Journal of General Virology (1998), 79, 825–830.
Printed in Great Britain
Mice lacking inducible nitric-oxide synthase are more
susceptible to herpes simplex virus infection despite enhanced
Th1 cell responses
Alasdair MacLean,1Xiao-Qing Wei,2Fang-Ping Huang,2Umaima A. H. Al-Alem,3Woon Ling Chan3
and Foo Y. Liew2
1,2Division of Virology1and Department of Immunology2, University of Glasgow, Glasgow G11 6NT, UK
3Department of Virology, St Bartholomew’s Hospital and Royal London Hospital School of Medicine, University of London, London EC1A
Mice deficient in the inducible nitric-oxide synthase
(iNOS), constructed by gene-targeting, were signifi-
cantly more susceptible to herpes simplex virus
(HSV)-1 infection, displayed a delayed clearance of
virus from the dorsal root ganglia (DRG) and
exhibited an increase in the frequency of virus
heterozygous mice. The infected iNOS-deficient
mice developed enhanced Th1-type immune re-
Inducible nitric-oxide synthase (iNOS) catalyses the syn-
thesis of high concentrations of nitric oxide (NO) from -
arginine and plays a role in microbicidal and tumoricidal
activities, and in immunopathology (Moncada & Higgs, 1993;
Nathan & Xie, 1994). It may also be important in immune
regulation (Albina et al., 1991; Liew, 1995). The role of NO in
virus infection is, however, controversial (Croen, 1993;
Karupiah et al., 1993; Burkrinsky et al., 1995; Mannick et al.,
1994; Rolph et al., 1996; Adler et al., 1997). Most of these
studies used -arginine analogues which are competitive
inhibitors of NO synthases. The discrepancy between the
results may be attributable to the non-isoform-selective nature
and the variation in the bioavailability of the inhibitors. By
gene targeting, we have constructed a mouse strain lacking
iNOS (Wei et al., 1995). These mice are highly susceptible to
intracellular parasitic infection. We have now tested the ability
of these mice to resist herpes simplex virus (HSV)-1 infection.
? Mice. iNOS-deficient mice were generated as described previously
(Wei et al., 1995). Disruption of the murine iNOS gene was achieved by
homologous recombination in 129sv embryonic stem cells. The re-
Author for correspondence: Foo Y. Liew.
Fax ?44 141 337 3217. e-mail f.y.liew?clinmed.gla.ac.uk
sponses and their spleen cells produced higher
concentrations of IL-12 than similarly infected het-
erozygous mice. This finding suggests that iNOS
plays an important role in resistance against HSV-1
infection. Furthermore, nitric oxide (NO) may block
the development of Th1 cells via inhibition of IL-12
synthesis and thereby play a role in immune regu-
combinant allele was passed through the germ line following mating of
embryonic stem cell chimaeras with MF1 (Harlan Olac). The homo-
zygous, heterozygous and wild-type mice thus generated were back-
crossed to MF1 for three generations. All the mice used were from
matings of littermates and should therefore have had a similar MF1 gene
background. Peritoneal cells from mutant mice did not produce iNOS
protein following activation by IFN-γ and LPS in vitro as judged by
Western blot. They also did not produce detectable amounts of NO
followingupto48 hculturewithIFN-γplusLPS.By72 h,however,alow
level of nitrite was detectable in the culture supernatant of cells from
mutant mice. This may reflect the accumulation of nitrite produced by
Female mutant mice and their heterozygous littermates were used at 3–4
weeks old. Extensive previous experiments showed no significant
phenotypic difference between the heterozygous and wild-type mice
(Wei et al., 1995); hence in the present study only heterozygous mice
were used as controls.
? Virus and infection. HSV-1 (Glasgow strain 17+) was grown and
titrated in baby hamster kidney cells (BHK21?C13) (MacPherson &
Stoker, 1962) propagated in Eagle’s medium as described previously
(MacLean et al., 1991). Mice were inoculated in the right hind footpad
with the appropriate doses of virus (Robertson et al., 1992), which were
titrated prior to inoculation.
? Peripheral virulence. Groups of five mice were inoculated with a
series of 10-fold dilutions of virus. Mice were examined daily and the
LD??calculated according to the formula of Reed & Muench (1938) on
the basis of the number of deaths up to 42 days post-infection.
