An in vitro bioassay for neurite growth using cryostat sections of nervous tissue as a substratum

Journal of Neuroscience Methods, 39 (1991) 193-202

193

cv) 1991 Elsevier Science Publishers B.V. All rights reserved 0165-0270/91/$03.5(I NSM 01282

An in vitro bioassay for neurite growth using cryostat sections of nervous tissue as a substratum R e b e c c a T u t t l e a n d W i l l i a m D. M a t t h e w

*

D~Tmrtment of Neurobiology, Harl'ard Medical School, Boston, MA 02115 (U.S.A.)

(Received 14 December 19911) (Revised version received 13 June 1991) (Accepted 14 June 199l)

K e y w o r d s : Cryosection; Cryostat section; N e u r i t e outgrowth; A x o n d e v e l o p m e n t ; Bioassay

An in vitro bioassay is described that can be used for studying neurite growth, cell adhesion, and cell migration, as well as other cellular behaviors. The bioassay, which uses tissue sections as substrata for either dissociated cell preparations or explants, offers several distinct advantages over other commonly used bioassays. In particular, this bioassay approximates in viw) cellular environments, and the preparation of large numbers of tissue sections as substrata is relatively straight forward and simple. The method can be employed for studying cell behavior on different ages or types of tissue, as a bioassay for screening antibodies for their ability to perturb a particular cellular behavior, and for assessing the biological activities of growth factors absorbed to specific tissue environments. We present a detailed description of the method, including the variety of options to be considered at each step in the procedure, with an emphasis on using the assay to study neurite growth.

Introduction In o r d e r to identify a n d c h a r a c t e r i z e the molecules that control axonal growth, r e s e a r c h e r s have o f t e n relied u p o n in vitro systems. Typically, e x p l a n t e d or dissociated p e r i p h e r a l ganglia or CNS tissue are p l a t e d o n t o a tissue c u l t u r e surface that has b e e n coated with a m o l e c u l e or cellular fraction suspected of i n f l u e n c i n g axon growth. Alternatively, the m o l e c u l e or fraction is simply a d d e d to the c u l t u r e m e d i u m . This relatively simple in vitro assay has led to the identification of several molecules that affect n e u r i t e

* Present address: Duke University Medical Center, Department of Neurobiology, Box 3209, Durham, NC 27710, U.S.A. Correspondence: R. Tunle, The Salk Institute, Molecular Neurobiology Lab., 10010 North Torrey Pines Rd., La Jolla, CA 92037, U.S.A. Tel.: (619) 453-4100, ext. 439; Fax: (619) 5586207.

growth ( L e v i - M o n t a l c i n i a n d A n g e | e t t i , 1963; H e n k e - F a h l e a n d Bonhoeffer, 1983; M a t t h e w a n d P a t t e r s o n , 1983; G u n d e r s e n , 1985; C o h e n et al., 1986). It is, however, difficult to c o n f i r m that a m o l e c u l e that affects n e u r i t e growth in these relatively simple in vitro systems plays a similar role in vivo. It is t h e r e f o r e necessary to develop in vitro assays which more closely a p p r o x i m a t e the in vivo condition. S a n d r o c k a n d M a t t h e w (1985, 1987a) have d e v e l o p e d o n e such assay. It involves p l a t i n g n e o n a t a l rat sympathetic ganglia o n t o cryostat sections of adult rat nerve that have b e e n dried o n t o coverslips. A f t e r two days, sympathetic n e u r i t e s exhibit extensive growth over sections of sciatic nerve, b u t fail to e x t e n d n e u r i t e s over sections of adult optic nerve. T h e bioassay therefore mimics the in vivo p h e n o m e n o l o g y ; that is, adult PNS s u p p o r t s axonal r e g e n e r a t i o n , b u t adult CNS does not (see also Fig. 1 in this paper). This n e u r i t e growth can be i n h i b i t e d by four m o n o clonal antibodies, whose a n t i g e n s are c u r r e n t l y

