Neurotrophic stimulation of fetal rat retinal explant neurite outgrowth and cell survival: Age-dependent relationships

DevelopmentalBrain Research, 18 (1985) 251-263

251

Elsevier BRD 50166

Neurotrophic Stimulation of Fetal Rat Retinal Explant Neurite Outgrowth and Cell Survival: Age-Dependent Relationships JAMES E. TURNER

Departmentof Anatomy, Bowman Gray School of Medicine of Wake Forest University, Winston-Salem, NC27103 (U.S.A.) (Accepted September llth, 1984)

Key words: CNS neurotrophic factor - - retinal explants - - neurite outgrowth - - cell survival

Serum-free tissue culture conditions have been defined where stimulation of neurite outgrowth from fetal rat retinal explants occurred only in the presence of an active fraction (BE) prepared from a pig brain extract purification procedure. Under these conditions, 18-20-day fetal retinal explants survived and continued to extend long radial neurites for at least 3 weeks in the presence of BE. However, if fibronectin was not equilibrated onto the basic collagen/poly-L-lysine substrate the neurite outgrowth was restricted to a short halo about the circumference of the explant. In addition, a dose-response relationship was demonstrated in the presence of increasing concentrations of BE with respect to the neurite growth index. The half-maximal response for BE was estimated to be between 5 and 10/~g/ml. In addition a number of important age-dependent relationships were observed with respect to BE stimulation of retinal neurite outgrowth and cell survival. An inverse relationship was demonstrated between increased developmental age and responsiveness to BE. After 1 week in culture, there was a 3-fold reduction in retinal neurite length measured from the 2-day neonatal explant when compared to that of the 18-day fetus. There was also a significant inverse relationship demonstrated between the length of time before BE was added to the culture medium and the ability of 20-day fetal explants to extend neurites onto the culture substrate. If BE was added as late as 2 weeks after initial explant culture, the various neurite outgrowth indices were significantly lower than in those situations where BE was added at the time of culture or 1 week later. These results imply that BE not only is required for stimulating neurite outgrowth from fetal rat retinal explants, but may be important in survival and maturation of developing retinal neurons. This hypothesis was confirmed when morphometric analysis was performed on 16- and 20-day explants cultured for a week in the presence or absence of BE. The number of necrotic cells in the developing retinal ganglion plexiform-cell layer of 20-day fetal explants was significantly lower when treated with BE. Conversely, the density of identifiable differentiating retinal ganglion-like cells was significantly greater in response to BE treatment in both 16- and 20-day retinal explants.

INTRODUCTION

vary with sequential stages of development2,11,12,21, 33. The successful regeneration of axonal connections in

It is believed that developing nerve cells are dependent upon their local e n v i r o n m e n t for survival

mature nervous systems may also rely partly upon similar neurotrophic activitiesSA3,1s,30,31,35.

and maturation 33. This r e q u i r e m e n t for neurotrophic

Until recently the only purified neurotrophic fac-

factors has been well established by both in vivo and

tor isolated for study has been the nerve growth fac-

in vitro studies 33. In addition, a large n u m b e r of abla-

tor ( N G F ) which is required by sympathetic and sen-

tion and transplantation experiments have demon-

sory neurons during their d e v e l o p m e n t 27. Purifica-

strated the epigenetic influence of target tissues on

tion of N G F and d e v e l o p m e n t of its antibody per-

the survival of their respective innervating neuronal

mitted a characterization of its structural proteins, its

populationsll, 15. Results from tissue extracts, conditioned media, and co-culture experiments have indi-

mechanism of action and physiological relevance. H o w e v e r , the actions of N G F are limited to the mam-

cated that different neuronal populations require

malian peripheral nervous system (PNS) 27.

specific factors, and that these r e q u i r e m e n t s may

Many tissue extracts and conditioned media have

Correspondence: J. E. Turner, Department of Anatomy, Bowman Gray School of Medicine of Wake Forest University, Winston-Salem, NC 27103, U.S.A. 0165-3806/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)

252 been used in attempts to isolate additional neurotrophic factors from different neuronal populations2,3, 9,21. Although many of these sources have been partially purified, no additional factors have been isolated for characterization especially from the central nervous system (CNS). However, a recent report has described the purification to homogeneity of a new neurotrophic factor (PF) from the mammalian brain 3. The purification of this factor was stimulated by previous results which had indicated that gliomaconditioned medium 4 and a rat brain extract 2 could support the survival of embryonic sensory neurons cultured in the absence of NGF. The biological activity of the newly isolated protein is similar to that previously reported for both gliomal-conditioned medium and the brain extract. The new growth factor is functionally and immunologically distinct from NGF, and it is a basic protein with a molecular weight of 12,300 and a PI > 10.13. Results also suggest that this protein from the mammalian brain may be produced by glial cells 2,4. It is now possible to study the effects and role in the CNS of this new neurotrophic factor and the intermediate active fractions derived from the purification process which has not been possible before. Subsequently, we have reported that a CM-cellulose column fraction (BE) derived from a pig brain extract during the purification procedure stimulated neurite outgrowth from fetal rat retinal explants 29. The BE effects were dose dependent and could not be altered by NGF or its antiserum. In a more recent abstract we reported further modification of the fetal rat retinal explant culture and substrate environment 33. In the present paper we will describe in greater detail the serum-free culture conditions in conjunction with a fibronectin substrate which allow for maximal, long radial neurite outgrowth in the presence of BE. The establishment of these new culture conditions now make it possible to explore more thoroughly without serum interference, the trophic effects of active fractions derived from the new purification process in both explant and dissociated cell environments. The present paper also reports a number of important age-dependent relationships observed with respect to BE stimulation of fetal rat retinal neurite outgrowth and cell survival in explant culture.

