J. gen. ViroL (2979), 42, 627-632
Printed in Great Britain
627
Rabies Virus Infection Selectively Impairs Membrane
Receptor Functions in Neuronal Model Cells
(Accepted 20 October I978)
SUMMARY
A persistent infection with rabies virus (HEP-Flury) was established in the
CNS-derived hybrid cell line IO8 CC 15 which possesses specific membrane receptors
for prostaglandins, catecholamines and acetylcholine. We report a differential
virus influence on the specific receptor response to PGE, isoproterenol and
acetycholine as indicated by typical changes of the intracellular cyclic AMP levels.
As the adenylate cydase activity was unchanged in infected cells in vitro, a
selective virus influence on specific receptors themselves or their coupling to the
cAMP synthesizing system must be considered.
The mouse neuroblastoma × rat glioma hybrid cell line Io8 CCI5 (NG-Io8-I5) has been
established as a nearly ideal model to study the basic functions of neuronal tissue in vitro
(Hamprecht, I974). Its characteristic properties include a variety of membrane receptors
for typical neuro-transmitters such as acetylcholine and catecholamines and hormone-like
substances such as prostaglandin E1 (for review, see Hamprecht, I977). Stimulation of these
receptors with their appropriate agonists is followed by well-defined alterations of intracellular cyclic adenosine monophosphate (cAMP) concentrations.
Rabies virus belongs to a group of neurotropic viruses which, after infection of animals,
usually persist in the neurons of the central nervous system (CNS) for a long period of time
before massive illness and finally death occur. At autopsy, in many cases, nearly all neuron
cells in the brain and spinal cord are found to be morphologically intact (Innes & Saunders,
I962; Perl, I975). Observed histopathological changes in the CNS do not match the drastic
dysfunction seen clinically (Murphy, I977). In clinically manifest rabies infection in man and
animals, neutralizing antibodies are often absent at the time of onset of symptoms even
though virus infection is then already widespread throughout the CNS. Antibody titres
often remain low until the terminal phases of illness. A possible involvement of cell-mediated
cytotoxicity or other cytotoxic immune reactions in the pathological events is uncertain (for
review, see Murphy, I977).
In order to study the possible direct action of rabies virus on typical neuronal functions
without complicating immunological reactions~ we investigated the membrane receptor
mediated changes of intracellular cyclic AMP levels in Io8 CCI5 hybrid cells infected with
rabies virus. The cells were cultured at 37 °C in plastic flasks in Dulbecco's modified
Eagle's medium (DMEM; GIBCO) containing Io ~ foetal calf serum, I × Io 4 m-hypoxanthine (Sigma), I × Io -5 M-aminopterine (Serva), I-6 × Io-4 M-thymidine (Boehringer;
Littlefield, I964), o.o2 M-glucose and 3 × ~o-2 M-NaHCO3 (osmolarity 33o to 34o mOsmol).
Cultures were maintained in a water saturated atmosphere of 9o ~ air, IO ~o COs and cells
were passaged by trypsinization (o'oo5 ~ trypsin). Rabies virus (HEP-Flury strain) was
propagated in BHK 2I A cells at 37 °C and isolated by the method described by Madore &
England (1977) for the ERA strain. Virus infection was performed at a multiplicity of 3
infectious units per cell. After virus adsorption for 45 min at 37 °C in a suspension of 3 × Io 7
oo22-1317/79/oooo-3488 $02.00 ~) I979 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:35:31
628
Short communications
cells]ml DMEM containing 1% FCS and 25/zg]ml DEAE-dextran (Pharmacia, Uppsala),
the cells were seeded on to Petri dishes and allowed to grow in fresh medium. After 24, 48,
72 h and after every passage, the percentage of infected cells in culture was estimated by
determination of intracytoplasmic ribonucleoprotein (RNP) and membrane-bound virus
antigens by direct immunofluorescence technique using FITC-conjugated rabbit antibody
against rabies RNP (a gift from Dr Schneider, Ttibingen) and FITC-conjugated rabbit
anti-rabies serum (Behringwerke, Marburg). Forty-eight hours after infection, 9° to lOO %
of cultured cells showed virus antigens in their cytoplasm and on their cell membranes and
were assayed for their response to receptor stimulation with prostaglandin El, isoproterenol
and acetylcholine.
