J. Embryol. exp. Morph. Vol. 23, 2, pp. 519-30, 1970
Printed in Great Britain
519
Dissociation of sensory ganglia from the embryonic
chick by pronase and other dispersing agents
By B A R B A R A E. C. BANKS 1 , D . V. B A N T H O R P E 2 ,
D . M A R G A R E T L A M O N T 2 , F . L. P E A R C E 2 ,
K A T H A R I N E A. R E D D I N G 2 AND C. A. V E R N O N 2
Departments of Chemistry and Physiology,
University College London
The gross effect of protein factors from mouse salivary glands and certain
snake venoms on the growth, in tissue culture, of expiants of sensory ganglia
from chick embryos is well established (Levi-Montalcini, 1964, 1965; Angeletti,
Calissano, Chen & Levi-Montalcini, 1967; Banks et al. 1968; Varon, Nomura
& Shooter, 1967a, b). The effects of these substances, called nerve growth
factors, on individual neurons in vitro have, however, been little investigated
and no quantitative studies have been reported. In order to make such a study
we required a method of making cell suspensions of high density from chick
embryonic sensory ganglia: the target tissue normally used for the assay of
nerve growth factors.
Several methods for obtaining dispersed cells from nervous tissue are recorded in the literature and, of these methods, micro-dissection of mammalian
central nervous tissue (Hyden, 1959; Lowry, 1962; Hyden & Lange, 1965)
appears to result in the least damage to cells. Suspensions of neurons suitable
for hand collection and gradient centrifugation have also been prepared from
mammalian central nervous tissue by a sieving technique (Roots & Johnston,
1964; Bocci, 1966; Satake & Abe, 1966; Rose 1967), but this method is only
suitable for small numbers of cells. In addition, centrifugation of cells isolated
by sieving caused readily discernible damage, particularly of embryonic tissue.
Various chemical methods, convenient for routine work, involving the breakdown of intercellular binding material have been described. As the binding
material is considered (Rinaldini, 1958) to be essentially collagenous with
elastic fibres embedded in a matrix containing metal ions, both proteases and
sequestering agents have been used to effect cell dispersion. These methods give
a number of relatively undamaged, separated cells. For example, neuroblasts
1
Author's address: Department of Physiology, University College London, Gower Street,
W.C.I.
2
Authors' address: Department of Chemistry, University College London, Gower Street,
W.C.I.
520
B. E. C. B A N K S A N D
OTHERS
from grasshopper have been dispersed by hyaluronidase (Armand & Tipton,
1954), chick neural retina by trypsin (Moscona & Moscona, 1966; Collins,
1966) and mouse brain by sodium tetraphenylboron (Rappaport & Howze,
1966a, b). Sensory ganglia from chick embryos have been dissociated by treatment with proteinase A (Nakai, 1956), by trypsin alone or in conjunction with
EDTA (Levi-Montalcini & Angeletti, 1963; Utakoji & Hsu, 1965) and by
pancreatin (Cohen, Nicol & Richter, 1964).
Cell suspensions from non-nervous tissues have been prepared with varying
degrees of success both from adult and embryonic sources by enzymic methods,
by change of pH (Townes & Holtfreter, 1955), by mechanical methods (Berry,
1962) and by treatment with sequestering agents (Anderson, 1953; Curtis &
Greaves, 1965; Rappaport & Howze, 1966/?). The use of sequestering agents
sometimes causes extensive changes in ultrastructure and function (Harris &
Leone, 1966; Friedmann & Epstein, 1967). Subcultures, too, may be affected
(Rinaldini, 1959) since the agents may have inhibitory effects and are, in any
case, difficult to remove from the suspension medium. It is generally felt that
enzymic methods are preferable (Rinaldini, 1959; Mateyko & Kopac, 1963).