MATERIALS AND METHODS

Dissection and culture of explants Rat fetuses (16-, 18- and 20-day) were removed from ether anesthetized pregnant females (Wistar strain) and placed in sterile phosphate-buffered saline (PBS). Neonates (2 day) were decapitated in preparation for eye dissection. Eyes were dissected, placed in PBS, opened and retinas were removed using watchmakers forceps under a dissecting microscope. At this stage of development, retinas can be cleanly stripped away from the surrounding tissue and the developing lens. It is also possible to remove the vascular net which begins to form on the vitreal surface of the retina and around the lens. Dissected retinas were collected in PBS and cut into small pieces (approximately 400-600 #m squares) using forceps. Culture dishes were coated first with a mixture of collagen (50 mg/ml, calf skin, Sigma) and poly-L-lysine (1 mg/ml, Sigma ) which was allowed to dry at room temperature incubation overnight after two sterile water rinses. The substrate was modified further by incubation with human fibronectin I (50 ~g/ml in culture medium, Biomedical Technologies) for 2-4 h in 5% CO2/95% air at 37 °C in an incubator followed by two media washes before explant culture. The culture media consisted of: (1) Eagles basal medium (GIBCO) with 26.4 mM NaHCO3, 33.3 mM glucose and 2.0 mM L-glutamine; (2) N l supplements6: insulin (5 ~g/ml), transferrin (5 ~g/ml), progesterone (2 × 10-s M), selenium (3 × 10-s M) and putrescine (1 × 10-3 M); (3) BSA (100/tg/ml); and (4) antibiotics; penicillin (10,000 U/ml) and streptomycin (10 mg/ml). Just prior to explantation the excess medium was aspirated from the dishes. The explants were then placed on the dish with a Pasteur pipette. One drop of medium was placed over the explants which were separated and positioned (5-10 explants/dish) by means of glass manipulators. Explants were allowed to equilibrate in a minimum amount of medium (0.2-0.3 ml/dish) overnight which allowed them to attach, after which 1 ml of medium was added. Only 0.2-0.3 mi of the medium was exchanged every 3-4 days.

253

Evaluation of neurite outgrowth Neurite outgrowth was measured in a manner previously described 30 by determining a neurite growth index (NGI) score which takes into consideration two aspects of neurite outgrowth, fiber density and fiber length. The NGI was defined as the product of the neurite density score, based on a scale of 0-4, and the length score. The length score was measured by an ocular micrometer (one unit equals 10~m using a 10 x objective and a 12.5 x ocular) where a length of 200/~m was assigned a score of 1,400gm a score of 2, etc. The length score for each explant was an average of four measurements, one from each side of the explant. The neurite density score was determined from a standard atlas of representative explants demonstrating various degrees of neurite outgrowth. Triplicate cultures were established for each assay point with a total of approximately 10-20 explants analyzed per point. For consistency, the longest fiber outgrowth from each side of the explant was measured. NGI scores were analyzed statistically by the Student's t-test and analysis of variance.

Preparation of pig brain extractfraction (BE) The pig brain extract fraction (BE) was a gift from Dr. Yves Barde and was prepared according to the procedure reported by Barde et al. 3. In comparison to the starting homogenate, the BE had 100 times higher specific activity when tested for survival activity on 20-day-old chick embryo sensory neurons. The BE preparation is equivalent to Step III of the purification procedure for PF 3.

Morphometric analysis of explants Control and BE (25 pg/ml) treated explants (16 and 20 fetal days old) were fixed after 1 week in culture in 3% glutaraldehyde prepared in 0.2 M cacodylate buffer, postflxed in 1% osmium, dehydrated through a graded ethanol series, infiltrated with Epon, removed to flat-embedding molds and the Epon polymerized at 60 °C for 3 days. Thick sections (1 pm) taken from the center of each expiant were mounted on slides, stained with toluidine blue and cover-slipped. Photomicrographs were taken with a Zeiss Universal microscope and used to measure explant cross-sectional area, neuroepithelial area, plexiform-cell layer area, intact retinal ganglion celllike density within the plexiform-cell layer and ne-

crotic cell density within the plexiform cell layer. Quantitation of the explant response to BE was accomplished by means of a Ladd Graphic digitizer interfaced to a Monroe 1830 Calculator. This analytical system allows one to follow specific structures with a cursor with resulting X and Y coordinates being continuously fed into the programmed calculator by the digitizer. The high accuracy allows for excellent evalution of the structure under investigation. Before processing with the digitizer some structures were first outlined to facilitate quantitation. The data were grouped for statistical analysis according to the experimental treatment and the significance of difference was tested with the Student's t-test. RESULTS

