Journal of Neuroscience Methods, 29 (1989) 261-265 261

Elsevier

NSM 00984

The use of c-fos as a metabolic marker

in neuronal pathway tracing
M. D r a g u n o w a n d R. Faull
Department of Anatomy, School of Medicine, The Unit~'ersi(vof Auckland, Auckland (New Zealand)
(Received 20 February 1989)
(Revised 10 April 1989)
(Accepted 13 April 1989)

Key words: Fos; Nuclear protein; Immunocytochemistry; Metabolic marker

The use of c-los protein (Fos) immunocytochemistry as a metabolic marker for tracing neuroanatomical connections, seizure
pathways and sites of action of neuroactive drugs is discussed in this report. Fos immunocytochemistry will be very useful for these
purposes providing that a number of potential problems are recognized and controlled. These include the observations that Fos exists
basally in neurons and can be non-specifically elevated after behavioural stress; neuronal bursting is required to elevate Fos in
neurons in anaesthetized animals; drugs such as ketamine can block Fos elevation in neurons: the time-course of Fos induction and
decay varies with different inducing stimuli and the brain region sampled; and some brain regions do not express Fos after any
treatments tried so far. To overcome these potential problems we list a number of steps that should be followed when using Fos
immunocytochemistry as a metabolic marker of brain activity.

Recently, we and others have reported that
activation of certain brain regions can lead to
activation of the nuclear protooncogene c-los
(Dragunow and Robertson, 1987; Hunt et al.,
1987; Morgan et al., 1987). More recently, it has
been suggested that c-los activation may be utilized
as a high resolution metabolic marker for polysyn-
aptic pathway tracing in the brain and spinal cord
(Sagar et al., 1988; Fig. 1). Ultimately, c-los pro-
tein (Fos) immunocytochemistry may be used
routinely in conjunction with (or instead of the
more expensive and less-well resolved) 2-de-
oxyglucose (2-DG) autoradiography for this pur-
pose. We believe that it will be a useful tool for
these purposes. For example, we have found that
Fos elevation can be used to map out the path-
ways involved in the spread of epileptic seizures in
Correspondence: M. Dragunow, Department of Anatomy,
School of Medicine, The University of Auckland, Private Bag,
Auckland, New Zealand.
0165-0270/89/$03.50 ~ 1989 Elsevier Science Publishers B.V. (Biomedical Division)
rats (Dragunow et al., 1988). Partial seizures
elicited by electrical stimulation of one amygdala
are confined to the ipsilateral amygdala and piri-
form area and this is reflected by Fos elevation in
these ipsilateral brain regions. However, gener-
alized seizures induced in amygdala-kindled rats
invade neocortex and hippocampus bilaterally as
well as the contralateral amygdala and piriform
and elevate Fos in these areas (Dragunow et al.,
1988). In these experiments we also found that
rats having generalized fits involving " w e t dog
shakes" (which are thought to be generalized by
dentate granule cells) showed Fos elevation in the
previously described areas as well as in the dentate
granule cells. However, rats having generalized
fits, but not " w e t dog shakes", showed Fos eleva-
tion in the previously described areas but not
dentate granule cells. Thus, Fos immunocy-
tochemistry provides a sensitive tool for mapping
out pathways involved in seizure propagation.
However, we would like to point out some
potential problems and also some points to con-
262