The enzymes used include a-chymotrypsin (Weiss & Kapes, 1966), collagenase
(Rinaldini, 1959; Hinz & Syverton, 1959), elastase (Rinaldini, 1959; Weiss &
Kapes, 1966), hyaluronidase (Rinaldini, 1959), neuraminidase (Weiss & Kapes,
1966), papain (Mateyko & Kopac, 1963), trypsin (Weiss & Kapes, 1966;
Moscona & Moscona, 1966, 1967), pronase (Mintz, 1962; Wilson & Lau, 1963;
Gwatkin & Thomson, 1964; Sullivan & Schäfer, 1966) and combinations of these
proteases (Rinaldini, 1959; Yamada & Ambrose, 1966). The relative effectiveness of dissociating agents in producing cell suspensions from chick heart
muscle (Rinaldini, 1959) and adult mouse tumor (Rappaport & Howze,
1966a, b) has been reported.
The preparation of cell suspensions from embryonic sensory ganglia by
microdissection on a large scale presents difficulties. Consequently, the dissociating agents listed above have been screened for suitability. We now
report a method of preparing cell suspensions from embryonic chick sensory
ganglia, heart, brain and adrenal tissue.
MATERIALS AND
METHODS
Glassware was sterilized by dry heat (120 °C). Solutions were either purchased
sterile or were passed through Millipore filters (0-22 ja pore size) and stored at
4 °C before use. Gentle shaking in a tube vibrator (Towers, Widnes) is referred
to as mechanical agitation. Pipetting operations were kept to a minimum and
stirring was avoided since these procedures resulted in damaged separated cells.
Dissociation of sensory
ganglia
521
Enzymes and other potential dispersing agents
Trypsin was obtained from Difco Labs, Detroit, Michigan or Sigma Chemical
Co. Ltd., London (pure grade). oi-Chymotrypsin from Sigma 'ex bovine pancreas', type 11, recrystallized and lyophilized. Pankreatin from Sigma 'porcine
type II'. Elastase from Sigma 'ex pancreas, recrystallized twice from suspension
in water'. Protease I from Sigma, 'ex S. griseus, type 6; repurified'. Protease II
from Sigma, 'ex pancreas crude'. Pronase from Calbiochem, Ltd., London, 'ex
S. griseus, Grade B'. Hyaluronidase from Sigma, 'ex Ovine testis, Grade 111'.
Pepsin from Sigma, 'twice recrystallized and lyophilized'. Papain from Sigma,
'ex papaya, crude type II', or from BDH Ltd., 'twice recrystallized'. Collagenase
from Koch Light Ltd., 'ex C. histolyticum, form 1'. Ethylene diamine tetraacetic acid (EDTA) as the disodium salt from BDH Ltd., London. Sodium
tetraphenylboron from BDH Ltd., London. Melittin from bee venom, prepared by Dr R. Shipolini.
Nerve growth factor (NGE)
NGF from Russell's viper venom was prepared as previously described
(Banks et al. 1968). The material was assayed on whole ganglia, which were
cultured on collagen in hanging drops, for each experiment. Only samples
which showed an activity of 3+ (see Banks et. al.) at a concentration in the
culture medium of less than 10-9 g ml -1 were used.
Dissociation conditions
Embryonic chick sensory ganglia (ca 75) from 7-, 8- or 9-day chick embryos
were dissected from the lumbosacral and thoracic regions. (Pronase was also
effective in dissociating ganglia from older embryos, e.g. 9-14 days.) The
ganglia were placed in either Pannet and Compton's saline or synthetic medium
no. 199 (Glaxo Ltd., Greenford, Middx.) at room temperature, for immediate
use. Dissociation was carried out by mechanical agitation at 37 °C in phosphatebuffered saline (Dulbecco & Vogt, 1954) at pH 7-2. The concentrations of
dissociating agents, times of incubation and nature of the culture medium are
noted in the results section. Enzyme solutions were made up immediately
before use to prevent autolysis. Sodium tetraphenylboron was used as specified
(Rappaport & Howze, 19666) except that the suspension was agitated mechanically rather than by pipetting.