Modified culture conditions Fetal rat retinal explants extend neurites onto the culture substrate under defined culture conditions (i.e. no serum) only if BE is present in the medium (Fig. 1A). BE stimulated neurite outgrowth was greatly enhanced if the collagen + poly-L-lysine substrate was modified further by fibronectin (Fig. 1C). There was approximately a 5-fold increase in the NGIs of BE treated (50 #g/ml) explants with fibronectin substrate modification (Fig. 2). However, fibronectin had no stimulatory effect on explant neurite outgrowth in the absence of BE treatment (Fig. 2). The dramatic BE stimulated increase in the NGIs of 18-day fetal explants was reflected mainly in the appearance of long radial fibers in the presence of fibronectin (compare Fig. 1A and C), whose length exceeded 1 mm in most cases after one week in culture. In contrast, without fibronectin a halo of short (< 0.5 mm) fibers formed about the explant periphery. Substrates consisting of collagen, collagen + poly-L-lysine, poly-L-lysine or polyornithine could not substitute for the flbronectin-mediated long radial fiber outgrowth. However, explants cultured on all these other substrates could be induced in the presence of BE to extend a short halo of fibers like that seen on the collagen + poly-L-lysine dishes (Fig. 1A). In addition, fibronectin coated on only the plastic culture dish surface was just as effective as coating on collagen + poly-L-lysine. Regardless of fibronectin substrate modification, retinal explants treated with BE (50/~g/ml) were vi-

254

Fig. 1. Phase-contrast micrographs of 18-day fetal rat retinal explants without fibronectin substrate modification after 7 days in culture. Mag. 160 x. A: BE (50/~g/ml)-treated explant showing short 'halo' fiber outgrowth characteristics. Note cell migration from explant edges (arrowheads). B: control without BE treatment demonstrating very little fiber outgrowth and no cell migration from explant edges. C: BE (50 ~g/ml)-treated explant showing long radial fiber outgrowth from the edges of the short 'halo' of fibers in contrast to the previous figure. Also note as in Fig. 1, a cell migration from the explant edges (arrowheads). D: control without BE treatment demonstrating no fiber outgrowth and no cell migration from explant edges. able under defined m e d i a conditions for at least 3 weeks in culture, using fiber length and density as criteria for explant survival. M o r e specifically, 20-day fetal explant neurite outgrowth increased significantly from 630 ~ m by the end of week one to 933 ktm by the third week (Table I). Similarly, without fibronectin t r e a t m e n t fiber outgrowth increased from 250 ktm at one week to 460 ktm by the third week of culture (Table I). Fibronectin alone had no effect on explant neurite outgrowth and controls exhibited no appreciable neurite outgrowth during this 3-week period. A d o s e - r e s p o n s e relationship was d e m o n s t r a t e d in the presence of increasing concentrations of B E (Fig. 3). B E t r e a t m e n t increased the N G I s of retinal explants over a range of 0 - 2 5 ktg/ml, p l a t e a u e d be-

tween 25-100 ktg/ml and exhibited a half-maximal response between 5 and 10ktg/ml (Fig. 3).

Age-dependent relationships There was an inverse relationship d e m o n s t r a t e d in vitro between increased retinal d e v e l o p m e n t a l age and responsiveness to B E using either fiber length, fiber density or N G I scores as criteria. The difference between the responses elicited from explants of the 18-day fetus, 20-day fetus and 2-day newborn are striking when m e a s u r e d after one week in culture (Table I). F o r example, B E - m e d i a t e d fiber length decreased significantly from 1106 ktm to 625 ktm between 18- and 20-day fetal explants and still further to 425 ~ m in the case of 2-day newborns (Table II).

255 17

(20)

24 22

15

Treatment K%_%-%_%_%_. (50 IJglml •BE )

2C

I

14

I Controls

13

le

z

12

x

11 ,o

-

14

l,o

o~

9

~

8

z

6

(

8

3

6

,~16)

0

4

S 1

1

I

5 10

I

25

1

I

50 100

BE Concentration ( jug / ml )

2 •

C -

I

Z

Fibronectin

7"

I

+'Fibronectin

Fig. 2. Graph demonstrating how substrate modification with fibronectin (50/zg/ml) significantly enhances the neurite outgrowth promoting ability of BE (50/xg/ml) from 18-day fetal rat retinal explants after 7 days in culture. Note that there is essentially no outgrowth from control explants without the presence of BE. Vertical bars represent S.E.M.