194

being characterized (Sandrock and Matthew, 1987a; Tuttle et al., 1989). Since this bioassay mimics the in vivo environment and presents cultured neurons with biologically relevant substrata, it is also suitable for assessing the biological activities of factors that are absorbed to the tissue section prior to culturing cells. This strategy was used to show that nerve growth factor (NGF) specifically binds to tissue sections of denervated sciatic nerve, but not normal nerve, and functions as a substrate-bound neurite growthpromoting activity for cultured sympathetic neurons (Sandrock and Matthew, 1987b). Other laboratories have also published studies of cell adhesion and neurite growth on cryostat sections (Stamper and Woodruff, 1976; Carbonetto et al., 1987; Covault et al., 1987; Crutcher, 1989; Savio and Schwab, 1989; Watanabe and Murakami, 1989). In recent years, we have continued to develop this cryosection bioassay for two main purposes: (1) to study the behavior of neurites on different regions of nervous tissue, and (2) to screen monoclonal antibodies for their ability to perturb neurite growth on these sections. While the cryosection method is now widely used, a detailed methodology has never been published.

Materials and methods

Preparation of coverslips All coverslips are acid washed in 3 M HCI for 30 min on a shaker (all steps on shaker at room temperature). Coverslips are rinsed twice (all rinses are 10 min each) with distilled water (dist. H 2 0 ) , twice with 95% alcohol, and finally 3 times with dist. H 2 0 . Coverslips are then autoclaved so that they are sterile and dry on removal, and stored under a sterile hood. Acid-washed eoverslips can either be treated with polylysine or reacted with glutaraldehyde according to a simplification of the method of Aplin and Hughes (1981). The latter method involves placing 12-mm circular coverslips in a large glass petri dish, washing briefly with 0.1 M NaOH, and then aspirating off the N a O H until the coverslips are dry. The surface of each coverslip is

reacted with approximately t00 /xl 3-aminopropyltriethoxysilane (APTS; Sigma, A3648) for 4 rain. Coverslips are then washed on a shaker three times (10 min each) with dist. H 2 0 , and then left in 1% gtutaraldehyde in dist. H 2 0 for 30 rain. After thoroughly washing 3 times with dist. H 2 0 , coverslips are sterilized and dried by placing them in the notches of plastic slide spacers (often provided in boxes of new glass slides) under the UV light in a sterile hood. To coat coverslips with potylysinc, approximately 100/~1 of a 100 ~ g / m l sterile solution of poly-D-lysine (Sigma, P-0899) in dist. H 2 0 is applied to sterile, acid-washed 12-mm coverslips. Coverslips are [eft in a sterile hood t~r at least 1 h, washed thoroughly with sterile dist. H 2 0 , and dried as above. Both types of coverslips can be stored for up to one week at room temperature.

Preparation of tissue explants and d~sociated cells Dorsal root ganglia ( D R G ) are removed from embryonic day 7 or 8 chicks (White Leghorn) and placed in calcium- and magnesium-free Hank's balanced salt solution (CMF-HBSS). D R G are treated with 0.1% collagenase (type CLS, Worthington Biochemical) in Dulbecco's modified Eagle's medium nutrient mixture F-12 HAM ( D M E M / F 1 2 ; ICN/Flow, 12-467) for 20 min in a 37°C, 3% CO 2 incubator. D R G are then rinsed thoroughly with D M E M / F I 2 , placed in complete culture medium, and returned to the incubator until plating. Superior cervical ganglia (SCG) are dissected from P0 Sprague-Dawley rats, treated as described above for chick DRG, and hemisected prior to plating. Cerebella are removed from postnatal day 8 or 9 rats (newborns are day 0), and placed in D M E M / F 1 2 at room temperature. The meninges are removed and 150 ~zm sagittal sections of cerebellum are cut with a tissue chopper (McIlwain tissue chopper; Mickle Laboratory Engineering Co., Ltd.). These sections are immediately placed in a large volume of D M E M / F 1 2 , and washed 2 or 3 times. The folia generally fall apart or may be easily teased apart, and pieces of appropriate size are cut for plating. The cerebellar pieces should be placed in complete culture medium and left in the incubator until plating.

