (E and F) enrichment of the motoneurons. Cells were photographed 3 h after plating. Arrows in C and E indicate cells containing the fluoresccnt dye rhodamine, as seen in D and F. Calibration bar indicates 100/~m for A and B and 50 ~m for C-F.

314 embryonic

\,

l

14 "iZ

///l

x

c h T c h:

A

20 10

[

i

.

months confralaferal

i

"" '., /

""

"

....

10 u~ % c2

ra,

neonatal

/ // '"" ""t

,,.

& 5

09 ¢9

•

/

&

0

/ •

/J

0

50

1O0

150

200

B 2o

25 months

BSA ------~ / / ~ - -

I

I

I

i

0 10

I

I I I I

50 protein

i

1 O0

I

i

500

Fig. 3. The effect of embryonic chick (O) and neonatal rat (El) extracts or BSA (x) on the percentage of cells exhibiting neurite outgrowth from motoneuron-enriehed fractions of embryonic chick spinal cord cells after 72 h in vitro. Values represent the mean (_+ S.E.M.) of duplicate wells from 4 experiments. increased and then declined in a dose-dependent fashion when incubated with either muscle extract, as shown in Fig. 3 (cf, refs. 17, 29, 65). In addition, rat muscle extract was significantly m o r e potent than chick muscle extract in promoting outgrowth from chick motoneurons (cf. ref. A ,o 10 r

n

o

n

f

h

s

~

'

comfralateral

(oJg/well)

concentration

7...... . . ~ ......................... _~

i

10 denervafed

,!

-£

0 protein

50 1 O0 concentra,ion

150 (Jig/well)

200

Fig. 5. The effect of muscle extract on the percentage of cells exhibiting neurite outgrowth after 72 h in vitro. Extract was derived from denervated (&) and contralateral (E]) limbs of unilaterally denervated rats and from unoperated (innervated) (0)limbs of (A) 10- and (B) 25-month rats. Values represent the mean (-+ S.E.M.) of duplicate wells from 6 experiments.

35

Innorvate~

30

denerv~

,~-'"

d:'

,

,"""""" "

'

~. e5 contraloferal 2O

15

0 25

50

1O0

150

200

months

35

x

&

Effects of denervation and aging on neurotrophic activity of muscle extract. To examine the effects of

30

=

Innervated 25

20 15

18), although the maximal response to both rat and chick extracts was not different statistically, Therefore, it was apparent that under these defined conditions embryonic chick m o t o n e u r o n s could respond not only to chick muscle extract but also to extract from rat muscle (cf. refs. 18, 55). As a control, B S A was also incubated with cells to ensure that the neurons were not simply responding to an increase in the protein concentration of the medium. As shown in Fig. 3, the B S A alone had no effect on neurite outgrowth (cf. ref: 56).

0 protein

50

1O0

concentration

150 200 (~lg/well)

Fig. 4. The effect of EDL muscle extract on the number of cells per well adherent to the collagen gel after 72 h in vitro. Forty thousand ceils were originally plated per well. Extract was derived from denervated (A) and contralateral (F1) limbs of unilaterally denervated rats and from unoperated (innervated) (0) limbs of (A) 10and (B) 25-month rats. Values represent the mean (+ S.E.M.) of duplicate wells from 6 experiments.

denervation and aging on the production of musclederived soluble factors, extracts prepared from 10- and 25-month rat denervated, contralateral and unoperated (innervated) E D L muscles were incubated with chick motoneurons. The mean number of cells per well which remained adherent to the collagen gel increased in a dose-dependent manner for all extracts, as seen in Fig. 4. The percentage of these cells possessing neurites was also affected by muscle extract, as shown in Fig. 5. In all cases, the number of cells with neurite outgrowth increased from 0 in the absence of extract to saturating values at protein concentrations between 125 and 200/xg protein/well.