Infection of these cells with the rabies virus HEP-Flury strain leads to the development of
a productive chronic virus infection with characteristic time-dependent oscillations of the
ratio of virus antigen-positive and antigen-negative cells in culture. This has also been
reported for rabies virus in other cell cultures and in other rhabdovirus infections (Holland
et al. 1976; Palma & Huang, 1974; Kawai et al. I975). As described for BHK-2I cells, the
rabies virus infected lO8 CCI 5 cells were not resistant to a virus challenge by heterologous
viruses. During the whole observation period of the infected IO8 CCI5 cell population
(56 passages over the course of 18o days) no differences in cell morphology, viability or
growth rate could be observed as compared to uninfected controls. For tests of the cellular
response to membrane receptor stimulation, only cells from an early phase of infection were
used in order to avoid possible complications resulting from changes in typical cell properties
by multiple passages of the infected cultures. Such changes cannot be excluded completely
in hybrid cells in general (Weiss & Ephrussi, 1966) and were in fact demonstrated with the
neuroblastoma x glioma hybrid cells (Heumann et aL 1977).
For receptor tests, rabies-infected cells and uninfected controls were freed from growth
medium, washed twice and incubated as monolayer cultures with the receptor activating
substances, dissolved in serum-free DMEM. After Io min, the incubation medium was
removed and displaced by 5 % trichloroacetic acid (TCA) for cell denaturation. The contents
of the dishes were carefully collected and centrifuged at 2000 g. The pellet was re-dissolved
in I M-NaOH and protein was determined according to the method of Lowry et al. (1951).
The supernatant was freed from TCA by ether extraction, lyophilized and cyclic AMP
determined using the method given by Gilman (197o). Specific cyclic AMP concentrations
were calculated as pmol cyclic AMP per mg protein.
Incubation of IO8 CCr5 ceils with prostaglandin E1 (PGE1) is followed by a rapid rise in
intracellular cAMP levels up to a maximum which is reached after about 15 min (Hamprecht
& Schultz, 1973 ; Sharrna et al. 1975). Fig. I shows that after incubation for IO min with
optimum concentrations of PGE~, the rise in intracellular cAMP concentrations in rabies
virus-infected lO8 CC 15 cells reached only 50 % of the levels observed in uninfected controls.
Blocking the degrading cyclic nucleotide phosphodiesterase activity by addition of 3isobutyl-I-methylxanthine (IBMX; Ashcroft et al. I973; Peytreman et aL I973; Schultz &
Hamprecht, 1973) resulted in higher cAMP concentrations in both virus-infected and
uninfected cells but did not alter their relative differences.
Simultaneous incubation of Io8 CCI5 cells with optimal concentrations of PGE~ and
catecholamines led to an inhibition of the cellular cAMP formation. L-Noradrenalin is
about IOO times more effective than L-isoproterenol. This is a typical a-adrenergic receptor
effect of the adrenergic agonists as shown by Traber et aL (1975b). Fig. 1 shows this c~adrenergic effect in uninfected lO8 CCI5 cells simultaneously incubated with PGE~ and
D,t.-isoproterenol and PGE1 alone. Under the experimental conditions used, the incubation
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:35:31
629
Short communications
U
?
o
X
1
1
U
//
//
/
/
/
/
//
-.4,.
/ i
/
//
o
U
/
/
/
PGEI
IBMX
L
PGE l
PGEI
IBMX
brot
1
PGEI
lprot
Fig. I. Specific intraceUular cAMP concentrations in IO8 C C I 5 mouse neuroblastoma × rat glioma
hybrid cells incubated with 2"5 × lO -5 M-prostaglandin E1 ( P G E 0 alone, with and without the
addition of I × IO -a M-3-isobutyl-x-methylxanthine (IBMX) and simultaneously with PGE1 and
2"5 x io -~ M-isoproterenol (Iprot), with and without the addition o f I B M X . PGE~ was a gift f r o m
the U p j o h n C o m p a n y ( K a l a m a z o o , Michigan, U.S.A.). Shaded columns represent rabies virus
( H E P - F l u r y ) infected cells, clear columns refer to uninfected controls. Each column delineates the
m e a n f r o m three independent parallel experiments and shows the standard deviation of the m e a n
(bars).
with D,L-isoproterenol together with PGE1 resulted in a cAMP level nearly 5o % lower than
with PGE 1 alone. This ~-adrenergic receptor effect was not observed in rabies virus infected
Io8 CCI5 cells. Addition of IBMX resulted in higher cAMP concentrations but did not
influence the ratio between levels in infected and uninfected cells.
