- Email: [email protected]
A convenient in vitro assay for the inhibition of neurite outgrowth by adult mammalian CNS myelin using immortalized neuronal cells
ELSEVIER
Journal of Neuroscience Methods 63 (1995) 23-28
A convenient in vitro assay for the inhibition of neurite outgrowth by adult mammalian CNS myelin using immortalized neuronal cells Andres M. Lozano a,b~c, * , Matthias Schmidt a-c,Arthur Roach a~’ a Division
of Immunology and Neurobiology, b Division of Neurosurgery, ’ Department of Molecular
Samuel Lunenfeld Research Institute, Mount Sinai Hospital, Toronto. Department of Surgery, University of Toronto, Toronto, Canada and Medical Genetics, University of Toronto, Toronto, Canada
Canada
Received 5 January 1995; revised 10 May 1995; accepted 4 June 1995 _._..-.---_---
Abstract The adult mammaiian CNS contains molecules which inhibit neurite outgrowth and which may be responsible for the lack of successful axonal regeneration after injuries in the brain and spinal cord. We describe an in vitro assay to measure the ability of primary and established lines of neuronal cells to produce neurites in the presence of CNS inhibitory molecules. The assay is suitable for identification of agents and treatments to overcome neurite growth inhibition. Assays are carried out in 96-well plates with CNS myelin substrates using NGlOS-15 cells, an immortalized cell line that can be induced to produce extensive neuritic growth. The inhibition of neurite outgrowth by CNS myelin observed in this assay is: (1) observed for NGlOS-15 cells and also PC12 cells and primary superior cervical ganglion neurons, (2) contact dependent, (3) half-maximal at 5 pg/cm* of myelin, and (4) trypsin-labile. This assay is quantitative, rapid, highly reproducible, convenient and can be used to test compounds which have the potential to overcome the growth inhibitory molecules present in CNS myelin. Keywords: NG108-15; PC12; Neuron; Tissue culture; Axon; Central nervous system
1. Introduction
In adult mammals, axonal regeneration following injury occurs spontaneously in the peripheral nervous system (PNS) but not in the central nervous system (CNS). This failure is believed to be an important determinant of the irreversible loss of neurological function associated with injuries to the brain, optic nerve or spinal cord. Recently, molecular approaches have been employed in attempts to understand the mechanisms underlying the poor regenerative potential of the adult mammalian CNS and to develop strategies to improve the repair and recovery of neurological function. Several recent observations suggest that the adult mammalian CNS environment contains specific molecules that inhibit axonal outgrowth (Schwab et al., 1993). In particular, CNS myelin contains
’ Corresponding author: Division of Neurosurgery, 2-433 McLaughlin Pavilion, Toronto Western Hospital, 399 Bathurst Street, Toronto M5T 2S8, Canada. Tel.: (416) 603-6200; Fax: (416) 603-5298; E-mail: [email protected]. 016%0270/95/$09.50 SSDI
0165-0270(95)00081-X
0 1995 Elsevier Science B.V. All rights reserved
two molecules, designated NI3.5 and N1250 (Caroni and Schwab, 1988a) that block neurite growth and fibroblast spreading, and myelin-associated glycoproteins, which can also inhibit neurite growth in vitro (Mukhopadhyay et al., 1994; McKerracher et al., 1994). Monoclonal antibodies directed against certain of these inhibitors have been reported to enhance CNS regeneration in a number of animal models (Schnell and Schwab, 1990; Schnell et al., 1994). However, because of the difficulties associated with the study of CNS regeneration in vivo, further study would benefit from the availability of a simple model system which retains the important elements of the inhibition defined previously with primary neurons or in vivo. Moreover, the identification of specific neurite growth inhibitors has prompted attempts to identify agents capable of reversing their inhibitory activities for which a rapid screening assay would be desirable. A useful model should have the following properties. (1) It should reproduce the previously described inhibitory properties of CNS molecules on neurite outgrowth. (2) The assay should permit the testing and screening of compounds that can potentially enhance neurite outgrowth on
24
A.M. Lozano
et al. / Journal
of Neuroscience
inhibitory substrates. (3) The assay should be fast, convenient, reproducible and allow the testing of a large number of samples. This report describes the development of an in vitro tissue culture system using substrate bound molecules and neuronal cell lines, for the study of the inhibition of neurite outgrowth due to environmental determinants in the CNS.
