Brain Research, 569 (1992) 14-25 (~) 1992 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/92/$05.00
14 BRES 17261
Cooperative regulation of nerve growth factor synthesis and secretion in fibroblasts and astrocytes by fibroblast growth factor and other cytokines Kazunari Yoshida* and Fred H. Gage Department of Neurosciences, M-024, School of Medicine, University of California, San Diego, La Jolla, CA 92123 (U.S.A.) (Accepted 30 July 1991)
Key words: Glia; Interleukin-lfl; Tumor necrosis factor-a; Transforming growth factor-i/l; Dibutyryl cyclic adenosine monophosphate
Acidic fibroblast growth factor (aFGF) enhances nerve growth factor (NGF) synthesis by astrocytes obtained from various brain regions. NGF secretion by fibrous-shaped astrocytes transformed by dibutyryl-cAMP (db-eAMP) pretreatment was less than that by untreated astroeytes. However, aFGF also enhanced NGF secretion by fibrous-shaped astrocytes. The effects of various kinds of intracellular signaling modulators on NGF synthesis were examined. None of the following second messenger effectors had an effect on NGF synthesis: protein kinase C (PKC) agonist (phorbol myristate acetate (PMA)) or antagonist (sphingosine (SP)). LiC1, and ionomycin (Iono). Further, increases of intracellular cAMP by forskolin (FK) or db-cAMP have no significant effect on NGF synthesis in astrocytes under a standard culture condition. However, NGF synthesis by astrocytes in the presence of aFGF was significantly enhanced by db-cAMP, but not by FK or sodium butyrate. These results indicate that an excessive amount of cAMP enhances the effect of aFGF on NGF synthesis in astrocytes. NGF synthesis in astroeytes was not affected by treatment with anti-aFGF or anti-bFGF neutralizing antibodies, indicating that FGFs are not involved in the autoerine regulation of NGF synthesis in astrocytes. Transforming growth factor-ill (TGF-fll), which inhibits some effects of FGFs, increased NGF synthesis in concert with aFGF. Furthermore, the highest NGF synthesis was observed when astrocytes were stimulated by all of the following cytokines: aFGF, interleukin-lfl (IL-lfl), tumor necrosis factor-a (TNF-a) and TGF-fll. The mechanism regulating NGF synthesis in fibroblasts obtained from prenatal rat skin was also investigated. Acidic FGF, basic FGF (bFGF), epidermal growth factor (EGF), platelet-derived growth factor (PDGF), transforming growth factor-a (TGF-a), TGF-fll, IL-lfl, and TNF-a were found to be regulators of NGF synthesis in skin fibroblasts. Among these cytokines, aFGF is the most potent regulator of NGF synthesis in fibroblasts. NGF synthesis by skin fibroblasts, either in the presence or absence of aFGF, was not modified by any of the following: FK, PMA, SP, LiCI, and Iono. However, db-cAMP significantly enhanced NGF synthesis in both conditions. Sodium butyrate enhanced NGF synthesis in the presence of aFGF, but not in the absence of aFGF. These results suggest that an excessive amount of cAMP and butyrate moiety regulate NGF synthesis in skin fibroblasts in different ways. Finally, the effects of cytokines, which can regulate NGF synthesis in skin fibroblasts, were found to be cooperative except in the following two combinations: aFGF+bFGF and EGF+TGF-a. These results demonstrate that NGF synthesis in both astrocytes and fibroblasts is cooperatively regulated by cytokines.
