An automated colorimetric microassay for neuronotrophic factors

Developmental Brain Research, 25 (1986) 191-198 Elsevier

191

BRD 50349

An Automated Colorimetric Microassay for Neuronotrophic Factors MARSTON MANTHORPE, ROBERTO FAGNANI*, STEPHEN D. SKAPER and SILVIO VARON

Department of Biology, School of Medicine, University of California at San Diego, La Jolla, CA 92093 (U. S. A. ) (Accepted August 13th, 1985)

Key words: nerve growth factor - - ciliary neuronotrophic factor - - neuronotrophic activity - neuronotrophic factor bioassay --tetrazolium salt - - dorsal root ganglion - - ciliary ganglion

A microassay is described for determining the number of neurons surviving after 24 h in response to added neuronotrophic factors. Neuronal cultures in 96-well microtiter plates are supplied with a yellow tetrazolium derivative, MTT (3-(4,5-dimethylthiazol-2-yl)2,5-diphenyl tetrazolium bromide), which is taken up selectively by viable neurons and converted to a blue formazan product. The amount of blue color development can be rapidly quantified using an automatic microplate spectrophotometer. The resulting optical density is directly proportional to the number of viable neurons. The spectrophotometer has been interfaced with a computer allowing a print out of individual absorbance values and calculation of half-maximal (one trophic unit) neuronal survival. The assay has been used for the quantification of the trophic activities of nerve growth factor and ciliary neuronotrophic factor using, respectively, dorsal root and ciliary ganglionic neurons from 8-day chick embryos. Assay parameters were optimized so that about 2000 individual cultures of ganglionic neurons can be set up and analyzed each day, thus allowing the serial titration in duplicate of 80-120 separate samples. The determination of neuronal number and titer calculation steps now requires about 2 min per microplate (96 cultures), a 50-fold reduction in time over existing methods. INTRODUCTION Neuronal survival and growth in culture require the addition of specific n e u r o n o t r o p h i c agents15,16. Such requirements are i n t e r p r e t e d as possible reflections of neuronal needs in vivo and as the basis for neuronal death during n o r m a l d e v e l o p m e n t , after injury and in association with certain diseases1,5,6,11,18. Several trophic substances have thus far been identified2,3,10,13,19,20, while m a n y others have been suggested and some partially characterizeda-6,11A5. In all cases, such n e u r o n o t r o p h i c agents are d e t e c t e d and quantitated using neuronal cell cultures17. Presently available quantitative assays9,17 of neuronotrophic factors such as nerve growth factor ( N G F ) and ciliary n e u r o n o t r o p h i c factor ( C N T F ) involve the presentation of serial dilutions of samples containing the factor to dissociated m o n o l a y e r neuronal cultures and subsequent microscopic counts of surviving neurons. These assays require (i) the deter-

mination of the n u m b e r of individual surviving neurons within a defined fractional a r e a of each culture and calculation of total neuronal n u m b e r p e r culture, (ii) manually plotting the n u m b e r of neurons p e r culture vs test sample dilution and (iii) calculation of the half-maximally effective sample dilutions and thus of the sample titer in trophic units ( T U ) p e r ml. The disadvantages of these assays are that they are timeconsuming (requiring a b o u t one hour to count the neurons in 96 wells of a microtiter plate), somewhat subjective (in scoring certain neurons as 'surviving' versus ' d e a d ' ) and tedious (in plotting individual sample titration curves and titer calculation). Mosmann12 has r e p o r t e d a rapid colorimetric assay for lymphocyte survival and growth in 96-well microplate cultures. This assay involves supplying the cultures with a tetrazolium salt, 3-(4,5-dimethylthiazol2-yl)-2,5-diphenyl tetrazolium b r o m i d e , (MTT), which is converted to an insoluble blue formazan product by living cells but not by dying cells or their

* Present address: Hybritech, 11095 Torreyana Road, San Diego, CA, U.S.A. Correspondence: M. Manthorpe, Department of Biology, School of Medicine, M-001, University of California at San Diego, La Jolla, CA 92093, U.S.A. 0165-3806/86/$03.50 (~) 1986 Elsevier Science Publishers B.V. (Biomedical Division)