Methods
63 (1995)
23-28
myelin basic protein confirmed a Cfold enrichment for myelin in the brain extract versus total brain homogenate. Protein concentrations were determined using a Bradford assay (Bio-Rad) and bovine serum albumin (Type IV; Sigma) as a standard. Extracts of adult rat sciatic nerve (peripheral nerve; PNS myelin) and muscle were prepared in a similar fashion. 2.3. Neurite outgrowth
2. Materials
assay
and methods
2.1. Cells
Rat pheochromocytoma PC12 cells were obtained from the American Type Culture Collection (Rockville, MD). Cells were grown in RPMI-1640 media (Gibco) with 15% fetal calf serum (FCS). PC12 cells were differentiated with 100 rig/ml of nerve growth factor (NGF) for 6-8 days (hereafter referred to as NGFPC12). Cells of the NG108-15 line (Christian et al., 19781 were a kind gift of Dr. G. Cheng (University of Manitoba). NG108-15 cells were grown in DME medium supplemented with 10% FCS and 1 X HAT (Gibco), and were induced to differentiate to the neuronal phenotype by reducing the serum to 5% concurrent with the addition of 1 mM dibutyryl cyclic adenosine monophosphate (dbcAMP) (Sigma) for 2-4 days (dbcAMPNG108-15). Primary superior cervical ganglion (SCG) neurons were obtained from newborn rats and cultured as previously described (Patterson and Chun, 1977; Hawrot and Patterson, 1979). NIH3T3 fibroblasts were grown in DME with 10% FCS. Penicillin (25 U/ml) and streptomycin (25 pg/ml) were added to all media. 2.2. Substrate preparation
CNS myelin was prepared from brains of Sprague-Dawley rats (250-300 g) using modifications of previously described procedures (Coleman et al., 1982; Caroni and Schwab, 1988a). Homogenization was carried out using for each gram of tissue, 10 ml of 0.25 M sucrose with 5 mM EDTA and 5 mM iodoacetamide (homogenization buffer) using a glass homogenizer. The homogenate was centrifuged at 2000 rpm in a Sorval HB-4 rotor for 3 min to pellet cell debris and nuclei. The supematant was layered over 20 ml of 0.85 M sucrose with 5 mM EDTA and 5 mM iodoacetamide in 38 ml SW-28 tubes (Beckman) and centrifuged at 4 C and 28 000 g for 1 h. The interface was collected, kept on ice and washed in 20 vol. of 30 mM Hepes pH 7.4 with 5 mM EDTA and 5 mM iodoacetamide. After centrifugation at 28000 g for 4 h, the pellet was resuspended in homogenization buffer and layered onto 0.85 M sucrose with protease inhibitors. The sample was recentrifuged (28 000 X g, 1 h), the resultant interface was again washed, pelleted at 28 000 X g for 4 h and resuspended in a small volume of 30 mM Hepes, pH 7.4. Western blot analysis using a monoclonal antibody against
Assays were done in 96-well plates (NunC). Substrate testing wells were precoated by filling with 100 pg/ml of poly-L-lysine (PLL) (Sigma) for 2 h at room temperature, followed by washing twice with phosphate buffered saline (PBS). Test wells were in addition coated by overnight drying of a suspension of containing 5-20 pg/cm’ of protein. Coating suspensions included bovine serum albumin (BSA Type IV; Sigma), adult rat brain myelin, sciatic nerve myelin, or muscle extracts. Substrate coated wells were UV light-treated and washed with PBS twice. One thousand to 2000 cells in 100 ~1 of appropriate medium were plated per well. After 24 h of culture at 37°C and 5% CO,, random fields were photographed with phase-contrast optics. Ten to 16 independent wells were scored for each substrate tested. The total number of cells per standardized photographic field was used as an index of attached cells. The percentage of cells bearing a process greater than 1 cell diameter in length was determined by a ‘blind’ scorer. Cells in close association with other cells, such that it could not be determined whether they had a process, were not counted. Neurite length was determined from phasecontrast photographs. In certain experiments the effect of substrate digestion with trypsin was studied. Substrate coated wells were treated with 0.25-0.00025% trypsin (Sigma T-2904) in PBS for 10 min at room temperature. Wells were washed twice with 10% FCS containing cell culture medium prior to the addition of cells.
3. Results To facilitate the study of the molecular basis for neurite growth inhibition in the CNS, we have focused on CNS myelin proteins as a model system. The influence of CNS myelin on cell attachment, fibroblast spreading and neurite outgrowth was tested in an in vitro assay. CNS myelin was prepared as a sucrose density fraction from adult rat brains following standard procedures for myelin isolation. We observed that CNS myelin impaired fibroblast s reading and neuritic outgrowth from NGFPC12 and d! CAMPNG108-15 neural cell lines and primary rat superior cervical ganglion neurons (Fig. 1). Cells on CNS myelin were generally round 24 h after plating while cells on PLL displayed greater spreading, more neurites and longer pro-
A.M. Lozano
et al. / Journal
of Neuroscience
Methods
63 (1995)
23--2X
PLL
SCG Fig. I, Cell growth on CNS myelin. Dissociated newborn tissue culture plastic pretreated with PLL (poly-L-lysine) after 74 h in culture.
rat SCG neurons, NGFPCl 2 cetls, “bcA”‘PNG108-15 cells or NIH3T3 fibroblasts. were plated onto alone (upper) or PLL followed by 20 wg/cm’ of CNS myelin (lower). Cells were photographed
cesses. Because of their ease of preparation, rapid elaboration of long neurites (within 24 h) and rapid cell proliferation we used NGIOS-15 cells for quantitative studies.
The propensity of NGlOS-15 cells grown in dibutyryl cyclic adenosine monophosphate CdbcAMPNG108- 15) to produce neurites was strongly influenced by their substrate
Fig. 2. Substrate influence on NG108-15 cell growth. Photomicrographs of representative fields of cultures of dbcAMPNG108-I.5 cells plated onto PLL alone (a), PLL followed by 20 pg/cm’ BSA (b) and PLL followed by 20 pg/cm2 CNS myelin cc). Panel (d) shows a single dhcAMPNG108 cell growing on a myelin-free patch. The border between the myelin-coated (CNS) and uncoated (‘pL’ for PLL) surfaces is emphasized with small arrowheads.
26
A.M. Lozano
et aL/Journal
of Neuroscience
Me?hods
60
63 (199s)
23-28
25
50 20 40
CELLS WITH
CELLS ATTACHE;’
30
NEURITES
10 20 5
PL
85
820
M5
M20
P5
P20
c5
0
c20
PL
85
SUBSTRATES Fig. 3. CNS myelin inhibits neurite growth by dbcAMPNG108 cells. Equal numbers of dbcAMPNG108 cells were plated in wells pretreated with PLL alone, or PLL followed by a coating of 5 or 20 pg/cm* of BSA (B5, B20), muscle membrane protein extract (M5, M20), PNS myelin proteins (PS, P20) or CNS myelin proteins (C5, C20). After 24 h, random fields were photographed and the proportion of cells with a process greater than 1 cell diameter was determined. Ten to 16 independent wells were scored for each substrate. The error bars represent the SEM. The differences the percentage of cells with processes on PLL versus other substrates was significant at P < 0.05 for B5, B20, M5, M20 and P5; P < 0.01 for P20 and P < 0.005 for C5 and P < 0.0005 for C20 using a Student’s t test.