INTRODUCTION Astrocytes, which are stromal cells peculiar to the nervous system, are known to possess exclusive neurotrophic effects 5'41'46. The neurotrophic effects of astrocytes have been investigated mainly in in vitro experiments and are considered to be m e d i a t e d by neurotrophic factors (NTFs), cell-surface molecules, and extracellular matrix p r o d u c e d by astrocytes. Nerve growth factor ( N G F ) , first identified as an N T F for the peripheral sensory neurons s'35, has since been shown to exist in the CNS 32 and to be acting as an N T F mainly for cholinergic neurons in the basal forebrain 2°'21'28. N G F is also the only well-characterized N T F p r o d u c e d by astrocytes 14'29'42'58'70'71. A l t h o u g h there is no direct evidence
of N G F production by astrocytes in vivo, the enhancement of neurotrophic activity in the lesioned brain has been d e m o n s t r a t e d as. F u r t h e r m o r e , we have shown that astrocytes play an important role in sprouting of N G F sensitive d a m a g e d cholinergic neurons in the basal forebrain 17. Thus, we speculate that astrocytes are activated to secrete N G F and o t h e r NTFs following brain injury to mitigate neurological and regenerative sprouting. N G F synthesis in astrocytes has been investigated in vitro. N G F secretion by astrocytes is not constant during a culture period and is highest just after passage 15'29'71. Mechanical stimulation during a cell-transfer procedure might activate astrocytes to produce NGF. Recently, many kinds of growth factors and lymphokines have been shown to regulate N G F synthesis in astrocytes 16'58'71.
* Present address: Department of Neurosurgery, School of Medicine, Keio University, 35 Shinanomachi, Shinjuku-ku, Tokyo 160, Japan. Correspondence: F.H. Gage, Department of Neurosciences, M-024, School of Medicine, University of California, San Diego, La Jolla, CA "92123, U.S.A. Fax: (1) (619) 534-5748.
15 Furthermore, we have demonstrated that acidic fibroblast growth factor (aFGF), which is the most potent regulator of NGF synthesis, enhances NGF synthesis in concert with the products of activated microglia, such as intedeukin-lfl (IL-lfl) and tumor necrosis factor-a (TNFa) 9'25. NGF secretion occurs immediately after synthesis, and FGF stimulates NGF synthesis and secretion from the transcription level 71. FGFs, which are abundant in the brain62, appear to be located in neurons in normal brain31'49 and might be released in the case of brain damage. FGFs are also known to be produced by astrocytes11'22. These findings suggest that FGFs regulate NGF synthesis in astrocytes according to an autocrine and paracrine manner. The first objective of this study was to reveal the precise mechanism regulating NGF synthesis in astrocytes by FGE In the peripheral nervous system (PNS), Schwann cells and fibroblasts are known to produce NGF26. Although we have demonstrated that fibroblasts derived from meninges can produce NGF and that FGFs stimulate NGF synthesis by meningeal fibroblasts 7:, the regulation of NGF synthesis by non-neuronal cells in sciatic nerve reported by Lindholm et al.3s is not exactly the same as that seen in astrocytes. The second objective of this study was, therefore, to elucidate the differences in the mechanisms of NGF regulation in the CNS and the PNS.
Immunocytochemistry For immunocytochemistry, the cells were seeded on poly-L-lysine coated coverslips (diameter 13 nun) placed in 24-well culture plates. The seeding cell number was 2 × 105 ceUs/well for astrocytes and 7 x 104 cells/well for skin fibroblasts. Immunocytochemistry with anti-cow GFAP (DAKO) and antihuman fibronectin (Cappel) was performed according to the avidinbiotin complex method, as previously described 71. For FGF receptor (FGF-R) immunocytochemistry, the cells were fixed with 3.7% formaldehyde in phosphate-buffered solution (PBS) for 5 min at room temperature. Subsequently the cells were treated sequentially with the following reagents: (i) 0.3% H20 2 in PBS for 30 min at room temperature; (2) 3% goat serum (GS) in PBS for 30 min at room temperature; (3) rabbit anti-chicken FGF-R IgG (Upstate Biotechnology Inc.; 1:400) in 3% GS-PBS for 16 h at 4°C; (4) biotinylated anti-rabbit IgG (Vector; 1:200) in 3% GS-PBS for 1 h at room temperature; (5) HRP-conjugated avidin-biotin complex (Vector; 1:100) in 3% GS-PBS for 1 li at room temperature; (6) 0.05% solution of diaminobenzidine, 0.01% H202, and 0.04% NiCI in 0.1 M TBS. Immunolabeled cells were dehydrated in graded ethanol and mounted on slideglasses with Permount (Fischer).