? Virus replication during acute infection. Mice were inoculated
with 10? p.f.u. of virus per mouse. Virus replicates in the footpad, enters
0001-5188 ? 1998 SGM
A. MacLean and othersA. MacLean and others
the sciatic nerve and travels by retrograde transport to the dorsal root
ganglion (DRG) where replication occurs prior to the establishment of a
latent infection. Groups of mice were sacrificed at regular intervals and
the right rear footpad together with the two lowest thoracic (T11, T12),
five lumbar (L1–5) and three sacral (S1–3) DRG were removed from the
inoculated side and stored at ?70 ?C. All ten ganglia from each mouse
were pooled, but the footpad and DRG from each mouse were processed
determined by titration in BHK cells (Robertson et al., 1992).
? Reactivation of latent virus from DRG. Mice were inoculated
with 10? or 10? p.f.u. of virus per mouse and examined daily for clinical
symptoms. Mice surviving 6 weeks were assayed for the presence of
latent virus. The mice were sacrificed and DRG removed from the
and screened for the release of infectious virus every second day by
transferring the culture medium to control BHK cells which were then
incubated for 2 days at 37 ?C before examining for the presence of virus
? Antibody titration. Serum (3–5? dilutions) was added into
microtitre plates coated with irradiated virus (2?10? p.f.u. equivalent?
ml) and incubated for 1 h. The plates were then washed and antibody
detected with peroxidase-conjugated rabbit anti-mouse IgG, IgG1 or
IgG2a (Dako) and developed with 2,2?-azino-bis(3-ethylbenzthiazoline-
6-sulfonic acid). The plates were read on a Bio-kinetics Reader (Bio-Tek
Instrument) at 405 nm. The end-point was calculated as the highest
dilution of serum which gave an absorbance reading twice that of the
background (with normal mouse serum).
nine or ten mice and analysed for CD3+, CD3+CD4+and CD3+CD8+
subsets using the appropriate monoclonal antibody (Becton Dickinson).
Labelled cells were analysed by flow cytometry (FACScan, Becton
? T cell proliferation. Draining lymph node cells from infected mice
were cultured (10? cells?ml) in medium (RPMI 1640, Gibco) containing
0?5% FCS and 0?5% normal mouse serum) in flat-bottom 96-well plates
(Nunclon) with graded concentrations (5?10?–5?10? p.f.u. equiv-
alent?ml) UV-irradiated HSV-1 for 3–4 days. The cells were then pulsed
with [?H]thymidine [1 µCi?well (37 kBq per well), Amersham] for a
further 16 h, harvested and the radioactivity counted in a Beta-plate
? Cytokine production and detection. Spleen cells from infected
mice were cultured (2?10? cells?ml) in medium (RPMI 1640 and 10%
FCS) in flat-bottom 24-well plates with graded concentrations (10?–10?
p.f.u. equivalent?ml) of UV-irradiated virus, or with concanavalin A
(ConA) (2?5 µg?ml) or LPS (5 µg?ml) for up to 7 days. Cell-free culture
supernatants were collected at 1, 2, 3, 4 and 7 days and cytokine
concentrations determined by ELISA in 96-well plates (Immulon 4). IL-12
was captured with a combination of rat monoclonal antibodies to mouse
IL-12 P40 (C15.1.2 and C15.6, kind gifts of the Genetic Institute,
Cambridge, Mass., USA), and detected with a rabbit anti-mouse IL-12
antibody (Rab.74.6, raised in our laboratory). IFN-γ was captured with a
rat monoclonal antibody to mouse IFN-γ (R46AT) and detected with a
rabbit anti-mouse IFN-γ antibody. IL-4 was captured with a rat
monoclonal anti-IL-4 antibody (TRFK4, PharMingen) and detected with
a biotinylated rat anti-mouse IL-4 antibody (BVD6-24G2, PharMingen).
ELISA was developed with HRP-conjugated donkey anti-rabbit IgG
(SAPU) for IL-12 and IFN-γ, or HRP-conjugated StrepAvidin (SAPU) for
IL-4, and followed using TMP HRP substrate (Dynatech, UK). The
absorbance was read on a multi-scan (MR5000, Dynatech) at 630 nm.
Recombinant cytokines were used as reference standards.
? Statistical analysis. Statistical significance (P?0?05) was cal-
culated by the Mann–Whitney test (Minitab software program).