315 In agreement with previous studies (cf. refs. 28, 30, 37, 52, 56), extract from 10-month denervated muscles induced more neurite outgrowth than extract from age-matched innervated tissue (Fig. 5A). This difference was significant and cannot be attributed to denervationinduced muscle atrophy because total extract protein concentrations before lyophilization were not different statistically (Table I). In contrast to the difference between the 10-month innervated and denervated muscle extract, there was no difference between 25-month innervated and denervated muscle extract (Fig. 5B). However, there was a significant difference between 10-month innervated and 25-month innervated muscle extract, as shown in Fig. 6A. This increase in neurite outgrowth in the presence of aged muscle extract cannot be attributed to significant differences in total soluble muscle protein (Table I). Following muscle denervation, however, there was no difference between 10- and 25-month muscle extract (Fig. 6B). Surprisingly, extracts from limbs muscles contralateral to the denervated limbs had an effect on neurite outgrowth similar to that from extract from the denervated muscles (cf. refs. 28, 30, 36). This result is consistent with the observation of transneuronal commu-

A

nication following unilateral denervation 45. However, the possibility that denervation results in a general systemic effect also exists. Although the soluble muscle extract influenced the initiation of neurite outgrowth, the extract did not affect neurite number or length. Of the cells exhibiting neurite outgrowth, neither the average number of neurites per cell (2.10 + 0.03) nor the mean neurite length (49.4 + 1.2 /~m) was affected by the source or the concentration of extract. The activity could be precipitated from the extract using 25-50% saturated ammonium sulfate. This resulted in a 4-fold increase in extract potency. The extracellular matrix protein, laminin, known to stimulate neurite outgrowth, precipitates within this window of ammonium sulfate saturation 1°'33. In addition, it has been found in substantial amounts in soluble muscle extract 17. Therefore, samples of the precipitate were examined following SDS-PAGE under reducing conditions and subsequent Western blot analysis with antilaminin IgG. As illustrated in Fig. 7, 2 bands from the muscle extract (lanes 2 and 4, a and b) comigrated with those of a laminin standard (lane la and b). However, immunoprecipitation of the laminin from the extract with antilaminin IgG, (lane 3), had no significant effect on the neurite-promoting activity of the extract.

2o in n e t - v o t e d

l

2

3

4

15 = 10 k c

a

& 5

0

B

50

100

150

200

b

2o

15 .~

denerv(~tecl

l /

t

25 months

O

0 protein

50 1 O0 concentration

150 200 (.ug/well)

Fig. 6, The effect of muscle extract on the percentage of cells exhibiting neurite outgrowth after 72 h in vitro. Extract was derived from 10- ( A ) and 25-month (O) innervated (A) and denervated (B) muscles. Data arc from Fig. 5 and were replotted to facilitate comparisons, Values represent the mean (4_ S.E.M.) of duplicate wells from 6 experiments.

Fig. 7. Detection of laminin in the muscle extract by Western blot analysis. Samples of (1) a laminin standard (200 ng) and (2 through 4) the protein fraction (10 keg) which precipitated from crude muscle extract between 25% and 50% saturated ammonium sulfate were subjected to SDS-PAGE under reducing conditions, electrophoretically transferred to Immobilon P, and probed with antilaminin lgG (see Materials and Methods). Lane 2 contains extract that was subjected to electrophoresis immediately upon being thawed. Lane 3 contains extract from which laminin had been previously immunoprecipitated (see Materials and Methods). Lane 4 contains extract that had been treated identically to that of the sample in lane 3, excluding the antilaminin lgG coincubation step. (a) and (b) indicate the laminin subunits with apparent mol. wts. of 440,000 and 200,000, respectively.

316

~i!:!il

i

. . . .

Fig. 8. Neurite outgrowth from motoneuron-enriched cell fractions on cryostat-cut muscle cross-sections. Cultures were photographed simultaneously with fluorescence and bright-field illumination to enable visualization of both nerve and muscle. Neur,ms were grown on 10-month denervated muscle in A, B, C and E, 10-month innervated muscle in D, E and G, and 25-month innervated muscle in H and I. A r r o w h e a d s in H mark the location of acetylcholinesterase-stained muscle endplates identified with bright-field illumination in l. Calibration bar indicates 50/~m.

317 125-

J

115~

. 3~

105

95

c

85

N 75

65

/ 1 O--inn

1 O--den

25--~nn

25--den

Fig. 9, The average neurite length from motoneuron-enriched cell fractions grown on 10- or 25-month innervated (inn) or denervated (den) cryostat-cut muscle sections. Values represent the mean (+ S.E.M.) neurite length from 30 neurons for each of 6 experiments. Neurite outgrowth on 10-month innervated muscle sections was significantly less (0.005 level) than outgrowth on sections from 10-month denervated or 25-month innervated or denervated muscles. The factor appears to be a soluble molecule since neurite outgrowth required the presence of muscle extract with the cells. Preincubation of wells with extract was not sufficient to induce neurite outgrowth, indicating that the activity could not be preadsorbed to the culture substratum. Likewise, extract preincubation did not diminish the activity of the extract when it was later incubated with cells in other wells.