Cytokines and chemicals The effects of cytokines and miscellaneous chemicals were examined at the following doses, except as indicated otherwise in the text: aFGF, 25 ng/ml (bovine brain; R and D Systems); bFGF, 25 ng/ml (bovine brain; R and D Systems); anti-aFGF, 100 #g/ml (rabbit IgG; R and D Systems); anti-bFGF, 100 #g/ml (rabbit IgG; R and D Systems); epidermal growth factor (EGF), 10 ng/ml (mouse submaxillary; Sigma); platelet-derived growth factor (PDGF), 10 ng/ml (human recombinant; G-enzyme); transforming growth factor-a (TGF-a), 25 ng/mi (human recombinant; Intergen);
250
MATERIALS AND METHODS
Cell cultures Astrocytes. Astrocyte-enriched cultures were obtained after two passages from 5 different sites of origin of newborn rat pups (Fischer 344): cerebral cortex, striatum, hippocampus, septum and cerebellum 71. For the measurement of NGF secretion, astrocytes were suspended in Dulbecco's modified Eagle's medium (DME), containing 10% fetal calf serum (FCS), 40 mg/l gentamycin, and 2.5 mg/l of Fungizone, and seeded in poly-L-lysine coated 24-well culture plates (1.4 x 105 cells/well), and incubated with 0.3 ml/well of 10% FCS-DME or serum-free defined medium (DME-DF) with or without testing materials for 9 or 24 h from 2 h or 3 days after passage. DME-DF was supplemented with 1% of GMS-S (Gibco) and 75 mg/l of bovine serum albumin (BSA) (Gibco) instead of FCS. DME-DF was used only to examine the effects of anti-FGF neutralizing antibodies. Skin-fibroblasts. Skin fragments obtained from rat fetuses (Fischer 344) were suspended in 10% FCS-DME and seeded in 25 cm 2 flasks. Ceils obtained after two passages with enzymatic dissociation (0.25% trypsin and 0.02% EDTA) were used. For the measurement of NGF secretion, the ceils were seeded in 24-well culture plates (5 x 104 cell/well), and incubated with 0.3 ml of the same culture medium for 9 h or 24 h from 2 h or 3 days after passage. The cultures used for NGF studies reached confluency within 2 days under the present culture condition. The cell numbers were counted by Coulter counter after enzymatic dissociation.
NGF protein measurement NGF protein levels were measured by 2-site enzyme immunoassay, as described previously71. The detection fimit of this assay is 5-10 pg/ml of fl-NGE
2OO
O) U.
0 z
150 100 50 0 Hlppocumpuo Septum
Cortex
Strlatum
Cerebellum
5.0(~
(A O
4.0(] 3.0(]
O
2.00 O
z
1.0(]
0
(J 0.00
Hlppo©empus
Septum
Cortex
Strlatum
Cerebellum
Fig. 1. The effects of aFGF on NGF secretion (a) and proliferation (b) of astrocytes obtained from 5 different sites of origins. The cells were incubated for 24 h beginning 3 days after passage, and NGF levels in conditioned medium (CM) and the final cell numbers were measured. Values represent means -+ S.D. of 3 determinations. *P < 0.01, **P < 0.001, Student's t-test.
16
:w
Fig. 2. Immunocytoehemistry for GFAP (a,b) and FGF-R (c,d) of astrocyte culture obtained from hippocampus, untreated (a,c) and treated with db-eAMP and Theo for 48 h (b,d). x260.
TGF-fll, 10 ng/ml (human platelet; Sigma); TNF-a, 100 ng/ml (human recombinant; Sigma); IL-lfl, 30 unit/ml (human recombinant; Genzyme); db-cAMP, 1 mM (Sigma); theophylline (Theo), 0.1 mM (Sigma); Forskolin (FK), 10 mM (Sigma); phorbol 12-myristate 13-acetate (PMA) 100 ng/ml (Sigma); Sphingosine (SP), 10 ~M (Sigma); LiCI, 1 mM (Sigma); ionomycin (Iono), 100 nM (gift from Dr. Joan Brown).
