THe JoURNAL oF BroLogicaL CuEMISTRY
Vol. 247, No. 10, Issue of May 25, pp. 3159-3169, 1972
Printed in U.S.A.

Markers for Gene Expression in Cultured Cells

from the Nervous System

(Received for publication, December 13, 1971)

SAMUEL H. Witson,* Bruce Kk. Scunien,t Joun L. Farsper,§ Epwarp J. Toompson, Roger N. RosEen-
BERG, ARTHUR J. BLUME,|| AND MARSHALL W. NIRENBERG

From the Laboratory of Biochemical Genetics, National Heart and Lung Institute, National Institutes of Health,

Bethesda, Maryland 20014

SUMMARY

Methods are presented for preparation of extracts from
cultured cells from the nervous system and for study of
choline acetyltransferase, acetylcholinesterase, glutamate
decarboxylase, and catechol O-methyltransferase activities.
These enzyme activities are markers that can be used for
studying gene expression in neurons. The methods are
sufficiently sensitive so that all assays can be performed with
protein harvested from one Petri dish. Activities of the
marker enzymes were assessed in surface cultures of new-
born mouse brain cells, and in glial and nonbrain cell lines.
Low activities of choline acetyltransferase, acetylcholines-
terase, and glutamate decarboxylase were detected in all
the cells tested. All of these activities, and particularly
glutamate decarboxylase, were higher in cultured brain cells
from newborn animals than in non-neuronal cell lines.
Glutamate decarboxylase activity in glial cells and in brain
cells was inhibited more than 95% by 1 mm amino-oxyacetic
acid.

 

Techniques have been developed in this laboratory and others
for culture of differentiated neurons (1-10). Activities of en-
zymes important in neuronal cell metabolism are useful param-
eters for following cell maturation and exploring steps in differ-
entiation in such cultures.

The purpose of this communication is to describe a set of
methods used to explore the expression of genes that determine
the metabolism of molecules involved in intercellular conmmuni-
vation in the nervous system. Simple, convenient methods are
presented for preparing cell-free extracts from surface cultures

* Present address, National Cancer Institute, Bethesda, Md.
20014.

t Present address, National Institute of Child Health and
Thiman Development, Bethesda, Md. 20014.

§ Present address, Department of Pathology, Health Science
Center, Temple University, Philadelphia, Pa. 19140.

© Present address, Division of Neurosciences, University of
California at La Jolla, San Diego, Calif. 92122.

| Present address, Roche Institute of Molecular Biology,
Nutley, N. J. 07110.

and for assays of acetylcholinesterase (EC 3.1.1.7), cutechol
O-methyltransferase (eC 2.1.1.1), choline acetyltransferase
(EC 2.3.1.6), and glutamate decarboxylase (EC 4.1,1.15)
activities. ‘The assays presented are extensive moclificutions of
previously published procedures.

Activities of the marker enzymes were assessed in a variety of
glial and nonbrain cell lines as well as in surface cultures of new-
born mouse brain cells. The results document properties and
usefulness of the methods presented and also show that newborn
brain cells in surface culture attained levels of glutamate de-
carboxvlase activity significantly higher than in non-neuronal
cell lines. A preliminary report of this work has been presented

(11).
METHODS

Isolation and Culture of Newborn Mouse Brain Cells

Whole brains from newborn Balb/c (National Institutes of
Health stock) mice were placed in Solution D (137 mm NaCl,
5.4 mM KCI, 0.17 mm Na2HPO,, 0.22 mm KHPOu, 5.5 mM elu
cose, and 5.9 mM sucrose) pH 7.2, 340 mosm (modified Puck's
D1 solution (12)) at 0-4°, weighed, washed several times with
Solution D, and minced to about 1-mm* pieces with iris scissors.
The minced tissue was subjected to two 15-min treatments with
0.25% erude trypsin (Nutritional Biochemicals 1:250 or Difeo
1:300) in Solution D (100 ml per g of tissue) at 37° with eon-
stant mild swirling. After each treatment, tissue pieces were
allowed to sediment and the dissociated cells that remained sus-
pended were collected by decantation. ‘Tryptic activity in both
the cell suspension and undissociated tissue was inhibited by the
addition of an equal volume of growth medium containing 106;
(v/v) fetal bovine serum (Colorado Serum Co.). The undis-
sociated tissue was then triturated five times by gravity flow
through a 10-ml serological pipette and cells were again recovered
by decantation. Cells were pooled and sedimented at 250 x
9max for 10 min at 3° and resuspended in Solution 1). Cell num-
ber and viability were determined by counting in a hemocvton-
eter with 0.540 (w/v) nigrosin (Allied Chemical, Morristown,
N. J... Only those cells which excluded the dve were vonsidered
viable. Usual cell recoveries for dissociation of newborn brain
were 90 to 100 < 108 cells per g of tissue (98 ta LOOC? viable).
Cells were inoculated at 107 viable cells per 150-mm Faleon
polystyrene Petri dish (145 ¢ny surface area) in 20 ml of DMEM

3159
3160

(906¢ Dulbecco-Vogt modification of Eagle’s medium, 10°; fetal
bovine serum with 50 units of sodium-penicillin and 10 yg of
streptomycin-SO, per ml). Dishes were maintained ut 37° im
an humidified atmosphere of 10 COs90¢% air. Changes of
medium were performed on the 3rd day of culture, every 2nd
day thereafter, and 16 to 20 hours before enzyme assay.

Other Cell Lines

The C-6 rat astrocytoma cell line wax obtained from the Amer-
ican Type Culture Collection (No. CCL 107). HeLa cells were
from Flow Laboratories, Rockville, Md. The 313 and Balb/C
373 cell lines were from Dr. George Todaro. The RG-179 line
was a permunent cell line obtained after multiple passages of
brain cells from 5-day-old Fisher rats. The human astrocytoma
CHB and rat glioma C2; were provided by Dr. S. Pfeiffer and
Dr. Hl. Schein, respectively. The C2, cell was originally de-
veloped by Dr. G. Benda.

Preparation of Homogenates

 

Surface Cultures—The procedure was designed to wash cells
free of serum protein and to recover cell enzymes reproducibly
in high yield. Growth medium was removed from Petri dishes
and discarded and the cell monolayer was gently washed twice
with 10 ml of Solution D containing 0.14 mm CaCl, and once
with 10 ml of Solution D. The Petri dish was drained for 90
sata 45° angle, then the dish was scraped with a spatula 3 em
wide (the end of a flexible plastic ruler) covered with disposable
Teflon tape (Scientific Specialties Services, Inc., Randallstown,
Md.). Cells were recovered by aspiration with a large bore
micropipette and transferred to a polyallomer tube, 3 x 0.5
inch. The surface of the dish was washed twice with 0.1 to
1.5-1n] portions of Buffer A (50 mm potassium phosphate buffer,
pIl 6.8; 1 mm EDTA, potassium salt) at 3°. The amount of
Buffer A added was adjusted so that the final protein concentra-
tion was 2 to 10 mg per ml of homogenate. Cells contained in
the washes were recovered and combined with the scraped cells.
The tubes were stoppered and placed in cold H.O in the chamber
of an ultrasonic oscillator (Raytheon, model No. DF101); cells
were lysed by sonication at 1° for 5 min. Homogenate volumes
were measured; each homogenate was divided into small por-
tions which were frozen quickly in a Dry Ice-acetone bath and
stored in the vapor phase of a liquid nitrogen freezer.

JAfouse Brain—Brain tissue from Balb/e mice was washed
with Solution D, blotted, minced in approximately 7 volumes of
Buffer A, and homogenized at 1° with a Potter-Elvehjem ho-
mogenizer and then by sonicution as described above. The
homogenate was centrifuged at 35,000 X gimax for 15 min at 1°;
the supernatant fraction (4 to 12 mg of protein per ml) was
divided into 0.05-ml portions, quick frozen, and stored at — 100°.