Fig. 1. Photomicrograph showing Fos immunoreactivity in a coronal section of rat brain. This rat was implanted with a stimulating
electrode in the left perforant-path and a recording electrode in the left dentate hilus as described in Douglas et al. (1988). The left
perforant path was stimulated at 10 Hz. This 10 Hz stimulation generated multiple population spikes in the dentate granule cells. As
can be seen from the figure, activation of the perforant-path with stimulation that induces granule cell bursting selectively elevates
Fos in these ceils (filled arrow). Note that electrode implantation through the neocortex and into the dentate hilus, without
stimulation, did not induce any increase in Fos in dentate granule cells (Douglas et at., 1988), but did lead to Fos elevation in the
piriform cortex on the electrode implanted side. We have previously argued that this cortical induction of Fos after brain surgery is
due to spreading depression (Dragunow and Robertson, 1988b). Thus, the elevation of Fos in the left piriform cortex in the figure
(open arrow) is due to electrode implantation. Thus, with sufficiently intense electrical stimulation parameters Fos immunocytoche-
mistry can be used to selectively trace the mono-synaptic connection from the entorhinal cortex (the axons of which form the
perforant-path) to the dentate granule cells. Bar = 2 ram. The immunocytochemical procedure that we used for detecting Fos in this
experiment has been detailed previously (Dragunow and Robertson, 1988). A number of different immunocytochemical methods
have been used to detect Fos in brain and spinal cord sections. Most researchers perfuse their animals first with saline and then with
4% paraformaldehyde made up in 0.1 M phosphate buffer (pH 7.4). Brains can then be cut directly on a vibratome or left overnight
in 30% sucrose in 0.1 M phosphate buffer and then sectioned on a microtome. Sections of 50-100 #m thickness are then generally
incubated with primary anti-Fos serum followed by washing and incubation in secondary biotinylated antisera using the avidin-bio-
tin method with horseradish peroxidase and diaminobenzidine as the chromagen (Dragunow and Robertson, 1988, Hunt et al., 1987;
Morgan et al., 1987; Popovici et al., 1988) or the indirect PAP method (Le Gal La Salle, 1988). A number of antisera to Fos are
available commercially, although we have mainly used an antibody directed against amino acids 127-152 of the Fos molecule
(generously provided by Tom Curran, Roche, N J), and have obtained excellent results with this antibody. Other antibodies available
commercially include ones that are directed against the N-terminal (CRB); residues 4-17 (Microbiological Associates), and a
beta-gal-Fos fusional protein (Medac Gentechnologie).

s i d e r w h e n u s i n g F o s as a m e t a b o l i c m a r k e r . O n e
p r o b l e m lies i n t h e m i s m a t c h b e t w e e n F o s e l e v a -
tion and 2-DG uptake in some brain regions. For
example, metrazol seizures induce 2-DG uptake in
t h e c e r e b e l l u m a n d s u b s t a n t i a n i g r a (as well as
o t h e r b r a i n r e g i o n s , B e n - A r i et al., 1981), h o w e v e r ,
m e t r a z o l s e i z u r e s d o not e l e v a t e F o s i n t h e s e t w o
b r a i n r e g i o n s ( D r a g u n o w a n d R o b e r t s o n , 1988;
M o r g a n et al., 1987, a l t h o u g h c-fos m R N A is
i n d u c e d i n t h e c e r e b e l l u m a f t e r m e t r a z o l seizures).
It m a y b e t h a t s o m e o f t h i s m i s m a t c h is d u e t o t h e
fact that 2-DG can detect axonal and dendritic
activity, whereas Fos immunocytochemistry stains
the neuronal nucleus and thus only detects neuro-
nal changes.
F u r t h e r m o r e , i n a r e c e n t s t u d y ( R o b e r t s o n et
al., 11989) w e f o u n d t h a t L - D O P A a c t i v a t e d F o s i n
the striatum, nucleus accumbens and parts of the
cortex of rats ipsilateral to a 6-OHDA-lesion of
the substantia nigra. We also showed that this