TABLE I Effect of rib-cAMP pretreatment Values represent means --- S.D. of 3 determinations. NGF Cell no. b concentra(×105 cells/ lion a (pg/mO well)
NGF secretion c (pg/24 h~ 105 cells)
RESULTS Effects o f a F G F on various types o f astrocytes
No treatment Control aFGF db-cAMP pretreated a Control aFGF
30--+8 107-.+17* 7_+3** 54-+10"
2.82-+0.19 3.45---0.03*
3.2-+0.8 9.3-+1.5 *
2.13-+0.18 ** 0.9-+0.3 ** 2.38+0.04 * 6.8-+1.2 *
a NGF concentrations in CM of hippocampal astrocytes were determined after 24 h incubation beginning 3 days after passage. b Cell numbers were counted at the end of experiments. c NGF secretions were standardized by the cell number determined at the end of experiment. d The cells were treated with db-cAMP and Theo for 48 h beginning 24 h after passage. * P < 0.05, one-way ANOVA followed by Dunnett's post hoc ~. t-test compared to control. ** P < 0.05, one-way ANOVA followed by Dunnett's post hoc t-test compared to control with no treatment.
T h e effects o f a F G F o n astrocytes o b t a i n e d f r o m various sites of origin - c e r e b r a l cortex, striatum, s e p t u m , h i p p o c a m p u s and c e r e b e l l u m - w e r e investigated. First, a F G F significantly i n c r e a s e d N G F p r o d u c t i o n and secretion by all of the astrocytes f r o m the 5 d i f f e r e n t sites of origin (Fig. l a ) . T h e cell n u m b e r s c o u n t e d after 24 h of i n c u b a t i o n with a F G F w e r e statistically h i g h e r t h a n t h o s e in controls in e a c h case (Fig. l b ) . N e x t , the effects of a F G F o n f i b r o u s - s h a p e d astrocytes i n d u c e d by p r e t r e a t m e n t with d b - c A M P and T h e o w e r e e x a m i n e d , b e c a u s e f i b r o u s - s h a p e d astrocytes i n d u c e d by d b - c A M P m i m i c so-called ' r e a c t i v e astrocytes' in vivo. T h e i m m u n o c y t o c h e m i s t r y for glial fibrillary acidic pro-
17 tein (GFAP) shows the morphology of these astrocytes (Fig. 2a,b). Pretreatment with these chemicals significantly suppressed proliferation and NGF secretion; however, aFGF also enhanced NGF synthesis by and proliferation of the fibrous-shaped astrocytes (Table I). The expression of FGF-R by both types of astrocytes was detected by immunocytochemistry (Fig. 2c,d).
140 120 E o)
IO0
o.
80
,, (3
60
Z
4O 20 0 None
IFGF
TGF.81
IFGF*TGF-81
None
IFGF
TGF-I
IFGF+TGF-I
lntracellular signaling and NGF synthesis in astrocytes To e v a l u a t e w h a t k i n d o f i n t r a c e l l u l a r signaling m e c h a n i s m is i n v o l v e d in N G F synthesis, t h e effects o f various s e c o n d m e s s e n g e r e f f e c t o r s o n N G F
synthesis in
astrocytes w e r e e x a m i n e d . T h e results, s u m m a r i z e d in Fig. 3a, s h o w t h a t n o n e o f t h e f o l l o w i n g i n t r a c e l l u l a r 140
e v e n t s plays an i m p o r t a n t r o l e in N G F synthesis in astrocytes: (1) i n c r e a s e o f c A M P ( d b - c A M P + T h e o , alone, and FK+Theo);
120
Theo
(2) s t i m u l a t i o n ( P M A ) o r inhi-
b i t i o n (SP) of P K C ; (3) d e c r e a s e o f inositol (LiCI); o r (4) i n c r e a s e o f C a 2+ ( I o n o ) . T h e s a m e e x p e r i m e n t was d o n e in t h e p r e s e n c e of
""
100
E O.
u. (..-1 ~,
60 4O 20
a F G F w i t h v e r y similar results, e x c e p t t h a t d b - c A M P a n d
0
Fig. 4. The effects of combinations of aFGF and TGF-fll (a) or aFGF and TGF-a (b) on NGF secretion by hippocampal astrocytes were examined under the same culture conditions described in the legend to Fig. 1. Values represent means --- S.D. of 3 determinations. *P < 0.05, one-way ANOVA followed by Dunnett's post hoe t-test compared to control. **P < 0.05, one-way ANOVA followed by Scheffe's post hoe F-test compared to others.
150
a E
loo
v IJ,.