Protein concentration was determined by a modification of
the method of Lowry eé al. (13) with 3 to 20 wg of protein per
reaction, DNA was determined by the spectrofluorophoto-
metric method of Kissane and Robbins (14). Histochemical
tests were kindly performed by Dr. Lloyd Guth.

inzyme Assays

Hfomogenates in Buffer A were thawed shortly before use and
portions were diluted with appropriate modifications of Buffer
A to adjust the homogenates to the specific conditions of each
assay. ILomogenates were added to reactions last. Each ho-

Markers for Neuronal Genes

Vol. 247, No. 10

mogenate was assayed at four concentrations; values were used
only if the rate of reaction was proportional to the homogenate
concentration. Triplicate homogenates were prepared and
assaved routinely; the average values are shown,

Acetylcholinesterase Assay

Acetylcholinesteruxe activity was assayed by a modification
of the methods of Reed et a/. (15) and Ehrenpreis e¢ a7. (16). Neu-
roblastoma clone N-18 homogenates were thawed immediately
before use and adjusted to the assay conditions by mixing +
volumes of homogenate with 1 volume of Buffer A containing
1.0 u NaCl and 2.66% (v/v) Triton X-100. The radioactive
substrate, [2-'H}acetvicholine chloride, 250 mCi per mmole
(Amersham-Searle) or [l-“Clacetylcholine iodide, 2.4 mCi per
mmole (New England Nuclear) was dissolved in HO and ly-
ophilized for 16 hours to remove possible volatile contaminants.
The cation exchange resin AG 50W-X8 (H+ form, 100 to 200
mesh, Bio-Rad Laboratories) was converted to the Nat form
with 2 x NaOH at 25° for 45 min and then washed with H.O
until the pH of the effluent was 6.0. Columns (0.5 x 5.0 em)
of resin were formed over small plugs of glass wool in 9-ineh
disposable Pasteur pipettes and washed with H.O.

Each reaction contained the following components in a final
volume of 50 pl unless stated otherwise: 2.8 mm [2-°H]acetvl-
choline chloride (0.15 wCi per reaction, 1.08 mCi per mmole),
200 mm NaCl, and 0.5°% Triton X-100 in Buffer A; and 0 to 40
ul of neuroblastoma homogenate, or Buffer A containing 200 mu
NaCl and 0.5% Triton X-100, or both. Reactions in disposable
glass tubes, 10 & 75 mm, were incubated for 10 min at 37°, then
transferred to an ice-water bath and diluted rapidly by the addi-
tion of 1.0 ml of HO at 1°. Each diluted reaction was immedi-
ately passed over a column of the cation exchange resin; the
tube was washed with two 1.0-ml portions of HO at 1°, and the
washes also were passed through the column. The column
effluent was collected in a glass scintillation vial, then 10 ml of
Triton X-100-toluene-Liquifluor (333 ¢:666 m1:55 ml) scintilla-
tion fluid were added and radioactivity was determined with a
scintillation counter (82% counting efficiency).

Since reactions are not deproteinized, the column step must be
performed rapidly. The rate of acetylcholinesteraxe activity
after dilution at 1° was approximately 1.5% that of undiluted
reactions at 37°, Therefore, the assay was performed in batches
of 20 tubes or less so that all reactions could be passed through
columns within 3 min. In some instances distinction was made
between “true” and “pseudo” cholinesterases by employing
1073 m BW 284C51 dibromide.

Catechol O-Methyltransferase Assay

The assay described was a modification of the method of
Nikodejevic e¢ al. (17). Neuroblastoma homogenates were ad-
justed to the conditions of the catechol O-methyltransferase
assay immediately before use by mixing 9 volumes of homogenate
with 1 volume of Buffer A containing 50 mm MgCl. The 1-S-
adenosy][methyl-“C]methionine, 58 mCi per mmole (obtained
from Amersham-Searle in approximately 0.001 n H.S0,), was
extracted twice with 10 ml of toluene; the aqueous phase was
lyophilized before use. Nonradioactive L-S-adenosylmethionine
iodide and 3,4-dihydroxybenzoic acid were obtained from Cal-
biochem Corp.

Each reaction contained the following components in a final
Issue of May 25, 1972

volume of 50 ul except where stated: 0.55 ma 1-S-adenosyl-
lnethyl-4C|methionine iodide (0.318 wCi per reaction, 11.55 mCi
per mmole), 2.5 mat dihydroxybenzoic acid, and 5 mm MgCl in
Buffer A; and 0 to 40 yl of neuroblastoma homogenate, or Buffer
A containing 5mm MgCl, or both. Each reaction was incubated
ina 12-ml conical glass centrifuge tube with ground glass stopper
for 20 min at 37°, then transferred to an ice-water bath and 0.2
ml of 1.0N HCl and 10 ml of toluene were added. ‘Tubes were
shaken vigorously for L min, and centrifuged for 5 min at 250 x
g. Then 9 ml of the toluene phase were removed and transferred
to a scintillation vial containing 5 ml of toluene-Liquifluor (958
ml:42 ml) scintillation solution and radioactivity was determined
(87¢¢ counting efficiency). A correction was applied so that
eich value reported represented the entire toluene phase. Au-
thentic 3-[4C]methoxy-4-hydroxybenzoic acid and 4-[4C]me-
thoxy-3-hydroxy benzoic acid were prepared with purified rat liver
cutechol O-methyltransferase obtained from Dr. C. R. Creveling.

Choline Acetyliransferase Assay

The assay described was a modification of the method of
Schrier and Shuster (18). Mouse brain extracts were thawed
immediately before use and adjusted to the choline acetyltrans-
ferase assay conditions by mixing 4 volumes of brain extract with
1 volume of Buffer A containing 1.0 mM NaCl and 2.5¢% Triton
X-100. The [1-"Clacety1-CoA (50 mCi per mmole) (New Eng-
land Nuclear) was lyophilized for 16 hours before use. Un-
labeled acetyl-CoA (trilithium salt, trihydrate, A grade) and
neostigmine methylsulfate were from Calbiochem, choline iodide
was from Schwarz. The anion exchange resin AG 1-X8, (C17
form, 100 to 200 mesh, Bio-Rad Laboratories) was washed with
HO until the effluent was pH 5.5, and columns of this resin for
each assay were prepared as described for the acetylcholines-
terase assay.

Kach reaction contained the following components in a final
volune of 50 wl, except where stated: 0.21 mat [1-4C]acetyl-CoA
(6 mCi per mmole), 2 mM choline iodide, 200 mm NaCl, 0.L mm
neostigmine methylsulfate, and 0.56 Triton X-100 in Buffer A;
and 0 to 40 wl of homogenate, or Buffer A containing 200 mu
NaCl and 0.46% Triton X-100, or both. Reactions were incu-
bated in glass tubes, 10 x 75 mm, for 10 min at 37°, then trans-
ferred to an ice-water bath and diluted by the addition of 1.0 ml
of H.O at 1°. The contents of each tube were passed through
an anion exchange column; the tube was washed with two 1.0-ml
portions of 1,0 at 1°, and the washes were also passed through
the column. The column effluent was collected in a glass scin-
tillation vial, then 10 ml of Triton-toluene-Liquifluor scintillation
mixture (see above) were added and radioactivity was deter-
mined at a counting efficiency of 87%.

Glutamate Decarboxylase Assay

Two modifications of the method of Wingo and Awapara (19)
were employed.

Method a—Homogenates were thawed prior to use and ad-
justed to the conditions of the glutamate decarboxylase assay by
mixing 9 volumes of homogenate with 1 volume of Buffer A con-
taining 10 mm 2-mercaptoethanol, 5% Triton X-100, and 5 mm
pyridoxal phosphate. The 1-[1-"“C]glutamic acid, obtained from
Calbiochem, was neutralized with KOH and lyophilized before
use. Anino-oxyacetic acid was obtained from Sigma.

