Download Interferon Induced within the Central Nervous System during

473
J. gen. ViroL (1989), 70, 473-478. Printed in Great Britain
Key words: interferon (Mu[FN-~/fl)/S R V/CNS
Interferon Induced within the Central Nervous System during Infection Is
Inconsequential as a Mechanism Responsible for Murine Resistance to
Street Rabies Virus
By D O N A L D L. L O D M E L L , ~* D A N N Y L. W I E D B R A U K
L A R R Y C. E W A L T 1
2 AND
~Laboratory of Persistent Viral Diseases, Rocky Mountain Laboratories, National Institute of
Allergy and Infectious Diseases, Hamilton, Montana 59840 and 2Difco Research and Development
Center, Ann Arbor, Michigan 48108, U.S.A.
(Accepted 17 October 1988)
SUMMARY
SJL/J mice are resistant, whereas A/WySnJ mice are susceptible to intraperitoneally
(i.p.) inoculated street rabies virus (SRV). In this report we determine whether
interferon (IFN) induced within the central nervous system (CNS) of these mice during
infection is associated with resistance. We show that the high concentration of type 1
interferon (IFN-c¢//~) within the CNS of A/WySnJ mice is ineffective in inhibiting SRV
replication in these tissues, and is unimportant in ameliorating disease. More
importantly, the 100~o survival of SRV-infected SJL/J mice following neutralization of
IFN within the CNS with anti-IFN-~/fl suggests that protection of target ceils by this
minimal amount of IFN is not the mechanism responsible for the innate resistance of
SJL/J mice to i.p. inoculated SRV.
The importance of interferon (IFN) as the earliest of the known host defence mechanisms
against viral infections has been well documented. The first evidence that a rabies virus
infection could induce an IFN response came from Stewart & Sulkin (1966) who showed that
large amounts of IFN and virus were present in brains of hamsters infected with rabies.
Subsequently, it was established that rabies virus is sensitive to IFN, in that rabies-infected
mice, hamsters, rabbits and monkeys are protected from disease by either passively
administering IFN, or by inducing IFN at or near the time of infection, but before viral invasion
of peripheral nerves and/or the central nervous system (CNS.) (Fenje & Postic, 1970; Postic &
Fenje, 1971 ; Janis & Habel, 1972; Wiktor et al., 1972; Hilfenhaus et al., 1975; Weinm~inn et al.,
1979). Less is known, however, about the I F N that is induced in target organs during rabies
virus infections, and what effect this virus-induced I F N has on the course of disease. We have a
unique model to study this effect.
In previous studies we found that murine resistance to intraperitoneally (i.p.) inoculated street
rabies virus (SRV) is genetically controlled by the concurrent presence of each of two
segregating genes. SJL/J mice are resistant, whereas A/WySnJ mice are susceptible (Lodmell,
1983). Additional studies showed that SRV seldom ascends from the spinal cord to the brain of
SJL/J mice that have high serum titres of anti-SRV neutralizing antibody (Lodmell & Ewalt,
1985). In this report we expanded our studies to determine whether virus-induced I F N is present
in the CNS of these two strains of mice following infection with SRV, and if so whether its
presence is associated with resistance.
To determine whether i.p. administered SRV induces a systemic I F N response, 20 8- to 15week-old female SJL/J and 20 8- t o 15-week-old female A/WySnJ mice were inoculated with a
10~o mouse brain suspension containing 5 x 107 mouse intracranial (i.c.) 5 0 ~ lethal doses
(MICLDs0) of a wild-type SRV that had been isolated from an adult bat (Eptesicus fuscus)
0000-8589 © 1989 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:08:37
474
Short communication
(Lodmell, 1983). All of the mice were bled from the retro-orbital plexus 2 and 6 h postinoculation, and at 24 h intervals for the next 7 days. Individual sera were assayed for antiviral
activity on outbred Swiss Webster mouse embryo (ME) cells by using a standard c.p.e, bioassay
(Armstrong, 1981), and a vesicular stomatitis virus (VSV) challenge. Negative control sera were
obtained from mice that had been inoculated with a 10~ normal mouse brain suspension;
positive control sera were obtained from mice that had been inoculated intravenously (i.v.) 2 h
previously with 50 ~tg of poly(I).poly(C) (NIAID Antiviral Substances Program). I F N titres
(units; U) were estabished as the reciprocal of the serum or tissue dilution that reduced the
number of VSV plaques by 50 ~ as compared to ME cell monolayers that were incubated with
normal mouse sera. In our assays, the N I H international reference mouse I F N (G002904511)
titred 1280 U. The sensitivity of the assay was 10 U of IFN.