Assessment of dispersion
After gentle agitation for the specified periods, cells were separated by
centrifugation (350 g, 5 min, room temperature) and the cell pellet was washed
twice, for 2 min with 20 % cockerel serum (Flow Ltd., Irvine, Ayrshire), in
buffered saline after protease treatment or with saline alone after treatment with
non-enzymic reagents. Treatment with serum is said to inhibit proteases
522
B. E. C. BANKS AND OTHERS
(Gwatkin & Thomson, 1964) but pronase was not completely inhibited by 20 %
cockerel serum. 50 % horse serum (Burroughs & Wellcome Ltd., London) was
not completely effective either, but 20 % calf serum (Glaxo Ltd., Greenford,
Middx.) which had been heated at 56 °C for 1 h prevented further action by
pronase.
The washed cell pellet was suspended in tissue culture medium no. 199
(ca. 1 ml) and the cell density was determined using a modified Fuchs-Rosenthal haemocytometer. The best dissociating agents gave an initial cell density,
in ca. 1 ml, of ca. 5 x 105 cells ml -1 .
Method of culture of dispersed cells
Glass rings (6 mm diameter, 4 mm high) were waxed on to coverslips coated
with a dried film of reconstituted rat tail collagen. Cell suspensions (ca. 0 1 ml)
were plated dropwise, from a Pasteur pipette, into the rings. Cockerel serum and
buffered saline (with and without NGF) were added to give a composition of
medium no. 199 : serum : saline of 2 : 1 : 1 by volume; (the amount of NGF
added, was that which gave a 3 ± response in the hanging drop assay.) Typically,
five pairs of rings were prepared for each cell suspension. The rings were placed
in Petri dishes and maintained at 37 °C in the dark, for 24 h, in a water-saturated
atmosphere of 5 % carbon dioxide in air. The medium was changed at 72 h
intervals when cultures were maintained over a longer period to permit the
development of a fibre network.
Fig. 1. Sensory cell culture prepared by a pronase dispersion of 8-day spinal ganglia.
Culture grown for 48 h in a medium containing purified nerve growth factor.
Methylene blue stain.
Dissociation of sensory ganglia
523
At the end of the incubation period, the rings were removed and the cultures
were stained with methylene blue (0-01 % w/v in saline) for 30-40 min at 37 °C,
fixed in ammonium molybdate solution (8 % w/v in glass distilled water) at
4 °C for 2-4 h and then washed with glass distilled water (10 min), rapidly
dehydrated and mounted in Canada balsam for microscopic examination.
Some cultures were stained with silver (Holmes, 1943), after fixation in a mixture of 10 % neutral formalin (5 % ) : glacial acetic acid (5 %): 80 % ethyl
alcohol (90 %).
Characterization of dispersed cells
A considerable proportion of the cells resulting from the most successful
dispersion methods were shown to be capable of sending out fibres in culture.
The cells of interest usually contained two visible nucleoli and had a characteristic elliptical perikaryon (ca. 5 x 9 / * ) , which was hyperchromic under phase
contrast. After staining with methylene blue, the cytoplasm became more intensely coloured than the nucleus while silver staining of cultured cells showed
clearly defined neurofibrils. Fig. 1 shows a portion of a methylene blue stained
culture.
Unsuccessful methods of dispersion and culture gave cells with spherical
perikaryons and granulated cytoplasm and such cells were more uniformly
stained with methylene blue. These cells produced stunted or completely
retarded fibre growth after 24 h culture.
RESULTS
The results of various dispersion methods were assessed after culture of
dispersed cells had been maintained for 24 h, with and without nerve growth
factor in the culture medium. Cell density was scored as nil, low, medium or
high, the proportion of neurons as 0, 1, 2 or 3 corresponding approximately
to zero, 5, 15 and 25 % of the total cell population and the condition of
neurons as 1, 2 or 3.
The dispersing agents which were relatively effective are listed in Table 1 with
the relevant conditions and the results in terms of the density of cells and the
proportion and condition of neurons.