Each reaction contained the following components in a final

Wilson et al,

3161

volune of 50 wl, except where stated: 5 mm L-{1-“C]glutamic aeid,
potassium salt (1.04 wCi per reaction, 4.16 mCi per mmole),
1 mM 2-mercaptoethanol, 0.5 mm pyridoxal phosphate monohy-
drate, and 0.5¢% Triton X-100 in Buffer A; and 0 to 40 ul of
homogenate, or Buffer A containing 0.5% Triton X-100, 1 mat
2-mereaptoethanol, and 0.5 mx pyridoxal phosphate, or both.
Reaction mixtures in glass tubes, 10 x 37 mm, were sealed with
tightly fitting rubber stoppers that could be easily perforated
with a needle. Reactions were incubated for 10 min at 37° and
then transferred to an ice-water bath. Each tube was then
placed in a glass scintillation vial containing 5 ml of Hyamine-
toluene-Liquifluor scintillation solution (288 ml of 1 Mm Hyamine
hydroxide in methanol to 640 ml to 42 ml). The vial was sealed
with the reaction tube stopper; then 0.2 ml of a solution of 10
mM acetic acid in methanol was injected through the rubber
stopper into the reaction in the tube with a hypodermic syringe.
The injection needle was used to dislodge the glass reaction tube
from the rubber stopper so that the tube fell to the bottom of the
seintillation vial. The stoppered vial was incubated for 30 min
at 24° to allow absorption of “COs by the Hyamine solution.
Then the reaction tube was removed from the scintillation vial
with forceps and the vial was capped with a scintillation vial top.
Radioactivity was determined by scintillation counting at 80%
efhciency.

Vethod b—This alternate method differed only in the technique
of #CO. collection. Each reaction was performed in the 6-mm
diameter center well of a 10-ml Erlenmeyer flask (Kontes)
equipped with a tightly fitting rubber stopper. Hydroxide of
Hyamine at 37° (Packard Instruments, Downer’s Grove, TL)
was placed in the outer portion of the flask and reaction compo-
nents were added to the center well. The flask was sealed with
the stopper and incubated at 37° for 11 min. The reaction was
terminated by injection of 0.2 ml of 10 mM acetic acid in meth-

TaBLe I

Range and sensitivity of assays

 

 

 

 

Usual
and amount
Source of Incu- | . : tivity, | eat
Assay homogenate ee | Radioactive product arts ' prod
| or 3H |
product Protein
reaction
min pmoles | pmoles
Acetyleho- | Neuroblas- 10 H]Acetate 1,000-|75 , 000
linester- toma clone 14,000
ase N-18 [4C]Acetate 75-|75 , 000
14,000
Choline ace- | Mouse brain . 10 | [#C]Acety!- 5-| 1,500
tyltrans- (age 35 choline 700
ferase days)
Glutamate Mouse brain 10 CO; 20- 150
decarbox- (newborn) 3,000
ylase
Catechol O- | Neuroblas- 20 | 3-[4C]Me- 10- 7d
methyl- toma clone thoxy-4-hy- | 1,200
transfer- N-18 droxybenzoic
ase acid?

 

 

 

 

* Approximately 5°, of the “C-product is 4-[!4C]methoxy-3-
hydroxybenzoie acid.
3162

anol through the stopper into the center well. After an addi-
tional 90 min at 37° an aliquot of the Hyamine-“CQ2 solution was
transferred to a scintillation vial containing 10 ml of toluene-
Liquifluor scintillation mixture. Radioactivity was determined
by scintillation counting at 87% efficiency.

RESULTS

Assays for Marker Enzymes

The usual range and sensitivity observed for the four enzyme
assays are shown in Table I. In evaluation of the assays neuro-
blastoma clone N-18 homogenate was used for acetylcholines-
terase and catechol O-methyltrausferase assays; mouse brain ex-
tracts were used for evaluation of the assays for choline acetyl-
transferase and glutamate decarboxylase, although these assays
were also applicable to cell culture extracts. The range of linear-
ity with protein for neuroblastoma choline acetyltransferase was
about half that obtained with mouse brain homogenates (20).

Conditions were selected so that a homogenate could be pre-
pared by harvesting a single Petri dish and stored frozen for
future use. About 30 mg of protein could be harvested from a

      
 
 
 
 
    

 

5 1,000 A z 20.000} B

- cL |

o 5 » ACETYL

a GLUTAMATE a CHOLINESTERASE

W750 DECARBOXYLASE WJ 15,000-

é ™~™

Lae E

S 8

a

8 500 § 'a000} )

& x | /
ol EHOLINE ACETYL- |

= TRANSFERASE |r /
=5) 250 7m’ 5,000

4

Q CATECHOL O-METHYL]

2 TRANSFERASE | =

    

Oo

10=—20.—CO30—s« 4 0 10 ©620.—30
VOLUME OF HOMOGENATE (yl) /REACTION

Fic. 1. The relation between concentration of enzyme extract
and reaction velocity for glutamate decarboxylase, choline acetyl -
transferase, and catechol O-methyltransferase (Panel A) and for
acetylcholinesterase (Panel B). Reactions contained the com-
ponents described under ‘Methods,’ and enzyme extract protein
at the following concentrations: 12.2, 1.06, 5.0, and 1.0 ug of pro-
tein per ul of extract for glutamate decarboxylase, choline acetyl-
transferase, catechol O-methyltransferase, and acetylcholinester-
ase, respectively. Glutamate decarboxylase was determined by
Method a.

 

  
  

 

 

"GLUTAMATE C. ue
oS TaMATE |
& 4,000 B20, 000" seerye- |
i FE ; CHOLINESTERASE
QO oO
a 4 ;
fe 3,000 1 ® 15,000} /
~ 5
b _o df
8 B 10,000 }
8 2,000 CATECHOL o- : Qu f
we METHYL~ ao
a TRANSFERASE
SF 1000 owe oF
a O00) TRANSFERASE | a 5,000 | ]
| iW
2 A 3
g iS
De i! 2 ve
a ° 30 60 90 0 30 60 930 oO 30 60 90
MINUTES MINUTES
hia. 2. The relation between length of incubation and amount

of product. formed. Hach glitamate deearboxylase, catechol
O-methylransferase, choline acetyltransferase, and acetyleholin-
esterase reaction contained the eomponents described under
“Methods” and 488, 200, 42.4, and 8.0 uz of protein, respectively.
The dashed line in Panel B represents the theoretical linear reac-
tion rate. Ghitamate decarboxylase was determined by Method
a.

Markers for Neuronal Genes

Vol. 247, No. 10

150-mm Petri dish (145 ecru?) containing 2 confluent monolayer of
neuroblastoma cells. Routinely, less than 1.0 mg of protein was
used to determine the specific activity of each enzyme studied,
and 40 to 100 assays were performed per day.

Kinetics of Reactions

As shown in Fig. 1, the rate of each reaction was proportional
to concentration of homogenate protein within the range studied.
Conditions were adjusted so that the proportion of radioactive
substrate converted to product was <5°; in the case of choline
acetyltransferase and <10% with the other three enzymes.

The relation between time of incubation and reaction rate is
shown in Fig. 2. Under the conditions employed, reactions
‘atalyzed by glutamate decarboxylase, catechol O-methyltrans-
ferase, choline acetyltransferase, and acetylcholinesterase were
linear for 60, 98, 20, and 90 min, respectively.

The relation between reaction mixture pI] and enzyme activity
is shown in Fig. 3. At pH 6.8 the activities of glutamate decar-
boxylase and catechol O-methyltransferase were essentially maxi-
mal (the latter was inhibited by Tris); however, choline acety1-
transferase and acetylcholinesterase activities were 60 to 65°¢ of
the observed maxima. Although these latter two activities were
optimal between pl 8 and 8.5, the four enzymes were routinely
assayed at pli 6.8 because acetyl-CoA and acetylcholine are more
stuble at pI16.8 than at pIL8. Inaddition, thesame homogenate
could be used to determine the activity of each enzyme without
adjustment of pI.

Enzyme Characteristics

Isnzyme stability and product recovery is shown in Table I.
Inzyme activity was not affected appreciably by freezing and
thawing homogenates or by keeping the enzyme extracts at 1°
for 2 to 3 hours prior to assay. However, each enzyme tested
was completely inactivated after incubation at 100° for 10 min.