192 lytic debris. The blue product is then solubilized by the addition of an alcoholic solution and color intensity is measured using a multiwell scanning spectrophotometer ('ELISA reader'). The optical density (OD) of the solution in the microwell is directly proportional to the number of viable lymphocytes present. Mosmann used the assay for the determination of cell numbers between the range of 102-105 cells/culture. Here, we have modified this technique for use in determining sensory and parasympathetic neuronal cell numbers below 500 neurons/culture in assays of NGF and CNTF and for facilitating the calculation of their neuronotrophic titers. MATERIALS AND METHODS

Materials NGF was prepared in its 7S form from male mouse submaxillary glandsi9 and CNTF was purified from chick embryo eye 2. MTT (M2128) and polyornithine hydrobromide (P-3655) were from Sigma and reagent grade isopropanol (A415) was from Fisher. Culture medium was Eagle's Basal Medium containing added D-glucose (33.3 mM), sodium bicarbonate (26.4 mM), L-glutamine (2 mM), penicillin (100 IU/ml) and 10% heat-inactivated fetal calf serum (Gibco). Microtiter plates (A/2, 3696) were from Costar. Fertilized white Leghorn chicken eggs were obtained from a local aviary (McIntyre Eggs, Lakeside, CA) and incubated for 8 days (to Stage 34) at 37.5 °C. Purified mouse laminin was purchased from Bethesda Research Labs.

Neuronotrophic factor bioassays Standard 4.5-mm diameter 96-well microtiter culture plates (A/2, Costar) were used throughout. Laminin-coated polyornithine substrata8 were prepared by adding to each well 25/~1 of polyornithine solution (0.1 mg/ml in borate buffer, pH 8.4) for 2 h at 25 °C, replacing it without washing the wells with 25/A laminin solution (10/~g/ml in phosphate-buffered saline, pH 7.4), incubating the plates overnight at 37 °C and then storing the plates at -20 °C individually wrapped in zip-lock plastic bags. For assays, the plates were thawed at 37 °C in a humidified incubator, the laminin aspirated and quickly replaced with 25/~1 culture medium again without washing and without allowing the surface to dry. All additions and aspi-

rations were performed using sterile solution troughs and a 12-channel micropipet (Titertek R) fitted with sterilized plastic tips. In the present study, samples to be assayed for trophic activity were set up as follows: the 96-well plates are normally arranged with 12 columns (labelled from 1 to 12) of 8 rows (labelled A - H ) of microwells. Culture medium (25 ~1) was added to each well. Four control wells were left undisturbed with 25 ktl culture medium without trophic factor (wells 1 A - D ) and 4 wells provided with 40 TU/ml of trophic factor (wells 1 E - H ) . Sample titrations were carried out in duplicate in the remaining wells starting in the column-2 wells and proceeded two-fold to the corresponding column-12 wells. Typically, a given sample was prediluted in culture medium in a plastic tube and 25 ~1 added to the wells of column 2 (A-B, C-D, E-F or G-H) to make 50 ~1, mixed, 25/~1 transferred and mixed in successive wells down each row and the final 25 ~1 (from wells 12 A - D ) discarded. Finally, 25 B1 of cell suspension was added to all wells. This serial dilution procedure allowed a sample to be titrated over a 210-fold range from 1/4 of its original dilution in the tube to a 1/4096 dilution. Dorsal root and ciliary ganglia were dissected and dissociated as described 10, the cells counted using a hemacytometer, diluted to 80,000/ml in culture medium and 8-12 ml seeded into 100-mm diameter uncoated tissue-culture dishes (Costar). Most of the non-neuronal cells were removed by differential attachment by culturing for 2.5 h at 37 °C in a 5% CO2/95% air incubator. The supernatants, enriched to 80-95% neurons, were collected from each plate into 12-ml plastic tubes and centrifuged at 500 g for 4 min. Cell pellets were resuspended in 1 ml culture medium by gentle trituration using a Pasteur pipet with a flame-constricted tip. Cell yields in these experiments were about 7-9000 neurons per ganglion of either type. Neurons were diluted to the desired cell densities with culture medium and 25 ~! seeded into each of the sample-containing microwells. After a period of culture (20 h unless otherwise indicated), 5 ,ul of sterile MTT solution (1.5 mg/ml in culture medium unless otherwise indicated, stored at -20 °C in aliquots and thawed immediately before use) was added to each well and the plates cultured up to a total in vitro period of 24 h. Replicate microcultures without added MTT were fixed with 2% glu-