(Fig. 2). As shown in Fig. 3, over 60% of cells plated onto PLL had long neurites at 24 h. Cells plated on BSA, muscle proteins or PNS myelin showed a propensity for neurite elaboration that was moderately lower than that seen on PLL (P < 0.05). Increasing the substrate concentration Cfold for BSA and muscle protein had no significant effect on the fraction of cells with neuritic processes. Increasing PNS myelin concentration 4-fold was associated with a small further decrease in the percentage of cells with neurites (P < 0.01). In contrast, the fraction of cells elaborating long neurites on CNS myelin was greatly reduced. The percentage of dbcAMPNG108-15 cells having a process greater than 1 cell body in diameter on 5, 10 or 20 @g/cm’ of CNS myelin was significantly reduced compared to PLL (Fig. mt
820
M5
M20
P5
P20
c5
c20
SUBSTRATES Fig. 5. Cell attachment in cultures of dbcAMPNG108 cells on myelin and control substrates. Equal numbers of dbcAMPNG108-15 cells were plated in wells pretreated with PLL alone, or PLL followed by BSA (B5, B20), muscle membrane protein extract (M5, M20), PNS myelin proteins (P5, P20) or CNS myelin proteins (C5, C20) at 5-20 pg/cm2. After 24 h, random fields were photographed and cells counted. Ten to 16 independent wells were scored for each substrate. Compared to PLL, only the CNS myelin protein substrate plated at 20 pg/cm2 had significantly less dbcAMPNG108-15 cell attachment P < 0.05. The number of attached cells on BSA, muscle proteins and PNS myelin at 5 +g/cm* was significantly higher than for PLL (P < 0.05). Attachment on P20 and C5 was not significantly different from PLL.
3). The half-maximal inhibition of neurite outgrowth observed at approximately 5 pg of protein/cm’ (Fig. 4). A similar concentration-dependent inhibition of neurite growth on CNS myelin was observed with NGF-treated PC12 cells and primary SCG neurons (not shown). On 20 pg/cm’ of CNS myelin, less than 6% of NG108-15 cells had processes at 24 h (P < 0.0005). This IO-fold difference in the proportion of process-bearing cells on CNS myelin versus PLL was not due to differential attachment, since the number of cells attached differed by less than 2-fold (Fig. 5). CNS myelin also decreased the length of neurites in cultures of dbcAMPNG108-15 cells (Table 1). On PLL, the average process length per cell was 20-fold greater than measured on 20 pg/cm2 CNS myelin. The inhibition of neurite growth did not appear to be due to a diffusible toxic agent since neurites grew apparently unimpeded on a small myelin-free patch in a well where Table 1 CNS myelin decreases neurite length in cultures of dbcAMPNG108-15 cells Mean measured neurite Cells counted substrate length/cell
6 MYELIN
10
16
60
(&xi)
Fig. 4. Concentration dependent inhibition of neurite outgrowth by CNS myelin. CNS myelin was applied at several concentrations onto PLLcoated wells. The fraction of process-bearing dbcAMPNG108-15 cells was determined at 24 h. Values represent the mean of 2-10 we&f SE.
Poly-L-lysine CNS myelin
pm 4wn
18
85 12
dbcAMPNG108-15 cells were plated on PLL alone or PLL coated with 10 pg/cm* CNS myelin proteins. After 24 h representative fields were photographed with phase-contrast microscopy. Cells were counted and neurites measured from photographs. This data is representative of that obtained in 5 independent experiments.
A.M. Lozano
et al. /Journal
of Neuroscience
greater than 90% of the surface was covered with CNS myelin (Fig. 2D). Cells placed at a CNS myelin/PLL boundary had neurites extending on PLL but which did not extend onto the myelin surface. Treatment of the CNS myelin substrate with trypsin in PBS for 10 min at 37°C resulted a concentration-dependent decrease in inhibitory activity, with 0.0025% trypsin reducing activity to 50% and 0.00025% trypsin to 90% of control.
4. Discussion Previous quantitative work on CNS myelin inhibition has relied on fibroblast spreading as a surrogate for neurite extension (Caroni and Schwab, 1988a, b), due to the difficulty of repeatedly generating reproducible cultures of primary neurons. We have explored the use of cell lines with neuronal properties that include inducible growth of neurites, in order to increase the relevance of the assay to CNS phenomena, while maintaining the advantages of established cell lines. These advantages are chiefly convenience, reproducibility and cost. Both NGFPC12 and dbcAMPNG 10X-15 cells were found to be inhibited in their neurite growth behaviour with parameters similar to those reported by others (Caroni and Schwab, 1988a) and observed here, for primary neurons. The inhibitory activity of CNS myelin substrate in our assay is similar to that reported by others in several ways: (1) the source of inhibitor from adult rat CNS myelin, (2) the method of myelin preparation, (3) the inhibitory activity per microgram of myelin protein substrate, (4) the contact dependence of the myelin inhibition of neurite growth, and (5) the trypsin lability of the inhibitory effects of myelin. In our hands, the inhibitory effects of CNS myelin were seen on both neuronal cells (neurite growth) and non-neuronal cells (spreading), as described previously (Caroni and Schwab, 1988a). The poor growth of dbcAMPNG108-15 neurites on CNS myelin substrates is due to inhibition rather than the lack of availability of trophic support, because the inhibitory properties of the substrate increase in direct relation to the amount of myelin plated. In wells where myelin substrates cover only a portion of the total surface area, we have observed that neural cells in direct contact with myelin have impaired neurite outgrowth. In contrast, other neurons in the same well attached on a myelin-free patch elaborated processes in a unimpeded fashion. Indeed neurons growing on PLL near a border with a CNS myelincoated region extend neurites onto the myelin only very rarely and for short distances. These observations indicate that the inhibition of neurite growth observed with CNS myelin is mediated through a non-diffusible, contact mediated mechanism. The assay we describe assesses neurite outgrowth in the ongoing presence of the inhibitor. The time scale of the
Methods
63 I 1995) 23 28
7-l
assay we present differs from the previously described acute growth cone collapse assays using solutions containing myelin components (Igarashi et al., 1993; Bandtlow et al., 1993), thrombin (Suidan et al., 1992) collapsin (Luo et al., 1993), lysophosphatidic acid or ATP. phorbol esters, trypsin and nocodazole (Smalheiser, 1993; Jalink et al., 1994). While there may be commonalities between the short-term (seconds-minutes) phenomenon of growth cone collapse and the long-term arrest of neurite growth, the molecular mechanisms underlying these behaviours remain largely uncharacterized. Because of their high proliferative activity and rapid elaboration of neurites, NGlOS-15 cells are well suited to study cell morphology and differentiation (Ethel et al., 1993), acute growth cone collapse (Smalheiser, 1993) and neurite outgrowth on myelin substrates as described here. The biological assay in this study can yield results within 24 h, is reproducible and is easily quantitated. The substrates used in the assay can be prepared in large quantities and stored frozen. The aliquoted frozen material retains its biological activity. In summary, by using established cell lines with inducible neurite growth properties we have shown that inhibition of neurite growth by CNS substrates can be conveniently studied in an in vitro model system. The method, described in detail here. has recently been used successfully to study the effects of rat myelin associated glycoprotein on neurite outgrowth (McKerracher et al., 1994) and to screen a panel of monoclonal antibodies for molecular reagents capable of modulating neurite growth in the presence of inhibitory CNS molecules (Lozano et al., 1995).