Results and Discussion
iNOS-deficient mice, together with their heterozygous
controls, were infected in the footpad with 10-fold dilutions of
HSV-1 Glasgow strain 17+(10?–10? p.f.u. per mouse). Disease
development and mortality were monitored for 42 days post-
infection. The LD??of the heterozygotes was 9?10? p.f.u. per
mouse, which is comparable to that of normal strain BALB?c
per mouse in the mutant iNOS-deficient mice. Compared with
similarly infected heterozygous control mice, the mutant mice
also displayed a significant delay in their ability to clear virus
from the footpad and the DRG (Fig. 1a, b). iNOS-deficient
mice that survived HSV-1 infection of 10?–10? p.f.u. per
mouse exhibited a marked increase in the frequency of virus
reactivation in the DRG compared with heterozygous mice
that had recovered from the same dose of virus (Fig. 1c, d). For
ganglia that are not directly supplying the sciatic nerve (T11,
T12, L1, L2, S2 and S3) and where virus is spread through the
spinal cord, the frequencies of virus reactivation were 85% for
the iNOS-deficient mice and 14% for heterozygous control
We next investigated the immune response of acutely
infected mice. Heterozygous mice infected with 10? p.f.u. of
HSV-1 per mouse produced HSV-1-specific antibody de-
tectable 5 days after infection and peaking by day 7. Mutant
mice produced significantly higher concentrations of specific
antibody from day 5 which progressed through to day 11 (Fig.
2a). All the antibody was of the IgG2a isotype; no IgG1
antibody was detected (data not shown). This is consistent
with a previous finding that cellular rather than humoral
(Nash & Wildy, 1983; Chan et al., 1985). During the latent
phase of infection by sublethal doses of virus, there was no
significant difference between mutant and heterozygous mice
per mouse, draining lymph node cells were collected and
from the heterozygous mice produced a modest but significant
proliferative response when cultured with 5?10?–5?10?
p.f.u.?ml of irradiated virus. This level was markedly elevated
in cells from the mutant mice (Fig. 2b).
CD4+T cells primed by glycoprotein B of HSV can protect
miceagainstacuteHSVinfection(Chan etal.,1985). Therefore,
spleen cells from the infected mice were examined for T cell
distribution and cultured with irradiated virus, ConA or LPS in
vitro, followed by determination of the concentations of IFN-
γand IL-4 in the culture supernatants.Flow cytometric analysis
from nine to ten mice per group) from the iNOS-deficient mice
HSV infection in iNOS-deficient mice HSV infection in iNOS-deficient mice
0 10 20
Time (days post-infection)
0 10 20
0 10 20
Time (days post-explantation)
300 10 2030
Virus titre (p.f.u. per tissue)
Fig. 1. Virus isolation and reactivation in mice infected with HSV-1. Mice were injected in the right rear footpad with 105p.f.u.
of virus per mouse and sacrificed at regular intervals. Virus titres in the infected footpad (a) and the DRG (b) were determined
by titration in BHK21/C13 cells. Each point represents the mean virus yield in p.f.u. per tissue sample of two to three mice.
Standard errors are shown as vertical bars. Typically, in this type of experiment, due to physical losses, a 100? decline in titre
immediately post-infection is observed. Thus the actual virus titre is probably 100? higher than shown. The pattern seen in the
heterozygous mice is typical of other normal strains of mouse where virus titre in the footpad peaks on day 2 and declines
thereafter, followed by a second peak on day 7 post-infection. No virus was detected by day 14. In the DRG, virus was first
detected on day 2, when virus would have travelled up the sciatic nerve, peaked on day 5 following replication in the DRG and
thereafter rapidly declined, with no virus detectable by day 14, when a latent infection would have been established. The
second peak in the footpad is believed to be due to virus travelling back down the sciatic nerve following replication in the
DRG. The time-course of explant reactivation in mice infected with 104(c) or 105(d) p.f.u. of virus per mouse was also
examined. Six weeks after inoculation of 4-week-old mice in the right rear footpad, ten DRG were explanted from the right side
of surviving mice (four to five mice per group) and assayed every second day for the release of infectious virus in BHK21/C13
cells. The percentage of total DRG reactivating at each time point is shown. Similar results were obtained in two additional
contained more CD3+cells (37?9% vs 31?2%, P?0?05) and
CD3+CD4+ cells (28?7% vs 20?91%, P?0?05) but similar
fractions of CD3+CD8+cells (7?9% vs 6?98%) compared with
heterozygous mice. Cells from infected iNOS-deficient mutant
mice produced significantly more IFN-γ and less IL-4 than
those from similarly infected heterozygous control mice (Fig.
3), indicating a preferential expansion of Th1 cells in the
absence of iNOS. These results are consistent with the
observation that NO inhibits the development of Th1 cells
(Wei et al., 1995; Taylor-Robinson et al., 1994). Cells from
mutant mice produced significantly more IL-12 than cells from
heterozygous mice at all time points tested (Fig. 4). Since IL-12
is a major inducer of Th1 cell development (Trinchieri, 1993)
and is predominantly produced by macrophages, these data
indicate that NO produced endogenously by iNOS in
macrophages could inhibit the production of IL-12, thereby
limiting the development of Th1 cells. The mechanism by
which NO inhibits IL-12 synthesis is at present unclear.