193 taraldehyde and the number of neurons determined by visual counting as described 9. MTT-containing cultures were terminated by adding 50/~1 isopropanol-0.08 N HC1. The amount of HCI added was increased to twice that reported by Mosmann12 to prevent the occasional precipitation of serum proteins. The isopropanol-acid and culture medium were mixed by 8 up and down aspirations using the 12channel pipet set at 50/~1 volume. After no longer than 1 h the plates were read on a Dynatech MR600 MicroELISA reader set at a test wavelength of 570 rim, reference wavelength of 630 nm, calibration setting of 1.99 and threshold of 1.50. The reader was interfaced with an Apple lie computer using a Dynatech software program in PASCAL, Immunosoft. This program saves the results on a data disk and allows the OD values, corrected for background, to be printed out in a 96-well format matching the original plate. Also, the program can be instructed to calculate the half-maximal OD value, based on the difference between the average OD values from trophic factor-supported (wells 1 E - H ) and unsupported cultures (wells 1 A - D ) . RESULTS Viable dorsal root and ciliary ganglionic ( D R G and CG) neurons in culture are readily able to convert the tetrazolium derivative, MTT, to a blue product. Fig. 1 compares the appearance of such cultures with or without added MTF. In the presence of the appropriate trophic factor, but absence of the dye, each neuronal type expresses a typical phase-bright soma and long thin neurites (Fig. 1B, F). In the presence of MTT the neurons appear to accumulate a blue material which forms elongated crystals that radiate from most of the cell bodies (Fig. 1D, H). In the absence of trophic support, very few neurons remain alive after 24 h (Fig. 1A, E) and very little blue product is present (Fig. 1C, G). The addition of isopropanol-acid to the trophic factor-supported cultures dissolves both the cells and their blue product and causes the solution to become faintly bluish. Similar addition of the alcohol-acid to non-supported cultures dissolves the remaining neurons and debris but does not elicit a visibly blue-colored mixture. Different numbers of D R G and CG neurons were provided to culture wells in the presence or absence

of a maximally supportive concentration of either NGF or CNTF. After 20 h one half of the cultures were provided with 5/xl of a 5-mg/ml MTT solution. At 24 h (4 h later) the non-MTT-containing cultures were fixed and the number of neurons counted. The MTT cultures were terminated with isopropanol-acid and the OD determined. The results are shown in Fig. 2. Three points can be made, namely: (i) for the cell densities examined, the number of neurons surviving in the presence or absence of the appropriate neuronotrophic factor linearly increases with the increase in the number of neurons seeded (Fig. 2A, C); (ii) the OD values are similarly increased with seeding density (Fig. 2B, D); and (iii) the amount of OD expressed per neuron in trophic factor-supported cultures remains relatively constant over the seeding densities examined and equal to about 1.5 x 10- 4 0 D units/neuron for both neuronal cell types (Fig. 2B, D, insets). Thus, color development in neuronal cultures exposed to MTT represents an accurate measurement of the number of cultured neurons scored as viable using light microscopy. Based on the OD profiles obtained (Fig. 2B, D) we arbitrarily chose to seed 1000 D R G neurons and 500 CG neurons per culture which typically gave for both neuronal types a differential OD value (i.e. ODfactor-ODno factor) of about 0.045. Using the chosen seeding densities we examined whether the Mq"F concentration and exposure times could be adjusted to increase further the 0.045 OD differential. Different concentrations of MTT were exposed for certain times to either D R G or CG neuronal cultures, with and without their respective trophic factor. The MTT was added to the cultures at selected times preceding their termination and measurement at 24 h and the ODfactor-ODno factor calculated. The results are shown in Table I. The exposure to 0.5 mg/ml MTT level (i.e. 5/xl of a 5-mg/ml solution) for the last 4 h of culture which was used in the previous experiments (Figs. 1, 2) rendered a differential OD value in this experiment of 0.054 (DRG) and 0.051 (CG; Table I, broken boxes). These values were considerably lower than those reached when lower MTT concentrations were included for longer culture periods. For both D R G and CG neurons the highest OD differentials (about 50% higher than the previous condition) were attained by using 0.06-0.25 mg/ml MTT during an 8-10 h exposure time (Table I,

