C
GLUTAHATE NEUROTOXICITY IN CNS CULTURES
OF HARINE SURFPERCH Cymstegester eggregale
Hindola DasGupta
June 9, 1989
Professor Stuart Thompson
Hopkins Harine Station
C
Abstract
Research on rat neocortical neurons in primary cell culture
has shown that the excitatory amino acid neurotransmitter, glutamate,
causes neurotoxicitu, a phenomenon associated with neurological
disesses, including Huntingdon's disesse (Cogle and Schwarz, 1976),
olivopontocerebellar atrophy (Plaitakis et al., 1962), stroke (Rothman,
1984), and hupoglucemic encephalopsthg (Wieloch, 1985).
In the present study, brief exposure to glutamate was found
to couse cell death of neurons cultured from the brain of adult
marine surfperch, Lyrsstagsster aggregste A dose-toxicity study
suggested an EDg of 100-500 uM glutamate for a 5-minute exposure.
A toxicity versus age in culture study showed that glutarate receptors
mey be expressed as earlu as 24 hours in culture. A dose response study
on Day 2 in culture indicated that glutamate neurotoxicity (GMT) is
curulative; the higher the glutamate concentration used the greater the
cell desth.
Some neurons survived short exposure to glutamate. This
resistance could be due to a lack of glutamate receptors on neurons from
certain regions of the brain. Cerebellar neurons exposed to glutamate,
for example, showed no deviation in viability from the control
percentage.
Introduction
The amino acid glutamate is present at concentrations of
several millimolar in the mammalian brain (Waelsch, 1951). Glutamate
has been shown to have an excitatory effect on neurons of the CNS, and
for these reasons, it is believed that glutamate is an excitatory amino
scid neurotransmitter (Dichiera and Gessa, 1981). It has also been
demonstrated, however, that glutamate is a neurotoxin (Choi et al.,
1987). Hormallu, protective uptake mechanisms prevent excessive
glutamate from causing neuronal damage, but under abnormal
circumstances, namely the neurological disease states mentioned above,
glutamate has plaged a role in neurotoxicitg. Huch research has been
undertaken recentlu in mammals as to the mechaniem of GHI, its
connection to hupoxia, and, of course, to answer the question of why a
neurotransmitter would serve the dual function of a neurotoxin.
The mammalian CHS, however, is too intricate and comples
to studu the effects of glutamate in viva The purpose of this study was
to determine whether neurons of marine surfperch Lymslagsser
sggregete possess glutamate receptors, and in addition, whether these
neurons are susceptible to GHT. Harine surfperch might then serve as a
simple model in which to studi the phenomenon.  report hère that
glutamate receptors mag be expressed on surfperch CNS neurons and that
glutamate is a potent neurotoxin in vitre.
Haterials and Hethods
Cell Culture. Several methods of tissue dissociation were attempted
before a satisfactoru method was found to culture dense, viable
monolayers of neurons with a minimum of contamination:
1) The protocol of Wolf and Ouimby was closely followed with a 16-hour
2.52 trupsin digestion at 4 °0.
2) The same protocol was followed woth a 1, 2 and 3-hour trypsin
digestion at room temperature.
3) Hanual dissociation was attempted. After dissecting the brain from
the fish, the brain lobes were probed with glass microelectrodes, and
dissocisted tissue was plated.
4) The lobes were manually minced with sterile forceps into small
fragments and transferred onto culture plates where the fragments
were further triturated.
5) The method for best results was as follows:
Hature marine surfperch were sacrificed in an icewater bath, submerged
in a 1:1O dilution of household bleach, and sterilized topicallg with 75
ethanol. The brain was then removed and preserved in a sterile solution
of Hanks salts (Sigma) in which the salt concentration was incressed to
200 mtl Macl. The salt solution (pH 7.2) was buffered with 10 mri Hepes.