Acknowledgements We thank Chaoying Li for expert technical assistance and Dr. G. Cheng (U. of Manitoba) for the gift of NGlO815 cells. We thank Dr. J. Roder for his support, advice. A.M.L. and A.R. were the recipients of Clinician-Scientist and Scholarship awards, respectively, from the Medical Research Council of Canada. This work was funded by the Canadian Network of Centres for Excellence for Neural Regeneration and Functional Recovery.
References Bandtlow, C.E., Schmidt, M.F., Hassinger, ‘T.D., Schwab, M.E. and Kater, S.B. (1993) Role of intracellular calcium in NT-35.evoked collapse of neuronal growth cones. Science, 259: 80-83. Caroni, P. and Schwab, M.E. (1988a) Two membrane protein fractions from rat central myehn with inhibitory properties for neurite growth and fibroblast spreading, J. Cell Biol., 106: 1281-1288. Caroni, P. and Schwab, ME. (1988bJ Antibody against myelin-associated inhibitor of neurite growth neutralizes nonpermissive properties of CNS white matter, Neuron, 1: 85-96.
28
A.M. Lozano
et al. /Journal
of Neuroscience
Christian, C.N., Nelson, P.G., Bullock, P., Mullinax, D. and Nirenberg, M. (1978) Pharmacologic response of cells of a neuroblastomax glioma hybrid clone and modulation of synapses between hybrid cells and mouse myotubes, Brain Res., 147: 261-276. Coleman, D.R., Kreibich, G., Fmi, A.B. and Sabatini, D.D. (1982) Synthesis and incorporation of myelin polypeptides into CNS myelin, J. Cell Biol., 95: 598-608. Ethel, D.W., Steeves, J.D., Jordan, L.M. and Cheng, K.W. (1993) Developmental transition by spinal cord plasma membranes of embryonic chick from permissive to restrictive substrates for the morphological differentiation of neuroblastoma X glioma hybrid NG 108- 15 cell. Dev. Brain. Res., 72: 1-8. Hawrot, E. and Patterson, P.H. (1979) Long-term culture of dissociated sympathetic neurons, Meth. Enzymol., 58: 574-585. Igarashi. M., St&matter, S.M., Vartanian, T. and Fishman, M.C. (1993) Mediation by G proteins of signals that cause collapse of growth cones. Science, 259: 77-79. Jalink, K., van Corven, E.J., Hengeveld, T., Morii, N., Narumiya, S. and Moolenaar, W.H. (1994) Inhibition of lysophosphatidate and thrombin-induced nemite retraction and neuronal cell rounding by ADP ribosylation of the small GTP binding protein Rho. J. Cell Biol., 126: 801-810 Lozano, A.M., Labes, M., Roder, J. and Roach, A.R. (1995) An antineuronal monoclonal antibody that reverses neurite growth inhibition by CNS myelin. J. Neurosci. Res., in press. Luo, Y., Raible, D. and Raper, J.A. (1993) Collapsin: a protein in brain that induces the collapse of neuronal growth cones. Cell, 75: 217-227.
Methods
63 (1995)
23-28
McKerracher, L., David, S., Jackson, D.L., Kottis, V., Dunn, R.J. and Braun, P.E. (1994) Identification of myelin-associated glycoprotein as a major myelin-derived inhibitor of neurite growth, Neuron, 13: 805-811. Mukhopadhyay, G., Doherty, P., Walsh, F.S., Cracker, P.R. and Filbin, M.T. (1994) A novel role for myelin-associated glycoprotein as an inhibitor of axonal regeneration, Neuron, 13: 757-767. Patterson, P.H. and Chun, L.L.Y. (1977) The induction of acetylcholine synthesis in primary cultures of dissociated rat sympathetic neurons, Dev. Biol., 56: 263-280. Schnell, L., Schnieder, R., Kolbeck, R., Barde, Y.-A. and Schwab, M.E. (1994) Neurotrophin-3 enhances sprouting of corticospinal tract during development and after adult spinal cord lesion, Nature, 7: 170- 173. Schnell, L. and Schwab, M.E. (1990) Axonal regeneration in the rat spinal cord produced by an antibody against myelin-associated neurite growth inhibitors, Nature, 343: 269-272. Schwab, M.E., Kapfhammer, J.P. and Bandtlow, C.E. (1993) inhibitors of neurite growth, Annu. Rev. Neurosci., 16:565-595. Smalheiser, N.R. (1993) Acute neurite retraction elicited by diverse agents is prevented by genistein, a tyrosine kinase inhibitor. J. Neurochem., 61: 340-343. Suidan, H.S., Stone, S.R., Hemmings, B.A. and Monard, D. (1992) Thrombin causes neurite retraction in neuronal cells through activation of cell surface receptors. Neuron, 8: 3-375.