(3 ~"
5O
31111
E 15o
O Z
100
~E 11111 U.
z
~
5o
z
2
o
z
--
I-
.oo •
14 BRES 17261
Cooperative regulation of nerve growth factor synthesis and secretion in fibroblasts and astrocytes by fibroblast growth factor and other cytokines Kazunari Yoshida* and Fred H. Gage Department of Neurosciences, M-024, School of Medicine, University of California, San Diego, La Jolla, CA 92123 (U.S.A.) (Accepted 30 July 1991)
Key words: Glia; Interleukin-lfl; Tumor necrosis factor-a; Transforming growth factor-i/l; Dibutyryl cyclic adenosine monophosphate
Acidic fibroblast growth factor (aFGF) enhances nerve growth factor (NGF) synthesis by astrocytes obtained from various brain regions. NGF secretion by fibrous-shaped astrocytes transformed by dibutyryl-cAMP (db-eAMP) pretreatment was less than that by untreated astroeytes. However, aFGF also enhanced NGF secretion by fibrous-shaped astrocytes. The effects of various kinds of intracellular signaling modulators on NGF synthesis were examined. None of the following second messenger effectors had an effect on NGF synthesis: protein kinase C (PKC) agonist (phorbol myristate acetate (PMA)) or antagonist (sphingosine (SP)). LiC1, and ionomycin (Iono). Further, increases of intracellular cAMP by forskolin (FK) or db-cAMP have no significant effect on NGF synthesis in astrocytes under a standard culture condition. However, NGF synthesis by astrocytes in the presence of aFGF was significantly enhanced by db-cAMP, but not by FK or sodium butyrate. These results indicate that an excessive amount of cAMP enhances the effect of aFGF on NGF synthesis in astrocytes. NGF synthesis in astroeytes was not affected by treatment with anti-aFGF or anti-bFGF neutralizing antibodies, indicating that FGFs are not involved in the autoerine regulation of NGF synthesis in astrocytes. Transforming growth factor-ill (TGF-fll), which inhibits some effects of FGFs, increased NGF synthesis in concert with aFGF. Furthermore, the highest NGF synthesis was observed when astrocytes were stimulated by all of the following cytokines: aFGF, interleukin-lfl (IL-lfl), tumor necrosis factor-a (TNF-a) and TGF-fll. The mechanism regulating NGF synthesis in fibroblasts obtained from prenatal rat skin was also investigated. Acidic FGF, basic FGF (bFGF), epidermal growth factor (EGF), platelet-derived growth factor (PDGF), transforming growth factor-a (TGF-a), TGF-fll, IL-lfl, and TNF-a were found to be regulators of NGF synthesis in skin fibroblasts. Among these cytokines, aFGF is the most potent regulator of NGF synthesis in fibroblasts. NGF synthesis by skin fibroblasts, either in the presence or absence of aFGF, was not modified by any of the following: FK, PMA, SP, LiCI, and Iono. However, db-cAMP significantly enhanced NGF synthesis in both conditions. Sodium butyrate enhanced NGF synthesis in the presence of aFGF, but not in the absence of aFGF. These results suggest that an excessive amount of cAMP and butyrate moiety regulate NGF synthesis in skin fibroblasts in different ways. Finally, the effects of cytokines, which can regulate NGF synthesis in skin fibroblasts, were found to be cooperative except in the following two combinations: aFGF+bFGF and EGF+TGF-a. These results demonstrate that NGF synthesis in both astrocytes and fibroblasts is cooperatively regulated by cytokines.
INTRODUCTION Astrocytes, which are stromal cells peculiar to the nervous system, are known to possess exclusive neurotrophic effects 5'41'46. The neurotrophic effects of astrocytes have been investigated mainly in in vitro experiments and are considered to be m e d i a t e d by neurotrophic factors (NTFs), cell-surface molecules, and extracellular matrix p r o d u c e d by astrocytes. Nerve growth factor ( N G F ) , first identified as an N T F for the peripheral sensory neurons s'35, has since been shown to exist in the CNS 32 and to be acting as an N T F mainly for cholinergic neurons in the basal forebrain 2°'21'28. N G F is also the only well-characterized N T F p r o d u c e d by astrocytes 14'29'42'58'70'71. A l t h o u g h there is no direct evidence
of N G F production by astrocytes in vivo, the enhancement of neurotrophic activity in the lesioned brain has been d e m o n s t r a t e d as. F u r t h e r m o r e , we have shown that astrocytes play an important role in sprouting of N G F sensitive d a m a g e d cholinergic neurons in the basal forebrain 17. Thus, we speculate that astrocytes are activated to secrete N G F and o t h e r NTFs following brain injury to mitigate neurological and regenerative sprouting. N G F synthesis in astrocytes has been investigated in vitro. N G F secretion by astrocytes is not constant during a culture period and is highest just after passage 15'29'71. Mechanical stimulation during a cell-transfer procedure might activate astrocytes to produce NGF. Recently, many kinds of growth factors and lymphokines have been shown to regulate N G F synthesis in astrocytes 16'58'71.