The results showed that no antiviral activity was detected in the sera of either mouse strain for
the first 5 days following the i.p. inoculation of SRV. Minimal titres of antiviral activity (10 to
20 U) were present in sera of three of 20 SJL/J mice and four of 20 A/WySnJ mice on the 7th day
post-inoculation. Furthermore, no I F N was detected in the peritoneal wash of similarly
inoculated mice of either strain for each of the 7 days following i.p. inoculation of SRV (data not
shown). Thus, i.p. inoculated SRV did not induce a prominent systemic antiviral response in
either the resistant or susceptible strain of mice. In contrast, sera from uninfected mice of each
strain contained high titres of I F N 2 h after an i.v. inoculation of 50 ~tg of the IFN inducer
poly(I), poly(C); the geometric mean I F N titre of sera from six SJL/J mice was 3980 (range 1280
to 5120), and that of sera from six A/WySnJ mice was 3670 (range 1280 to 5120). These data
indicate that the IFN-inducing systems of the susceptible and resistant mice are essentially
similar, in that (i) minimal amounts of systemic antiviral activity are induced following i.p.
inoculation of SRV and (ii) high titres of systemic I F N are induced after i.v. inoculation of
poly(I), poly(C).
To determine whether antiviral activity was present in the CNS following i.p. inoculation of
virus, 56 susceptible A/WySnJ and 56 resistant SJL/J mice were inoculated with 5 x 107
MICLDs0 of SRV. At various times post-inoculation eight mice of each strain were
exsanguinated and cerebrospinal fluids (CSFs), spinal cords and brains were assayed for
infectious virus (i.c. inoculation of 21-day-old mice) and I F N (Fig. 1 a, b). The data in Fig. 1 (a)
show that virus was present in spinal cords and brains of A/WySnJ mice 5 days post-inoculation.
Thereafter, virus titres steadily increased in brains until day 10 and in spinal cords until day 8;
virus was not detected in CSFs.
IFN was initially detected on day 7 in CSFs, spinal cords and brains of A/WySnJ mice.
Thereafter, IFN titres in brains steadily increased. IFN titres in CSFs and spinal cords
increased until day 8 and then decreased slightly on day 10. Little, if any, I F N was detected in
the sera of these mice before day 8. IFN titres ranging from 40 to 320 U were detected in sera on
the 10th day (data not shown). Furthermore, there was a direct correlation between the
concentrations of infectious virus and IFN titres in brains and spinal cords of A/WySnJ mice; in
general, the higher the virus titre, the higher the I F N titre. Likewise, the concentration of I F N
in CSFs correlated directly with the virus concentration in spinal cords.
In contrast to SRV-susceptible A/WySnJ mice, SRV was first detected in spinal cords of
resistant SJL/J mice on day 6 (Fig. 1 b). Thereafter, low concentrations of virus were detected in
spinal cords on days 7 and 8 but not on day 10; trace amounts of virus were detected in brains of
only four of 24 (17 ~ ) SJL/J mice on days 7 and 8. Only minimal amounts of I F N were detected
in spinal cords at these intervals; I F N was only detected in brains on day 8. Minimal I F N titres
(10 units) were detected in sera of two SJL/J mice 8 days post-inoculation, but no I F N was
detected in CSFs of any mice of this strain. Thus, our data suggest that the high concentrations
of IFN in the CNS of SRV-susceptible A/WySnJ mice were induced by replicating virus. More
importantly, this high concentration of IFN did not appear to have a direct effect on inhibiting
virus replication or limiting the infection, because SRV infections consistently resulted in the
death of A/WySnJ mice by the 12th to 14th day after i.p. inoculation of virus (Lodmell & Ewalt,
1985). Furthermore, the minimal amounts of IFN that were present in the CNS of SJL/J mice
correlated with the minimal amount of viral replication that occurred in these tissues.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:08:37
475
Short communication
/I i,,, II
/
5-0',1--(a)
II___~
,
i
4"0F[
i
,,
/t(:
i
-r-r4i,
5
rn- 10
5.0 (b)
i
II
,
105
104 k3
110'~.~4.0"~
~-
103 ,~
Z
"7- 2.0i
0
..