Journal of Neuroscience Methods 63 (1995) 23-28
A convenient in vitro assay for the inhibition of neurite outgrowth by adult mammalian CNS myelin using immortalized neuronal cells Andres M. Lozano a,b~c, * , Matthias Schmidt a-c,Arthur Roach a~’ a Division
of Immunology and Neurobiology, b Division of Neurosurgery, ’ Department of Molecular
Samuel Lunenfeld Research Institute, Mount Sinai Hospital, Toronto. Department of Surgery, University of Toronto, Toronto, Canada and Medical Genetics, University of Toronto, Toronto, Canada
Canada
Received 5 January 1995; revised 10 May 1995; accepted 4 June 1995 _._..-.---_---
Abstract The adult mammaiian CNS contains molecules which inhibit neurite outgrowth and which may be responsible for the lack of successful axonal regeneration after injuries in the brain and spinal cord. We describe an in vitro assay to measure the ability of primary and established lines of neuronal cells to produce neurites in the presence of CNS inhibitory molecules. The assay is suitable for identification of agents and treatments to overcome neurite growth inhibition. Assays are carried out in 96-well plates with CNS myelin substrates using NGlOS-15 cells, an immortalized cell line that can be induced to produce extensive neuritic growth. The inhibition of neurite outgrowth by CNS myelin observed in this assay is: (1) observed for NGlOS-15 cells and also PC12 cells and primary superior cervical ganglion neurons, (2) contact dependent, (3) half-maximal at 5 pg/cm* of myelin, and (4) trypsin-labile. This assay is quantitative, rapid, highly reproducible, convenient and can be used to test compounds which have the potential to overcome the growth inhibitory molecules present in CNS myelin. Keywords: NG108-15; PC12; Neuron; Tissue culture; Axon; Central nervous system
1. Introduction
In adult mammals, axonal regeneration following injury occurs spontaneously in the peripheral nervous system (PNS) but not in the central nervous system (CNS). This failure is believed to be an important determinant of the irreversible loss of neurological function associated with injuries to the brain, optic nerve or spinal cord. Recently, molecular approaches have been employed in attempts to understand the mechanisms underlying the poor regenerative potential of the adult mammalian CNS and to develop strategies to improve the repair and recovery of neurological function. Several recent observations suggest that the adult mammalian CNS environment contains specific molecules that inhibit axonal outgrowth (Schwab et al., 1993). In particular, CNS myelin contains
’ Corresponding author: Division of Neurosurgery, 2-433 McLaughlin Pavilion, Toronto Western Hospital, 399 Bathurst Street, Toronto M5T 2S8, Canada. Tel.: (416) 603-6200; Fax: (416) 603-5298; E-mail: [email protected]. 016%0270/95/$09.50 SSDI
0165-0270(95)00081-X
0 1995 Elsevier Science B.V. All rights reserved
two molecules, designated NI3.5 and N1250 (Caroni and Schwab, 1988a) that block neurite growth and fibroblast spreading, and myelin-associated glycoproteins, which can also inhibit neurite growth in vitro (Mukhopadhyay et al., 1994; McKerracher et al., 1994). Monoclonal antibodies directed against certain of these inhibitors have been reported to enhance CNS regeneration in a number of animal models (Schnell and Schwab, 1990; Schnell et al., 1994). However, because of the difficulties associated with the study of CNS regeneration in vivo, further study would benefit from the availability of a simple model system which retains the important elements of the inhibition defined previously with primary neurons or in vivo. Moreover, the identification of specific neurite growth inhibitors has prompted attempts to identify agents capable of reversing their inhibitory activities for which a rapid screening assay would be desirable. A useful model should have the following properties. (1) It should reproduce the previously described inhibitory properties of CNS molecules on neurite outgrowth. (2) The assay should permit the testing and screening of compounds that can potentially enhance neurite outgrowth on
24
A.M. Lozano
et al. / Journal
of Neuroscience
inhibitory substrates. (3) The assay should be fast, convenient, reproducible and allow the testing of a large number of samples. This report describes the development of an in vitro tissue culture system using substrate bound molecules and neuronal cell lines, for the study of the inhibition of neurite outgrowth due to environmental determinants in the CNS.