* Present address: Department of Neurosurgery, School of Medicine, Keio University, 35 Shinanomachi, Shinjuku-ku, Tokyo 160, Japan. Correspondence: F.H. Gage, Department of Neurosciences, M-024, School of Medicine, University of California, San Diego, La Jolla, CA "92123, U.S.A. Fax: (1) (619) 534-5748.
15 Furthermore, we have demonstrated that acidic fibroblast growth factor (aFGF), which is the most potent regulator of NGF synthesis, enhances NGF synthesis in concert with the products of activated microglia, such as intedeukin-lfl (IL-lfl) and tumor necrosis factor-a (TNFa) 9'25. NGF secretion occurs immediately after synthesis, and FGF stimulates NGF synthesis and secretion from the transcription level 71. FGFs, which are abundant in the brain62, appear to be located in neurons in normal brain31'49 and might be released in the case of brain damage. FGFs are also known to be produced by astrocytes11'22. These findings suggest that FGFs regulate NGF synthesis in astrocytes according to an autocrine and paracrine manner. The first objective of this study was to reveal the precise mechanism regulating NGF synthesis in astrocytes by FGE In the peripheral nervous system (PNS), Schwann cells and fibroblasts are known to produce NGF26. Although we have demonstrated that fibroblasts derived from meninges can produce NGF and that FGFs stimulate NGF synthesis by meningeal fibroblasts 7:, the regulation of NGF synthesis by non-neuronal cells in sciatic nerve reported by Lindholm et al.3s is not exactly the same as that seen in astrocytes. The second objective of this study was, therefore, to elucidate the differences in the mechanisms of NGF regulation in the CNS and the PNS.
Immunocytochemistry For immunocytochemistry, the cells were seeded on poly-L-lysine coated coverslips (diameter 13 nun) placed in 24-well culture plates. The seeding cell number was 2 × 105 ceUs/well for astrocytes and 7 x 104 cells/well for skin fibroblasts. Immunocytochemistry with anti-cow GFAP (DAKO) and antihuman fibronectin (Cappel) was performed according to the avidinbiotin complex method, as previously described 71. For FGF receptor (FGF-R) immunocytochemistry, the cells were fixed with 3.7% formaldehyde in phosphate-buffered solution (PBS) for 5 min at room temperature. Subsequently the cells were treated sequentially with the following reagents: (i) 0.3% H20 2 in PBS for 30 min at room temperature; (2) 3% goat serum (GS) in PBS for 30 min at room temperature; (3) rabbit anti-chicken FGF-R IgG (Upstate Biotechnology Inc.; 1:400) in 3% GS-PBS for 16 h at 4°C; (4) biotinylated anti-rabbit IgG (Vector; 1:200) in 3% GS-PBS for 1 h at room temperature; (5) HRP-conjugated avidin-biotin complex (Vector; 1:100) in 3% GS-PBS for 1 li at room temperature; (6) 0.05% solution of diaminobenzidine, 0.01% H202, and 0.04% NiCI in 0.1 M TBS. Immunolabeled cells were dehydrated in graded ethanol and mounted on slideglasses with Permount (Fischer).