A. MacLean and othersA. MacLean and others
Time (days post-infection)
Antigen dose (p.f.u./ ml)
10–4 × Anti-HSV antibody titre
(reciprocal of dilutions)
10–3 × [3H ]Thymidine uptake (c.p.m.)
Fig. 2. (a) Virus-specific antibody
response. Mice were infected as
described in the legend to Fig. 1. Sera
were collected when mice were
sacrificed, and titrated for anti-HSV
antibody by ELISA. Each point represents
the mean of triplicate wells ?1 SD from
two to three mice. (b) Proliferative
response of draining lymph node cells
from mice infected 5 days previously with
105p.f.u. of HSV-1 per mouse. For
details see Methods. Each point
represents the mean of triplicate cultures
?1 SD from a pool of cells from three
mice. Similar results were obtained in two
10–3 × IFN–γ concn (pg/ ml)
10–2 × IL–4 concn (pg/ ml)
Fig. 3. IFN-γ (a) and IL-4 (b) produced
by spleen cells from mice infected 40
days previously with 104p.f.u. of HSV-1
per mouse and cultured in vitro with
optimal dose of antigen (106p.f.u.
equivalent/ml of irradiated HSV-1) for 4
days, or with ConA (2?5 µg/ml) or LPS
(5 µg/ml) for 3 days. Data for peak time
points are shown. Culture supernatants
were titrated for the cytokines by ELISA.
Data shown are means ?1 SD of
triplicate cultures. Similar results were
obtained in two additional experiments.
However, it is unlikely to be the consequence of higher virus
load, since IL-12 is induced rapidly after infection (Kanangat et
al., 1996; Scott, 1993; Ma et al., 1996) and the virus replication
rate was similar in both iNOS-deficient and control mice for
phenomenon was observed for Leishmania major infection
Using inhibitors for NO synthase, NO has been shown to
inhibit the in vitro replication of ectromelia virus, HSV-1,
vaccinia virus, vesicular stomatitis virus and human immuno-
et al., 1995; Mannick et al., 1994; Bi & Reiss, 1995). These
inhibitors have also been shown to significantly exacerbate
ectromelia virus infection (Karupiah et al., 1993), HSV-1-
induced pneumonia (Adler et al., 1997) and influenza virus-
induced pneumonia (Akaike et al., 1996). However, treatment
of mice with NOS inhibitors failed to influence the course of
vaccina virus (Rolph et al., 1996), influenza virus (J. P. Tite & F.
Y. Liew, unpublished), or lymphocytic choriomeningitis virus
(R. M. Zinkernagel & F. Y. Liew, unpublished) infections.
Furthermore, NO appears to protect mice from fatal en-
cephalitis induced by Sindbis virus by a mechanism that does
notdirectlyinvolvetheimmune responseor inhibitionof virus
growth (Tucker et al., 1996). The present study provides direct
in vivo evidence that iNOS plays a role in inhibiting HSV-1
replication and in host protection. The mechanism by which
NO limits virus replication is at present unclear. It is likely that
NO acts as a direct effector molecule rather than indirectly
through the enhancement of host immune responses because
in spite of enhanced Th1 response, which is known to be host-
by lysing target cells or by direct damage to viral particles as
in determining the outcome of HSV infection, and that
resistance correlates with the ability of macrophages to restrict
HSV replication and dissemination (Johnson, 1964). This
mechanism predicts that NO is important in infections with
against virus infections in which CD8+killer cells are the main
effector mechanism. iNOS-deficient mice contain similar levels
of CD8+ T cells as heterozygous mice following HSV-1
HSV infection in iNOS-deficient miceHSV infection in iNOS-deficient mice
10–3 × IFN–γ concn (pg/ ml)
10–2 × IL–12 concn (pg/ ml)
Time in culture (days)
5 days post-infection10 days post-infection 10 days post-infection
Fig. 4. IFN-γ (a, b, c) and IL-12 (d, e, f) produced by pooled spleen cells from mice infected at 5 and 10 days (105p.f.u. per
mouse) or 40 days (104p.f.u. per mouse) with HSV-1. Cells were cultured with or without UV-irradiated virus for the periods
indicated. Data for optimal cytokine production are shown. Optimal concentrations of antigen for cytokine production were
107p.f.u./ml for cells from day 5 post-infection, and 106p.f.u./ml for cells from 10 and 40 days post-infection. Cytokine
concentrations were determined by ELISA. Each point represents the mean ?1 SD, n?3. Some of the error bars are obscured
by the symbols. Similar results were obtained in two additional experiments.
infection. The ability of iNOS-deficient mice to control low
dosesof infectionindicatesthat mechanisms in additiontoNO
are involved in the resistance against HSV-1 infection.