Although little or no activities of the other enzymes were de-

tected at 1°, acety]cholinesteraxe activity at 1° was approxi-
; 2 : VI

 

 

1200p 12,000; ;
3 Neel ACETYLCHOLINESTERASE
5 4000 “(3 4 510,000; ax
i | o °
ied fra ' s
~ 800 7X 8,000, {
8 6 8 : /
8 600 caT + S 60007 sf
a x
a a.
7, ‘| .
#4090 | eo 4,000 s
n wn /
4 uw /
dS 200 + 82,000 6
= =
a comt | &
Pee eer po
OS a a
pH py

Fro. 3. The effect of pH upon reaction rates. Each glutamate
decarboxylase, choline acetyltransferase, catechol O-methyl-
transferase, and acetylcholinesterase reaction contained the com-
ponents described under ‘Methods,’ the specified buffer, and 122,
26.5, 50, and 10 ug of protein, respectively. Glutamate deear-
boxylase was determined by Method @. The formation of radio-
active products was determined with standard 50-ul reactions;
pH was determined in 0.5-ml reactions that were identical with
the 50-zl reactions except that unlabeled substrates were em-
ployed. Solid symbols represent reactions with 50 mat potassium
phosphate buffers; open symbols represent 50 mM Tris-HCl buffers.
The abbreviations used are: GAD, glutamate deearboxylase; CAT,
choline acetyltransterase; COMT, catechol O-methyltransferase.
Issue of May 25, 1972

Tasur II
Enzyme stabtlily and product recovery

Each reaction contained the components deseribed under
“Methods” and 40, 15, 366, or GO wg of protein for choline acetyl-
transferase, acetylcholinesterase, glutamate decarboxylase, and
catechol O-methyltransferase reactions, respectively. The
amount of radioactive product formed (picomoles) per complete
reaction corresponding to 100°; were 610, 11,300, 630, and 114 for
choline acetyltransferase, acetylehclinesterase, glutamate de-
rarboxvlase, and eatechol O-methyltransferase reactions, re-
spectively. Glutamate decarboxylase was determined by Method
a (see “‘Methods’’). Product recovery was tested by adding the
folowing eompounds to reactions in place of radioactive sub-
strate: choline acetyl transferase assay, 28 nmoles of [1-!C]acetyl-
choline iodide (1.58 & 10° dpm); aecetylcholinesterase, 208 1moles
of sodium [4CJacetate (8.05 & 106dpm); glutamate decarboxylase,
35 umoles of NaH'CO; (3.97 « 105 dpm); catechol O-methyl-
transferase, 0.133 nmole of 3-[4C]methoxy-t-hydroxybenzvie acid
(3.4 * 10% dpm) and 0.015 nmole of 4-[4C]methoxy-3-hydroxy-
benzoic acid (350 dpm).

1
'Catechol

 

Choline Acetyl- Gluta-
Modification acetyl cholin- mate O-meth-
Forase [P'S posettse "feuce
re Ve % % %
Enzyme stability | |
Complete reaction. .... ; 100 | 100 100 100
Minus enzyme. | 2 | 9 8 9
fougzvme frozen and “thawed | i |
three times... 0. .0.0.0.0.0..0...... 98 | 99 | 109 108
Inzyme held at 1° for 2-3 hrs.. | 100 | 100 | 116 104
Iunzyme held at. 100° for 10 min.: 25 9 | 8 8
Reaction incubated at 19.0.0... 9 31 | 10 7
Product recovery
tadioactive product added in- |
stead of substrate. .......... 95 110 | OF 83

 

 

mately 20¢¢ of that found at 37°. For this reason acetylcholines-
terase reactions were performed in batches (<20 tubes) so that
reactions could be chilled immediately after incubation, diluted,
and passed through the ion exchange columns in <3 min.

Radioactive product was added to reactions instead of sub-
strate to determine the percentage of product recovered. Greater
than 94% of the radioactive products of the choline acetyltrans-
feriuse, acety lcholinesterase, and glutamate decarboxylase reac-
tions and 836% of the radioactive product of the catechol O-meth-
yitransferase reaction were recovered.

The effect of reaction components and other compounds on the
rate of enzyme activity is shown in Table IIT. Choline acetyl
transferase from mouse brain was completely dependent upon
choline for activity and was stimulated by 200 mm NaCl En-
zyme activity was not affected appreciably by omission of Triton
X-100 or by the addition of 2mm MgCh.

Neuroblastoma acetylcholinesterase activity was reduced in
the absence of Triton X-100; however, omission of NaClor EDTA
in the presence or absence of MgCl was without effect. Marked
inhibition was observed by 107° m BW 284C41, a potent inhibitor
of acetylcholinesterase (EC 3.1.1.7).

The activity of glutamate decarboxylase from mouse brain was
reduced slightly in the absence of pyridoxal phosphate but was
not iffected appreciably by omission of EDTA or 2-mereapto-
ethanol, The reaction was stimulated slightly by omission of

Wilson et al.

31638

TasLE III
elssay conditions

Each reaction contained the components described under
“Methods,’’ except as noted, and the following: 40, 15, 366, and 60
ve of homogenate protein for the choline acetyltransferase,
acetylcholinesterase, glutamate decarboxylase, and catechol O-
methyltransferase reactions, respectively. Iuzyme — specific
activities, in the order stated above were: 1,480, 64,400, 140, and
88 pmoles of product per min per mg of protein. Glutamate
decarboxylase was measured by Method @ with {1-"C]glutamic
ac id as substrate.

 

 

 

: smoun
: Tadioactive
Modification product centae
per min
i pales re
Choline acetyltransferase
Complete. 0000 eee | G4 ; 100
Minus enzyme (mouse brain)...............43 2 3
Minus choline... .....0.0..0.0.0.000.0002 000% 2 :
Minus NaCl... 00.000. ee 40 62
Minus Triton X-100..000...0.0....00....... 61 » OF
Plus 2mm MgCls..0.0000000000000.0000.0... 59 91
Acetylcholinesterase
Complete. 1130 * 100
Minus enzyme Cneurobl astoma). veveeeet 105 th
Minus Triton X-100.....0.0..0.0...0........... - 750 » 66
Minus NaCl... 2.0.02 ee ; 1140 101
Minus EDTA.. . Le 1160 102
Minus EDTA, plus 1 mm uw MgCle. eee 1130 100
Plus 107° w BW 284C51. 0 eee. j 135 12
Glutamate decarboxylase i
Complete... 0.2.0.0. | 63 » 100
Minus enzyme (mouse brain)... ........-.. | 8
Minus pyridoxal phosphate... ............... | 55 87
Minus EDTA.. ol ittttecseeeee.f 68 | 110
Minus 2-mere aptoethanol So | 60 > 96
Minus Triton X-100...000.000...............1 87 » 140
Plus l mu MgCly. o.oo eee : 76 i 120
Plus 10 mm iodoacetamide... ............... J 18 » 29
Plus 10 mm hydroxylamine.................. 8 . 2
Plus 1 mm amino-oxyacetie acid. 2.0.0.0... i 8 12
Catechol O-methyltrausferase !
Complete.....0.0.0.000.0000 000 c cee | 5.7; 100
Minus enzyme (neuroblastoma)............. 0.48 9
Minus dihydruxybenzoie acid... ............. 0.69 12
Minus dihydroxybenzoic¢ acid, plus 1 mm p- |
hydroxy phenviacetic acid. .......0.0..0... 0.76 | 13
Plus 1 mm p-hydroxyphenylacetic acid...... 6.5 115
Minus MgCly.. 0000 ence 0.98 | 17
Minus EDTA... 00.000. 0c een 5.7; 100
Minus MgCl, and EDTA... 000. 3.0 } 52
Plus 0.52 (v/v) Triton N-100............... , 6.6 5 115
Plus 1 mm 2-mercaptoethanol............... 7.9 140
Plus 0.25 mm dihvdroxymethoxybenzoic acid. 1.3 j 28
Plus 1mm tropolone.. 0. 0.945 17

Triton X-100 or by the addition of 1 mm MgCl. Marked in-
hibition was observed in the presence of iodoacetamide, hyvdrox-
ylamine or amino-oxyacetic acid, as expected (21).