Effects of denervation and aging on neurite outgrowth along muscle cell surfaces. To examine the effect of denervation and aging on muscle membrane- or matrixassociated molecules, motoneuron cell fractions were plated onto cryostat-cut muscle cross-sections from 10and 25-month innervated and denervated E D L muscles. The majority of neurons adhering to the muscle sections after 72 h in vitro appeared to attach to muscle fiber membranes or extracellular matrix, collectively referred to as the muscle cell surface, as opposed to the interior (sarcoplasm) of the muscle. Examples are shown in Fig. 8. Therefore, despite a relatively high initial plating density, actual cell attachment was sparse and neurons were usually located 2 - 4 muscle fibers apart. Neuron processes generally followed the contours of the muscle cell surfaces, although neurite outgrowth did occassionally extend across the interior of a muscle fiber (Fig. 8). These observations are similar to those reported by Covault et al.:5 Neurite outgrowth on 10-month denervated muscle sections was significantly (0.005 level) greater than on 10-month innervated tissue; average neurite lengths were 114.4 + 6.8 u m and 76.9 +__ 4.6 ~m for 10-month denervated and innervated muscle, respectively, as shown in Fig. 9 (cf. ref. 15). In contrast, the average

neurite length on 25-month denervated muscle sections (118.5 4-_ 3.3 ~m) was not different from 25-month innervated muscles (106.1 _ 6.6/~m). In addition, the response to 25-month muscle was significantly (0.02 level) greater than 10-month innervated muscle, but was essentially the same as 10-month denervated tissue. The average number of neurites per cell (1.82 + 0.04) was not affected by muscle type, however. Approximately 50% of all muscle cross-sections were classified as either endplate-poor (few-to-no endplates per muscle section) or endplate-rich (numerous-to-many endplates per section), by visual observation of acetylcholinesterase staining (see Fig. 8). Neither the average neurite number nor the length was affected by the abundance or paucity of synaptic regions. This differs from the observations of others using diaphragm muscle ~. The discrepancy may be explained by the fact that extreme examples of endplate-rich vs endplate-poor regions can be more easily obtained from the fiat diaphragm muscle than from the cylindrical E D L muscle. Furthermore, of 21 examples in which neurons were located within 100 ktm or less of endplates, 86% failed to extend processes toward those endplates. This suggests that a gradient of soluble factors as well as cell surface molecules may be necessary to direct axons to synaptic sites (cf. refs. 17, 48). DISCUSSION These results show that both soluble and cell surfaceassociated factors from aged rat skeletal muscles are more effective in promoting neurite outgrowth from embryonic motoneurons in vitro than factors from younger muscle. Moreover, the effects of aged muscle are similar to those following muscle denervation of younger animals. These observations in vitro suggest that the age-related reduction in the number of nerve terminals within E D L endplates is probably not due to a decline in the production of muscle-derived neurite-promoting factors, since these factors appear to increase with age. The decrease in nerve terminal number may instead be a consequence of alterations in the synaptic input to the motoneuron or changes in the relationship between the motoneuron and the accompanying Schwann cell. The increase in ultraterminal sprouting seen in the E D L muscle during aging is, however, consistent with elevated levels of muscle-derived growth-promoting factors. These sprouts are characteristically associated with partial denervation 6'23'4~. It has been postulated that they are induced to grow from endplates by a signal released from denervated muscle fibers. In aged animals ultraterminal sprouts may grow in response to increased levels of