Other reagents used without success are listed in Table 2.
Studies on pronase-dispersed cells
The optimum conditions for cell dispersion by pronase were found to be
mechanical agitation and incubation with 0 01 % w / v pronase solution for
15 min at 37 °C either in Pannet and Compton's saline or phosphate-buffered
saline. In the resulting suspensions typically fewer than 10 % of cells were
aggregated. In plated cell cultures which had been grown on collagen films for
4 h up to 30 % of the cells were nerve cells.
524
B. E. C. BANKS AND OTHERS
Table l. Dispersion of cells of ganglia by different treatments
Not. Compound
1
Trypsin
Conen
(w/v) %
Time
(min)
NGF
0-5
30
+
-
2
a-Chymotrypsin
0-5
30
+
3
Pancreatin
1
30
+
015
30
+
001
30
-
+
-
0 005
30
+
-
4
Elastase
01
30
+
-
5
Protease I
0-5
30
+
-
6
Protease II
0-2
10
+
Med.
Med.
Med.
Low
Low
Low
Med.
Low
Low
Low
Negligible
Negligible
Med.
Med.
High
High
15
+
0-2
10
+
-
001
15
+
High
-
High
Low
Low
0 05
15
+
-
0 005
15
+
-
Pronase
Proportion
neurons Condition
Negligible
Negligible
High
High
Med.
Med.
High
High
High
High
-
7
Density
0-25
0 001
20
+
—
3
2
3
2
2
0
3
1
3
2
0
0
3
2
3
2
3
2
0
0
3
2-3
3
2-3
3
2-3
0
0
3
1
3
1
2
3
1
3
2-3
2
2
3
2
3
2-3
3
2-3
3
2-3
3
2-3
-
Culture of pronase-dispersed cells on collagen
Microscopic examination ( x 100 or x 400 final magnification) of cells grown
on collagen showed that within 1 h of plating, a considerable proportion of
neurons (elliptical in shape) had settled and attached to the substrate. Neurons,
particularly from embryos which were more than 11 days old, sometimes
possessed fibre stumps. Within 2 h of plating, some anchored neurons had
started to develop fibres while appreciable fibre development was apparent at
the end of a further hour. Microscopic examination, necessarily intermittent
since the dispersed neurons were light sensitive, also indicated that both fibre
outgrowth and withdrawal occurred and that some neurons, often with welldeveloped processes, rotated or migrated in the culture medium. Within 24 h
Dissociation of sensory ganglia
525
of plating, a fibre network had developed and cell aggregation was apparent.
Most fibres were linked either to other fibres or to perikaryons but a few appeared to terminate on other types of cells or cell debris. A considerable number
of neurons appeared to be bipolar. Supporting cells could be recognized from
their characteristic morphology in cell cultures.
Table 2. Unsuccessful treatments
Reagent
Hyaluronidase
Pepsin
EDTA
Papain
Sodium
tetraphenylboron
Collagenase
Melittin
Conditions
Comments
0-5 %w/v, 30 min
0-5 %w/v, 30 min, pH 7-2
0001 %w/v, in Ca-Mg-free
saline, pH 8 (NH4C1)
0-2 %w/v, 0 02 % cysteine hydrochloride. 0 0 5 %w/v, 0 0 2 %
cysteine hydrochloride
0 1 %, 10-12 min, 20-37°C
No effect
No effect
No effect
0 0 5 or 0 0 1 % w/v, in buffered
saline 20-25 min
0 1 %w/v, 10-30 min
Cell debris plus fibroblasts produced
No effect
Partially or totally disintegrated
the ganglia but no intact cells
Destruction of ganglia
Culture of pronase-dispersed cells on plasma clot
Fibre growth is more readily assessed in supporting than in liquid media.