The activity of neuroblastoma catechol O-methyltransferase
was almost completely dependent on the substrate, dihydroxy-
benzoie acid. p-Hydroxyphenylacetie acid did not serve as a
substrate for the enzyme and did not inhibit methyl transfer to
3164

dihydroxybenzoic acid. Hence, the enzyme was capable of dis-
tinguishing between mono- and dihydroxy substrates. Omission
of MgCl, markedly reduced the rate of reaction; omission of
EDTA was without effect. Enayme activity was markedly in-
hibited in the presence of compounds known to inhibit catechol
O-methyltransferase, such as dihydroxy methoxybenzoic acid (17)
and tropolone.

Product [dentification

The radioactive products of the choline acetyltransferase, ace-
tylcholinesterase, und catechol O-methyltransferase reactions
were characterized by thin layer or paper chromatography as
shown in Table TV. Virtually all of the radioactive product of
the acetylcholinesterase reaction wax chromatographically indis-

Markers for Neuronal Genes

Vol, 247, No. 10

tinguishable from authentic sodium acetate added as carrier.
Greater than 73¢¢ of the applied radioactive produet was recov-
ered after chromatography.

Similarly, virtually all of the radioactive product formed in the
choline acetyltransferuse reactions with a mouse brain homoge-
nate was chromatographically indistinguishable from authentic
acetylcholine chloride. It has been shown (20), however, that
with some homogenates possessing low choline acety|transferase
activity, values may be spuriously high due to the formation of
radiouctive products other than acetylcholine.
as described is valuable as a rapid screening technique, but must
be validated by product identification.

Most of the radioactive product formed in the presence of
neuroblastoma catechol O-methyltransferase preparations co-

Thus, the assay

Taste IV

 

Chromatographic —systems—Solvent 1,  ethanol-HzO-concen-
trated NH.OH (8:1:1) (Reference 22), thin layer chromatography
with MN Polygram Cel 300; Solvent 2, 1-butanol-cyclohexane-
ethylene glycol-H,O-coneentrated NH,OH-cyclohexylamine (30:
30:10:3.7:0.07:0.05), thin layer chromatography with MN Poly-
gram Cel 300; Solvent 3, 1-propanol-0.1 N acetic acid (8:1) (Ref-
erence 23), Whatman 3MM paper; Solvent 4, 1-propanol-formie
acid-H2,0 (8:1:1) (Reference 23), Whatman 3MM paper; Solvent
5, L-butanol-1-propanol-H2O (4:2:1) (Reference 23), MN Polygram
Cel 300 thin layer; Solvent 6, 1-propanol-benzyl alcohol-H.O
(5:2:2) (Reference 23), MN Polygram Cel 300 thin layer;
Solvent 7, 2-propanol-coneentrated NHsOH-H.O (8:1:1), Silica
Gel G thin layer; Solvent 8, 1-butanol-formie acid-benzene-H.O
(15;25:5:1:1.5), Silica Gel G thin laver; Solvent 9, benzene-gla-
cial acetic acid-methanol (10:1:5), Silica Gel G thin layer.

Acetyleholinesterase— The reaction contained the components
described under ‘‘Methods” except that 2.78 mm [2-°H Jacetyl-
choline chloride (0.43 Ci) and 48 yg of neuroblastoma clone N-18
protein were present (61,200 pmoles per mg of protein per min).
The column effluent. was collected and NaOH was added to a final
concentration of 0.18 mM. Authentic sodium ["C]acetate was
added and portions were applied to thin layer chromatograms.
The chromatograms were developed by ascending chromatogra-
phy, dried, covered with cellophane tape, cut into 0.5-em hori-
zontal strips, and each was placed in a scintillation vial with 20
ml of toluene-Liquifluor scintillation solution (see “Methods’’)
for determination of radioactivity at efficiencies of 54% (4C) and
16% @H).

Choline Acelyltransferase—The reaction contained the eompo-
nents described under ‘“Methods” except that 8.1 nmoles of
(2-8H]acetyl-CoA (1.24 Ci per mmole) were used as substrate,
choline chloride was added in place of choline iodide, 168 ug of
homogenate protein from mouse brain were added, and the reac-
tion was incubated for 20 min at 37° in a final volume of 0.1 ml.
Authentic [“Clacetyleholine chloride was added to the reaction
after incubation. Acetic acid was added to the column effluent
to a final concentration of 0.06m. Portions of the column effluent
were subjected to ascending paper or thin layer chromatography.
leveloped chromatograms were cut into 1.0- or 0.5-em strips and
each was placed in a scintillation vial containing 20 ml of tuluene-
Liquifluor scintillation solution for determination of radioactivity
as described above.

Catechol O-Methyllransferase—The reaction mixture contained
the components described under “Methods” in a final volume of

0.25 ml with 750 yg of neuroblastoma N-18 homogenate protein.
The reaction was incubated for 60 min and extracted with toluene;
the toluene phase was evaporated to dryness under a stream of
nitrogen at 24° and the residue then was dissolved in 0.35 ml af
methanol. Of the radivactivity originally present in the taluene
extract, 81% was recovered in the methanol solution. Authentic
3-methoxy-4-hydroxybenzoic acid and 4-methoxy-3-hydroxy-
benzoic acid were added and portions of the extract were subjected
to ascending thin layer chromatography. Chromatograms were
dried at 24° and marker compounds were located by exposure to
wtraviolet light. Chromatograms were cut into strips, 1 x 3
em, and each was placed in a scintillation vial containing 10 ml
of Triton-toluene-Liquifluor scintillation solution for derermina-
tion of radioactivity.

 

 

Percentage |
of radio- Percentage
Chromato- active of applied
Enzyme graphic product radio- Re
solvent with Rp of| activity
authentic | recovered
compound
re | % % :
Sodium
| (§HJacetate
Acetylcholinesterase 1 72 73 0.70
2 90 90 0.25
(3HjAcetyl-
choline
chloride
Choline acetyl! transferase 3 80 2 0.66
+ 80 81 0.77
5 86 87 0.49
6 103 104 0.89
0,65¢
3[MC]-
Methoxy-
4-hydroxy-
benzoic acid
Catechol O-methyltrans- 7 84. 92 0.38
ferase 8 73 78 0.91
9 76 81 0.86
4-[UC]}-
Methoxy-
3-hydroxy- :
benzoic acid |
Catechol O-methylirans- 7 5 92 0.76
ferase 8 4 78 0.75
9 i 5 81 0.75

 

 

 

 

¢ With Solvent 6 two peaks of radioactivity were found, both with authentic (4CJacetylcholine chloride and with the tritiated
reaction product. This phenomenon has beeu attributed to the effect of salts on the chromatographic mobility of acetylcholine

(23).
Issue of May 25, 1972

chromatographed with authentic 3-methoxy-4-hydroxybenzoie
acid; 4+ to 5% of the reaction product exhibited the echromato-
graphic mobility of 4-methoxy-38-hydroxybengoie acid. Greater
than 78° of the applied radioactive material was recovered alter
chromatography. No discrete contaminants were detected.

Characterization of the products of the glutamate decarboxyl-
ase reaction is shown in Table V. In addition, the amounts of
y-aminolU-4C butyric acid and “CO. formed during incubation
with L-[U-“C glutamic acid as substrate are compared. The re-
sults show that y-amino|U-4C]butyrie acid was formed during
Incubation and also that similar amounts of y-aminobutyric acid
and CO: were formed. However, ax shown below, this 1:1 ratio
of products was not found with all tissues assessed. Because of
the possibility that #CO. may not accurately reflect production
of y-uninobutyric acid, quantitative identification of the latter
product is required to validate the glutamate decarboxylase
assay.