Fig. 3. A dose-response curve demonstrating the BE-induced neurite outgrowth expressed as a neurite growth index (NGI) in 18-day fetal rat retinal exlants after seven days in culture. Vertical lines represent the S.E.M. All points are different significantly (P < 0.001) from the control (i.e. no BE in the medium). However, the three highest concentrations are not different significantly from each other (P > 0.05). The numbers in parentheses indicate the total number of explants evaluated.

D e n s i t y scores do not c h a n g e in the fetal e x p l a n t response to B E ; h o w e v e r , n e w b o r n retinas s h o w a dram a t i c d e c r e a s e f r o m 4.0 to 2.6. C o r r e s p o n d i n g l y , the

TABLE I

Viability of BE (50 ~g/ml)-treated 20-day fetal retinal explants as measured by fiber outgrowth

N G I s significantly d e c r e a s e f r o m 22 to 13 and finally to 6.0 with i n c r e a s e d d e v e l o p m e n t a l age in r e s p o n s e to B E t r e a t m e n t ( T a b l e II).

Weeks in culture

Fiber outgrowth measurement Fibronectin presenta Fiber length (/xm)

1 2 3 a b c d e f

Density Fiber length score (l~m)

630 + 86c,d (10) e 4 + 0 836+31 f (10) 4 + 0 933+77 (10) 4 + 0

T h e a d d i t i o n of B E to the culture m e d i u m at p r o -

Fibronectin absent

gressively later times p r o d u c e d s o m e i n t e r e s t i n g re-

Density score

sults. M o r e specifically, t h e r e w e r e no significant dif-

250 + 0.01g (26) 3 + 0.2 360+0.01 h(26) 4 + 0 460+0.01 (26) 4 + 0

(50/~g/ml) was a d d e d as late as o n e w e e k after initial

Fibronectin concentration was 50/~g/ml. Density score is based on a scale from 0-4 with 4 maximum. Mean value + S.E.M. Value significantly lower than that at 2 weeks (P < 0.001). Total number of explants evaluated. Value not significantly different from that at 3 weeks (P > 0.05). g Value significantly different from that at 2 weeks (P > 0.001). h Value significantly different from that at 3 weeks (P > 0.001).

f e r e n c e s in fiber length, density score o r N G I s if B E e x p l a n t a t i o n w h e n c o m p a r e d to explants t r e a t e d at the t i m e of culture ( T a b l e III). H o w e v e r , if t h e r e was a t w o - w e e k p e r i o d b e t w e e n e x p l a n t a t i o n and B E t r e a t m e n t , all the various scores w e r e significantly l o w e r w h e n m e a s u r e d a w e e k later ( T a b l e III). R e t i n a l e x p l a n t s c u l t u r e d at 1 6 - 2 0 fetal days in the p r e s e n c e of B E (25/~g/ml) for o n e o r m o r e w e e k s exhibited a b u n d a n t l o n g radial fiber o u t g r o w t h as well as a d e n s e h a l o of fibers a b o u t the p e r i p h e r y (Fig. 1C). In contrast, a g e - m a t c h e d c o n t r o l s s h o w e d

256 TABLE II The relationship o f rat retinal development to the in vitro neurite growth promoting effects o f BE (50 l~g/ml) after 7 days in culture Developmental

Neurite outgrowth measurement

age

Fiber length (l~m)

18-day fetus 20-day fetus 2-daynewborn

1106 + 76c,a (16)e 4 _+0 629 + 85 (10) 4+0 429 + 88 (10) 2.6 + 0.4 f

Density score~

NGI b

22 + 1.5b 13 + 1.7 6 + 2.0

a Based on a scale of 0-4 with 4 being the maximum. b NGI = neurite growth index. c Meanvalue + S.E.M. All values in this column differ significantly from one another (P < 0.001). e Indicates total number of explants evaluated. f This value is significantly lower (P < 0.001) than others in this column. g All values in this column differ significantly from one another (P < 0.001). no fiber outgrowth (Fig. 1D). In addition, both 16and 20-day fetal retinal explants d e m o n s t r a t e d a similar histological laminar organization consisting of a basal (i.e. with respect to the culture substrate) layer occupied p r e d o m i n a n t l y by undifferentiated neuroepithelial cells and an outer plexiform-cell layer consisting of ganglion-like neurons (Figs. 4 and 5). G a n glion-like neurons could be clearly differentiated from neuroepithelial cells by their larger d i a m e t e r e d , oval, pale staining nuclei with 1 or 2 p r o m i n e n t nucleoli (Fig. 6). In contrast, neuroepithelial cells had smaller, m o r e eliptical, darkly staining, heterochroTABLE III Ability of 20-day fetal retinal explants to extend neurites in response to BE treatment at progressively later time periods in culture Time in culturea when BE added (weeks)

Fiber outgrowth measurement Fiber length (~m)

Fiber density N G I b score

0 1 2

629 + 86c (7)d 639 + 33 (14) 430 + 54e (10)