Culture of cell suspensions in plasma clot resulted in an average fibre growth of
ca. 70 jLt, compared with an average outgrowth of ca. 300 JLL from whole ganglia,
in hanging drop cultures, under the same conditions. In the presence of nervegrowth factor, the fibre growth in plasma clot averaged ca. 100 JLL after 24 h
culture with some exceptional growths to ca. 400 fi. In addition, culture in
plasma clots in the presence of nerve-growth factor resulted in the development
of processes by a higher proportion of the neurons present. About 90 % of all
neurons in cultures on plasma clot were unipolar.
Removal of cultured cells from collagen and glass surfaces
Trypsin (003 %, 10 min, 37 °C) is commonly used to remove cultured cells
from collagen and glass surfaces. Pronase (0-001 %) was also found to be highly
effective ; 5 min gentle agitation of sensory cell cultures grown on collagen and
glass supports yielded suspensions of single cells.
Dissociation of other tissues by pronase
Under the optimum conditions described for treatment of embryonic sensory
ganglia, pronase was found to be effective in giving single cell suspensions from
heart and brain tissue (cut into 1 mm cubes) of 7-16 day chick embryos and from
526
B. E. C. B A N K S A N D
OTHERS
adrenal tissue of 13-16-day-old embryos. More concentrated pronase (005 %
w/v) was effective in the dissociation of spinal cord from 7-16-day-old embryos.
The cell suspensions were not examined in detail, but neurons were readily
discernible in the suspensions from central nervous tissue and adrenals while
morphologically characteristic cells were seen in the suspensions from heart
tissue. Little debris was found in these experiments.
Sympathetic spinal ganglia, pooled from six 7-13-day embryos, when treated
with pronase (0-2 % w/v for 35 min) did not give true single cell suspensions.
The cells tended to clump but individual cells behaved in tissue culture in a
manner similar to that observed in the culture of cells from sensory ganglia.
DISCUSSION
The successful use of a pronase as a dissociating agent has been reported for a
number of tissues including chick embryo muscle (Wilson & Lau, 1963), whole
chick embryo (Gwatkin & Thomson, 1964), adult and embryonic mouse tissue
(Mintz, 1962; Gwatkin and Thomson, 1964), various epidermal tissues (Sullivan
& Schäfer, 1966) and Hela cells (Niitsu & Handa, 1965). On the basis of these
results, pronase has come to be regarded as a better dissociating agent than
trypsin for these tissues. However, a report that this same enzyme causes
severe damage to neurons from ox brain (Bocci, 1966) confirms the fact that
the choice of dissociating agent and the optimal conditions for dissociation
depends critically on the particular tissue being studied.
In our hands, pronase has been used successfully to obtain suspensions of
single cells from chick embryo sensory ganglia, heart, brain, spinal cord and
adrenal tissue and, with somewhat less success, from sympathetic spinal ganglia.
The cell suspensions from sensory ganglia could be cultured, and maintained
for at least 7 days, on collagen films in a medium containing 25 % fresh cockerel
serum (at the high cell densities used). Well-developed fibre networks were
observed within 24 h of plating, both with and without the addition of NGF if
the culture medium was supplemented with 25 % cockerel serum or 10 % horse
or calf serum. However, although large numbers of sensory nerve cells survive to
regenerate a network of neurites when cultured under these especially favourable
conditions, without the addition of purified NGF, it must be remembered that
NGF is present in the serum which constitutes a large part of the culture
medium.
Further, the similarity between dissociated cell cultures treated with NGF
and their controls is apparent only at the favourably high cell densities recorded.
At lower cell densities appreciably fewer nerve cells survive to regenerate nerve
fibres in untreated control cultures. Omission of serum from the culture medium
leads to a marked irreproducibihty in the resulting cultures. These cultures may
be maintained for at least 5 days in a medium 199 containing NGF and nerve
cells survive to regenerate the extensive network of nerve fibres described above.