195

The culture medium for DRG, SCG, and cerebellar explants consists of D M E M / F 1 2 , 5 mM Hepes buffer, 10/xM cytosine arabinoside (araC), 2 mM glutamine, 100 units/ml penicillin, 100 /xg/ml streptomycin, and 1% Nutridoma SP (Boehringer Mannheim Biochemicals, 100104). Culture medium for DRGs and SCGs also contains 10 n g / m l 2.5 S NGF (Boehringer Mannheim Biochemicals, 100700). Dissociated cells were prepared from CNS tissue (spinal cord, whole brain, visual cortex, and cerebellum) by several methods - all yielding good neurite growth. In particular, the method of Heuttner and Baughman (1986) was employed. For neuron-enriched cultures of cerebellum and DRG, the protocols of Keilhauer et al. (1985) and Seilheimer and Schachner (1988) were followed.

Preparation of tissue for cryostat sections Tissue used for making cryostat sections may be either fresh or chemically fixed. The fixative used is a 1.5% solution of paraformaldehyde in calcium- and magnesium-free Dulbecco's phosphate buffered saline (CMF-PBS); a 1% fixative solution yields comparable results. For rats younger than about embryonic day 20, embryos are removed from anaesthetized mothers and decapitated. The tissue is rapidly removed and immersed in cold fixative for 1 to 2 h on a shaker in a cold room. The tissue is then rinsed 3 times (30-60 min total) on a shaker at 4°C with Dulbecco's phosphate-buffered saline (PBS). Rats older than about embryonic day 20 are perfusion-fixed at room temperature for 10 min (roughly 10-15 ml of fixative). Before perfusionfixation, rats less than one week old are anaesthetized by cooling, while older rats are given i.p. injections of 5% chloral hydrate solution at 350 m g / k g of body weight. After the perfusion, the rat is washed with 70% alcohol, and the tissue is removed and rinsed 3 times in cold PBS on a shaker. If tissue is not to be fixed, animals are anaesthetized as above, rinsed with alcohol, and the tissue is quickly dissected out and rinsed briefly in cold PBS before freezing. Fixed or unfixed tissue is next frozen onto metal chucks for sectioning in a cryostat. If the orientation of the tissue is critical (e.g., when

exact longitudinal sections are desired through a nerve), then the chucks are coated with "Tissuetek O.C.T. Compound" (OCT; Miles Laboratory), frozen, and the surface of the frozen OCT is sectioned in a cryostat to form a flat surface; before removing the chuck from the cryostat, a notch is cut in the OCT so that the chuck can later be repositioned in the cryostat in the same orientation. To mount tissue onto these flattened chucks, excess PBS is carefully dabbed from the tissue, and using a spatula and fine forceps, the tissue is placed onto the surface of the OCT and rapidly frozen by immediately surrounding the chuck with powdered dry ice. The tissue should be mounted so that it is securely anchored in the OCT, however most of its mass should project above the OCT surface; in this way, only the tissue is sectioned (the OCT should not be added to the culture system). To achieve these contradictory aims, the OCT should be just beginning to soften at the surface as the tissue is mounted. If orientation of the tissue is not critical, then OCT can be added to the chuck, and the chuck placed in powdered dry ice. When the OCT is almost entirely frozen, the tissue is placed on the surface. After about 1 min in dry ice, the chuck with frozen tissue should be wrapped tightly in parafilm to prevent the tissue from dehydrating. If tissue is sectioned immediately, sectioning tends to be easier, sections (especially unfixed) may attach better, and neurite growth is occasionally greater. Sections can be cut at thicknesses ranging from 8 to 20 /.tm. Sections are picked up with sterile coverslips to which they should readily attach and dry within 2-5 min, although neurite growth was often quite good on sections left to dry for 1 h. Coverslips are then placed in 35-mm dishes. If one plans to plate explants, then the sections can be rehydrated by adding at least 1 ml of culture medium to each dish, and then placing the dishes in the incubator until plating. When plating dissociated cells, the sections must be rehydrated without wetting the bottom of the culture dish; if the bottom of the culture dish gets wet, then the cell dissociate spreads over the dish instead of being restricted to the 12-mm coverslip, thereby losing control of cell plating density. To avoid this, only

196

the section is wet with a small volume (roughly 10 ~1) of the appropriate culture medium. The dishes are placed in the incubator until plating.