"
102 ~
102 =
.~ 2-0
""
i 1.0
Oo1= 3
<lO'
oo
1 3 '5
6
7
8
" 10
<10~
Time post-inoculation (days)
Fig. 1. Virus and IFN titres in CNS tissues of susceptible A/WySnJ (a) and resistant SJL/J (b) mice
after i.p. inoculation of 5 x 107 MICLDso of SRV. At various times after inoculation mice were
exsanguinated. At each interval CSFs, spinal cords and brains of eight mice of each strain were
harvested and assayed. The CNS tissues were prepared as 10% suspensions. IFN titres [in CSF ([3),
spinal cord ( i ) and brain ( [])] are expressed as U in CSF or tissue suspension as mentioned in the text.
Virus titres [in spinal cord (
) and brain (...)] are expressed as MICLDs0/0-03 ml of a 10% tissue
suspension. The minimal concentration of virus necessary for calculation of SRV titres is 100.7
MICLD50/0.03 ml. The virus-positive SJL/J brains are shown only to indicate that a few brains were
positive. The minimal concentrations of virus in these brains did not permit determination of titres.
To determine whether the I F N detected in these assays was induced in the C N S during
infection and was not produced by the M E cells used in the in vitro assay, brains and spinal cords
from eight SRV-infected A / W y S n J mice were pooled, prepared as a 10% suspension and
assayed for I F N and infectious virus. It was shown that all virus infectivity was pelleted from
this C N S suspension following centrifugation at 50000 r.p.m, for 1 h. More importantly, the
remaining supernatant fluid still contained 100% of the I F N (titre 1280), whereas the pelleted
SRV induced only a minimal amount of I F N from the M E cells (titre 20) (data not shown). Thus,
the I F N present in the C N S of A/WySnJ-infected mice was induced in vivo during infection and
was not induced from the M E cells that were used in the in vitro assay.
A pool o f brains and spinal cords from A / W y S n J SRV-infected mice also was used to
characterize the I F N . The data in Table 1 indicate that the antiviral activity was a typical virusinduced p H 2,0-stable type 1 I F N with ~ and fl specificity. In addition, the I F N in C N S tissues of
SJL/J mice also was neutralized by the rabbit anti-murine IFN-ct/fl. These d a t a support the work
of other investigators who described a similar I F N in brain extracts and plasma of mice that
were infected with the challenge virus standard fixed rabies virus rather than an SRV
(Marcovistz et al., 1984a, 1986).
To this point, our data suggest that neither virus-induced systemic I F N nor virus-induced
I F N in the C N S is important in the resistance of SJL/J mice to i.p. inoculated SRV. There were,
however, low titres of I F N present in the C N S of SJL/J mice 7 and 8 days post-inoculation.