Cytokines and chemicals The effects of cytokines and miscellaneous chemicals were examined at the following doses, except as indicated otherwise in the text: aFGF, 25 ng/ml (bovine brain; R and D Systems); bFGF, 25 ng/ml (bovine brain; R and D Systems); anti-aFGF, 100 #g/ml (rabbit IgG; R and D Systems); anti-bFGF, 100 #g/ml (rabbit IgG; R and D Systems); epidermal growth factor (EGF), 10 ng/ml (mouse submaxillary; Sigma); platelet-derived growth factor (PDGF), 10 ng/ml (human recombinant; G-enzyme); transforming growth factor-a (TGF-a), 25 ng/mi (human recombinant; Intergen);
250
MATERIALS AND METHODS
Cell cultures Astrocytes. Astrocyte-enriched cultures were obtained after two passages from 5 different sites of origin of newborn rat pups (Fischer 344): cerebral cortex, striatum, hippocampus, septum and cerebellum 71. For the measurement of NGF secretion, astrocytes were suspended in Dulbecco's modified Eagle's medium (DME), containing 10% fetal calf serum (FCS), 40 mg/l gentamycin, and 2.5 mg/l of Fungizone, and seeded in poly-L-lysine coated 24-well culture plates (1.4 x 105 cells/well), and incubated with 0.3 ml/well of 10% FCS-DME or serum-free defined medium (DME-DF) with or without testing materials for 9 or 24 h from 2 h or 3 days after passage. DME-DF was supplemented with 1% of GMS-S (Gibco) and 75 mg/l of bovine serum albumin (BSA) (Gibco) instead of FCS. DME-DF was used only to examine the effects of anti-FGF neutralizing antibodies. Skin-fibroblasts. Skin fragments obtained from rat fetuses (Fischer 344) were suspended in 10% FCS-DME and seeded in 25 cm 2 flasks. Ceils obtained after two passages with enzymatic dissociation (0.25% trypsin and 0.02% EDTA) were used. For the measurement of NGF secretion, the ceils were seeded in 24-well culture plates (5 x 104 cell/well), and incubated with 0.3 ml of the same culture medium for 9 h or 24 h from 2 h or 3 days after passage. The cultures used for NGF studies reached confluency within 2 days under the present culture condition. The cell numbers were counted by Coulter counter after enzymatic dissociation.
NGF protein measurement NGF protein levels were measured by 2-site enzyme immunoassay, as described previously71. The detection fimit of this assay is 5-10 pg/ml of fl-NGE
2OO
O) U.
0 z
150 100 50 0 Hlppocumpuo Septum
Cortex
Strlatum
Cerebellum
5.0(~
(A O
4.0(] 3.0(]
O
2.00 O
z
1.0(]
0
(J 0.00
Hlppo©empus
Septum
Cortex
Strlatum
Cerebellum
Fig. 1. The effects of aFGF on NGF secretion (a) and proliferation (b) of astrocytes obtained from 5 different sites of origins. The cells were incubated for 24 h beginning 3 days after passage, and NGF levels in conditioned medium (CM) and the final cell numbers were measured. Values represent means -+ S.D. of 3 determinations. *P < 0.01, **P < 0.001, Student's t-test.
16
:w
Fig. 2. Immunocytoehemistry for GFAP (a,b) and FGF-R (c,d) of astrocyte culture obtained from hippocampus, untreated (a,c) and treated with db-eAMP and Theo for 48 h (b,d). x260.
TGF-fll, 10 ng/ml (human platelet; Sigma); TNF-a, 100 ng/ml (human recombinant; Sigma); IL-lfl, 30 unit/ml (human recombinant; Genzyme); db-cAMP, 1 mM (Sigma); theophylline (Theo), 0.1 mM (Sigma); Forskolin (FK), 10 mM (Sigma); phorbol 12-myristate 13-acetate (PMA) 100 ng/ml (Sigma); Sphingosine (SP), 10 ~M (Sigma); LiCI, 1 mM (Sigma); ionomycin (Iono), 100 nM (gift from Dr. Joan Brown).