4.0 + 0 4.0 + 0 3.3 + 0.2e

a

b c d e

12.6 + 1.7 12.9 + 0.7 7.0 + 0.8e

BE (50/~g/ml) added at time indicated and measurements made 7 days later. NGI = neurite growth index. Meanvalue + S.E.M. Indicates total number of exlants evaluated. These values are significantly lower (P < 0.001) than the two previous measurements in each of their respective columns. In turn the first two values in each column are not significantly different from one another (P > 0.05).

matic nuclei with no prominent nucleoli (Fig. 6). Explants cultured at 16 fetal days in the presence of BE for one week d e m o n s t r a t e d no significant difference in their total cross-sectional areas when comp a r e d to controls; however, the plexiform-cell areas were significantly increased and the neuroepithelial areas significantly reduced over controls (Table IV). Accordingly, both the plexiform-cell and neuroepithelial layers in BE-treated explants demonstrated a significantly higher density of differentiating ganglion-like neurons (Table IV). M o r e specifically, control explants had 68% of their total cross-sectional areas occupied by the neuroepithelial layer with very few ganglion-like neurons present (i.e. 359 cells/mm2). In contrast, B E - t r e a t e d explants showed only 53% of the explant area occupied by the neuroepithelial layer and the ganglion-like neuron density was 4-fold greater than controls at 1448 cells/ram 2. In addition, the control plexiform-cell layer occupied only 32 + 2% of the total explant area with a density of 4120 cells/mm 2. H o w e v e r , 47% of the B E - t r e a t e d explant area was occupied by the plexiform-cell layer with a correspondingly greater density of 5460 cells/mm2. The response of 20-day fetal retinal explants cultured for 1 week was dramatically different from that of 16-day fetal explants. The main differences were observed in the plexiform-cell layer. Photomicrographic observation clearly indicated a general lack of organization and lamination in control when compared to B E treated explants (Figs. 5 and 6). In addition, a p r o n o u n c e d increase in the degeneration of presumed ganglion-like neurons was observed in control explants when c o m p a r e d to those treated with B E (Figs. 5 and 6). M o r e specifically, at this stage of t r e a t m e n t with B E there was no significant change in the amount of area occupied by the plexiform-cell layer as well as in the ganglion-like cell density (Table V). H o w e v e r , the striking difference was that 26% of the control plexiform cell layer was completely necrotic with no such areas a p p a r e n t in BEtreated explants (Table V). These observations were further c o r r o b o r a t e d when m o r p h o m e t r i c analyses revealed that 56% of cells in the control explant plexiform-cell layer were necrotic c o m p a r e d to only 23% in the B E - t r e a t e d group.

A

Fig. 4. Photomicrographs of thick sections from 16-day fetal rat retinal explants after one week in culture. Note at this stage of development retinas consist of an undifferentiated neuroepithelial region (i.e. below long arrows) with a differentiating plexiform-cell region (i.e. above long arrows) consisting of ganglion-like neurons (GC) separated by a fiber layer. A: control explant demonstrating a small plexiform-cell region with fewer ganglion-like cells. Note the absence of ganglion-like ceils in the neuroepithelial region. B: BE (25 pg/ml)-treated explant showing a larger plexiform-cell region with many prominent ganglion-like neurons found in the neuroepithelial region (small arrow). Mag. 600 ×.

A

D,

GC

Fig. 5. Light micrographs of thick sectioned 20-day fetal rat retinal explants after one week in culture. A: control explant demonstrating extensive cell degeneration (i.e. dark staining areas) in the poorly organized plexiform cell layer (PCL) in contrast to the circular rosettes (R) composed primarily of neuroepithelial cells (NE). B: dramatic enhancement of cell survival and organization within the PCL in response to BE (25/~g/ml) treatment. Note that a laminar organization has emerged consisting of NE cells and a plexiform layer (PL) containing differentiating ganglion-like cells (GC). Mag. 600 x.

259

Fig. 6-. Higher magnifications of thick sectioned 20-day fetal retinal explants after one week in culture. Mag. 1600 ×. A: control explant showing cell degeneration area in greater detail. Note the large number of necrotic cells (arrows) with no evidence of an intact plexiform layer. B: BE (25/2g/ml)-treated explant showing an area comparable to that seen in A. Note the appearance of predominantly intact ganglion-like cells (arrows) with prominent nucleoli and an intact interposing plexiform layer (PL). Also note a rosette formation (R) of neuroepithelial cells.