Substrate Concentration

The relations between substrate concentrations and enzyme
activities are shown in Fig. 4. The apparent Michaelis con-
stants, determined by the method of Lineweaver and Burk (24),
were us follows: acetylcholinesterase, 9.1 x 1074 M acetylcholine-
chloride, choline acetyltransferase, 1.5 107° M acetyl-CoA
(with 2 x 107* m choline iodide), and 9.1 x 107+ m choline iodide
(with 2.1 x 1074 M acetyl-CoA); glutamate decarboxylase, 3.7 x
107? M L-glutamic acid; and catechol O-methyltransferase, 2.3 x
10™ wi dihydroxybenzoie acid (with 5.5 x 1074 Mm L-S-adenosyl-
methionine), and 4.0 x 107°  L-S-adenosylmethionine (with
2.5 < 107? Mm dihydroxybenzoie acid). Substrate inhibition was
detected only in the case of acetylcholinesterase, similar to ob-
servitions that have been reported with that enzyme from other
tissues (25).

In standard reactions (Fig. 1) substrates were present in ¢on-
centrations well above their respective enzyme Km values and
less than 10% of the radioactive substrates were converted to
product during reactions. High substrate concentrations were
used to maintain zero order kineties and to reduce error due to
endogenous substrates or inhibitors and other reactions compet-
ing for substrate.

Recovery of Protein and Enzyme Activity from Petri Dishes

The effectiveness of the harvesting procedure was examined
by comparing the protein yield obtained by harvesting replicate
plates by the scrape-wash method with the amount of protein
recovered by the addition of 0.1 x NaOH to dishes (Table VD.
The results show that 95% of the protein was recovered by the
standard harvesting procedure. In experiments with a variety
of cells in culture the reproducibility between replicate plates was
+15% with respect to the specific activity of the four enzymes
and +25% with respect to the yield of protein recovered per
plate.

The recovery of enzyme activity and cell protein also was
studied by adding to Petri dishes known quantities of a mixture
of neuroblastoma N-18 and brain homogenates which had previ-
ously been characterized with respect to enzyme activity and
protein concentration. The homogenate was then recovered by
the usual procedure. A portion was set aside for future assay of
enzyme activity and protein concentration; the remainder was
added to a fresh Petri dish and recovered again. ‘The harvesting
cycle was performed four times. The results, presented in Fig. 5,

Welson et al.

3165

Tasin V
BCO2 and y-amino[hC]bulyric acid synthesis in glutamic acid
decarboxylase reaction

The reaction contained 2.44 mg of protein from mouse brain
(156 pmoles of 4CO. formed per min per mg of protein) and the
components described under ‘Methods’? except that u-[(U-4C]-
ghitamate was employed as substrate and the final volume was
0.25 ml. Tneubation was for 45 min at 37°. The reaction was
terminated by the addition of 0.175 ml of 10 mat acetie acid in
methanol. The evolved “CO. was collected for 60 min by Method
a. The acidified reaction then was centrifuged at 15,000 x g
for 10 min and the supernatant fraction was collected. Unlabeled
y-aminobutyrie acid, glutamie acid, and glutamine were added
and portions were applied to thin layer chromatograms of cellulose
MN Polygram Cel 300 and subjected to ascending thin layer
chromatography with Solvent Systems 1, 2, and 3 (two-dimen-
sional chromatography with System 3) and to Whatman 3MM
paper for high voltage electrophoresis (45 min at 61 volts per em
and 180 ma) with Solvent 4. Solvent systems were: Solvent 1,
2-propanol-methanol-concentrated NH.OH (9:7:4); Solvent 2
phenol saturated with 6 m NH,OH; Solvent 3, 2-propanol-1-
butanol-l n HCI (6:1.5:2.5) for the first dimension and phenol
saturated with H,O for the second dimension; and Solvent 4,
pyridine-acetic acid-H.0, pH 3.95 (1.8:5.0:144.5). Chromato-
grams and electropherograms were dried, stained with 0.25%
ninhydrin in aeetone, and cut into 0.5- er 1.0-cm strips. Each
strip was placed in a scintillation vial with 10 ml of Triton-tolu-
ene-Liquifluor seintillation solution and radioactivity was deter-
mined. Specific activities of “CO, and y-amino[U-"C butyric
acid reaction products were 1,580 and 6,320 dpm per nmole, respec-
tively.

 

 

Amount of |
Solvent system ach, lau Amant of | yutobu
recovered acy
—___. .-—. ne re i... ]— | —_—_ | ___. —
mmoles | ninoles
1 2% | 2 0.42
2 33 | 22 | 0.88
3 (1st dimension) 28 | 22 0.77
(2nd dimension) | | 0.89
4 | 33 22 oo

show that recovery for each harvest cycle was 96% for protein,
87° for glutamate decarboxylase, 95% for choline acetyltrans-
ferase, 83% for catechol O-methyltransferase, and 88% for acety-
cholinesterase. Thus, enzyme activity released from cells dur-
ing the scraping procedure would be recovered in high yield and
with little alteration of enzyme-specific activity.

sarker Enzymes in Cell Culture

Non-neuronal Cells in Culture—In order to determine the
specificity of the marker enzymes, established cell lines from
brain and other tissues were grown in surface culture until several
days after confluency and then tested for presence of the enzyme
activities. As shown in Table VII, choline acetyltransferase,
acetylcholinesterase, and glutamate decarboxylase activities were
present in all of the cell lines examined. Activities of the three
enzymes in established cell lines were considerably lower than
corresponding activities in newborn brain.

Glutamate decarboxylase specific activity in 1-929 cells and
in glial cell lines Cs, C21, and RG-179 when measured by -y-amino-
[4C]butyric acid production was 10° or less of that measured by
4COz production. The presence of 1 ma y-aminobutyric acid in
3166

Markers for Neuronal Genes

Vol. 247, No. 10

 

 

   

 

 

 

 

 

 

 

 

; + + 14 1 1 ' , + TF
CATECHOL O-METHYLTRANSFERASE
2.5| ACETYLCHOLINESTERASE |
z zile
: :
° b 4
2 20 L10 J
° °
‘e 2
x ] a
r r 4
5 5 z10 5
3 9 >
8 b GB a
© 8 < | & 6 4
a 1.0 Bes &
a“ ~ I
g 4 70 4 IS 4 °
3 ot, = ol w tT DIHYOROXYBENZOIC ACIDS J
3 16 1? oO S2
O@ ACETYCHOLINE MOLARITY a
0 2 3 4 56 7 po

|
ACETYLCHOLINE MOLARITY x !0*

 

CHOLINE ACETYLTRANSFERASE

on

uN 4

t

CHOLINE

ow
+

  

PRODUCT x 107/10 MIN

    

ACETYL-COENZYME A

4,

PMOLES

 

1 1 4 4

a a a

 

 

{
SUBSTRATE MOLARITY x10"

Fig. 4. Relation between reaction velocity and substrate
concentration. Reactions for acetylcholinesterase, choline acetyl-
transferase, glutamate decarboxylase, and catechol O-methyl-
transferase contained the components described under ‘‘Meth-
ods”’ except for substrate concentrations as indicated and 15,
42.4, 488, and 200 ug of protein, respectively. In addition acetyl-

Taste VI
Recovery of cell protein from Petri dishes

Petri dishes (145 em? surface area) containing neuroblastoma
clone N-18 cells were incubated for 4 days and then harvested
by the procedure specified in the table. Hach method was tested
in triplicate; values shown are averages. Harvesting Method
la is the standard procedure deseribed under “Methods.’? For
Method 1b, 2.5 ml of 0.1 N NaOH was added to each dish that had
been seraped and washed. The dish then was incubated at 4°
for 15 min to dissolve protein. For Method 2, 10 mi of 0.1 N
NaOH was added to each dish containing a confluent monolayer
of eells and the dish was again incubated for 15 min at 4°.