Plating tissue explants and dissociated cells Explants are plated in a sterile hood, using a dissecting microscope with dark field optics (dark field optics are necessary when working with translucent sections of developing CNS). Explants are transferred onto coverslips in the 35-mm dishes with a 9-inch Pasteur capillary pipet (explants rarely stick to the pipet unless they are sucked up past the narrow capillary region of the pipet). With a pipet in one hand and fine forceps in the other, the explant is gently nudged into place with the forceps while the level of the culture medium is slowly lowered, taking care not to damage either the substratum or the explant. The final level of the medium will depend on the size and nature of the explant. The surface tension of the culture medium helps hold the explant in place; the level of the medium must therefore be low enough to prevent movement of the explant, yet high enough to prevent dehydration. This achieved, cultures can generally be maintained for two days in a 37°C, 3% CO 2, humidified incubator without the further addition of any culture medium. To plate dissociated cells, the density of the cell suspension must be determined, and adjusted so that it is in a range appropriate for the particular experiment. Cells can then be plated with polypropylene pipet tips and a pipetter, plating roughly 30 ~1 of cell suspension per 12-mm coverslip. Cultures should be returned to the incubator for 2-12 h, before adding 1-2 ml of culture medium.

Analysis of neurite growth Neurite growth is generally examined between 40 and 48 h for explants, and usually earlier for dissociated cells. To visualize axonal processes, as well as all living cells, cultures are stained with the vital dye, 5 (and -6) carboxyfluorescein diacetate, succinimidyl ester (Bronner-Fraser, 1985). A 6.15 mg/ml stock solution (in dimethyl sulfoxide) is diluted 1:300 in PBS. To inhibit photobleaching, a second solution of roughly 5 mM

p-phenylenediamine (PPDA; Kodak, P394) is prepared by adding about 10 mg of PPDA to 25 ml of PBS and immersing the solution in the tank of an ultrasonic cleaner for tess than 1 rain to dissolve the PPDA. Both solutions should be protected from light and used immediately. Culture medium is removed from the 35-ram dish, and about 1 ml of the vital dye solution added. After roughly 5 min, this solution can be removed, and about 1 ml of the PPDA solution is added. Cells can now be observed and photographed with fluorescein optics. This is most easily done by placing the dish containing the coverslip in PPDA solution under a dry lens with a long working distance (the Zeiss Neofluar 6.3 × /0.20 lens is well suited for this purpose). Exposure times of about 2-8 s, using a 400 ASA film, are recommended. The cryosection is weakly fluorescent, and, if desired, can be photographed with the longer exposure times. The type of fluorescent filter set can also affect imaging of the cryosection.

Results and discussion

The type of coverslip used, one treated with either polylysine or APTS/glutaraldehyde, depends on the age and type of the cryosectioned tissue. Cryostat sections of tissue from neonatal or adult PNS attach securely to coverslips treated only with polylysine. Sections of embryonic or neonatal CNS tissue (particularly unfixed) tend to break up when attached to polylysine-treated coverslips, but attach well to coverslips treated with APTS/glutaraldehyde. A second consideration is whether to use fixed or unfixed tissue sections. If explants are being plated, then either type of section can be used, although neurite growth is often longer~and less fasciculated on unfixed sections. In fact, there is little or no neurite growth from explants onto either white or gray matter of fixed adult CNS sections (spinal cord, brainstem, or cerebellum) (Fig. 1), but neurites grow on unfixed sections of adult CNS; this growth is, however, stunted and tangled, as previously reported (Savio and Schwab, 1989). It would therefore appear that

197

D

Fig. 1.48-h culture of E8 chick D R G on 21) p~m longitudinal sections of adult rat spinal cord fixed by perfusion with 1.5% paraformaldehyde. The cryosection is through both the spinal cord and an attached DRG. A dashed line delineates the border between the CNS (cns) and PNS. Although the E8 chick D R G (D) has been placed in contact with both PNS and CNS, it extended long neurites onto the PNS (double arrows), but only s t u n t e d and tangled p r o c e s s e s onto CNS (arrowheads). Cell somata, probably non-neuronal cells, that have emigrated from the ganglion (single arrows) are indicated. Living cells and their processes are stained with 5 (and -6) carboxyfluorescein diacetate, succinimidyl ester. The culture medium contained 5% rat serum.