318 muscle-derived diffusible factors and a more permissive substrate, which may result from progressive disuse or partial denervation. The failure of denervation to significantly increase the effectiveness of aged muscle supports this idea. Similar results have been obtained in studies of the central nervous system. For example, Needels et al. 36 examined the effects of aging on neurite-promoting activity prior to and following brain injury. These investigators found that the activity in unoperated tissue from aged animals was significantly greater than that from young adult tissue. Likewise, an enhancement of activity following denervation was only seen in the tissue from the younger animals. The precise composition of the soluble extract is not known, The factor initiating neurite outgrowth was effective as a diffusible molecule since cells had to be cultured in the presence of muscle extract to elicit outgrowth. Other laboratories have obtained results similar to these 19"32'36'57. In this respect the soluble factor resembles nerve growth factor (NGF) 11'35 A major difference between the effects of the soluble and the cell surface-associated factors was that although the muscle extract could initiate neurite outgrowth, it had no effect on neurite length (cf. refs. 14, 55, 61). Thus, the main effect of the soluble extract may be to influence cell survival. In contrast, the cell surface-associated molecules primarily affected neurite length. This latter result confirms the work of Covault et al. 1~ Laminin, which also stimulates neurite outgrowth in vitro 9"17'26'33'34"42, was present in the muscle extract. However, the laminin did not appear to contribute to the observed outgrowth since immunoprecipitation of laminin had no effect (cf. refs. 27, 36, 58). This may be due to the overriding influence of the collagen gel matrix on which the cells were grown. Laminin, in conjunction with other attachment factors, such as a heparan sulfate proteoglycan, fibronectin and N-CAM, is most likely associated with the outgrowth observed on the muscle cross-sections ~5 (cf. ref. 4). In these experiments, 2 relatively novel in vitro techniques were used to distinguish between the influence of soluble and cell surface-associated muscle factors on neurite outgrowth from motoneurons. The advantages of both of these techniques have been well documented 9' 12.15.17.21,39,46,49,5o However, the limitations of drawing quantitative conclusions from these experiments to interpret in vivo phenomena are also apparent. For example, REFERENCES 1 Barde, Y.-A., Edgar, D. and Thoenen, H., Sensory neurons in culture: changing requirements for survival factors during embryonic development, Proc. Natl. Acad. Sci. U.S.A.. 77

only a relatively small percentage of the cells in either group of experiments exhibited neurite outgrowth (cf. ref. 17). In addition, the behavior of embryonic cells was used to explore the potential effects of skeletal muscle factors on motoneurons, although the requirements for motoneurons may be quite diverse at different ages (cf. refs. 1, 24). Furthermore, when using extracts, it must be assumed that the factors triggering neurite outgrowth are those same factors released by intact muscle cells. It is also important for the interpretation of these experiments that the rate of secretion follows the rate of synthesis. Organ culture experiments currently being performed in this laboratory confirm that an outgrowth factor is indeed released from the intact muscles (unpublished observations). A further caveat in the context of these experiments is that it is impossible to determine whether the same components are responsible for eliciting neurite outgrowth following denervation and during aging. While denervation may trigger the up-regulation of neuriteoutgrowth factors, aging may be accompanied by the suppression of inhibitory activity. This appears to be involved in the enhanced sprouting observed in Alzheimer's disease brains 63. A plausible hypothesis consistent with the present results is that in vivo, aging may be accompanied by a gradual decline in the frequency of synaptic input onto motoneurons. This could cause a subsequent reduction in nerve terminal number, thereby affecting communication between motoneurons and their associated muscle fibers. In turn, the muscle fibers may respond as if they were partially denervated. One response might be the release of greater amounts of a diffusible neurotrophic factor. Another consequence might be the development of a more favorable terrain beyond the boundary of the endplate along which the axon could advance. It has been postulated that a diffusible factor could spread approximately 200/~m from its source 51'52 (cf. ref. 7). Endplates within this distance might then extend ultraterminal sprouts towards the source of the diffusible factor. A potential outcome would be the reestablishment of complete innervation of the muscle, although in the aged muscle, this may never occur. Acknowledgements. This study was supported by NIH Grants AG01572 and AG05340. Technical assistance provided by Kristine Bork, Lisa Trotter and Yan Ping Xu, helpful discussions with Drs. Mark Emmerling and Kevin Williams, and critical perusal of the manuscript by Dr. Katherine Kalil are greatly appreciated.

(1980) 1199-1203. 2 Berg, D.K., New neuronal growth factors, Annu. Rev. Neurosci., 7 (1984) 149-170. 3 Betz, W.J., Caldwell, J.H. and Ribchester, R.R., Sprouting of active nerve terminals in partially inactive muscles of the rat, J.