194

~ i '~

Fig. 1. Identification of viable sensory and parasympathetic neurons m culture using MTT. Dissociated neurons from 8-day chick embryonic dorsal root ( A - D ) and ciliary ( E - H ) ganglia were cultured for 24 h on a laminin-coated polyornithine substratum as described in Materials and Methods. Cultures were either not provided with trophic factors (A, C, E, G) or provided with 20 TU/ml mouse submaxillary 7S N G F (B, D) or chick eye CNTF (F, H). At 20 h some cultures (C, D, G, H) were provided with 0.5 mg/ml MTT for 4 h. All cultures were fixed with 2% glutaraldehyde and photographed using phase-contrast microscopy. Note that a dark crystalline material appears to emanate from most, but not all (see arrow in H) those surviving neurons provided with MTI'. Bar, 50pro.

195 boxed values). We chose to use 0.15 mg/ml MTT (final concentration in the cultures) and 9-h exposure times (15-24 h of culture) in the remaining experiments (along with a 1000- or 500-DRG or -CG neuronal seed/well, respectively).

DRG

A.

Plus NGF O N

~

•

1000-

Separate samples of purified N G F or CNTF were titrated and the number of D R G or CG neurons surviving after 24 h compared in replicate cultures by direct microscopic counts or by the colorimetric measurement using MTT. The results are shown in Fig. 3.

:l xxx x

0.12

y

/o.. .

/

~FN seed L~ o

0.06

500'

o,. o

o

0

Z p.

8

• o

16

CG

>

'

C.

24

,~

CO

I o r~ u3 a

e

0

0 4

Plus CNTF

z.

0.12

1000'

8

<

16

/

XxX×x x

6 " lo

24

J

/

..Q E Z

500"

0.06

0

5

10 Hundreds

0

15 of

Neurons

Seeded

5 per

1'0

1"5

Microwell

Fig. 2. Correlation between the number of surviving neurons and OD after staining with MTI'. Microwells were seeded with the indicated numbers of D R G (A, B) or CG (C, D) ganglion neurons and cultured for 24 h in the absence (open circles) or presence (closed circles) of 40 TU/ml of the appropriate neuronotrophic factor (NGF or CNTF). Replicate cultures were either fixed at 24 h and the number of surviving neurons counted directly using phase-contrast microscopy (A, C), or stained for 4 h (between 20 and 24 h of culture) using MTI', the dye dissolved with isopropanol and the OD difference at 570 and 630 nm (A OD) determined using a microplate spectrophotometer (see Materials and Methods). The ratio of AOD and neuronal number (N) for each of the seeding densities is shown as insets in B and D. Error bars represent S.D.s of 3 separate culture seedings.

196 TABLE I

Optimization of color development by viable neurons during neuronotrophic factor bioassays Each well in microtiter plates (A/2, Costar) was coated with polyornithine and laminin and 25 gl culture medium with or without 40 TU/ml NGF (for DRG cultures) or CNTF (for CG cultures) was added, followed by 25/A culture medium containing 1000 DRG or 500 CG neurons. The cultures were adjusted to the indicated MTT concentration and developed with isopropanol-HCl at 24 h following the indicated exposure time. Absorbance differences from cultures in the absence of the trophic factor were subtracted from those with the factor (ODplusNTF--ODnoNTF)"The highest OD differences are enclosed in a box while the OD values obtained using the conditions previously reported 12for lymphocyte cytotoxic assay are enclosed in broken boxes.

MTT exposure (h)

Final M T T concentration (mg/ml) 0.03

0.06

0.12

0.25

0.50

1.00

DRG 1 2 4

0.006 0.010 0.023

0.011 0.018 0.045

0.019 0.046 0.064

0.029 0.058 0.070

0.035 0.051 0.0541

0.032 0.049 0.054

8 10 20

0.050 0.060 0.026

10.079 0.081 t).014

0.082 0.084 0.024

0.089 0.086[ 0.016

0.080i 0.067 0.020

0.055 0.047 0.007

1 2 4 8 10 20

0.001 0.008 0.028 0.041 0.055 0

0.006 0.018 0.045 0.059 [0.070 0

0.019 0.041 0.063 0.061 0.07~ 0.010

0.024 0.042 I 0.051 /).053 11.056 0.007

0.017 0.1)29 0.030 0.1132 0.032 0

CG

__

0.013 0.033 0.062 i0.071 0.074 ~ 0.014

It is c l e a r t h a t i d e n t i c a l titers a r e o b t a i n e d b y t h e t w o