Because I F N was present in infected tissues at these intervals, although at low levels, it was reevaluated for having a potential role in resistance. Thus, at various times after virus inoculation,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:08:37
476
Short communication
T a b l e 1. Characterization of the IFN-ct/fl induced by SR V within the CNS of susceptible
A~ WySnJ mice*
Treatment
Interferon titre (U)t
None
pH 2.0 (60 h at 4 °C):~
56 °C for 1 h
Trypsin (1 mg/ml)§
RNase (50 rtg/ml)ll
Dialysis (14000 Mr cut-off)
Centrifugation (100000 g for 4 h)
Neutralization with 3000 units of antiserum to
Mouse IFN-ct//~
Mouse IFN-/~
Normal rabbit globulin
1280
1280
1280
< 10
1280
1280
1280
< 10
80
1280
* Brains and spinal cords of eight SRV-infected A/WySnJ mice were pooled, prepared as a 10~ suspension, and
the antiviral activity of the supernatant fluid was characterized.
t The titre (U) represents the reciprocal dilution that reduces the number of VSV plaques by 50~ as compared
to ME cell monolayers incubated with a similar 10~ mouse brain suspension from uninfected mice.
:~ Samples were dialysed against 0.15 M-HCI-KCI (pH 2.0) for 24 h at 4 °C. The pH was neutralized by dialysis
against Hanks' balanced salt solution (pH 7.4) for 16 h at 4 °C.
§ Gibco trypsin (1:250) was added to the IFN to a final concentration of 1 mg/ml. Following incubation for 1 h
at 37 °C, 1.0 mg of soybean trypsin inhibitor was added.
IIBovine pancreatic RNase A in 1 mM-EDTA was added to 1 ml of IFN to give a final concentration of 50 Ixg/ml.
After incubation for 1 h at 37 °C, the pH was adjusted to 7.4 with 0.1 M-NaOH.
¶ Lee Biomolecular Research Laboratories.
T a b l e 2. I.c. inoculated rabbit anti-mouse IFN-ct/fl does not decrease the resistance of SJL/J
mice to SR V*
Time after SRV infection of
anti-IFN inoculation (days)
Test
3
5
6
7
8
5, 6
7, 8
6, 7, 8
6, 7, 8, 9
Controlst
6, 7, 8, 9 (NRS)
SJL/J
A/WySnJ
Survivors/total
4/4
4/4
4/4
6/6
4/4
4/4
4/4
4/4
4/4
4/4
6/6
0/6
* Eight- to 12-week-old SRV-resistant SJL/J and SRV-susceptible A/WySnJ mice were inoculated i.p. with
5 x 107 MICLDs0 of SRV. At various times thereafter single or multiple inoculations of 3000 units of rabbit antimouse IFN-c~/flwere administered i.c. The anti-mouse IFN (Catalog no. 21032, Lot no. 87021) was quantified by
Lee Biomolecular Research Laboratories by comparative assays with the anti-IFN reference agents provided by
the Antiviral Substances Program, NIAID, NIH. One neutralizing unit completely neutralizes 3 to 10 U of
corresponding IFN activity. Assays in our laboratory produced similar results, i.e. 10 neutralizing units
neutralized 50 U of IFN-~/fl.
t Control SJL/J mice were inoculated with SRV and normal rabbit serum (NRS) or SRV only. Control
A/WySnJ mice were inoculated with SRV only. The experiment was terminated 21 days after SRV inoculation.
single or multiple i.c. inoculations of 3000 units of r a b b i t anti-mouse I F N - ~ / f l (Lee B i o m o l e c u l a r
R e s e a r c h Laboratories, San Diego, Ca., U . S . A . ) were a d m i n i s t e r e d before, concurrently or after
the intervals w h e n I F N had been previously d e t e c t e d in S J L / J C N S tissues. C o n t r o l S J L / J m i c e
were inoculated i.p. w i t h S R V alone, or w i t h S R V i.p. and n o r m a l rabbit s e r u m i.c. C o n t r o l
A / W y S n J m i c e r e c e i v e d only S R V i.p. T h e d a t a in T a b l e 2 illustrate that n o n e o f the S R V -
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:08:37
Short communication
477
infected SJL/J mice died, regardless of the interval or the quantity of anti-IFN that was
administered. Furthermore, spinal cords and brains removed from other SJL/J mice that also
received 3000 units of anti-IFN i.c. 7 and 8 days post-infection did not contain I F N activity
8 days after SRV infection (data not shown). These data suggest that the i.c. inoculated anti-IFN
neutralized the I F N in the CNS of the SJL/J mice that survived in the in vivo experiment (Table
2). SJL/J mice that received only SRV had spinal cord I F N titres of 20 at this time (Fig. 1 b).