Fig. 2 demonstrates the typical shifts in intracellular cAMP levels induced by simultaneous
incubation with PGE1 and acetylcholine. The latter transmitter binds to the specific
muscarinic membrane receptor of the hybrid cells inhibiting the PGErinduced stimulation
of AMP synthesis (Traber et al. I975a ). It is obvious that in rabies virus-infected cells,
acetylcholine had quantitatively the same effect on cAMP levels, inhibiting PGEl-induced
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:35:31
Short communications
63o
U
,p
2
o
X
,1)
o
-6
ra
U
.<
//1
/ll
/A
II
//1
/A
//1
U
U
1
PGE l
IBMX
PGE l
PGEI
PGEt
1BMX
A('II
ACH
F i g . 2. Specific cAMP levels in lO8
C C I 5 cells incubated with 2. 5 × i o -5 M - P G E 1 alone, with and
without addition of t mM-IBMX and simultaneously with PGE1 and I × ~0 5 M-acetylcholine
( A C H ) , with and without addition of t mM-IBMX. Shaded columns represent rabies virus ( H E P F l u r y ) infected cells, clear columns refer to uninfected controls. Each column delineates the mean
from three independent parallel experiments and shows the standard deviation of the mean ( b a r s ) .
rises of this cyclic nucleotide by approx. 50 ~ . Incubation in the presence of IBMX
illustrates this situation on a larger scale.
These findings indicate that the normal action of PGE 1 and isoproterenol on the cyclic
AMP levels of Io8 CCT 5 cells is impaired following rabies virus infection whereas the
cellular response to acetylcholine seems to be unaffected. These effects cannot be caused by
different cyclic AMP phosphodiesterase activities under stimulation with different agonists
since addition of IBMX, in suitable concentrations, does not alter the relative cAMP levels
found under various conditions in the absence of IBMX. The possibility that there could
be differences in membrane permeability in uninfected and infected cells for cAMP under
the influence of different agonists binding to their specific receptors was excluded (data not
shown). Analysis of the fluoride-stimulated adenylate cyclase activity in membrane preparations of rabies virus-infected and uninfected io8 CC~5 cells demonstrated that cAMP
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:35:31
Short communications
63 I
synthesizing enzyme activity was not altered under virus influence (data not shown). A
different persistent infected cell system has been described in the previous paper in which an
alteration of the specific activity of the adenylate cyclase itself has been observed (Halbach
& Koschel,~I979).
From our data on rabies virus-infected Io8 CCI5 cells, we have to assume a direct virus
influence on specific membrane receptor functions. Whether this virus-induced impairment
of normal receptor function is due to a loss in the number of specific receptor molecules, to
a decrease in the receptor affinity for the agonists used, or to the obstruction in functional
coupling of specific receptors to the adenylate cyclase units, cannot be decided at present.
Experiments are in progress to increase our understanding of these mechanisms of virus
action.
Many investigators have reported that, especially in cells of the CNS, hormone or transmitter action is followed by characteristic shifts in the intracellular cAMP levels of the
target cells. Here, we present evidence that this normal and supposedly important cellular
response can be considerably hampered by a persistent non-cytocidal virus infection. The
specificity of the viral attack affecting the signal transfer of one agonist but leaving another
unchanged provides a new aspect of virus-cell interaction worth considering in the pathogenesis of certain diseases of the nervous system.
We thank Drs E. Wecker and V. ter Meulen for helpful discussions and Mrs Wagner and
Miss G. Wicht for excellent technical assistance. This work was supported by the Deutsche
Forschungsgemeinschaft, SFB IO5.
Institut fiir Virologie und Immunbiologie
KLAUS KOSCHEL
der Universitdt Wiirzburg
MICHAELHALBACH
Versbacher Strasse 7
D-87oo Wiirzburg, West Germany
REFERENCES
ASHCROrT, S. J. ~., RANDLE, ~'. J. & TXLJEDAL, J.-B. (I973). Cyclic nucleotide phosphodiesterase activity in
normal mouse pancreatic islets. FEBS Letters 20, 263-266.
CILMAN, a. ~. (1970). A protein binding assay for adenosine 3': 5'-cyclic monophosphate. Proceedings of the
National Academy of Sc&nces of the United States of America 67, 3o5-312.
I~ALBACH,M. & KOSCHEL, K. (I979). Impairment of h o r m o n e dependent signal transfer by chronic SSPE-virus
infection. Journal of General Virology 4z (previous paper).
HAMPRECnT, ~. (1974). Cell cultures as model systems for studying the biochemistry of differentiated
functions of nerve cells. Hoppe-Seyler's Zeitschriftfi~r Physiologische Chemie 355, IO9-~ I t .