some of the neurite growth-promoters in these sections are fixation sensitive. If an experiment aims to optimally preserve the native molecular environment of the section, and if explants are being used in the bioassay, then unfixed sections should be employed. There are however instances in which fixed sections might be prefered: for example, if the cryosection bioassay is being used to test the function blocking capacity of an antibody, then it might be desirable to simplify the substrata by eliminating certain fixation sensitive epitopes, thereby making an inhibitory effect mediated by the antibody more apparent. Fixed tissue sections

are always preferred when dissociated cells are used, since survival of dissociated cells and neurite growth from dissociated cells are poor on unfixed sections. This is true for platings that include dissociated neuronal and nonneuronal cells, as well as platings that are highly enriched for neuronal cells. This might be partly due to the toxicity of the dead tissue sections, since unfixed sections are more likely to disintegrate and in some way exert a toxic effect. Explants might be better protected from such toxic effects since the cell somata are sequestered within the natural tissue environment of the explant. If fixed sections are used, it is preferable that, whenever possible, tissue is perfused with fixative as opposed to being immersed in it. One reason for this is that neurite growth is generally more robust on sections that are fixed by perfusion. The concentration of the fixative also effects the resulting neurite growth: lower concentrations of fixative consistently produce sections that are better substrata for neurite growth. One possible explanation is that some of the neurite growthpromoting molecules in the section are fixationsensitive; as the concentration of the fixative is increased, the ability of the tissue to support neurite growth is diminished. Fixed tissue had previously been found to be a good substratum for neurite growth in vivo (Ramon y Cajal, 1968). In vitro, fixed cell cultures are also good substrata for neurite growth (Hawrot, 1980). The decision to use either dissociated cells or explants depends on the question being addressed in the experiment. For studies of neurite growth, the use of explants excludes the complication of cell adhesion phenomena. If, however, cell adhesion phenomena are being studied, then dissociated cells should be used. One advantage of using explants in this bioassay is that the neuronal cell somata are maintained in a cellular environment similar to their in situ environment. Secondly, in explant cultures containing AraC, the effects on neurite growth mediated by living non-neuronal cells are minimized, since under such conditions, few non-neuronal cells emigrate from the explant and those that do, remain close to the explant itself (Figs. 1, 2, and 3). Alternatively, dissociated cell preparations that are highly

198

Fig. 2. E7 chick D R G (D) on 1.5% paraformatdehyde, perfusion-fixed 2 0 / x m thick P0 rat spinal cord sections. Few cells emigrate from the ganglion and those that do. remain close to the ganglion itself (arrows). All neurites are on the section. The stain is 5 (and -6) carboxyfluorescein diacetate, succmimidyl ester. The culture medium contained 5% fetal calf serum and 2.5% rat serum.

enriched for neurons could be used. Neurons in these enriched cultures die if A r a C is added before about 16 h postplating - a period during which the small fraction of non-neuronal cells can

Fig. 3 . 4 8 - h culture of P0 rat superior cervical ganglion (S) on 20 /zm thick, unfixed sections of adult rat sciatic nerve. Neurite growth on the perineurial connective tissue (single arrows), which borders this section, is m u c h shorter than that on neural parenchyma (double arrows). Growth on the A P T S / g l u t a r a l d e h y d e - t r e a t e d coverslip is stunted and tangled (arrowheads). Few cells have emigrated from the ganglion. Dashed lines delineate the section edge. Cells are stained with 5 (and -6) carboxyfluorescein diacetate, succinimidyl ester. The serum-free culture m e d i u m contained Nutridoma-SP.