260 TABLE IV Morphometric analysis of brain extract (BE) effects on 16-day fetal rat retinal explants after I week in culture Treatment

Explant area (ktm2)

Controla (16)c 110,078 + 5913b BE (25~g/ml)a (17)a 111,647+ 5720~

% Neuroepithelial area/explant

Neuroepithelial layer ganglion-like cell density/mm 2

% Plexiform-cell area/explant

Plexiform area cell density~ram 2

68 _+2 53 __+2~

359 _+54 1448 + 126e

32 + 2 47 + 2e

4120 + 226 5460 _+227e

One ml of medium added on day 4 of culture (total volume = 2 ml). b Represents average value + S.E.M. c Number of explants analyzed d Control and experimental values not significantly different (P > 0.05). e Control and experimental values significantly different (P < 0.001). TABLE V Morphometric analysis of brain extract (BE) effect on 20-day fetal rat retinal explants after I week in culture Treatment

Controla (6)c BE (25/~g/ml)a (8) a b c d e

% Plexiform-cell area/explant

% Plexiform-cell area necrotic

Cells in plexiform layer Total density (mm 2)

Intact cell density (mm 2)

Necrotic cell density (mm 2)

% Necrotic cells

54 _ 2b 52 + 3~

26 + 7 0e

7647 _+588 6899 + 349d

3293 -+ 334 5314 + 293e

4354 _+626 1585 _+228e

57 _+5 23 + 2e

One ml of medium added on day 4 of culture (total volume = 2 ml). Represents average value + S.E.M. Number of explants analyzed. Control and experimental values not significantly different (P > 0.2). Control and experimental values significantly different (P < 0.001).

DISCUSSION The present results represent a further modification of our previously established fetal rat retinal explant culture system in which the m e d i u m contained 5% rat serum 29. It has been the concern of many investigators that the presence of serum in the culture media presents m a n y u n k n o w n variables and may even mask the effects of neurotropic factors6,18. Indeed with serum-supplemented media, we observed some degree of spontaneous neurite outgrowth from rat retinal explants 29. This made assay procedures very difficult because of the inconsistencies, particularly in the control situations. We have reported here that the use of a serum-free supplemented media has eliminated these problems. More specifically, conditions were defined in this study where stimulation of neurite outgrowth from fetal rat retinal explants occurred only in the presence of a CM-cellulose column eluted active fraction (BE). Therefore, results u n d e r these new culture conditions were highly consistent in that there was an 'all-or-none' response with respect to B E stimulation of neurite out-

growth in 100% of the experimental retinal explants, as opposed to no appreciable outgrowth from untreated controls. It was determined from our previous studies that the B E did not contain a NGF-like molecule with respect to its immunological or functional characteristics 29. While experimenting with our serum-free supplem e n t e d media conditions, it became apparent that further substrate modification was essential for achieving long radial fiber outgrowth. With just collagen, poly-L-lysine or a mixture of both substrates, BE extract stimulated only a dense halo of short fibers ( < 0.5 mm). However, if fibronectin (50 ktg/ml) was b o u n d to the collagen + poly-L-lysine substrate, BE was able to stimulate long radial fiber outgrowth in addition to the short halo of fibers seen previously. Fibronectin has also been reported to be essential for long radial fiber outgrowth in aggregates of 7-day embryonic chick neural retina cultured in a serumfree medium1. However, in our system, unlike that reported for the chick retinal aggregates, fibronectin by itself does not affect neurite outgrowth. This may be another strong indication of the need by fetal rat

261 retinal explants for trophic stimulation of neurite outgrowth. In addition, with N 1 supplementation of the media, neurite outgrowth in our study was maintained by BE for at least 3 weeks even in the absence of fibronectin, as opposed to the chick retinal aggregates which survived only 3-5 days in culture 1. The reason for such short-term survival in the chick system may be due to lack of proper media supplementation and/or the absence of a neurotrophic factor(s) like BE. Dose-response curves from this and our previous study 29 indicate a reasonable similarity especially in the activity generated over a comparable BE concentration range. In vitro, the fetal and early neonatal rat retinal explants showed a distinct age-dependent response to BE. We reported a significant inverse relationship between increased developmental ages and decreased responsiveness to the BE. This age-dependent relationship between developing nervous tissue and BE elicited responses has been demonstrated convincingly in NGF-sensitive peripheral ganglia. For example, it has been previously reported that explants of dorsal root ganglia from chick embryos older than E13-E16 no longer respond to N G F by extending processes16, 34, which is correlated with a marked decrease of N G F receptors during this period 16. Also, it has been previously reported that neurons dissociated from older ganglia lose their responsiveness to N G F 2,12,14,25. Unlike the response to NGF, the survival of dissociated chick sensory neuron cultures with glioma-conditioned medium or rat brain extract increased when those neurons were isolated from older embryos 2. These data taken together indicate the multiple and changing requirements of developing PNS neurons, particularly with respect to trophic factors. For the first time, our studies verify this phenomenon for a developing CNS tissue. If BE is not added to the culture medium of the fetal rat retinal explants until week 2 of culture, the neurite outgrowth response is significantly diminished. This may imply that BE is not only required for stimulation of neurite outgrowth but is also necessary for support of the survival of the neurite producing retinal cells. It has been recently established that developing retinas of this age undergo considerable ganglion cell death during the first postnatal week 10A7,24,26. It may be that if BE is added early enough to the culture medium, it may rescue the po-