Methods
63 (1995)
23-28
myelin basic protein confirmed a Cfold enrichment for myelin in the brain extract versus total brain homogenate. Protein concentrations were determined using a Bradford assay (Bio-Rad) and bovine serum albumin (Type IV; Sigma) as a standard. Extracts of adult rat sciatic nerve (peripheral nerve; PNS myelin) and muscle were prepared in a similar fashion. 2.3. Neurite outgrowth
2. Materials
assay
and methods
2.1. Cells
Rat pheochromocytoma PC12 cells were obtained from the American Type Culture Collection (Rockville, MD). Cells were grown in RPMI-1640 media (Gibco) with 15% fetal calf serum (FCS). PC12 cells were differentiated with 100 rig/ml of nerve growth factor (NGF) for 6-8 days (hereafter referred to as NGFPC12). Cells of the NG108-15 line (Christian et al., 19781 were a kind gift of Dr. G. Cheng (University of Manitoba). NG108-15 cells were grown in DME medium supplemented with 10% FCS and 1 X HAT (Gibco), and were induced to differentiate to the neuronal phenotype by reducing the serum to 5% concurrent with the addition of 1 mM dibutyryl cyclic adenosine monophosphate (dbcAMP) (Sigma) for 2-4 days (dbcAMPNG108-15). Primary superior cervical ganglion (SCG) neurons were obtained from newborn rats and cultured as previously described (Patterson and Chun, 1977; Hawrot and Patterson, 1979). NIH3T3 fibroblasts were grown in DME with 10% FCS. Penicillin (25 U/ml) and streptomycin (25 pg/ml) were added to all media. 2.2. Substrate preparation
CNS myelin was prepared from brains of Sprague-Dawley rats (250-300 g) using modifications of previously described procedures (Coleman et al., 1982; Caroni and Schwab, 1988a). Homogenization was carried out using for each gram of tissue, 10 ml of 0.25 M sucrose with 5 mM EDTA and 5 mM iodoacetamide (homogenization buffer) using a glass homogenizer. The homogenate was centrifuged at 2000 rpm in a Sorval HB-4 rotor for 3 min to pellet cell debris and nuclei. The supematant was layered over 20 ml of 0.85 M sucrose with 5 mM EDTA and 5 mM iodoacetamide in 38 ml SW-28 tubes (Beckman) and centrifuged at 4 C and 28 000 g for 1 h. The interface was collected, kept on ice and washed in 20 vol. of 30 mM Hepes pH 7.4 with 5 mM EDTA and 5 mM iodoacetamide. After centrifugation at 28000 g for 4 h, the pellet was resuspended in homogenization buffer and layered onto 0.85 M sucrose with protease inhibitors. The sample was recentrifuged (28 000 X g, 1 h), the resultant interface was again washed, pelleted at 28 000 X g for 4 h and resuspended in a small volume of 30 mM Hepes, pH 7.4. Western blot analysis using a monoclonal antibody against
Assays were done in 96-well plates (NunC). Substrate testing wells were precoated by filling with 100 pg/ml of poly-L-lysine (PLL) (Sigma) for 2 h at room temperature, followed by washing twice with phosphate buffered saline (PBS). Test wells were in addition coated by overnight drying of a suspension of containing 5-20 pg/cm’ of protein. Coating suspensions included bovine serum albumin (BSA Type IV; Sigma), adult rat brain myelin, sciatic nerve myelin, or muscle extracts. Substrate coated wells were UV light-treated and washed with PBS twice. One thousand to 2000 cells in 100 ~1 of appropriate medium were plated per well. After 24 h of culture at 37°C and 5% CO,, random fields were photographed with phase-contrast optics. Ten to 16 independent wells were scored for each substrate tested. The total number of cells per standardized photographic field was used as an index of attached cells. The percentage of cells bearing a process greater than 1 cell diameter in length was determined by a ‘blind’ scorer. Cells in close association with other cells, such that it could not be determined whether they had a process, were not counted. Neurite length was determined from phasecontrast photographs. In certain experiments the effect of substrate digestion with trypsin was studied. Substrate coated wells were treated with 0.25-0.00025% trypsin (Sigma T-2904) in PBS for 10 min at room temperature. Wells were washed twice with 10% FCS containing cell culture medium prior to the addition of cells.
3. Results To facilitate the study of the molecular basis for neurite growth inhibition in the CNS, we have focused on CNS myelin proteins as a model system. The influence of CNS myelin on cell attachment, fibroblast spreading and neurite outgrowth was tested in an in vitro assay. CNS myelin was prepared as a sucrose density fraction from adult rat brains following standard procedures for myelin isolation. We observed that CNS myelin impaired fibroblast s reading and neuritic outgrowth from NGFPC12 and d! CAMPNG108-15 neural cell lines and primary rat superior cervical ganglion neurons (Fig. 1). Cells on CNS myelin were generally round 24 h after plating while cells on PLL displayed greater spreading, more neurites and longer pro-
A.M. Lozano
et al. / Journal
of Neuroscience
Methods
63 (1995)
23--2X
PLL
SCG Fig. I, Cell growth on CNS myelin. Dissociated newborn tissue culture plastic pretreated with PLL (poly-L-lysine) after 74 h in culture.
rat SCG neurons, NGFPCl 2 cetls, “bcA”‘PNG108-15 cells or NIH3T3 fibroblasts. were plated onto alone (upper) or PLL followed by 20 wg/cm’ of CNS myelin (lower). Cells were photographed
cesses. Because of their ease of preparation, rapid elaboration of long neurites (within 24 h) and rapid cell proliferation we used NGIOS-15 cells for quantitative studies.
The propensity of NGlOS-15 cells grown in dibutyryl cyclic adenosine monophosphate CdbcAMPNG108- 15) to produce neurites was strongly influenced by their substrate
Fig. 2. Substrate influence on NG108-15 cell growth. Photomicrographs of representative fields of cultures of dbcAMPNG108-I.5 cells plated onto PLL alone (a), PLL followed by 20 pg/cm’ BSA (b) and PLL followed by 20 pg/cm2 CNS myelin cc). Panel (d) shows a single dhcAMPNG108 cell growing on a myelin-free patch. The border between the myelin-coated (CNS) and uncoated (‘pL’ for PLL) surfaces is emphasized with small arrowheads.
26
A.M. Lozano
et aL/Journal
of Neuroscience
Me?hods
60
63 (199s)
23-28
25
50 20 40
CELLS WITH
CELLS ATTACHE;’
30
NEURITES
10 20 5
PL
85
820
M5
M20
P5
P20
c5
0
c20
PL
85
SUBSTRATES Fig. 3. CNS myelin inhibits neurite growth by dbcAMPNG108 cells. Equal numbers of dbcAMPNG108 cells were plated in wells pretreated with PLL alone, or PLL followed by a coating of 5 or 20 pg/cm* of BSA (B5, B20), muscle membrane protein extract (M5, M20), PNS myelin proteins (PS, P20) or CNS myelin proteins (C5, C20). After 24 h, random fields were photographed and the proportion of cells with a process greater than 1 cell diameter was determined. Ten to 16 independent wells were scored for each substrate. The error bars represent the SEM. The differences the percentage of cells with processes on PLL versus other substrates was significant at P < 0.05 for B5, B20, M5, M20 and P5; P < 0.01 for P20 and P < 0.005 for C5 and P < 0.0005 for C20 using a Student’s t test.