 

 

| Amount of

 

 

protein
Method of harvesting protein | recovered
; per Petri
dish
| we
|
1. a, Dish scraped and washed (standard method)..... 6.14
b, Protein remaining in dish that had been scraped |
and washed was recovered by the addition of 0.1 |
N NaOH. eee | 0.30
Total... ee eee | 6.44
2 0.1 nN NaOH added to Petri dish containing a cell |
monolayer... 0... cece eee eee 6.00

reaction mixtures had little influence on activity by either meas-
wrement. The presence of 1 mm amino-oxyacetic acid reduced
CO: production by only 10 to 30%, vet y-aminobutyric acid pro-
duction was reduced by at least 95¢°¢. Thus, it appeared that

 

 

 

SUBSTRATE MOLARITY x 10>

 

L-GLUTAMATE DECARBOXYLASE

   
   

  

C}] PRODUCT xiO /IOMIN

   

  

 

 

 

*
06 ]
a
3
ogo
ot 2 3 4 8
a

L-GLUTAMATE MOLARITY x 10°

cholinesterase and catechol O-methyltransferase reactions were
incubated for 7 and 15 min, respectively. The data were plotted
according to the Lineweaver and Burk method (24). Glutamate
decarboxylase was determined by Methods a and 6 (Method a is
shown).

 

100

@
oO

 
 

GLUTAMATE
DECARBOXYLASE

 
   
  

Bey
oO

 
  

i! CATECHOL ~O-METHYL-

 
     

% RECOVERY OF ACTIVITY AND PROTEIN
a
Oo

 

| TRANSFERASE ACETYL —
| CHOLINESTERASE
20 ;
ul
0 } 2 3.40 | 2 3.440 | 2] 3.4

TIMES HARVESTED

Fig. 5. Recovery of homogenate protein and total enzyme sac-
tivities in multiple cycles of the scrape-wash procedure. Stand-
ard homogenates for assays of glutamate decarboxylase, catechol
O-methyltransferase-acetylcholinesterase, and choline. acetyl-
transferase were mixed in the proportion 10:4:1, added to culture
dishes, and repeatedly cycled through the serape-wash recovery
procedure described in the text. The starting homogenate con-
tained 20.3 mg of protein and enzyme specific activities of 145,
AL7, 31,850, and 118.4 pmoles of product formed per min per mg
of protein for glutamate decarboxylase, catechol O-methyltrans-
ferase, acetylcholinesterase, and choline acetyltransferase, re-
spectively. Glutamate decarboxylase was determined by Method
a.

“CO, was being produced via pathways other than glutamate
decarboxylase in the L-929 and glial cell lines tested.

Normal Brain Cells in Culture—Growth characteristics and
marker enzyme activities were assessed in surface cultures of

 
Issue of May 25, 1972

Weson et al.

3167

 

1OG

©

Fic. 6. Growth characteristics and
morphology of cultured mouse brain
cells. Mixed cells from brains of
newborn mice were obtained and cul-
tured us described under ‘‘Methods.”’
A, (©) cell number (determined by
eounting trypsirized cells); (A, A)
protein and (0) DNA eontents of
dishes at various times in culture.
Values at zero time are those deter- ec
mined for the inoeula. Open symbols,
average of determinations on eight
eulture dishes (four from each of two
separate, simultaneous time curves);
closed symbols, average of determina- O 5
tions on four culture dishes from a
third time eurve. B to D, Phase-
contrast photomicrographs of living
cultures from the time curves. Mag- ..:
nifteations of the 35-mm negatives |
were & 50. 2B, 3 days in culture.
C. 11 days in culture, arrow indicates
phase-dark cell (see text). D, 11 days
in culture, arrow indicates phase-
bright) cell (see text). #, Bodian
stain Gnodified protargol) of similar
cells grown on cover glasses for 42
days. Bright field microscopy; mag-
nifieation of the 4 5 inch negative
was X 600.

10° CELLS/DISH
oO)

f

 

 

newborn mouse brain cells. The effect of time in culture upon
viable cell number, protein, and DNA content is shown in Fig.
6.1. ach Petri dish was inoculated with 1 x 107 viable cells;
however, after 3 days only 1.2 x 108 viable cells remained.
Viable cell number and protein content increased during the 3rd
to the 16th day, although the mixed population had formed a

  
    
  

 

B

i
i)

i
oO

Protein

>
oo
PROTEIN (mg) or DNA(O.Img)/ DISH

A

L
o

!
i

i
NM

 

 

  

Jl l L

IS 20 25
CULTURE

lO
DAYS IN

“ty
*

confluent monolayer by the 7th day in culture. The morpho-
logical appearance of the cultures is shown in Fig. 6, B through #.
After 3 days of culture, cells with relatively small cell bodies and
short processes were present along with a variety of large flat
cells without processes (Fig. 68). The confluent layer of cells
present after 11 days of culture was composed of large flat cells,
3168

Tasie VII
Enzyme specific activities in newborn mouse brain and
cultured cells

Specific activities of glutamate decarboxylase, choline acetyl-
transferase, and acetylcholinesterase were determined on extracts
of cultured cells as described under ‘‘Methods.’’ An uncen-
trifuged homogenate of newborn Balb/c mouse brain was used
for comparison. The maximum content of homogenate protein
in the assays was: 237 pg for mouse brain homogenates, 110 ug
for C-6, 906 ug for C2, 852 pg for CHB, 258 yg for RG-179, 1306 ug
for L-929 (B-82 clone), 936 ug for HeLa, 533 xg for 3T3-S (Swiss
mouse 3T3), and 580 yg for mouse brain cells cultured 30 days.
Formation of y-aminobutyric acid was determined in assays with
1-(U-“4C]glutamate as substrate, followed by electrophoresis and
chromatography as described above, with both Methods a and
b for collection and determination of evolved CQs,. All deter-
minations on cultured cells were performed when cells were 7 to
20 days postconfluency. Glutamate decarboxylase activity
shown for C-6 represents the highest activity (at 31.5 mg of cell
protein per 150-mm dish) found among four separate points on a
growth curve.

Cultured cells

 

 

 

a
orn at ~ .
Determination | ain iia eabiienieibiaas __Nonbrain  ewborn
homog- a | Touse
enate e _ 3 in i 2 3 2 cells
3|8 S|”! 3 |e\&
pmoles product/min/mg protein
Glutamate decar-
boxylase |
By CO: produc-
tion... eee, 487) 79] 71) 84] 43° 55 | 13] 8) 98
By y-Aminobutyr- |
ate production. . 496 8 5, 1 3 0.2 6 86
Choline acetyltrans- |
ferase........... 75 9 ; 2; 10 3) 5. 16
Acetylcholinester- |
AS@. 2. ee 25, 900/732/701 280 625 12991102 1,120

 

 

 

 

 

 

 

on top of which were smaller cells that were dark in the phase
microscope (phase-dark cells) (Fig. 6C). Also present were
phase-bright cells with large cell bodies and long processes (Fig.
6D). As seen in Fig. 6£, silver impregnation was detected (with
a modified Bodian stain) in cells with multiple long processes and
large cell bodies.

The effect of time in culture on marker enzyme activities is
shown in Fig. 7. Catechol O-methyltransferase activity did not
reach measurable levels during culture. Glutamate decarboxyl-
ase specific activity (Fig. 7.4) did not begin to rise until the 15th
day in culture, when increase in cell number had ceased. In
contrast to glial cell lines (Table VIZ), cultured brain cell extracts
produced y-aminobutyric acid and CO. in equal amounts.
Choline acetyltransferase specific activity (Fig. 7B) fell in the
first 3 days to less than 7 units (the minimum that could have
been detected). The specific activity then increased from the
3rd to the 15th day; this rise was coincident in time and rate with
the increase in cell number. After the culture had renched a
stationary cell number, choline acetyltransferase specifie activity
continued to rise, but at «a slower rate.

The specific activities of choline acety]transferase and gluta-
mate decarboxylase attained in these cultures were 47 and 43°%,
respectively, of the levels in dissociated newborn brain cells.