L-DOPA activation of Fos only occurred in
animals where L-DOPA induced contralateral ro-
tation. L-DOPA did not elevate Fos in the sub-
stantia nigra. However, 2-DG studies of L-
DOPA-induced turning in rats show an increase in
2-DG uptake in the ipsilateral striatum and sub-
stantia nigra pars reticulata (Robertson and Ro-
bertson, 1987). Thus, there seems to be a mis-
match for some brain regions (although very good
matches in others) for Fos and 2-DG.
The second problem (which may account for
the first) is that neurons in certain brain regions
do not show Fos elevation no matter what the
stimulus. This is the case for the substantia nigra.
It may be that future studies will reveal induction
in these brain regions by drugs or other types of
stimulation. However, we would like to suggest
that the reason neurons in some brain regions
never show Fos elevation despite being activated
is that they lack the required biochemical mes-
sengers regulating Fos activation in neurons. What
these messengers may be in neurons is unclear. In
dividing cells a variety of hormones and second
messengers can induce c-fos production (Morgan
and Curran, 1986). Is this also true of adult CNS
neurons or are specific messengers involved? At
present we cannot answer this question although
N M D A / c a l c i u m / c a l m o d u l i n are involved in c-fos
induction in PC12 "neuronal" cell lines and
primary cultured neurons (Morgan and Curran,
1986, Szekely et al., 1987).
Another potential problem is to what extent
Fos exists basally in neurons; because basal ex-
pression may mask any changes related to the
activation of specific pathways. Using one anti-
serum (rabbit anti-c-fos peptide, amino acids
127-152) we and others (Dragunow and Robert-
son, 1987, 1988; Morgan et al., 1987; Sagar et al.,
1988) have found Fos-positive neurons scattered
in neocortex and limbic system of control rat
brains. However, others (Hunt et al., 1987; Le Gal
La Salle, 1988; Popovici et al., 1988) found essen-
tially no basal Fos in brain and spinal cord using
an antibody directed against the N-terminal of
Fos (although both anti-sera show the same pat-
tern of staining after seizure activity). Because the
antibody used in our studies recognizes Fos as
well as Fos-related nuclear antigens (Franza et al.,
263
1987) it is possible that the basal levels we are
detecting are Fos-associated proteins, not c-los
itself. However, this does not explain the observa-
tion of low but detectable constitutive expression
of c-fos m R N A in adult neurons (Gubits et al.,
1988). A more likely explanation derives from the
observation that activity, handling and stress ap-
pear to elevate neuronal Fos levels. Therefore, the
basal levels being measured in control brain might
be greatly influenced by the past history (1-12 h~
of the subject.
Furthermore, it should be recognized that for
Fos to be induced in neurons, they must be
activated strongly. In the dentate granule cells of
urethane-anaesthetized rats we have found that
bursting of granule cells is sufficient and obliga-
tory for Fos elevation (Douglas et al., 1988; Fig.
1). Dendritic depolarization sufficient to induce
long-term potentiation but not cell bursting, does
not elevate Fos in granule cells of anaesthetized
rats (Bliss et al., 1988; Douglas et al., 1988; Hunt
et al., 1987b), although in unanaesthetized rats
bursting is not required (Dragunow et al., 1988).
We have also found that anaesthetics can interfere
with Fos elevation, and the analgesic ketamine, in
particular, blocks c-fos m R N A and protein induc-
tion elicited in neocortical neurons after mechani-
cal damage (Dragunow et al., submitted).
Finally, one of the most important method-
ological considerations to consider when using
Fos immunocytochemistry is that of the time-
course of Fos elevation and decay. This may vary
both with the brain region and the type of induc-
ing agent (e.g. drug vs seizure). In general Fos
elevation in neurons after seizures decays more
slowly than in cell culture systems (Dragunow and
Robertson, 1987; Morgan et al., 1987). The reason
for this longer-lasting induction in neurons in vivo
is presently unclear. In neurons after a seizure Fos
levels have largely returned to baseline by 24 h.
The onset latency and the time of maximal Fos
accumulation in neurons after seizures varies
according to the brain region. For example, maxi-
mal Fos levels are observed in dentate granule
cells within 30 min of kindled seizures, metrazol
seizures or kainic-acid seizures, whereas at this
time other brain regions also activated by the
seizures do not show elevated Fos levels (Dragu-
264
now a n d R o b e r t s o n , 1987; Le G a l La Satle, 1988;
M o r g a n et al., 1987). M a x i m a l i n d u c t i o n in the
neocortex a n d h i p p o c a m p a l p y r a m i d a l cells o n l y
occurs 4 h after seizure onset. T h e reason(s) for
this differential i n d u c t i o n rate is u n k n o w n , al-
t h o u g h it c o u l d be that p a t t e r n s of electrical activ-
ity evolve over time. Also, the d e c a y r a t e m a y
differ d e p e n d i n g u p o n the i n d u c i n g stimulus. F o r
e x a m p l e , F o s a c c u m u l a t i o n after k i n d l e d seizures