Anti-IFN was not given to A/WySnJ mice. Thus, neutralization of the minimal amount of I F N
that was present in CNS tissues of SJL/J mice did not affect their resistance to SRV, suggesting
that protection of target cells by I F N was not the mechanism responsible for the innate
resistance of this mouse strain to SRV.
There is no doubt that rabies mortality can be markedly reduced by the local injection of IFN,
potent vaccines that induce IFN prior to neutralizing antibody, and/or IFN inducers such as
poly(I), poly(C) or poly(IC), poly(LC). The exogenously transferred or the endogenous I F N that
is induced in injected muscles appears to limit the minimal extraneural replication of virus in
striated muscle cells at the site of infection before virus infects peripheral nerves and spreads to
the CNS (Baer, 1981). Additional evidence supporting the importance of I F N in rabies
infections has been documented by Marcovistz et al. (1986) and GuiUon & Tsiang (1980), who
showed that infected mice pretreated with anti-IFN-ct/fl globulin had a significantly increased
sensitivity to infection compared to normally infected mice. Nonetheless, the present study, as
well as studies of others (Stewart & Sulkin, 1966; Smith, 1981 ; Marcovistz et al., 1984b), have
shown that IFN was not protective after virus reached the CNS because rabies virus-infected
animals died with very high levels of I F N in CNS tissues.
The ineffectiveness of locally produced I F N to protect mice following virus invasion of the
CNS could be due to several different circumstances. It may be that I F N is produced in
insufficient amounts or produced too late in the infection. Alternatively, the total amount of
virus in CNS tissues may be too high for death to be prevented by IFN, or damage to CNS cells
may have reached a stage at which inhibition of viral replication by I F N no longer protects the
host (Postic & Fenje, 1971). It also may be that few I F N receptors are present on the surface of
CNS cells. If so, these cells would be less amenable to the protective effect of IFN. Furthermore,
the lack of antiviral effectiveness of IFN may be due to the failure of the oligo-2',5'-adenylate
synthetase and protein kinase IFN-dependent enzymes to elicit an antiviral state in the CNS
(Marcovistz et al., 1984a, b, 1986). There may also be genetic control of the sensitivity to I F N
action, as has been previously shown by the poor inhibition of influenza A virus replication by
I F N in mouse cells lacking the gene M x (Hailer, 1981).
It has been shown that primary cultures of murine astrocytes, but not neurons, produce I F N
that is biologically and antigenically similar to IFN-ct/fl (Tedeschi et al., 1986). Furthermore,
there appears to be compartmentation of IFN in rabies virus-infected brains in that the highest
level of IFN is detected in the brain stem and the lowest level is in the striatum (Marcovistz et at.,
1984a). Compartmentation of IFN also has been shown in patients dying from St Louis
encephalitis; I F N is present in frontal lobes, cerebellum and basal ganglia but not brain stem
(Luby et al., 1969). Likewise, marked differences in I F N content between certain areas of
human brains infected with Western equine encephalitis virus have been reported (Luby et al.,
1971). Thus, it may be that astrocytes in different locations of the brain, and possibly the spinal
cord, produce different concentrations of IFN. If so, these concentration differences throughout
the CNS may make it impossible for I F N to reach all of the cells that need protection, and the
host will die.
We are still uncertain why SJL/J mice are resistant to i.p. inoculated SRV. Nonetheless,
because immunosuppression of these mice with cyclophosphamide converts an asymptomatic
infection into a lethal one (Lodmell & Ewalt, 1985), we are of the opinion that resistance is
controlled by the immune response. Our present studies are focusing on cytotoxic T cells and
their interaction with the humoral immune response.
We thank Irene Cook Rodriguez for typing the manuscript and Robert Evans for graphic arts assistance.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:08:37
478
Short communication
REFERENCES
ARMSTRONG, J. A. (1981). Cytopathic effect inhibition assay for interferon: microculture plate assay. Methods in
Enzymology 78, 381-387.