tentially dying retinal ganglion cells in much the same manner as the actions of N G F in the developing PNS 27. It seems likely that this is not just simply a matter of the age-dependent decreased responsiveness to the tophic factor, since retinal explants respond maximally to BE when added after one week in culture. Indeed, this hypothesis was confirmed when thick sections of 20-day fetal retinal explants were observed and analyzed morphometrically one week later after BE treatment. The number of intact differentiating ganglion-like neurons was significantly greater and the histotypic organization of the developing retina was maintained in response to BE treatment. Similar results have been recently reported by Carri and Ebendal8 who have demonstrated that brain extracts prepared from chick embryos stimulated neurite outgrowth from 6 day avian embryonic retinas. Most importantly, in the same study retinal thick sections examined after 3 days in culture revealed that the layered structure of the chick retina was less well preserved in the absence of the brain extracts. There was a 3-fold increase in the density of necrotic cells in the rat plexiform cell layer of control 20-day explants. Therefore, it appears that the BE can reduce neuronal necrosis in the plexiform-cell layer at a comparable time in vitro which coincided with the peak of naturally occurring retinal ganglion cell death 17. However, it was demonstrated earlier in this paper that explants can achieve optimal outgrowth even if BE was not added until the end of the first week in culture. This appears to contradict the 20-day fetal explant morphometric results; however, it has been noted that neuroepithelial cells continue to survive after one week in culture even in the absence of BE. Although ganglion-like cells may degenerate during the first week in culture under these conditions, the later addition of BE may stimulate further maturation and/or rescue of continually differentiating retinal neurons from the surviving neuroepithelium thus allowing for neurite outgrowth to be initiated. Consequently, since BE is derived from the mammalian brain it may possess some molecule(s) which is target tissue-specific and normally regulates at later fetal ages retinal ganglion cell survival as is thought to be the case with NGF 2°,27. In contrast, explants treated with BE at an earlier embryonic age which coincides with ganglion cell matu-

262 ration (i.e. 16 fetal days) 7 showed a significant increase in the density of differentiating ganglion-like cells. Consequently, the plexiform-cell layer was also significantly increased in BE-treated explants, most likely due to increased cell density and/or sprouting of processes from retinal ganglion-like cells. It is possible at the earlier fetal age that BE is accelerating the process of maturation in much the same m a n n e r as N G F in the developing PNS23,28, especially since

existing culture conditions, BE is not only required for stimulating neurite outgrowth from fetal rat retinal explants, but is also important in survival and maturation of developing retinal neurons. Since BE contains a heterogeneous mixture of molecules it will be important to determine which one(s) is responsible for the above-reported neurotrophic activity. ACKNOWLEDGEMENTS

there was a 4-fold increase in the presence of differentiating retinal ganglion-like neurons in the neuroepithelial layer where they are rarely seen 7,9,22. The

This work was supported by a grant from the Na-

identification of ganglion-like neurons and the histo-

tional Eye Institute (EY 04377). The author wishes to thank Drs. Yves Barde and Hans T h o e n e n for

typic organization of our explants was consistent with that reported in other in vitro and in vivo rat retinal studies7.19.22.

nical assistance of Angela Tesh is greatly appreciated.

their generous supply of brain extract. Also, the tech-

In conclusion, our results strongly imply that under

REFERENCES 1 Akers, R. M., Mosher, D. F. and Lilien, J. E., Promotion of retinal neurite outgrowth by substratum-bound fibronectin, Develop. Biol., 86 (1981) 179-188. 2 Barde, Y. E., Edgar, D. and Thoenen, H., Sensory neurons in culture: changing requirements for survival factors during embryonic development, Proc. nat. Acad. Sci. U.S.A., 77 (1980) 1199-1203. 3 Barde, Y. A., Edgar, D. and Thoenen, H., Purification of a new neurotrophic factor from mammalian brain, EMBO J., 1 (1982) 549-553. 4 Barde, Y. A., Lindsay, R. M., Monard, D. and Thoenen, H., New factor released by cultured glioma cells supporting survival and growth of sensory neurons, Nature (Lond.), 274 (1978) 818. 5 Bjorklund, A. and Stenevi, V., Nerve growth factor: stimulation of regenerative growth of central noradrenergic neurons, Science, 175 (1972) 1251-1253. 6 Bottenstein, J. E., Skaper, S. D., Varon, S. S. and Sato, G. H., Selective survival of neurons from chick embryo sensory ganglionic dissociates utilizing serum-free supplemented medium, Exp. CellRes., 125 (1980) 183-190. 7 Braekevelt, C. R. and Hollenberg, M. J., The development of the retina of the albino rat, Amer. J. Anat., 127 (1970) 281-302. 8 Carri, N, G. and Ebendal, T., Organotypic cultures of neural retina: neurite outgrowth stimulated by brain extracts, Develop. Brain Res., 6 (1983) 219-229. 9 Coughlin, M. D., Bloom, E. M. and Black, I. B., Characteriza'tion of neuronal growth factor from mouse heart cell conditioned media, Develop. Biol., 82 (1981) 56-68. 10 Cunningham, T. J., Mohler, I. M. and Giodano, D. L., Naturally occurring neuron death in the ganglion layer of the neonatal rat: morphology and evidence for regional correspondence with neuron death in superior colliculus, Develop. Brain Res., 2 (1982) 203-215. 11 Dibner, M. D. and Black, I. B., The effect of target organ removal on the development of sympathetic neurons, Brain