Markers for Neuronal Genes

Vol. 247, No. 10

3 ©

‘oO
10° CELLS/ DISH

UNITS OF ACTIVITY

D

b

 

 

   

9 10 20 30
DAYS IN CULTURE

Fic. 7. Development of marker enzyme spccifie activities in
surface cultures of newborn mouse brain cells. Each homogenate
of Fig. 6A was evaluated for the marker enzyme activities as
described under “Methods.’? Changes in specifie activities
(©, @) and cell number (dashed lines) with time are shown for:
A, glutamate decarboxylase (assessed by "CO: evolution by
Method 6); B, choline acetyltransferase; and C, acetylcholines-
terase. One unit of activity is defined as 1 pmole of product
formed per min per mg of homogenate protein. Open and closed
symbols are explained in the legend to Fig. 6.

From the 3rd to the 30th day in culture, total activities of these
enzymes (picomoles per min per culture dish) increased by 48-
fold for choline acety|transferase and 21-fold for glutamate de-
carboxylase. Specific activity of glutamate decarboxylase was
at least 10-fold higher in these cultures than in the nonbrain and
glial cell lines tested (Table VID.

Acetylcholinesterase activity (Fig. 7C) decreased until the 6th
day in culture, and remained constant for the remainder of the
culture period. In all cultures, including the established cell
lines, esterase activities were inhibited 80 to 95@ by 107° uw BW
284C51 dibromide, an inhibitor (25) of acetvlcholinesterase
(EC 3.1.1.7). Hydrolysis of acety!choline by mouse brain was.
also inhibited 95% by this compound.

DISCUSSION

The results show that the procedure for preparation of ex-
tracts from cells in surface culture is reproducible and gives ex-
cellent recovery of protein and the enzyme activities measured.
Various other procedures for preparation of extracts were not ex-
tensively studied; however, trypsinization, FTA treatment, or
scraping followed by washing of cells resulted in lower recoveries
than the scraping procedure.
and easily performed and are applicable for routine use with a
large number of samples. Important features in applying the
assays are identification of reaction products and determination
of the specificity of the enzyme activities. In the glutamate de-
carboxylase assay, evolution of MCO. was not always a meusure
of concomitant production of y-aminobutyric acid by glutamate
decarboxylase. In the choline acetyltransferase assay, enayme
systems utilizing acetyl-CoA may produce products other than
acetylcholine that are recoverable in ion exchange column efflu-
ents (20).
cholinesterase and catechol] O-methyltransferase assays. Hence,
methods for product identification and enzyme characterization
were presented as integral parts of the assays.

The enzyme activities studied were selected because they are
important in neurotransmitter metabolism and also because acet-

The enzyme assays are rapidly

There are similar potential difficulties with the acetyl-
Issue of May 25, 1972

yleholinesterase, choline acetyltransferase, and glutamate de-
carboxylase activities are higher in brain than in most other
tissues. Relatively low levels of these enzyme activities were
found, however, in several nonbrain and glial cell lines. Thus,
the enzyme activities are markers for neurons only when specific
activities are considerably higher than those of the non-neuronal
cell lines shown in Table VII. The low levels of the activities
exhibited by these non-neuronal cell lines serve as a useful base-
line for assessment of nerve cell cultures.

Properties of growth of normal newborn brain cells in surface
culture were investigated as a first step toward establishing clonal
cell lines of mammalian neurons. Primary surface cultures of
newborn brain cells contained cells that appeared to be differenti-
ated neurons both morphologically and histochemically. Activi-
ties of choline acetyltransferase, acetylcholinesterase, and cate-
chol O-methyltransferase did not reach levels that were signifi-
eantly higher than non-neuronal cell lines. In contrast, brain
cultures attained levels of glutamate decarboxylase activity that
were at least 10-fold higher than non-neuronal cells, and the in-
crease in activity occurred after increase in cell number in the
cultures had ceased. Thus, primary brain cell cultures possess
several properties of differentiated neurons.

It is important to note that with improved culture techniques
and with cells from younger animals, brain cell cultures have been
obtained with activities of these enzymes which exceed those of
newborn brain homogenates! (26). In addition, other sensitive
radiochemical methods recently developed by Hildebrand e al.
(27) also may be applicable to cells in culture.

Acknowledgments—We are grateful to Doctors M. Wileox,
D. R. Creveling, and J. W. Daly for helpful suggestions and to
Dr. J. Minna for providing cells used in some of the experiments.

REFERENCES

1. Wor, M. k. (1964) J. Cell Brol. 22, 259
2. Kunse, R. J., anp Ruppie, F. H. (1969) J. Cell Biol. 43, 69
3. Varon, 8., aND Rargorn, C. W., JR. (1969) Brazn Res. 12, 180

 

'B. K. Schrier, submitted for publication.

Wilson et al.

12.
13.
14.

16.
17.

18.
19.

20.
21.
22.

23.

24.
25.

26.
27.

3169

. Aveusti-Tocco, G., anp Sato, G. (1969) Proc. Nat. Acad. Se7.

U.S. A. 64, 311

. Scuusert, A., HumMpHReys, 8S., Baroni, C., anp Coun, M.

(1969) Proc. Nat. Acad. Sci. U. S. A. 64, 316
Nexson, P., Rurrner, B. W., ano Nrrenperc, M. W. (1969)
Proc. Nat. Acad. Sct U. 8. A. 64, 1004

. Werner, I., Peterson, G. R., anp Suuster, L. (1971) J.

Neurochem. 18, 141

. Minna, J., Netson, P., Peacock, J., Guazer, D., AnD NtrEN-

BERG, M. (1971) Proc. Nat. Acad. Sci. U. 8. A. 68, 234

. Frscuspacu, G. D. (1970) Science 169, 1831-1333
. SHimapDs, Y., Frscuman, D. A., anp Moscona, A. A. (1970)

Proc. Nat. Acad. Sci. U. S. A. 62, 715

. Scurier, B., Rosenserc, R., Tompson, E., anp Farser, J.
a , y , ? ?

(1970) Fed. Proc. 29, 480

Ham, R. G., ann Puck, T. T. (1962) Methods Enzymol. 5,
90-119

Lowry, O. H., Rosrsrouayu, N. J., Farr, A. L., Ranpau,
R. J. (1951) J. Biol. Chem. 193, 265-275

Kissane, J. M., anp Rossgins, E. (1958) J. Biol. Chem. 238, 184-
188

. Reep, D.J., Goto, K., anp Wana, C. H. (1966) Anal. Biochem.

16, 59-64

EHRENPREIS, S., Cutesa, A., Brao-GuLurno, M., anp Patrick.
P, (1967) Fed. Proc. 26, 296

NIKODEJEVIC, B., SENOH, 8., Daty, J. W., AND CREVELING,
C. R. (1970) J. Pharmacol. Exp. Ther. 174, 88

Scurier, B. K., anp Swuster, L. (1967) J. Neurochem. 14, 977

Winao, W. J., aNnD AWaAPaARA, J. (1950) J. Biol. Chem. 187, 267-
271

Amano, T., RicHetson, L., anp Nrrenpere, M. (1971) Fed.
Proc. 30, 1085

Roserts, E., anp SIMONSEN, D. G. (1963) Biochem. Pharmacol.
12, 113

Jones, A. R., Dowtine, E. J., anp Skrasa, W. J. (1953) Anal.
Chem. 25, 394

WuitTaker, V. P. (1963) in O. ErcHurr anp A. Farau (Edi-
tors), Handbuch der experimentellen pharmakologie, pp. 2-34,
Springer-Verlag, Berlin

Linewraver, H., ano Burk, D. (19384) J. Amer. Chem. Soc.
56, 658

Aveustinsson, K-B. (1963) in D. Guickx (Editor), Methods in
biochemical analysis, p. 1, John Wiley and Sons, New York

Sreps, N. W. (1971) Proc. Nat. Acad. Sci. U. S. A. 68, 1858

Hitpesranp, J. G., Barker, D. L., Hersert, E., anp
Kravitz, E. A. (1971) J. Neurobiol. 2, 231