proliferate. Finally, neurRe growth from explants is simpler t o assess both qualitatively and quantitatively (Sandrock and Matthew, 1987a). The ability of non-neuronal cells to influence neurite growth onto tissue sections must be considered carefully, particularly if mitotic inhibitors are not employed in the culture. Non-neuronal cells could affect neurite growth even when they do not extend to the leading edge of the neuritic halo. Schwann cells, for example, secrete a variety of molecules including laminin and NGF, and these molecules could bind locally to the section as well as diffuse beyond the front of non-neuronal cells and thereby influence neurite growth. In dissociated cell cultures, the potential influence of non-neuronal cells, through cell-cell contact, is even greater. In these cultures, the effect of nonneuronal cells can be minimized by using dissociated cell preparations enriched for neurons, employing mitotic inhibitors in the medium, and controlling cell density so that one is certain of studying interactions between the cells of interest and the sections, and not between the cells themselves. Therefore. we feel it is essential to minimize the number of non-neuronal cells when one plans to assess neurite growth. The composition of the culture medium can affect the type of neurite growth one observes on tissue sections. In dissociated cerebellar cell cultures with fetal calf serum (Hazelton Biologics. Inc.) present in the culture medium, neurite growth was often not observed on the cryostat sections of developing CNS. In contrast, in these same cultures, neurite growth on the polylysinc adjacent to the section was robust. The reason for this is not known, but it is interesting to note that a potent inhibitory effect of fetal calf serum on neural crest cell outgrowth in three-dimensional gels has been observed (Bilozur and Hay, 1988). In addition. Caroni and Schwab (1988) note greater inhibitory properties of CNS myelin fractions in the presence of fetal calf serum. In contrast to dissociated cells, neurite growth from explants was robust in the presence of fetal calf serum, although subtle variations in neurite growth were observed from explants depending on the protein source of the culture medium: explants exhibited good neurite growth in fetal

199

calf serum, rat serum, or in Nutridoma-SP (Fig. 1). Nutridoma-SP is preferred since it provides more defined culture conditions and neurite growth in these cultures was good. Cell survival in the explants is poor when protein-free culture medium is used. Recently, a completely defined medium has been used with success for the D R G explants that substitutes an i n s u l i n / t r a n s f e r r i n / selenium mix (1TS; Boehringer Mannheim Biochemicals, 1074547) for the Nutridoma-SP. Problems with contamination were rarely experienced even in cultures maintained for 4 - 6 days. In previous studies, cryosections were exposed to sterilizing UV irradiation for 20 rain (under UV light in Baker laminar flow clean bench) before plating explants (Sandrock and Matthew, 1987a); this seems not to be necessary. In fact, sections irradiated for 70 min have minimal neurite growth (Fig. 4). This is in general agreement with previous data (Covault et al., 1987; Watanabe and Murakami, 1989), and is probably due to denaturation of neuritc growthpromoting molecules in the section ( H a m m e r b a c k et al., 1985).

The methodology described here recommends the use of 35 mm culture dishes. When attempts were made to adapt the bioassay to either 24-well plates or modified culture dishes (holes drilled in the base of a dish and coverslips affixed to the bottom surface; Bray 1970), the cultures always died. It seems that the size and, in the case of the modified dishes, the configuration of the well led to a concentration of cryosection debris, that, due in part to the resulting meniscus of the culture medium, remained in the center near the explant. This raises the issue of the optimal size of the cryosection; the advantage of a thinner and smaller section is that less potentially toxic material is being added to the culture system. In general, the length of the culture period should not exceed 2 days since neurite growth is already extensive by 2 days. Second, some data suggest that the substratum may somehow be modified by the culture conditions. For example, it was found that while neurite growth on sections of 4% paraformaldehyde-fixed brain was poor at 2 days, neurite growth on these same sections, without the further addition of fresh culture

Fig. 4. Dissociated P8 rat cerebellum on 1.5% paraformaldehyde perfusion-fixed 20 p,m thick sections of (A) P8 cerebellum, (B) P8 cerebellum that was UV-irradiated for 70 rain, and (C) adult brainstem. Only unirradiated P8 cerebellum (A) is a good substratum for neurite growth. The cells in this photo were visualized with fluorescein diacetate (Rotman and Papermaster, 1966), a vital dye that is not as bright as 5 (and -6) carboxyfluorescein diacetate, succinimidyl ester. Also, PPDA was not used and therefore the fluorescein has bleached somewhat making the thin neurites less apparent. The culture medium contained 5¢~;- rat serum.