TABLE I Effect of rib-cAMP pretreatment Values represent means --- S.D. of 3 determinations. NGF Cell no. b concentra(×105 cells/ lion a (pg/mO well)
NGF secretion c (pg/24 h~ 105 cells)
RESULTS Effects o f a F G F on various types o f astrocytes
No treatment Control aFGF db-cAMP pretreated a Control aFGF
30--+8 107-.+17* 7_+3** 54-+10"
2.82-+0.19 3.45---0.03*
3.2-+0.8 9.3-+1.5 *
2.13-+0.18 ** 0.9-+0.3 ** 2.38+0.04 * 6.8-+1.2 *
a NGF concentrations in CM of hippocampal astrocytes were determined after 24 h incubation beginning 3 days after passage. b Cell numbers were counted at the end of experiments. c NGF secretions were standardized by the cell number determined at the end of experiment. d The cells were treated with db-cAMP and Theo for 48 h beginning 24 h after passage. * P < 0.05, one-way ANOVA followed by Dunnett's post hoc ~. t-test compared to control. ** P < 0.05, one-way ANOVA followed by Dunnett's post hoc t-test compared to control with no treatment.
T h e effects o f a F G F o n astrocytes o b t a i n e d f r o m various sites of origin - c e r e b r a l cortex, striatum, s e p t u m , h i p p o c a m p u s and c e r e b e l l u m - w e r e investigated. First, a F G F significantly i n c r e a s e d N G F p r o d u c t i o n and secretion by all of the astrocytes f r o m the 5 d i f f e r e n t sites of origin (Fig. l a ) . T h e cell n u m b e r s c o u n t e d after 24 h of i n c u b a t i o n with a F G F w e r e statistically h i g h e r t h a n t h o s e in controls in e a c h case (Fig. l b ) . N e x t , the effects of a F G F o n f i b r o u s - s h a p e d astrocytes i n d u c e d by p r e t r e a t m e n t with d b - c A M P and T h e o w e r e e x a m i n e d , b e c a u s e f i b r o u s - s h a p e d astrocytes i n d u c e d by d b - c A M P m i m i c so-called ' r e a c t i v e astrocytes' in vivo. T h e i m m u n o c y t o c h e m i s t r y for glial fibrillary acidic pro-
17 tein (GFAP) shows the morphology of these astrocytes (Fig. 2a,b). Pretreatment with these chemicals significantly suppressed proliferation and NGF secretion; however, aFGF also enhanced NGF synthesis by and proliferation of the fibrous-shaped astrocytes (Table I). The expression of FGF-R by both types of astrocytes was detected by immunocytochemistry (Fig. 2c,d).
140 120 E o)
IO0
o.
80
,, (3
60
Z
4O 20 0 None
IFGF
TGF.81
IFGF*TGF-81
None
IFGF
TGF-I
IFGF+TGF-I
lntracellular signaling and NGF synthesis in astrocytes To e v a l u a t e w h a t k i n d o f i n t r a c e l l u l a r signaling m e c h a n i s m is i n v o l v e d in N G F synthesis, t h e effects o f various s e c o n d m e s s e n g e r e f f e c t o r s o n N G F
synthesis in
astrocytes w e r e e x a m i n e d . T h e results, s u m m a r i z e d in Fig. 3a, s h o w t h a t n o n e o f t h e f o l l o w i n g i n t r a c e l l u l a r 140
e v e n t s plays an i m p o r t a n t r o l e in N G F synthesis in astrocytes: (1) i n c r e a s e o f c A M P ( d b - c A M P + T h e o , alone, and FK+Theo);
120
Theo
(2) s t i m u l a t i o n ( P M A ) o r inhi-
b i t i o n (SP) of P K C ; (3) d e c r e a s e o f inositol (LiCI); o r (4) i n c r e a s e o f C a 2+ ( I o n o ) . T h e s a m e e x p e r i m e n t was d o n e in t h e p r e s e n c e of
""
100
E O.
u. (..-1 ~,
60 4O 20
a F G F w i t h v e r y similar results, e x c e p t t h a t d b - c A M P a n d
0
Fig. 4. The effects of combinations of aFGF and TGF-fll (a) or aFGF and TGF-a (b) on NGF secretion by hippocampal astrocytes were examined under the same culture conditions described in the legend to Fig. 1. Values represent means --- S.D. of 3 determinations. *P < 0.05, one-way ANOVA followed by Dunnett's post hoe t-test compared to control. **P < 0.05, one-way ANOVA followed by Scheffe's post hoe F-test compared to others.
150
a E
loo
v IJ,.
(3 ~"
5O
31111
E 15o
O Z
100
~E 11111 U.
z
~
5o
z
2
o
z
--
I-
.oo •