The brain lobes were then minced and transferred into a solution of
1.252 dispase (30 mg. dispase in 2.5 ml. Hanks) and stirred at 25 ?. for
1 hour. Subsequentlu the cells were plated for 30 minutes onto
autoclaved coverslips which had been costed with polg-L-lysine for 3
hours and rinsed with distilled water. After 172 hour, the plated cells
were placed in a 3 ml. bath of growth medium and incubated overnight at
17 C. The growth medium consisted of L-15 Leibovitz Hedium (Gibco),
5 fetal bovine serum, 12 L-glutamine Penicillin-Streptomgcin, and
20OmH Hacl at pH 7.4. The medium did not initially contain ang
glutamate, except in the serum, which according to the suppliers
information, contained less than 20 uf.
Neuronal Viabilitg.
1) Cell cultures were loaded with 25 ul fura-2-AH and 10 ul pleuronie
acid in a 3 ml. medium bath for 2 hours at 17 *C. and 1 hour stirring
at room temperature. The cells were then transferred into 3 ml.
Hanks and observed on a low-light video microscope 400K. 1 um
jonomucin was added. Fura flourescence was recorded on wideo at
340 Hz and 360 Hz before and after ionomycin addition at 10 second
intervals. Then 1 uM manganese was added, after which
fluprescence frequencies at 340 Hz and 380 Hz were agsin recorded.
Cell cultures were also loaded with 25 ul fluo-3-AH and 10 ul
pleuronic acid in a 3 ml. medium bath for 30 minutes at room
temperature. The cells were then placed in 3 rl. Hanks and observed
with the low-light video microscope 400K. Fluorescence
frequencies were recorded on videotape continuously before and
after addition of 2 uH KCl.
Glutamate Exposure. Prior to glutamate exposure, the cultured celle
were examined under phase contrast 6OÖK and photographed after various
lengths of time in culture.
During glutsmate studies, the cells were bathed in the
control Hanks solution, to which various concentrations of glutamate
were added for S minutes at room temperature. In a triple exchange, the
cells were then washed in sterile Hanks (effective dilution 2 300),
placed in 3 ml. growth medium, and returned to incubation at 17 *C.
To assess the effects of glutamate exposure, the cells were
stained for 2-3 minutes at room tempersture with O.43 trupan blue, a
due normally excluded bu healthy cells, and examined under phase
contrast 600X. Meurons were identified by their relativelg large size and
dark cell bodies and processes. In contrast, the glial cells in culture
were 20-502 the size of neurons, flat, devoid of color, and
interconnected in dense mats. Counts of viable cells in regions of the
plate were calculated as a percentage of all cells on the plate and
compared to percent viability of control plates, which showed damage to
15-252 of neurons and glia.
Results
The protocol of Wolf and Quimby (1976) for general
dissociation of fish tissue for primary culture vis a trupsin digestion,
even for onlu 1 hour, led to excessive dissociation and cell death in this
studu. Then a manual dissociation was attempted using mocroelectrodes,
but these cultures were too sparsely plated, due to a lack of sufficientlg
dissociated tissue. Then manual trituration with forceps was attempted,
but the cultures were too densely plated with undissociated tissue
clusters. Finallu, brain tissue, which was minced and stirred in 1.252
dispsse for i hour, produced homogeneous monolagers of neurons and
dlial cells with sufficient densitg. In addition, 3 hours of polg-L-lgsine
costing of glass coverslips was necessary to assure proper adhesion of
cells in plating. If this process is carried out for less time, the cells
detsch during medium changes.
In fish brain cell cultures prepared as described, neurons
could be seen as small circular or oval cell bodies from which processes
extended. Heurons were either isolated, interconnected with other
neurons, or connected to a glial network. The glial cells were clear and
flat in morpholoqu, and half the size or less than the neurons present.
Cells were examined for glutamate neurotoxicity as early as several
hours in culture and as late as S days, after which time cultures were
discarded.
Heuronal visbility was tested vis low-light video
microscopy. When loaded with fura-2-AH followed by ionomycin and
manganese, a substantial incresse in cellular calcium influx occurred.