d e n s i t i e s u s e d in Fig. 3 (1000 D R G a n d 500 C G n e u -

m e t h o d s for e i t h e r t r o p h i c f a c t o r . B o t h t r o p h i c fac-

r o n s p e r well). A t all t h e s e n e u r o n a l cell d e n s i t i e s t h e

tors w e r e also t i t r a t e d u s i n g t h e s a m e o p t i m u m M T T

t r o p h i c f a c t o r s a m p l e s c o n t i n u e d to e x h i b i t t h e s a m e

c o n d i t i o n s (0.15 m g / m l , 9 h ) b u t u s i n g n e u r o n a l s e e d -

titers with either the neuronal count or colorimetric

ing d e n s i t i e s o f 5 0 % , 7 5 % , 1 2 5 % a n d 1 5 0 % o f t h e

methods (data not shown).

TABLE II

Computer output of OD values from a microplate spectrophotometer and estimation of neuronotrophic factor titers Four separate samples (I-IV) containing different amounts of CNTF activity were titrated serially two-fold (from wells 2-12) and assayed using the MTT colorimetric technique for their ability to support 8-day chick embryonic CG neurons (see Materials and Methods). Values of OD differences (570 rim-630 nm) are printed out in the format of the original 96-well plate along with the computercalculated average values from unsupported (1A-D) and maximally supported (1E-H) wells and their half-maximal difference (0.0374). Broken histogram bars progress from the left where the sample was first introduced and the serial dilution began and end at the approximate location of the half-maximal value. The titers (in TU/ml) of samples I - I V are estimated by interpolation to be, respectively, 40, 190, 10 and 2 × the original sample dilution. Background = average of 1 A - D = 0.0005; plateau = average of 1 E - H = 0.0743; half-maximal = (0.0743-0.0005/2) + 0.0005 = 0.0374.

Dilution 1 1

A

0.001

B

0.000

C D E F G H

0.003 -0.002 0.077 0.070 0.071 0.079

2 0.076 I 9.077 ', 0.075 0.076 O.-077 0.074 ()_067 0.0_59

Sample 2

4

8

16

3 4 5 6 0.078 0.069 0.070 0.058 0.07_2 _0.078 _ 0_.0_70 _ 0.055.. . . . 0.072 0.071 0.075 0.079 0.075 0.076 0.073 0.077 0.07-7---0_072 --0_0-39 ] 0.023 0.069 0.071 0.043 J 0.027 6.0~5 0.617 01609 0.003 0.099 0.021 0.009 0.008

32

64

7 8 0.0401 0.021 I 0.043L_0_.0_22 0.071 0.073 0.071 0.078 0.017 0.010 0.015 0.007 0.003 0.000 0.001 0.002

128

256

512

1024

9 0.014 0.018 0.0481 0.050 I 0.004 0.003 0.000 0.001

10 0.007 0.009 0.025 0.027 0.002 0.000 0.000 0.000

11 0.004 0.003 0.014 0.017 0.001 0.000 0.000 0.000

12 0.002 0.001 0.008 0.009 0.000 0.000 0.000 0.000

I

II III IV

197

5oo~

o

A.DRG

01

4O0

008

3O0

0.06

• ~

200

~

0.04

i

o cL c o

5

oo Z

>

100

0.02

10 (10 3)

1 0 (104 )

ng/ml

o o~

•0 0 1 (10 5 )

dl £3 0

7S NGF (sample dilution)

5oo~° ~~

I

B.CG 0.06

400 dl E Z

rn O

= • " ~11~ \'0 0.04

iii

0.02

100 1.0 ( 1 0 2 )

o. E t3

~:ao 0.1 ('10 4 )

~

0

0.01 (10 5

ng/ml CNTF (sample dilution)

Fig. 3. Comparison of neuronotrophic factor titration curves using direct neuronal counts (closed circles) and OD (open circles) generated by the MTT calorimetric method (see Materials and Methods). DRG (A) and CG (B) cultures were used to titrate purified mouse submaxillary 7S NGF or chick eye CNTF, respectively. The half-maximal responses generated by neuronal cell counting (unbroken arrows) or optical density (broken arrows) were similar for each neuronotrophic factor.