Dissociation of sensory ganglia
527
Appreciable numbers of supporting cells also survive, although markedly fewer
than when serum is included in the medium. Sensory cell cultures plated in
medium 199 in buffered saline alone typically show massive degeneration within
72 h but small numbers of neurons can survive at least 5 days. (The role played
by NGF in dissociated sensory nerve cell cultures has been investigated in some
detail and the results of these experiments will be published later.) The high
quality of cultures prepared by pronase dispersion and grown under our
conditions without the addition of NGF might be contrasted with previous
reports (Levi-Montalcini & Angeletti, 1963; Cohen, Nicol & Richter, 1964;
Utakoji & Hsu, 1965) that cell suspensions obtained by the dissociation of
embryonic sensory ganglia by other enzymes only developed fibre networks
when cultured in a liquid medium to which NGF had been added. Even in
culture media containing NGF, fibre networks were developed more slowly from
trypsin-dispersed cells than from the pronase-dispersed cells described in the
present work. The effect of trypsin as a dissociating agent has been largely
confirmed in the present work, the cell suspensions from embryonic sensory
ganglia being cultured reproducibly only in the presence of NGF in addition to
serum. Without NGF in culture medium, trypsin-dispersed cell cultures progressively deteriorated over a period of 24 h although occasionally excellent
fibre networks were established within this period of culture.
These results are consistent with the view that the dissociation of chick embryo sensory ganglia by pronase gives cells which are less damaged than those
obtained by carrying out the dissociation with trypsin.
Most proteases tested in the present work gave cell suspensions from sensory
ganglia of varying degrees of viability. Hyaluronidase and pepsin had little
effect on the same tissue while collagenase and papain were too destructive.
Sodium tetraphenylboron was also found to be too destructive and EDTA was
without effect.
SUMMARY
1. The effectiveness of a number of proteolytic enzymes and metal chelating
agents in promoting the dissociation of chick embryo sensory ganglia has been
examined.
2. The enzyme pronase was found to be the most satisfactory dispersing
agent.
3. Sensory nerve-cell cultures prepared by a pronase dissociation can be
maintained on collagen supports in a serum supplemented-liquid medium for at
least 7 days, during which time an extensive network of nerve fibres develops.
4. The nerve-growth factor requirement for the maintenance of nerve-cell
viability and the regeneration of nerve fibres is briefly discussed.
5. Conditions under which pronase treatment gives single cell suspensions
from chick embryo heart, brain, adrenal tissues and spinal cord are described.
528
B. E. C. BANKS AND OTHERS
RÉSUMÉ
Dissociation de ganglions sensitifs d'embryons de poulet par la pronase et autres
agents de dispersion
1. L'efficacité d'un certain nombre d'enzymes protéolytiques et d'agents
chélateurs pour dissocier les ganglions sensitifs d'embryons de poulet a été
examinée.
2. L'enzyme pronase s'est montrée l'agent de dispersion le plus satisfaisant.
3. Des cultures de cellules nerveuses sensibles préparées par une dissociation
à la pronase peuvent être maintenues sur des supports de collagène dans un
milieu liquide contenant du sérum pendant au moins 7 jours, période pendant
laquelle se développe un important réseau de fibres nerveuses.
4. L'exigence d'un facteur de croissance nerveuse pour le maintien de la
viabilité de la cellule nerveuse et pour la régénération de fibres nerveuses est
brièvement discutée.
5. Les conditions dans lesquelles un traitement à la pronase donne des suspensions de cellules isolées en provenance de coeur d'embryon de poulet, de
cerveau, de tissu surrénal et de moelle épinière sont décrites.
The authors wish to thank The Whitehall Foundation of New York for generous financial
support for this work and are particularly grateful for the interest shown by Mr J. Moffett.
The award of a Science Research Council grant to KAR and University of London Postgraduate Studentships to FLP are gratefully acknowledged. Thanks are due to Miss Janet
Cook and Miss Janet Drew for skilled technical assistance and to Dr H. ff. S. Davies and
Mr A. R. Berry for supplying purified NGF.