This change confirmed previous membrane impermesbility to calcium,
and suggested cell viability. In addition, cells from other cultures were
similarlu loaded with fluo-3-AM followed bu KCl. After KCl addition, an
immediate cellular calcium influx occurred, which indicated that the
cells had been depolarized and were viable.
Control cultures were studied for longevity (see Fig. 1).
After S hours in culture, 852 of plated cells were undamaged, according
to trupan blue tests. This viability percentage decreased somewhat until
Dau 5, at which time only 5OS of the cells were still alive.
In a series of experiments, cultures of various ages were
exposed to Imti glutamate for S minutes, stained with trypan blue the
following dau (see Fig. 1), and examined under phase contrast. Although
no immediate visible change was apparent during the glutamate
exposure, latent effects of neuronal degeneration resulting in death were
seen within 24 hours. After this lapse of time, neurons in some plates
were replaced bu debris (See Fig. ?) and stained with O.42 trupan blue. In
addition, only a fraction of the original culture remained plated i dag
after exposure; a considerable number of cells had become detached
during the process of degeneration. Cultures not exposed to glutamate
remained completelu plated. Desth rates of cells which had remained
plated after imf glutamate exposure were as high as 6 over control
cultures (See Fig. 1). Interestinglu, neurons benesth the outer lager of
undissociated clumps tended to survive exposure more often than those
which were sparselu dispersed in plating.
Glial cells were as heavily affected by 1 mil glutamate
exposure as neurons. In regions of culture plates containing glis and
neurons, both cell tupes were stained equally by trupan blue 1 dag after
glutamate exposure.
Quantitative determination of neuronal injurg or loss versus
olutamate concentration, measured in several cultures of a single
plating, is represented in Fig. 2. Although most neurons were destroged
bu exposure to 1 mM glutamate, a considerable number of apperentig
viable neurons were found 1 day after exposure. These surviving cells
excluded trupan blue and remained morphologically similar for several
daus. Cultures of cerebellar tissue, in particular, did not differ
substantially in cell visbility percentiles from control cultures(see Fig.
2, Day 2).
Discussion
The present studies indicste that glutamate receptors exist
on the cell membranes of marine surfperch neurons, and that glutamate
is a neurotosin in cell culture. In the concentration-effect relationship
for GMT on CHS neurons, as little as 100 uH glutamate can cause 452
cell death. This concentration of glutamate is only 12 of the 10 mri
concentration normally found in whole mammalien brain (Waelsch, 1951).
Few studies have been conducted on concentrations of glutamate in
marine vertebrate cell culture, but presumably, longer exposure times
with lower concentrations than those used in this study could be toxic.
This study merely touches on the existence of glutamate receptors and
the prospects of GMT in marine surfperch, and on the use of
Cumetegsster eggregate as a model for medical research on GNT in
hupoxiafischemia.
GMT in cell culture in vitra may be exaggerated in
comparison to neurotoxicity in vive, because the uptake mechanisme
which normallu maintain homeostasis by removing excess extracellular
amounts are absent in primary cell culture (Schwarcz et al.,1978). This
absence may also explain why gliotoxicity was equally prevalent in
culture, but is not seen when low doses of glutamate are used in viva
(Olney et al., 1971).
It is interesting that some neurons regularly survive in
cultures exposed to glutamate. This glutamate resistance could be
similar to the phenomenon seen in mamnmals, whereby some, not all,
regions of the brain are susceptible to GMT. In most of the experiments
described, all regions of the brain were dissociated and plated together,
without regard to origin. In addition, even within a single brain region,
some neurons do appear to be more vulnerable than others to injurg by
excitatori amino acids (Hadler et al., 1980; Kohler and Schwarcz, 1983).
A subject for further study would be to determine whether these
resistant neurons were a case of random sampling from the original
population, or what characteristics deem these neurons invulnerable to
GNT.