319

Physiol. (Lond.), 303 (1980) 281-297. 4 Bixby, J.L., Pratt, R.S., Lilien, J. and Reichardt, L.E, Neurite outgrowth on muscle cell surfaces involves extracellular matrix receptors as well as Ca2+-dependent and -independent cell adhesion molecules, Proc. Natl. Acad. Sci. U.S.A., 84 (1987) 2555-2559. 5 Bottenstein, J.E., Culture methods for growth of neuronal cell lines in defined media. In D.W. Barnes, D.A. Sirbasku and G.H. Sato (Eds.), Cell Culture Methods for Molecular and Cell Biology, Methods for Serum-free Culture of Neuronal and Lymphoid Cells, Vol. 4, Alan R. Liss, New York, 1984, pp. 3-13. 6 Brown, M.C. and Ironton, R., Sprouting and regression of neuromuscular synapses in partially denervated mammalian muscles, J. Physiol. (Lond.), 278 (1978) 325-348. 7 Brown, M.C., Holland, R.L., Hopkins, W.G. and Keynes, R.J., An assessment of the spread of the signal for terminal sprouting within and between muscles, Brain Research, 210 (1980) 145151. 8 Brown, M.C., Holland, R.L. and Hopkins, W.G., Motor nerve sprouting, Annu. Rev. Neurosci., 4 (1981) 17-42. 9 Calof, A.L. and Reichardt, L.F., Motoneurons purified by cell sorting respond to two distinct activities in myotube-conditioned medium, Dev. Biol., 106 (1984) 194-210. 10 Calof, A.L. and Reichardt, L.F., Response of purified chick motoneurons to myotube-conditioned medium: laminin is essential for the substratum-binding, neurite outgrowth-promoting activity, Neurosci. Lett., 59 (1985) 183-189. 11 Campenot, R.B., Local control of neurite development by nerve growth factor, Proc. Natl. Acad. Sci. U.S.A., 74 (1977) 4516-4519. 12 Carbonetto, S.D., Evans, D. and Cochard, P., Nerve fiber growth in culture on tissue substrata from central and peripheral nervous systems, J. Neurosci., 7 (1987) 610-620, 13 Chapman, M.R. and Smith, D.O., Differential kinetics of acetylcholine synthesis and choline uptake in functionally different muscles of mature adult and aged rats, Soc. Neurosci. Abstr., 12 (1986) 1581. 14 Collins, F. and Dawson, A., Conditioned medium increases the rate of neurite elongation: separation of this activity from the substratum-bound inducer of neurite outgrowth, J. Neurosci., 2 (1982) 1005-1010. 15 Covault, J., Cunningham, J.M. and Sanes, J.R., Neurite outgrowth on cryostat sections of innervated and denervated skeletal muscle, J. Cell Biol., 105 (1987) 2479-2488. 16 Doherty, P., Dickson, J.G., Flanigan, T.P. and Walsh, F.S., Human skeletal muscle cells synthesise a neurotrophic factor reactive with spinal neurons, J. Neurochem.. 46 (1986) 133-139. 17 Dohrmann, U., Edgar, D., Sendtner, M. and Thoenen, H., Muscle-derived factors that support survival and promote fiber outgrowth from embryonic chick motor neurons in culture, Dev. Biol., 118 (1986) 209-221. 18 Dribin, L.B., On the species and substrate specificity of conditioned medium enhancement of neuritic outgrowth from spinal cord explants, Dev. Brain Res., 3 (1982) 300-304. 19 Dribin, L.B. and Barrett, J.N., Two components of conditioned medium increase neuritic outgrowth from rat spinal cord explants, J. Neurosci. Res., 8 (1982) 271-280. 20 Duchen, L.W., Changes in motor innervation and cholinesterase localization induced by botulinum toxin in skeletal muscle of the muscle: differences between fast and slow muscles, J. Neurol. Neurosurg. Psychiatry, 33 (1970)40-54. 21 Flanigan, T.P., Dickson, J.G. and Walsh, F.S., Cell survival characteristics and choline acetyltransferase activity in motor neurone-enriched cultures from chick embryo spinal cord, J. Neuroehem., 45 (1985) 1323-1326. 22 Giulian, G.G., Moss, R.L. and Greaser, M., Improved methodology for analysis and quantitation of proteins on onedimensional silver-stained slab gels, Anal. Biochem., 129 (1983) 277-287.