REFERENCES
ANDERSON,
N. G. (1953). The mass isolation of whole cells from rat liver. Science, N.Y. 117,
627-8.
P., CALISSANO, P., CHEN, J. S. & LEVI-MONTALCINI, R. (1967). Multiple molecular forms of the nerve growth factor. Biochim. biophys. Acta 147, 180-2.
ARMAND, G. S. ST. & TIPTON, S. R. (1954). The separation of neuroblasts and other cells from
grasshopper embryos. Science, N. Y. 119, 93-4.
ANGELETTI,
BANKS, B. E. C, BANTHORPE, D. V., BERRY, A. R., DAVIES, H. ff. S., DOONAN, S., LAMONT,
D. M., SHIPOLINI, R. & VERNON, C. A. (1968). The Preparation of nerve growth factor
from snake venoms. Biochem. J. 108, 157.
M. N. (1962). Metabolic properties of cells isolated from adult mouse liver. J. Cell
Biol. 15, 1-8.
Bocci, V. (1966). Enzyme and metabolic properties of isolated neurons. Nature, Lond. 212,
826-7.
COHEN, A. I., NICOL, E. C. & RICHTER, W. (1964). Nerve growth factor requirement for the
development of dissociated embryonic sensory and sympathetic ganglia in culture. Proc.
Soc. exp. Biol. Med. 116, 784-9.
COLLINS, (1966). Electrokinetic properties of dissociated chick embryo cells. Calcium ion
binding by neural retinal cells. /. exp. Zool. 163, 39-47.
CURTIS, A. S. G. & GREAVES, M. F. (1965). The inhibition of cell aggregation by a pure
serum protein. /. Embryol. exp. Morph. 13, 309-26.
DULBECCO, R. & VOGT, M. (1954). Plaque formation and isolation of pure lines with poliomyelitis viruses. /. exp. Med. 99, 167-82.
BERRY,
Dissociation of sensory ganglia
529
3
FRIEDMANN, T. & EPSTEIN, e . J. (1967). The incorporation of leucine- H into protein by
tetraphenylboron and citrate-dispersed rat liver parenchymal cells. Biochim. biophys.
Acta 138, 622-4.
GWATKIN, R. B. L. & THOMSON, J. L. (1964). A new method for dispersing the cells of
mammalian tissues. Nature, Lond. 201, 1242-3-.
HARRIS, C. C. & LEONE, C. A. (1966). Some effects of EDTA and tetraphenylboron on the
ultrastructure of mitochondria in mouse liver cells. / . Cell Biol. 28, 405-8.
HINZ, R. W. & SYVERTON, J. T. (1959). Mammalian cell cultures for study of influenza virus.
I. Preparation of monolayer cultures with collagenase. Proc. Soc. exp. Biol. Med. 101,
19-21.
HOLMES, W. (1943). Silver staining of nerve axons in paraffin sections. Anat. Rec. 86,
157-87.
HYDEN, H. (1959). Quantitative assay of compounds in isolated, fresh nerve cells and glial
cells from control and stimulated animals. Nature, Lond. 184, 433-5.
HYDEN, H. & LANGE, P. W. (1965). The steady state and endogenous respiration in neurone
and glia. Actaphysiol. scand. 64, 6-14.
LEVI-MONTALCINI, R. (1964). The nerve growth factor. Ann. N.Y. Acad. Sei. 118, 149-70.
LEVI-MONTALCINI, R. (1965). Morphological and metabolic effects of the nerve growth
factor. Archs Biol. Liège 76, 387-417.
LEVI-MONTALCINI, R. & ANGELETTI, P. U. (1963). Essential role of N G F in the survival and
maintenance of dissociated sensory and sympathetic nerve cells in vitro. Devi Biol. 7, 653-9.
LOWRY, O. H. (1962). The chemical study of single neurones. Harvey Lect. 58, 1-19.
MATEYKO, G. M. & KOPAC, M. J. (1963). Cytophysical studies on living normal and neoplastic cells. Part I. Separation into isolated populations. Ann. N.Y. Acad. Sei. 105, 183—
218.