SELECTED REFERENCES
1. Barker, SJ (1956) Neture 321, 519-522.
2. Choi, D.W., Haulucci-Gedde, M.A., Kriegstein, A.R. (1987)
Neurescience 7, 357-360.
3. Choi, D.W. (1987) J Neurascience 7,369-379.
4. Choi, D.W. (1908)  Neuroscience 8, 165-195.
5. Cogle, JT.; Schwarcz, R. (1976) Nature 263, 244-246.
5. Davidson, Neil. Heurotransmitter Amino Acids. Hew Vork:
Academic Press, 1976. pp. 6-36.
7. Dichiers, G.D.; Gessa, Gl. (1981) Adv Bischem, Fsuchopherm. 27.
S. Kim, JP., Choi, D.M. (1967) Neurascience 23, 423-432.
9. Koh, J., Choi, D.W. (1988) d Neurascience 2164-2171.
10. Mattson, Mark P.; Guthrie, Peter B., Hages, Barbars C., Kater, S.B.
(1989) d Neurescience 9(4), 1223-1232.
11. Murphu, S.N., Thager, S.A., Miller, R.J. (1987) dNeurascience 7,
4145-4156.
12. Hadler, J.V., Perru, B.W., Cotman, C.W. (1978) Nature 271.
676-677.
13. Ulneg, J.W., Ho, OL.; Rhee, V. (1971) Exp. Srain Fes. 14,61-75.
14. Plaitakis, A., Berl, S. Vahr, HD. (1982) Science 216, 193-196.
--—
15. Rothman, SM. (1963) Seience 220,536-9
16. Rothman, SM. (1954)  Neurescissce 4. 1534-1591.
17. Rothman, S.M., Thurston, J.H., Hauhart, R.E. (1987) Neurescience
22, 471-430.
16. Schwarcz, R., Scholz, D., Coyle, J.T. (1978) Neurophermacolagy
17, 145-151.
19. Simon, R.P.; Swan, J.H., Griffiths, T., Heldrum, B.S.; (1984) Science
226,850-852.
20. Sladeczek, R.; Pin, J.P.; Recasens, M., Bockaert, J.; Weiss,S.; (1985)
Neture 317, 717-719.
21. Waelsch, H. (1951) Adv Fratein Chem 6,299-341.
22. Wieloch, H. (1965) Science 230,661-653.
23. Wolf, Ken, Quimby, H.C. (1976) TCA Hanual. 2, 453-456.
EIGURE LEGENDS
Figure 1. Glutamate receptors are expressed by neurons as early as Dag
in culture. Percentages of cell viability of cultures exposed to imM
glutamate (lower line, *) were plotted against percentages of the control
(upper line, D). Cell viabilities were already considerably lower than
control values on Dag 1. The relativelg high cell viability for Day 2 of
the culture exposed to glutamate could be explained by a high proportion
of cerebellar neurons, which are known not to express glutamate
receptors in mammals.
Eigure 2. Glutamate neurotoxicity is cumulative. Control cultures had
viabilities of 803 on Day 2 in culture. The higher the concentrations
Used, the greater the cell death observed.
Figure 39. Neurons in cell culture were identified bu their circular or
oval cell bodies, from which processes extended.
b. Neurons were found in isolated areas, connected in groups, or,
as in this figure, connected to a mat of glial celle.
Figure 4. Control cells in cultures stained with trupan blue excluded the
dye.
Figure 5a. Cell cultures exposed to glutamate for 5 minutes were tested
for viebility with trupan blue the following dag. After exposure to 5 mM
glutamate, as in this figure, only 23 of cells were undarnaged.
b. In some cultures exposed to 1 mil glutamate, viable
morphollogically intact neurons were replaced by debris within 24 hours.
This culture was stained with 0.4 2 trupan blue.
11
7 Cell Vability of Culture
ktatatatakaaaaa-


o +
o +
Cell Viability of Culture Exposed
skataa-
5
o 8



3
—
80
fique 3b.
Hu