Considerable time and effort must be spent in plotting neuronal numbers and determining the halfmaximal (1 TU) responses for each trophic sample (cf. Fig. 3). A very simple and rapid estimation of sample titers can be performed using a computer interfaced with the microplate reader (see Materials and Methods). One of several possible formats is presented in Table II. In this example the microplate reader was blanked against well B1 and the OD value for each well was printed out in the same format as the original culture plate. The half-maximal OD value was calculated by the computer to be 0.0374

based on average OD values from the sets of 4 wells containing unsupported neurons (wells 1 A - D ) or neurons provided with 40 TU/ml of trophic factor (wells 1 E - G ) . Since the samples ( I - I V ) are titrated in duplicate series of two-fold dilutions from columns 2 to 12, it is a simple matter to scan the wells and locate those expressing the approximate half-maximal values. In Table II samples I - I V had titers of, respectively, about 40, 190, 10 and two times the sample dilution originally placed in the wells comprising column 2. The entire procedure of reading each 96well plate, printing out all 96 OD values and estimating the titer requires about two min. In our hands the previous method involving direct neuronal counts using microscopy, calculation of averages and graphic determination of titers requires about 100 min per plate. DISCUSSION The presently described calorimetric microassay for the biological activities of neuronotrophic factors is more rapid and convenient than previously described techniques. Usefulness of the technique is not restricted to determining titers Of samples containing purified NGF or CNTF as presented here. We have used it just as successfully to monitor NGF or CNTF" activity in (i) crude extracts and derivative fractions from a variety of tissues of different species (ii) gel eluates after isoelectric focusing 7 or sodium dodecylsulfate polyacrylamide gel electrophoresis during CNTF purificationej0 and (iii) fractions generated during high-pressure liquid chromatography. The optimized assay works equally well for the determination of NGF and CNTF titers on 10-day chick embryo D R G neurons which respond to both factors (data not shown). The calorimetric microbioassay offers additional advantages as well. We have used A/2 microplates which have one-half of the culture area of the 6-mm diameter 96-well plates previously used for neuronotrophic assays 9. These smaller wells require one-half the neuronal seed and culture media volume to produce an OD value similar to 6-mm plates since the sample path-lengths measured spectrophotometrically are the same for the wells in both plate types. Thus, half as many neurons are now needed to set up the same number of assay cultures. In view of the re-

198 maining limitation on the n u m b e r of embryonic ganglia that can be obtained for each dissociation, we

cifically activate the dehydrogenases. In addition, any colored substance, such as hemoglobin or mel-

can set up about 2000 individual D R G or CG neuro-

anin, which may absorb in the 570- or 630-mm re-

nal cultures at a time, thus allowing the duplicate

gion, can interfere with the colorimetric determination. Basic substances may neutralize the acid in the

two-fold serial titration (over a concentration range of 1-2 li) of 80-120 samples concurrently. A n o t h e r advantage is that, unlike neuronal counts over a 'rep-

isopropanol and convert the phenol red in the culture media to a purple color. Finally, we have found that

resentative' culture area, the O D value sums up the

certain microplate readers are not adequate to meas-

contributions from all the viable neurons in the cul-

ure reproducibly the ODs generated u n d e r the pres-

ture and thus may be useful for determining cell num-

ent assay condition. Nevertheless, with the above

ber in clumped, reaggregate or even explant cultures. Despite the above advantages the colorimetric microassay has a few potential limitations. The tetrazolium derivative M T T is acted on by various dehydrogenases within active mitochondria 14. Thus, M T T can be converted to the blue formazan product and 'measured' as a t r o p h i c response if the culture becomes contaminated with bacterial dehydrogenases or when foreign substances are presented that spe-