BAER, G. M. (1981). The combination of interferon inducer and vaccine in rabies post-exposure prophylaxis. In Cell
Culture Rabies Vaccinesand TheirProtectiveEffect in Man, pp. 221 226. Edited by E. K. Kuwert, T. J. Wiktor
& H. Koprowski. Geneva: International Green Cross.
FENJE, P. & POSTIC, B. (1970). Protection of rabbits against experimental rabies with poly(I)-poly(C). Nature,
London 226, 171-172.
GUmLON, J. C. & TSIANG,H. (1980). R61e de l'interferon et du t h y m u s dans la pathogenese de l'infection rabique chez
la souris. Annales de Virologic 131E, 229 245.
HALLER, O. (1981). Inborn resistance of mice to orthomyxoviruses. Current Topics in Microbiologyand Immunology
92, 24-52.
HILFENHAUS, J., KARGES, H. E., WEINM,gtNN, E. & BARTH, R. (1975). Effect of administered h u m a n interferon on
experimental rabies in monkeys. Infection and lmmunity 11, 1156-1158.
JANIS, B. & HABEL, K. (1972). Rabies in rabbits and mice: protective effect of polyriboinosinic-polyribocytidylic
acid. Journal of Infectious Diseases 125, 345 352.
LODMELL,D. L. (1983). Genetic control of resistance to street rabies virus in mice. Journalof Experimental Medicine
157, 451-460.
LODMELL, D. L. & EWALT, L. C. (1985). Pathogenesis of street rabies virus infections in resistant and susceptible
strains of mice. Journal of Virology 55, 788-795.
LUBY, J. P., STEWART,W. E., II, SULKIN, S. E. & SANFORD, J. P. (1969). Interferon in h u m a n infections with St. Louis
encephalitis virus. Annals of Internal Medicine 71, 703-710.
LUBY, J. P., SANDERS,C. V. & SULKIN, S. E. (1971). Interferon assays and quantitative virus determinations in a fatal
infection in m a n with Western equine encephalomyelitis virus. American Journal of Tropical Medicine and
Hygiene 20, 765-769.
MARCOVISTZ,R., HOVANESSIAN,A. G. & TSIANG, H. (1984a). Distribution of rabies virus, interferon and interferonmediated enzymes in the brains of virus-infected rats. Journal of General Virology 65, 995-997.
MARCOVISTZ,R., TSIANG, H. & HOVANESSIAN, A. G. (1984b). Production and action of interferon in mice infected with
rabies virus. Annales de Virologie 135E, 19-33.
MARCOVISTZ,R., GALABRU,J., TSIANG, H. & HOVANESSIAN,A. G. (1986). Neutralization of interferon produced early
during rabies virus infection in mice. Journal of General Virology 67, 387-390.
POSTIC, a. & FENJE, P. (1971). Effect o f administered interferon on rabies in rabbits. Applied Microbiology 22,
428-431.
SMITH, J. (1981). Mouse model for abortive rabies infection of the central nervous system. Infection and lmmunity
31,297-308.
STEWART,W. E. & SULKIN,S. E. (1966). Interferon production in hamsters experimentally infected with rabies virus.
Proceedings of the Society for Experimental Biology" and Medicine 123, 650-654.
TEDESCHI, B., BARRET,J. N. & KEANE, R. W. (1986). Astrocytes produce interferon that enhances the expression of
H-2 antigens on a subpopulation of brain cells. Journal of Cell Biology 102, 2244-2253.
WEINMANN, E., MAJER, M. & HILFENHAUS,J. (1979). Intramuscular and/or intralumbar postexposure treatment of
rabies virus-infected cynomolgus monkeys with h u m a n interferon. Infection and Immunity 24, 24-31.
WIKTOR, T. J., POSTIC, B., HO, M. & KOPROWSKI,H. (1972). Role of interferon induction in the protective activity of
rabies vaccines. Journal of Infectious Diseases 126, 408-418.
(Received 26 July 1988)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:08:37