200

medium, was quite good after 4 days. In any experiment, some of the cultures should be examined at least every 24 h. In this way, the sequence of developmental events that lead to the final pattern of neurite growth begin to be apparent. The method one uses to visualize the living cells on the tissue sections is very important. Recent studies have used glyoxylic acid histofluorescence to stain neurites from sympathetic neurons, as well as silver and methylene blue staining methods. These methods range from selectively staining only the neurons to, at best, resolving the non-neuronal cells poorly. In light of our arguments that it is crucial to consider the effects of non-neuronal cells on neurite growth, we recommend the use of methods that allow one to clearly visualize both neuronal and non-neuronal cells. The vital dye recommended for use in this paper, 5 (and -6) carboxyfluorescein diacetate, stains all living cells and their processes intensely, drawing attention to even a small number of non-neuronal ceils (Fig. 2). It is often desirable to visualize the cryosection and cells simultaneously. While this vital dye clearly distinguishes living cells from the nonliving substrate, the cryosection is weakly fluorescent and can be photographed by increasing the exposure time (Fig. 1). Alternatively, one can illuminate the preparation with low intensity visible light and UV light simultaneously. A method for fixing cells stained with this vital dye has been described (Bronner-Fraser, 1985): this method could be applied to the cryosection culture system. In summary, although this bioassay is straightforward and versatile, there are a variety of parameters to consider carefully. We hope that a detailed description of the method, the options to consider, and the precautions to take will facilitate the use of this bioassay in other laboratories. Our laboratory, and several others, has used this system to assess the endogenous neurite growthpromoting and -inhibiting activities that reside within neural tissues. We have shown that this bioassay provides biologically relevant substrata for absorbing growth factors or tissue extracts in order to assess their functional activities in an environment that closely mimics the in vivo situation. The method has also proven to be valuable

for screening monoclonal antibodies for their ability to perturb neurite growth. There are many possible strategies for using this bioassay and future applications will certainly include an assessment of a tissue's ability to regulate gene expression within cultured neurons and cells.

Acknowledgements We are grateful to Drs. Pradeep Bhide, Michael Bilozur, and Rick Born for their thoughtful comments on the manuscript. This work was supported by a grant to R.T. from the Paralyzed Veterans of America Spinal Cord Research Foundation, and by a Fellowship in Neurobiology awarded to R.T. by the F. Hoffmann-La Roche and Co., Ltd. The work was also aided by grant 1-1117 from the March of Dimes Birth Defects Foundation and Research Grant NS02253 from the National Institutes of Health.

References Aptin, J.D. and Hughes, R.C. (1981) Protein-derivatised glass coverstips for the study of cell-to-substratum adhesion. Anal. Biochem.. 113: 144-148. Bilozur, M.E. and Hay, E.D. (1988l Neural crest migration in 3D extracellular matrix utilizes laminin, fibronectin, or collagen. Dev. Biol., 125: 19-33. Bray, D. (1970) Surface movements during the growth of single explanted neurons. Proc. Natl. Acad. Sci. USA. 65: 905-910. Bronner-Fraser, M. (1985) Alterations m neural crest migration by a monoclonal antibody that affects cell adhesion. J. Cell Biol.. 101: 6t0-617. Carbonetto. S.. Evans. D. and Cochard. P. (1987)Nerve fiber growth in culture on tissue substrata from central and peripheral nervous systems. J. Neurosci.. 7: 610-620. Caroni. P. and Schwab. M,E. (1988) Two membrane protein fractions from rat central myelin with inhibitory propentes for neurite growth and fibroblast spreading. J. Cell Biol.. 106: 1281-1288. Cohen, J., Burne, J.F., Winter. J. and Bartlett. P. (19861 Retinal ganglion cells lose response to laminin with maturation. Nature, 322: 465-467. Covault, J., Cunningham, J.M. and Sanes, J.R. (1987) Neurite outgrowth on cryostat sections of innervated and denervated skeletal muscle. J. Cell Biol., 105: 2479-2488. Crutcher, K.A. (1989) Tissue sections from the mature rat brain and spinal cord as substrates for neurite outgrowth