MINTZ, B. (1962). Experimental study of the developing mammalian egg: removal of the
zona pellucida. Science, N. Y. 138, 594-5.
MOSCONA, A. A. & MOSCONA, M. H. (1966). Inhibition of cell aggregation in vitro by puromycin. Expi Cell Res. 41, 703-6.
MOSCONA, A. A. & MOSCONA, M. H. (1967). Comparison of aggregation of embryonic cells
dissociated with trypsin or versene. Expl Cell Res. 45, 239-43.
NAKAI, J. (1956). Dissociated dorsal root ganglia in tissue culture. Am. J. Anat. 99, 81-129.
NIITSU, Y. & HANDA, T. (1965). Protease of Streptomyces griseus in tissue culture. Sei. Rep.
Res. Inst. Tohuku Univ. Ser. C. 12, 90-8. Cited in Chem. Abstr. 63, 13680.
RAPPAPORT, C. & HOWZE, G. B. (1966a). Dissociation of adult mouse liver by sodium tetraphenylboron, a potassium complexing agent. Proc. Soc. exp. Biol. Med. 121, 1010.
RAPPAPORT, C. & HOWZE, G. B. (19666). Further studies on the dissociation of adult mouse
tissue. Proc. Soc. exp. Biol. Med. 121, 1016.
RINALDINI, L. M. J. (1958). The isolation of living cells from animal tissues. Int. Rev. Cytol. 7,
.587-647.
RINALDINI, L. M. J. (1959). An improved method for the isolation and quantitative cultivation of embryonic cells. Expl Cell Res. 16, 477-505.
ROOTS, B. I. & JOHNSTON, P. V. (1964). Neurones from ox-brain nuclei—their isolation and
appearance by light and electron microscopy. / . Ultrastruct. Res. 10, 350-61.
ROSE, S. R. P. (1967). Preparation of enriched fractions from cerebral cortex containing
isolated, metabolically active neuronal and glial cells. Biochem. J. 102, 33-43.
SATAKE, M. & ABE, S. (1966). Preparation and characterisation of nerve cell perikaryon from
rat cerebral cortex. / . Biochem., Tokyo 59, 72-5.
SULLIVAN, J. C. & SCHÄFER, I. A. (1966). Survival of pronase treated cells in tissue culture.
Expl Cell Res. 43, 676.
TOWNES, P. L. & HOLTFRETER, J. (1955). Directed movements and selective adhesion of
embryonic amphibian cells. / . exp. Zool. 128, 53-120.
UTAKOJI, T. & Hsu, T. C. (1965). Nucleic acid and protein synthesis of isolated cells from
chick embryonic spinal ganglia in culture. / . exp. Zool. 158, 181-202.
VARON, S., NOMURA, J. & SHOOTER, E. M. (1967a). The isolation of the mouse nerve growth
factor protein in a high molecular weight form. Biochemistry, N. Y. 6, 2202-9.
34
E M B 23
350
B.E. C.BANKS AND OTHERS
VARON, S., NOMURA, J. & SHOOTER, E. M. (19676). Subunit structure of a high-molecular-
weight form of the nerve growth factor from mouse submaxillary gland. Proc. natn. Acad.
ScL U.S.A. 57, 1782-9.
WEISS, L. & KAPES, D. L. (1966). Observations on cell adhesion and separation following
enzyme treatment. Expl Cell Res. 41, 601-8.
WILSON, B. W. & LAU, T. (1963). Dissociation and cultivation of chick embryo cells with an
actinomycete protease. Proc. Soc. exp. Biol. Med. 114, 649-51.
YAMADA, T. & AMBROSE, E. J. (1966). Preparation of cell suspensions from tissues using
collagenase and hyaluronidase, especially for measuring the electrical charges of surface
membranes of mammalian cells. Expl Cell Res. 44, 634-6.
{Manuscript received 23 April 1969)