23 Gurney, M.E., Suppression of sprouting at the neuromuscular junction by immune sera, Nature (Lond.), 307 (1984) 546-548. 24 Gurney, M.E., Heinrich, S.P., Lee, M.R. and Yin, H., Molecular cloning and expression of neuroleukin, a neurotrophic factor for spinal and sensory neurons, Science. 234 (1986) 566-574. 25 Hamburger, V. and Hamilton, H.L., A series of normal stages in the development of the chick embryo, J. Morphol., 88 (1951) 49-92. 26 Hantaz-Ambroise, D., Vigny, M. and Koenig, J., Heparan sulfate proteoglycan and laminin mediate two different types of neurite outgrowth, J. Neurosci., 7 (1987) 2293-2304. 27 Hayashi, Y. and Miki, N., Purification and characterization of a neurite outgrowth factor from chicken gizzard smooth muscle, J. Biol. Chem., 260 (1985) 14269-14278. 28 Henderson, C.E., Huchet, M. and Changeux, J.-P., Denervation increases a neurite-promoting activity in extracts of skeletal muscle, Nature (Lond.), 302 (1983) 609-611. 29 Henderson, C.E., Huchet, M. and Changeux, J.-P., Neuritepromoting activities for embryonic spinal neurons and their developmental changes in the chick, Dev. Biol., 11)4 (1984) 336-347. 31) Hill, M.A. and Bennett, M.R., Motoneurone survival activity in extracts of denervated muscle reduced by prior stimulation of the muscle, Dev, Brain Res., 24 (1986) 305-308. 31 Johnson, G.D. and Araujo, G.M. de C,N., A simple method of reducing the fading of immunofluorescence during microscopy, J. lmmunol. Methods, 43 (1981) 349-350. 32 Kligman, D., Isolation of a protein from bovine brain which promotes neurite extension from chick embryo cerebral cortex neurons in defined medium, Brain Research, 250 (1982) 93-100. 33 Lander, A.D., Fujii, D.K. and Reichardt, L.F., Purification of a factor that promotes neurite outgrowth: isolation of laminin and associated molecules, J. Cell Biol., 101 (1985) 898-913. 34 Lander, A.D., Fujii, D.K. and Reichardt, L.E, Laminin is associated with the 'neurite outgrowth-promoting factors' found in conditioned media, Proc. Natl. Acad. Sci. U.S.A.. 82 (1985) 2183-2187. 35 Levi-Montalcini, R. and Angeletti, P.U., Ncrvc growth factor, Physiol. Rev., 48 (1968) 534-569. 36 Needels, D.L., Nieto-Sampedro, M. and Cotman, C.W., Induction of a neurite-promoting factor in rat brain following injury or deafferentation, Neuroscience, 18 (1986) 517-526. 37 Nurcombe, V., Hill, M.A., Eagleson, K.L. and Bennett, M.R., Motor neuron survival and neuritic extension from spinal cord explants induced by factors released from dencrvated muscle, Brain Research, 291 (1984) 19-28. 38 Nurcombe, V., Tout, S. and Bennett. M.R., Motoneurone survival and outgrowth promoted by different cell types in embryonic muscle, Dev. Brain Res., 21 (1985) 49-611. 39 O'Brien, R.J. and Fischbach, G.D., Isolation of embryonic chick motoneurons and their survival in vitro, J. Neurosci., 6 (1986) 3265-3274, 40 Pestronk, A. and Drachman D.B., A new stain for quantitative measurement of sprouting at neuromuscular junctions, Muscle and Nerve, 1 (1978)7(I-74. 41 Pockett, S. and Slack, J.R.. Source of the stimulus for nerve terminal sprouting in partially denervated muscle. Neuroscience, 7 (1982) 3173-3176. 42 Rogers, S.L., Letourneau, P.C., Palm, S.L., McCarthy, J. and Furcht, L.T., Neuritc extension by peripheral and central nervous system neurons in response to substratum-bound fibronectin and laminin, Dev. Biol., 98 (1983) 212-221/. 43 Rosenheimer, J.L. and Smith, D.O., Differential changes in the end-plate architecture of functionally diverse muscles during aging, J. Neurophysiol., 53 (1985) 1567-1581. 44 Rosenheimer, J.L., Trotter, L.A. and Smith, D.O., Age-related changes in ultraterminal sprouting within different muscles, Soc. Neurosci. Abstr., 14 (1988) 1171. 45 Rotshenker, S. and Tal, M.. The transneuronal induction of