REFERENCES 1 Appel, S.H., A unifying hypothesis for the cause of amyotrophic lateral sclerosis, Parkinsonism and Alzheimer disease, Ann. Neurol., 10 (1981) 499-505. 2 Barbin, G., Manthorpe, M. and Varon, S., Purification of the chick eye ciliary neuronotrophic factor (CNTF), J. Neurochem., 43 (1984) 1468-1478. 3 Barde, Y.A., Edgar, D. and Thoenen, H., Purification of a new neurotrophic factor from mammalian brain, Eur. Mol. Biol. Org. J., 1 (1982) 549-553. 4 Berg, D.K., New neuronal growth factors, Annu. Rev. Neurosci., 7 (1984) 149-170. 5 Cowan, W.M., Fawcett, J.W., O'Leary, D.D. and Stanfield, B.B., Regressive events in neurogenesis, Science, 225 (1984) 1258-1265. 6 Gurney, M.E. and Apitalo, B., Activity of muscle-derived growth factor for spinal neurons in vitro and in vivo, Soc. Neurosci. Abstr., 10 (1984) 1051. 7 Manthorpe, M., Barbin, G. and Varon, S., Isoelectric focusing of the chick eye ciliary neuronotrophic factor, J. Neurosci. Res., 8 (1982) 233-239. 8 Manthorpe, M., Engvall, E., Ruoslahti, E., Longo, F.M., Davis, G.E. and Varon, S., Laminin promotes neuritic regeneration from cultured peripheral and central neurons, J. Cell Biol., 97 (1983) 1881-1890. 9 Manthorpe, M., Skaper, S.D. and Varon, S., Neuronotrophic factors and their antibodies: in vitro microassays for titration and screening, Brain Res., 230 (1981) 295-306. 10 Manthorpe, M., Skaper, S.D., Williams, L.R. and Varon, S., Purification of adult rat sciatic nerve ciliary neuronotrophic factor, Brain Res., in press. 11 Manthorpe, M. and Varon, S., Regulation of neuronal survival and neuritic growth in the avian ciliary ganglion by trophic factors. In G. Guroff (Ed.), Growth and Maturation Factors, Vol. 3, Wiley, new York, 1985, pp. 77-117.

qualifications, the presented neuronotrophic activity bioassay affords a great improvement over existing assays.

ACKNOWLEDGEMENTS We wish to thank Ms. E l e a n o r e Hewitt for expert technical assistance. This work was supported by NIH G r a n t NS16349 and NSF G r a n t BNS 82-18366.

12 Mosmann, T., Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxic assays, J. lmmunol. Meth., 65 (1983)55-63. 13 Selak, I., Skaper, S.D. and Varon, S., Pyruvate participation in the low molecular weight trophic activity for CNS neurons in glia-conditioned media, J. Neurosci., 5 (1985) 23-28. 14 Slater, T.F., Sawyer, B. and Stravli, U.D., Studies on succinate-tetrazolium systems. III. Points of coupling of four different tetrazolium salts, Biochem. Biophys. Acta, 77 (1963) 383-393. 15 Varon, S., Factors promoting the growth of the nervous system, Discuss. Neurosci., in press. 16 Varon, S., Adler, R., Manthorpe, M. and Skaper, S.D., Culture strategies for trophic and other factors directed to neurons. In S.E. Pfeiffer (Ed.), Neuroscience Approached through Cell Culture, Vol. 2, CRC Press, Boca Raton, Florida, 1983, pp. 53-77. 17 Varon, S., Manthorpe, M., Skaper, S.D. and Adler, R., Neuronotrophic factors: problems and perspectives. In B. Haber, J.R. Perez-Polo and J.D. Coulter (Eds.), Proteins of the Nervous System-Structure and Function, Progress in Clinical and Biological Research, Vol. 79, Liss, New York, 1982, pp. 225-242. 18 Varon, S., Manthorpe, M. and Williams, L.R., Neuronotrophic and neurite promoting factors and their clinical potentials, Dev. Neurosci., 6 (1984) 73-100. 19 Varon, S., Nomura, J., Perez-Polo, J.R. and Shooter, E.M., The isolation and assay of the nerve growth factor proteins. In R. Fried (Ed.), Methods and Techniques of Neurosciences, Dekker, New York, 1972, pp. 203-229. 20 Walicke, P., Varon, S. and Manthorpe, M., Purification of a human red blood cell protein supporting the survival of cultured CNS neurons, and its identification as catalase, J. Neurosci., in press.