has largely d i s s i p a t e d b y 24 h, whereas i n d u c t i o n
in neocortical n e u r o n s after stab w o u n d injury can
last for as long as 72 h ( D r a g u n o w a n d R o b e r t s o n ,
1988b). Thus, care is r e q u i r e d when using Fos
i m m u n o c y t o c h e m i s t r y in d e t e r m i n i n g the ap-
p r o p r i a t e t e m p o r a l p a r a m e t e r s for m e a s u r i n g any
c h a n g e in F o s a c c u m u l a t i o n .
In conclusion we believe that the following
p o i n t s s h o u l d be b o r n e in m i n d when using F o s
i m m u n o c y t o c h e m i s t r y as a m e t a b o l i c m a r k e r :
(1) Because a n t i b o d i e s d i r e c t e d against the
N - t e r m i n a l o f F o s a p p e a r to selectively detect F o s
(in c o n t r a s t to o t h e r antisera that recognize F o s
a n d F o s - r e l a t e d antigens, F r a n z a et al., 1987)
N - t e r m i n a l d i r e c t e d a n t i s e r a are p r e f e r a b l e for use
in tracing studies. A l t h o u g h , in a recent s t u d y
c o m p a r i n g the C u r r a n a n t i s e r u m with one against
the N - t e r m i n a l of F o s ( C R B ) we ( D e C a s t r o a n d
D r a g u n o w , in p r e p a r a t i o n ) f o u n d that b o t h anti-
sera gave identical p a t t e r n s of staining in neurons,
glia a n d e p e n d y m a l cells after b r a i n injury. The
o n l y difference b e t w e e n the two antisera was the
g r e a t e r b a s a l F o s d e t e c t e d b y the C u r r a n anti-
s e r u m as p r e v i o u s l y discussed.
(2) Stress to the a n i m a l s 1 - 2 4 h b e f o r e sacrifice
s h o u l d be a v o i d e d to reduce " n o n - s p e c i f i c " F o s
elevation.
(3) Electrical s t i m u l a t i o n in p a t h w a y m a p p i n g
s h o u l d be i n t e n s e a n d if the a n i m a l s are a n a e s t h e -
tized (versus using c h r o n i c indwelling electrodes)
cell b u r s t i n g m a y be r e q u i r e d to elicit F o s eleva-
tion
(4) K e t a m i n e can block c-los i n d u c t i o n in s o m e
n e u r o n s a n d s h o u l d be avoided. Also, b a r b i t u r a t e s
m a y interfere with F o s i n d u c t i o n .
(5) Because of p o i n t s 3 a n d 4 n e u r o n a l p a t h w a y
tracing studies where electrical s t i m u l a t i o n is
d e l i v e r e d are best p e r f o r m e d on a w a k e rats with
c h r o n i c i n d w e l l i n g electrodes. These a n i m a l s
s h o u l d n o t be used for at least two weeks after
surgery to allow i n j u r y - i n d u c e d F o s levels to re-
turn to baseline. U s i n g " c h r o n i c " p r e p a r a t i o n s
will also o v e r c o m e the p o t e n t i a l p r o b l e m of F o s
elevation in n o n - n e u r a l b r a i n cells (e.g. glia,
e p e n d y m a , D r a g u n o w a n d R o b e r t s o n , 1988b)
which occurs a r o u n d an injury site.
(6) T h e time of sacrifice after the s t i m u l a t i o n ,
seizure or d r u g should be chosen to m a x i m i z e the
F o s a c c u m u l a t i o n . As a rule of t h u m b we r e c o m -
m e n d a 4 - 6 h time interval, a l t h o u g h it is better to
use a s p r e a d of t i m e - p o i n t s to s t u d y the evolving
p a t t e r n of F o s elevation in different b r a i n regions.
(7) N e g a t i v e results (i.e. no F o s elevation)
should not be a u t o m a t i c a l l y t a k e n to m e a n that
the structure has n o t been a c t i v a t e d b y the d r u g or
stimulation. H o w e v e r , positive signals (i.e. F o s
elevation) with g o o d c o n t r o l s will p r o v i d e v a l u a b l e
insights into the a n a t o m i c a l a n d p h y s i o l o g i c a l
o r g a n i z a t i o n of the b r a i n a n d will be i n v a l u a b l e in
localizing the sites of a c t i o n of drugs a n d m a p p i n g
out seizure p a t h w a y s .
References
Ben-Aft, Y., Tremblay, E., Riche, D,, Ghilini, G. and Naquet,
R. (1981) Electrographic, clinical and pathological alter-
ations following systemic administration of kainic acid,
bicuculline or pentetrazole: metabolic mapping using the
deoxyglucose method with special reference to the pa-
thology of epilepsy, Neuroscience, 6: 1361-1391.
Bliss, T.V.P., Errington, M.L., Evan, G. and Hunt, S.P. (1988)
Induction of c-los-like protein in rate hippocampus follow-
ing electrical stimulation, J. Physiol. (Lond.), 398: 96.
Douglas, R.M., Dragunow, M. and Robertson, H.A. (1988)
High-frequency discharge of dentate granule cells, but not
long-term potentiation, induces c-fos protein, Mol. Brain
Res., 4: 259-262.
Dragunow, M. and Robertson, H.A. (1988) Localization and
induction of c-fos protein-like immunoreactive material in
the nuclei of adult mammalian neurons, Brain Res,, 440:
252-260.
Dragunow, M. and Rohertson, H.A. (1987) Generalized seizures
induce c-los protein(s) in mammalian neurons, Neurosci.
Lett., 82: 157-161.
Dragunow, M. and Robertson, H.A. (1987) Kindling stimula-
tion induces c-fos protein(s) in granule cells of the rat
dentate gyrus, Nature (Lond.), 329: 441-442.
Dragunow, M. and Robertson, H.A. (1988b) Brain injury
induces c-fos protein(s) in nerve and glial-like cells in adult
mammaliam brain, Brain Res, 455: 295-299.