320

46

47

48 49

50

51

52

53 54

55

56

sprouting and synapse formation in intact mouse muscles, J. Physiol. (Lond.), 360 (1985) 387-396. Sandroek, A.W. and Matthew, W.D., Identification of a peripheral nerve neurite growth-promoting activity by development and use of an in vitro bioassay, Proc. Natl. Acad. Sci. U.S.A.. 84 (1987) 6934-6938. Sanes, J.R., Schachner, M. and Covault, J., Expression of several adhesive macromoleeules (N-CAM, L1, J1, NILE, uvomorulin, laminin, fibronectin, and a heparan sulfate proteoglyean) in embryonic, adult, and denervated adult skeletal muscle, J. Cell Biol., 102 (1986) 420-431. Sanes, J.R., Extracellular matrix molecules that influence neural development, Annu, Rev. Neurosci., 12 (1989) 491-516. Sehnaar, R.L. and Schaffner, A.E., Separation of cell types from embryonic chicken and rat spinal cord: characterization of motoneuron-enriched fractions, J. Neurosci., l (1981) 204-217. Schwab, M.E. and Thoenen, H., Dissociated neurons regenerate into sciatic but not optic nerve explants in culture irrespective of neurotrophie factors, J. Neurosci., 5 (1985) 2415-2423. Slack, J.R. and Pockett, S., Terminal sprouting of motoneurones is a local response to a local stimulus, Brain Research, 217 (1981) 368-374. Slack, J.R. and Pockett, S., Motor neurotrophic factor in denervated adult skeletal muscle, Brain Research, 247 (1982) 138-140. Slack, J.R., Hopkins, W.G. and Pockett, S., Evidence for a motor nerve growth factor, Muscle and Nerve, 6 (1983) 243-252. Smith D.O., Muscle-specific decrease in presynaptic calcium dependence and clearance during neuromuscular transmission in aged rats, J. Neurophysiol., 59 (1988) 1069-1082. Smith, R.G. and Appel, S.H., Extracts of skeletal muscle increase neurite outgrowth and cholinergic activity of fetal rat spinal motor neurons, Science, 219 (1983) 1079-1081. Smith, R.G., McManaman, J. and Appel, S.H., Trophic effects of skeletal muscle extracts on ventral spinal cord neurons in

vitro: separation of a protein with morphologic activity from proteins with cholinergic activity, J Cell Biol., 101 (1985J 1608-1621. 57 Smith, R.G., Vaca, K., McManaman, J. and Appcl, S . H . Selective effects of skeletal muscle extract fractions on motoneuron development in vitro, J. Neurosci., 6 (1986) 439-447. 58 Steele, J.O. and Hoffman, H., Neurite-promoting activity from fetal skeletal muscle: partial purification of a high-molecularweight form, J. Neurosci. Res., 15 (1986) 323-339. 59 Steele, J.G. and Dalton, B.A., Neurite-promoting activity from fetal skeletal muscle: immunological comparison with laminin, J. Neurosci. Res., 17 (1987) 119-127. 60 Tanaka, H., Chronic application of curare does not increase the level of motoneuron survival-promoting activity in limb muscle extracts during the naturally occurring motoneuron cell death period, Dev. Biol., 124 (1987) 347-357. 61 Thompson, J.M. and Thompson, E.G., Developmental changes in spinal cord neurite-promoting activity from chick muscle extracts, Dev. Brain Res., 40 (1988) 158-160. 62 Towbin, H.T., Staehelin, T. and Gordon, J., Electrophoretic transfer of proteins from polyaerylamide gels to nitrocellulose sheets: procedure and some applications, Proc. Natl. Acad. Sci. U.S.A., 76 (1979) 4350-4354. 63 Uchida, Y. and Tomonaga, M., Neurotrophic action of Alzheimer's disease brain extract is due to the loss of inhibitory factors for survival and neurite formation of cerebral cortical neurons, Brain Research, 481 (1989) 190-193. 64 Varon, S. and Manthorpe, M., Trophic and neurite-promoting factors for cholinergic neurons. In I.B. Black (Ed.), CeUularand Molecular Biology of Neuronal Development, Plenum Press, New York, 1983, pp. 251-275. 65 Whittemore, S.R., Nieto-Sampedro, M., Needels, D.L. and Cotman, C.W., Neurotrophic factors for mammalian brain neurons: injury induction in neonatal, adult and aged rat brain, Dev. Brain Res., 20 (1985) 169-178.