Res., 103 (1978) 93-102. 12 Edgar, D., Barde, Y. A. and Thoenen, H., Subpopulations of cultured chick sympathetic neurons differ in their requirements for survival factors, Nature (Lond.), 289 (1981) 294-295. 13 Glaze, K. A. and Turner, J. E., Regenerative repair in the severed optic nerve of the newt (Triturus viridescens). The effect of nerve growth factor antiserum, Exp. Neurol., 58 (1978) 500-510. 14 Greene, L. A., Quantitative in vitro studies on the nerve growth factor (NGF) requirement of neurons. II. Sensory neurons, Develop. Biol., 58 (1977) 106-113. 15 Hamburger, V., Trophic interactions in neurogenesis: a personal historical account, Ann. Rev. Neurosci., 3 (1980) 269-278. 16 Herrup, K. and Shooter, E. M,, Properties of the fl-nerve growth factor receptor in development, J. Cell Biol., 67 (1975) 118-125. 17 Jeffery, G. and Perry, U. H., Evidence of ganglion cell death during development of the ipsilateral retinal projection in the rat, Develop. Brain Res., 2 (1981) 176-180. 18 Johnson, J. E. and Turner, J. E., Growth from regenerating goldfish retinal cultures in the absence of serum or hormonal supplements: tissue extract effects, J. Neurosci. Res., 8 (1982) 315-329. 19 Lavail, M. M. and Hild, W., Histotypic organization of the rat retina in vitro, Z. Zellforsch., 114 (1971) 557-579. 20 Levi-Montalcini, R. and Angeletti, P. O., Nerve growth factor, Physiol. Rev., 48 (1968) 534-569. 21 Manthorpe, M., Adler, R. and Varon, S., Cholinergic neurotrophic factor. VI. Age-dependent requirements by chick embryo ciliary ganglionic neurons, Develop. Biol., 85 (1981) 156-163. 22 McLoon, L. K., McLoon, S. C. and Lund, R. D., Cultured embryonic retinae transplanted to rat brain: differentiation and formation of projections to host superior collicu:us, Brain Res., 226 (1981) 15-31. 23 Mobley, W. C., Server, A. C., lshii, D. N., Riopelle, R. J. and Shooter, E. M., Nerve growth factor, New Engl. J.

263

Med., 297 (1977) 1149-1158. 24 Potts, R. A., Dreher, B. and Bennett, M. R., The loss of ganglion cells in the developing retina of the rat, Develop. Brain Res., 3 (1982) 481-486. 25 Rohrer, H. and Barde, Y. A., Presence and disappearance of nerve growth factor receptors on sensory neurons in culture, Develop. Biol., 89 (1982) 309-315. 26 Sangelaub, D. R. and Finlay, B. L., Cell death in the mammalian visual system during normal development: I. Retinal ganglion cells, J. comp. Neurol., 204 (1982) 311-317. 27 Thoenen, H. and Barde, Y. A., Nerve growth factor, Physiol. Rev., 60 (1980) 1284-1335. 28 Thoenen, H., Barde, Y. A. and Edgar, D., Factors involved in the regulation of the survival and differentiation of neurons. In J. G. Nicholls (Ed.), Repair and Regeneration of the Nervous System, Springer-Verlag, New York, 1982, pp. 173-185. 29 Turner, J. E., Barde, Y. A., Schwab, M. E. and Thoenen, H., Extract from brain stimulates neurite outgrowth from fetal rat retinal explants, Develop. Brain Res., 6 (1983) 77-83.

30 Turner, J. E. and Glaze, K. A., Regenerative repair in the severed optic nerve of the newt (Triturus viridescens). The effect of nerve growth factor, Exp. Neurol., 57 (1977) 687-691. 31 Turner, J. E., Schwab, M. E. and Thoenen, H., NGF stimulates neurite outgrowth from goldfish retinal explants: the influence of a prior lesion, Develop. Brain Res., 4 (1982) 59-66. 32 Turner, J. E., Barde, Y. A. and Thoenen, H., A new neurotrophic factor isolated from mammalian brain: in vitro studies on fetal rat retina, Soc. Neurosci. Abstr., (1983). 33 Varon, S. and Bunge, R. P., Trophic mechanisms in the peripheral nervous system, Ann. Rev. Neurosci., 1 (1978) 327-361. 34 Winik, M. and Greenberg, R. E., Chemical control of sensory ganglia during a critical period of development, Nature (Lond.), 205 (1967) 180-181. 35 Yip, H. K. and Grafstein, B., Effect of nerve growth factor on regeneration of goldfish optic axons, Brain Res., 238 (1982) 329-339.