HAMPRECHT, R. (1977). Structural, electrophysiological, biochemical and pharmacological properties o f
neuroblastoma-glioma cell hybrids in cell culture. International Review of Cytology 49, 99-I7O.
HAMPRECrtT, R. & SCrIULTZ, J. (I973). Influence of noradrenalin, prostaglandin E~ and inhibitors o f phosphodiesterase activity on levels of cyclic adenosine 3 ' : 5 ' - m o n o p h o s p h a t e in somatic cell hybrids.
Hoppe-Seyler's Zeitschrift f~r Physiologische Chemie 354, 1633-1641.
HEUMANN, R., VALET, G., MAISON, D., KEMPER, J., REISER, G. & HAMPRECHT, B. (I977). Influence o f the time in
culture on cellular and neuronal properties of neuroblastoma × glioma hybrid cells. With an a p p e n d i x Mathematical description of the kinetics of the loss in cell volume. Journal of Cell Sc&nce 27, 141-I55.
HOLLAND, J. J., V1LLARREAL~L. P., WELSH, R. M.~ OLDSTONE, M. B. A.~ KOHNE~D.~ LAZZARINI, R. & SCOLNICK~E.
(1976). Long-term persistent vesicular stomatitis virus and rabies virus infection o f cells in vitro.
Journal of General Virology 33, I93-211.
1NNES, 1. R. M. & SAUNDERS,L. Z. (1962). In Comparative Neuropathology, pp. 384-394. New York: Academic
Press.
KAWAI, A., MATSUMOTO,S. & TANABE, K. (1975)- Characterization o f rabies virus recovered f r o m persistently
infected B H K cells. Virology 67, 520-523.
Ln~LEFIELD, J. W. 0964)- Selection of hybrids from matings of fibroblasts in vitro and their presumed
recombinants. Science x45, 7o9-7IO.
LOWRY, O. H., ROSEBROUGI-I,N. N., FARR, A. L. & RANDALL, a. J. (1951). Protein measurement with folin phenol
reagent. Journal of Biological Chemistry x93, 265-275.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:35:31
63 2
Short communications
MADORE, H. P. H ENGLAND, J. M. (1977). Rabies virus protein synthesis in infected BHK-21 cells. Journal of
Virology 22, Io2-I ~2.
MURPHY, r. A. (I977). Rabies pathogenesis. Archives of Virology 54, 279-297.
PALMA, E. L & HUANG, A. S. (1974). Cyclic production of vesicular stomatitis virus caused by defective interfering particles. Journal of Infectious Diseases 129, 4o2-41o.
PERL, D.P. (I975). In The Natural History of Rabies, pp. 236-272. Edited by G. M. Baer. New York:
Academic Press.
PEYTREMAN, A., NICHOLSON, W.E., LIDDLE, A. W., HARTMAN, J. G. & SUTHERLAND, E.W. (I973). Effects of
methylxanthines on adenosine 3': 5'-monophosphate and corticosterone in the rat adrenal. Endocrinology
92, 525-530.
SCHUL-rZ, J. & HAM~RECn'r, B. 0973)- Adenosine 3":5"-monophosphate in cultured neuroblastoma cells:
effect o f adenosine phosphodiesterase inhibitors. Naunyn-Schmiedeberg'sArchivfi~r Pharrnakologie 278,
215-225 .
SHARMA, S. K., NIRENBERG, M. H KLEE, W. A. (1975). Morphine receptors as regulators ofadenylate cyclase
activity. Proceedingsof the National Academy of Sciences of the United States of America 72, 590-594.
TRABER, J., FISCHER, K., BUCHEN, C. H HAMPRECHT, B. (I 975 a). Muscarinic response to acetylcholine in neuroblastoma × glioma hybrid cells. Nature, London 255, 558-560.
TRABER, J., REISE~, G., fISCHER, K. H 8AMPRECHT, B. (I975b). Measurement of adenosine 3 ' : 5 ' cyclic m o n o phosphate and m e m b r a n e potential in neuroblastoma x glioma hybrid cells: opiates and adrenergic
agonists cause effects opposite to those of prostag]andin. FEBS Letters 52, 327-332.
WEISS, M. C. H EPHR.SSI, a. (I966). Studies of interspecific (rat × mouse) somatic hybrids. I. Isolation, growth
and evolution o f the karyotype. Genetics 54, lO95-11o9.
(Received 27 September I 9 7 8 )
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:35:31