Dragunow, M., Robertson, H.A. and Robertson, G.S. (1988)
Amygdala kindling and c-los protein(s), Exp. Neurol., 102:
261-263.
Dragunow, M., Abraham, W.C., Mason, B. and Faull, R i . M .
(1988) c-los and LTP: is there a link?, Soc. Neurosci.
Abstr., 14: 565.
Franza, B.R., Sambucetti, L.C., Cohen, D.R. and Curran, T.
(1987) Analysis of Fos protein complexes and Fos-related
antigens by high-resolution two-dimensional gel electro-
phoresis, Oncogene, 1 : 213-221.
Gubits, R.M,, Hazelton, J.L. and Simantov, R. (1988) Varia-
tions in c-los gene expression during rat brain develop-
ment, Mol. Brain Res., 3: 197-202.
Hunt, S.P., Pini, A. and Evan, G. (1987) Induction of c-los-like
protein in spinal cord neurons following sensory stimula-
tion, Nature (Lond.), 328: 632.
Hunt, S.P., Pini, A., Errington, M.L. and Bliss, T.V.P. (1987b)
Induction of c-fos protein in neurons of the central nervous
system, Soc. Neurosci. Abstr., 13: 1597.
Le Gal La Salle, G. (1988) Long-lasting and sequential in-
crease of c-fos oncoprotein expression in kainic acid-in-
duced status epilepticus, Neurosci. Lett., 88: 127-130.
265
Morgan, J.1. and Curran, T., (1986) Role of ion flux in the
control of c-fos expression, Nature (Lond.), 322:552 555.
Morgan, J.l., Cohen, D.R., Hempstead, J.L. and Curran, l .
(1987) Mapping patterns of c-los expression in the central
nervous system after seizure, Science, 237: 192--197.
Popovici, Th., Barbin, G. and Ben-Ari. Y. (1988t Kainic acid-
induced seizures increase c-los-like protein in the hippo-
campus, Eur. J. Pharmacol., 150: 405-406.
Robertson, G.S. and Robertson, H.A. (1987) 1) I and D,
dopamine agonist synergism: separate sites of action'.'
Trends Pharmacol. Sci., 8: 295-299.
Robertson, G.S., Herrera, D.G., Dragunow, M. and Robertson,
H.A. (1989) I,-DOPA activates c-los in the striatum ipsi-
lateral to a 6-hydroxydopamine lesion of the substantia
nigra, Eur. J. Pharmacol., 159: 99-100.
Sagar, S.M., Sharp, F.R. and Curran, T. (1988) Expression of
c-fos protein in brain: metabolic mapping at the cellular
level, Science, 240: 1328-1331.
Szekely, A.M., Barbaccia, M.L. and Costa, E. (1987) Activa-
tion of specific glutamate receptor subtypes increases c-los
proto-oncogene expression in primary cultures of neonatal
rat cerebellar granule cells, Neuropharmacology, 26:
1779--1782.