(Fig. 2). As shown in Fig. 3, over 60% of cells plated onto PLL had long neurites at 24 h. Cells plated on BSA, muscle proteins or PNS myelin showed a propensity for neurite elaboration that was moderately lower than that seen on PLL (P < 0.05). Increasing the substrate concentration Cfold for BSA and muscle protein had no significant effect on the fraction of cells with neuritic processes. Increasing PNS myelin concentration 4-fold was associated with a small further decrease in the percentage of cells with neurites (P < 0.01). In contrast, the fraction of cells elaborating long neurites on CNS myelin was greatly reduced. The percentage of dbcAMPNG108-15 cells having a process greater than 1 cell body in diameter on 5, 10 or 20 @g/cm’ of CNS myelin was significantly reduced compared to PLL (Fig. mt
820
M5
M20
P5
P20
c5
c20
SUBSTRATES Fig. 5. Cell attachment in cultures of dbcAMPNG108 cells on myelin and control substrates. Equal numbers of dbcAMPNG108-15 cells were plated in wells pretreated with PLL alone, or PLL followed by BSA (B5, B20), muscle membrane protein extract (M5, M20), PNS myelin proteins (P5, P20) or CNS myelin proteins (C5, C20) at 5-20 pg/cm2. After 24 h, random fields were photographed and cells counted. Ten to 16 independent wells were scored for each substrate. Compared to PLL, only the CNS myelin protein substrate plated at 20 pg/cm2 had significantly less dbcAMPNG108-15 cell attachment P < 0.05. The number of attached cells on BSA, muscle proteins and PNS myelin at 5 +g/cm* was significantly higher than for PLL (P < 0.05). Attachment on P20 and C5 was not significantly different from PLL.
3). The half-maximal inhibition of neurite outgrowth observed at approximately 5 pg of protein/cm’ (Fig. 4). A similar concentration-dependent inhibition of neurite growth on CNS myelin was observed with NGF-treated PC12 cells and primary SCG neurons (not shown). On 20 pg/cm’ of CNS myelin, less than 6% of NG108-15 cells had processes at 24 h (P < 0.0005). This IO-fold difference in the proportion of process-bearing cells on CNS myelin versus PLL was not due to differential attachment, since the number of cells attached differed by less than 2-fold (Fig. 5). CNS myelin also decreased the length of neurites in cultures of dbcAMPNG108-15 cells (Table 1). On PLL, the average process length per cell was 20-fold greater than measured on 20 pg/cm2 CNS myelin. The inhibition of neurite growth did not appear to be due to a diffusible toxic agent since neurites grew apparently unimpeded on a small myelin-free patch in a well where Table 1 CNS myelin decreases neurite length in cultures of dbcAMPNG108-15 cells Mean measured neurite Cells counted substrate length/cell
6 MYELIN
10
16
60
(&xi)
Fig. 4. Concentration dependent inhibition of neurite outgrowth by CNS myelin. CNS myelin was applied at several concentrations onto PLLcoated wells. The fraction of process-bearing dbcAMPNG108-15 cells was determined at 24 h. Values represent the mean of 2-10 we&f SE.
Poly-L-lysine CNS myelin
pm 4wn
18
85 12
dbcAMPNG108-15 cells were plated on PLL alone or PLL coated with 10 pg/cm* CNS myelin proteins. After 24 h representative fields were photographed with phase-contrast microscopy. Cells were counted and neurites measured from photographs. This data is representative of that obtained in 5 independent experiments.
A.M. Lozano
et al. /Journal
of Neuroscience
greater than 90% of the surface was covered with CNS myelin (Fig. 2D). Cells placed at a CNS myelin/PLL boundary had neurites extending on PLL but which did not extend onto the myelin surface. Treatment of the CNS myelin substrate with trypsin in PBS for 10 min at 37°C resulted a concentration-dependent decrease in inhibitory activity, with 0.0025% trypsin reducing activity to 50% and 0.00025% trypsin to 90% of control.
4. Discussion Previous quantitative work on CNS myelin inhibition has relied on fibroblast spreading as a surrogate for neurite extension (Caroni and Schwab, 1988a, b), due to the difficulty of repeatedly generating reproducible cultures of primary neurons. We have explored the use of cell lines with neuronal properties that include inducible growth of neurites, in order to increase the relevance of the assay to CNS phenomena, while maintaining the advantages of established cell lines. These advantages are chiefly convenience, reproducibility and cost. Both NGFPC12 and dbcAMPNG 10X-15 cells were found to be inhibited in their neurite growth behaviour with parameters similar to those reported by others (Caroni and Schwab, 1988a) and observed here, for primary neurons. The inhibitory activity of CNS myelin substrate in our assay is similar to that reported by others in several ways: (1) the source of inhibitor from adult rat CNS myelin, (2) the method of myelin preparation, (3) the inhibitory activity per microgram of myelin protein substrate, (4) the contact dependence of the myelin inhibition of neurite growth, and (5) the trypsin lability of the inhibitory effects of myelin. In our hands, the inhibitory effects of CNS myelin were seen on both neuronal cells (neurite growth) and non-neuronal cells (spreading), as described previously (Caroni and Schwab, 1988a). The poor growth of dbcAMPNG108-15 neurites on CNS myelin substrates is due to inhibition rather than the lack of availability of trophic support, because the inhibitory properties of the substrate increase in direct relation to the amount of myelin plated. In wells where myelin substrates cover only a portion of the total surface area, we have observed that neural cells in direct contact with myelin have impaired neurite outgrowth. In contrast, other neurons in the same well attached on a myelin-free patch elaborated processes in a unimpeded fashion. Indeed neurons growing on PLL near a border with a CNS myelincoated region extend neurites onto the myelin only very rarely and for short distances. These observations indicate that the inhibition of neurite growth observed with CNS myelin is mediated through a non-diffusible, contact mediated mechanism. The assay we describe assesses neurite outgrowth in the ongoing presence of the inhibitor. The time scale of the
Methods
63 I 1995) 23 28
7-l
assay we present differs from the previously described acute growth cone collapse assays using solutions containing myelin components (Igarashi et al., 1993; Bandtlow et al., 1993), thrombin (Suidan et al., 1992) collapsin (Luo et al., 1993), lysophosphatidic acid or ATP. phorbol esters, trypsin and nocodazole (Smalheiser, 1993; Jalink et al., 1994). While there may be commonalities between the short-term (seconds-minutes) phenomenon of growth cone collapse and the long-term arrest of neurite growth, the molecular mechanisms underlying these behaviours remain largely uncharacterized. Because of their high proliferative activity and rapid elaboration of neurites, NGlOS-15 cells are well suited to study cell morphology and differentiation (Ethel et al., 1993), acute growth cone collapse (Smalheiser, 1993) and neurite outgrowth on myelin substrates as described here. The biological assay in this study can yield results within 24 h, is reproducible and is easily quantitated. The substrates used in the assay can be prepared in large quantities and stored frozen. The aliquoted frozen material retains its biological activity. In summary, by using established cell lines with inducible neurite growth properties we have shown that inhibition of neurite growth by CNS substrates can be conveniently studied in an in vitro model system. The method, described in detail here. has recently been used successfully to study the effects of rat myelin associated glycoprotein on neurite outgrowth (McKerracher et al., 1994) and to screen a panel of monoclonal antibodies for molecular reagents capable of modulating neurite growth in the presence of inhibitory CNS molecules (Lozano et al., 1995).