2(I I in vitro: extensive growth on gray matter but little growth on white matter. Exp. Neurol., 104: 39-54. Gundersen, R.W. (1985) Sensory neurite growth cone guidance by substrate adsorbed nerve growth factor. J. Neurosci. Res., 13: 199-212. Hawrot, E. (1980) Cultured sympathetic neurons: effects of cell-derived and synthetic substrata on survival and development. Dev. Biol., 74: 136-151. Hammarback, J.A., Palm, S.L., Furcht, L.T. and Letourneau, P.C. (1985) Guidance of neurite outgrowth by pathways of substratum-adsorbed laminin. J. Neurosci. Res., 13: 213220. Henke-Fahle, S. and Bonhoeffer, F. (1983) Inhibition of axonal growth by a monoclonal antibody. Nature, 303:65 67. Hucttner, J.E. and Baughman, R.W. (1986) Primary culture of identified neurons from the visual cortex of postnatal rats. J. Neurosci., 6: 3044-3060. Keilhauer, G., Faissner, A. and Schachner, M. (19851 Differential inhibition of neurone-neurone, neurone-astrocyte and astrocyte-astrocyte adhesion by LI, L2 and N-CAM antibodies. Nature, 316:728 730. Levi-Monlalcini, R. and Angeletti, P.U. (1963) Essential role of the nerve growth factor in the survival and maintenance of dissociated sensory, and sympathetic embryonic nerv'e cells in vitro. Dev. Biol., 7: 653-659. Matthew. W.D. and Patterson, P.H. (1983) The production of a monochmal antibody which blocks the action of a neurite outgrowth promoting factor. Cold Spring Harbor Symp. Quant. Biol., 48:625 631. R a m o n y Cajal, S. (1968) Degeneration and Regeneration of the Nervous System. Facsimile of the 1928 edition translatcd and edited by R.M. May. Hafner Publishing Co., London, 1: 360. Rotman. B. and Papermaster. B.W, (1966) Membrane proper-

ties of living m a m m a l i a n cells as studied by enzymatic hydrolysis of fluorogenic esters. Proc. Natl. Acad. Sci. USA, 55: 134-141. Sandrock, A.W. and Matthew, W.D. (19851 A bioassay for the analysis of neurite-promoting factors in the extracellular matrix of rat peripheral nerve. Soc. Neurosci. Abstr., I I: 592. Sandrock, A.W. and Matthew. W.D. (1987a) Identification of a peripheral nerve neurite growth-promoting activity by development and use of an in ~'itro bioassay. Proc. Natl. Acad. Sci, USA, 84: 6934-6938. Sandrock, A.W. and Matthew, W.D. (1987b) Substrate-bound nerve growth factor promotes neurite growth in peripheral nerve. Brain Res., 425:360 363. Savio, T. and Schwab, M.E. (1989) Rat CNS white matter, but not gray matter, is nonpermissive R)r neuronal cell adhesion and fiber outgrowth, J. Neurosci., 9:112(3 1133. Seilheimer, B. and Schachner, M. (1988) Studies of adhesion molecules mediating interactions between cells of peripheral nervous system indicate a major role for LI in mediating sensory neuron growth on Schwann cells in culture. J. Cell Biol., 107: 341-351. Stamper, H.B. and Woodruff, J.,l. (19761 Lymphocylc homing into lymph nodes: in vitro demonstration of the selective affinity of recirculating lymphocyles for high-endothelial venules. J. Exp. Med., 144: 828-833. Tuttle, R., Sandrock, A.W. and Matthew, W.D. (1989) Analysis of complex matrices functional in neuronal process extension using monoclonal antibodies m t i t r o and its rit'o. Dev. Neurosci., 11:289 299. Watanabe, E. and Murakami, F. (19891 Preferential adhesion of chick central neurons to the gray matter of the central nervous system. Neurosci. Letl., 97: ¢~t) 74.