Acknowledgements We thank Chaoying Li for expert technical assistance and Dr. G. Cheng (U. of Manitoba) for the gift of NGlO815 cells. We thank Dr. J. Roder for his support, advice. A.M.L. and A.R. were the recipients of Clinician-Scientist and Scholarship awards, respectively, from the Medical Research Council of Canada. This work was funded by the Canadian Network of Centres for Excellence for Neural Regeneration and Functional Recovery.
References Bandtlow, C.E., Schmidt, M.F., Hassinger, ‘T.D., Schwab, M.E. and Kater, S.B. (1993) Role of intracellular calcium in NT-35.evoked collapse of neuronal growth cones. Science, 259: 80-83. Caroni, P. and Schwab, M.E. (1988a) Two membrane protein fractions from rat central myehn with inhibitory properties for neurite growth and fibroblast spreading, J. Cell Biol., 106: 1281-1288. Caroni, P. and Schwab, ME. (1988bJ Antibody against myelin-associated inhibitor of neurite growth neutralizes nonpermissive properties of CNS white matter, Neuron, 1: 85-96.
28
A.M. Lozano
et al. /Journal
of Neuroscience
Christian, C.N., Nelson, P.G., Bullock, P., Mullinax, D. and Nirenberg, M. (1978) Pharmacologic response of cells of a neuroblastomax glioma hybrid clone and modulation of synapses between hybrid cells and mouse myotubes, Brain Res., 147: 261-276. Coleman, D.R., Kreibich, G., Fmi, A.B. and Sabatini, D.D. (1982) Synthesis and incorporation of myelin polypeptides into CNS myelin, J. Cell Biol., 95: 598-608. Ethel, D.W., Steeves, J.D., Jordan, L.M. and Cheng, K.W. (1993) Developmental transition by spinal cord plasma membranes of embryonic chick from permissive to restrictive substrates for the morphological differentiation of neuroblastoma X glioma hybrid NG 108- 15 cell. Dev. Brain. Res., 72: 1-8. Hawrot, E. and Patterson, P.H. (1979) Long-term culture of dissociated sympathetic neurons, Meth. Enzymol., 58: 574-585. Igarashi. M., St&matter, S.M., Vartanian, T. and Fishman, M.C. (1993) Mediation by G proteins of signals that cause collapse of growth cones. Science, 259: 77-79. Jalink, K., van Corven, E.J., Hengeveld, T., Morii, N., Narumiya, S. and Moolenaar, W.H. (1994) Inhibition of lysophosphatidate and thrombin-induced nemite retraction and neuronal cell rounding by ADP ribosylation of the small GTP binding protein Rho. J. Cell Biol., 126: 801-810 Lozano, A.M., Labes, M., Roder, J. and Roach, A.R. (1995) An antineuronal monoclonal antibody that reverses neurite growth inhibition by CNS myelin. J. Neurosci. Res., in press. Luo, Y., Raible, D. and Raper, J.A. (1993) Collapsin: a protein in brain that induces the collapse of neuronal growth cones. Cell, 75: 217-227.
Methods
63 (1995)
23-28
McKerracher, L., David, S., Jackson, D.L., Kottis, V., Dunn, R.J. and Braun, P.E. (1994) Identification of myelin-associated glycoprotein as a major myelin-derived inhibitor of neurite growth, Neuron, 13: 805-811. Mukhopadhyay, G., Doherty, P., Walsh, F.S., Cracker, P.R. and Filbin, M.T. (1994) A novel role for myelin-associated glycoprotein as an inhibitor of axonal regeneration, Neuron, 13: 757-767. Patterson, P.H. and Chun, L.L.Y. (1977) The induction of acetylcholine synthesis in primary cultures of dissociated rat sympathetic neurons, Dev. Biol., 56: 263-280. Schnell, L., Schnieder, R., Kolbeck, R., Barde, Y.-A. and Schwab, M.E. (1994) Neurotrophin-3 enhances sprouting of corticospinal tract during development and after adult spinal cord lesion, Nature, 7: 170- 173. Schnell, L. and Schwab, M.E. (1990) Axonal regeneration in the rat spinal cord produced by an antibody against myelin-associated neurite growth inhibitors, Nature, 343: 269-272. Schwab, M.E., Kapfhammer, J.P. and Bandtlow, C.E. (1993) inhibitors of neurite growth, Annu. Rev. Neurosci., 16:565-595. Smalheiser, N.R. (1993) Acute neurite retraction elicited by diverse agents is prevented by genistein, a tyrosine kinase inhibitor. J. Neurochem., 61: 340-343. Suidan, H.S., Stone, S.R., Hemmings, B.A. and Monard, D. (1992) Thrombin causes neurite retraction in neuronal cells through activation of cell surface receptors. Neuron, 8: 3-375.