Furthermore, macrophages are not the only cell type HSV-1
A major finding in this report is the preferential induction
of Th1 cells in mice deficient in iNOS. This is likely to result
from inhibition of IL-12 synthesis by NO. It is now generally
agreed that the balance between Th1 (producing IFN-γ) and
Th2 (producing IL-4) subsets of CD4+T cells determines the
outcome of many infectious and autoimmune diseases (Sher &
Coffman, 1992; Liew, 1992; Mason & Fowell, 1992). Our data
suggest that iNOS is important in maintaining a state of
immunological balance, preventing the overexpansion of Th1
cells which have been implicated in a range of immuno-
pathologies. They also suggest that iNOS inhibitors may be
useful adjuvants for vaccination where an enhanced Th1 cell
response is essential for protective immunity.
We thank the Medical Research Council of Great Britain, the Joint
Research Board of the Barts NHS Trust and the Wellcome Trust for
Adler, H., Beland, J. L., Del-Pan, N. C., Kobzik, L., Brewer, J. P.,
Martin, T. R. & Rimm, I. J. (1997). Suppression of herpes simplex virus
type I (HSV-1)-induced pneumonia in mice by inhibition of inducible
Akaike, T., Noguchi, Y., Ijiri, S., Setoguchi, K., Suga, M., Zheng, Y. M.,
Dietzschold, B. & Maeda, H. (1996). Pathogenesis of influenza virus-
induced pneumonia: involvement of both nitric oxide and oxygen
radicals. Proceedings of the National Academy of Sciences, USA 93,
Albina, J. E., Abate, J. A. & Henry, W. L., Jr (1991). Nitric oxide
production is required for murine resident peritoneal macrophages to
suppress mitogen-stimulated T cell proliferation: role of IFN-γ in the
inductionof the nitric oxide-synthesizing pathway. Journal of Immunology
A. MacLean and othersA. MacLean and others Download full-text
Bi, Z. & Reiss, C. S. (1995). Inhibition of vesicular stomatitis virus
infection by nitric oxide. Journal of Virology 69, 2208–2213.
Burkrinsky,M. I.,Nottet,H. S. L. M.,Schmidtmayerova,N.,Dubrovsky,
L., Mullins, M. E., Lipton, S. A. & Gendelman, H. E. (1995). Regulation
of nitric oxide activity in HIV-infected monocytes: implication for HIV-
associated neurological disease. Journal of Experimental Medicine 181,
Chan, W. L., Lukic, M. L. & Liew, F. Y. (1985). Helper T cells induced by
an immunopurified herpes simplex virus type I (HSV-1) 115 kilodalton
glycoprotein (gB) protect mice against HSV-1 infection. Journal of
Experimental Medicine 162, 1304–1318.
Croen, K. D. (1993). Evidence for an antiviral effect of nitric oxide.
Journal of Clinical Investigation 91, 2446–2452.
Johnson, R. T. (1964). The pathogenesis of herpes virus encephalitis. II.
A cellular basis for the development of resistance with age. Journal of
Experimental Medicine 120, 359–369.
Kanangat, S., Thomas, J., Gangappa, S., Babu, J. S. & Rouse, B. T.
(1996). Herpes simplex virus type 1-mediated up-regulation of IL-12
(p40) mRNA expression. Journal of Immunology 156, 1110–1116.
Karupiah, G., Xie, Q.-w., Buller, R. M. L., Nathan, C., Duarte, C. &
MacMicking, J. D. (1993). Inhibition of viral replication by interferon-γ-
induced nitric oxide synthase. Science 10, 1445–1448.
Liew, F. Y. (1992). Induction, regulation and function of T-cell subsets
in leishmaniasis. Chemical Immunology 54, 117–135.
Liew, F. Y. (1995). Interactions between cytokines and nitric oxide.
Advances in Neuroimmunology 5, 201–209.
Ma, X., Chow, J. M., Gri, G., Carra, G., Gerosa, F., Wolf, S. F., Dzialo, R.
& Trinchieri, G. (1996). The interleukin 12 p40 gene promoter is primed
by interferon γ in monocytic cells. Journal of Experimental Medicine 183,
MacLean, A. R., Ul-Fareed, M., Robertson, L., Harland, J. & Brown, S.
M. (1991). Herpes simplex virus type 1 deletion variants 1714 and 1716
pinpoint neurovirulence-related sequences in Glasgow strain 17+ be-
tween immediate early gene 1 and the ‘a’ sequence. Journal of General
Virology 72, 631–639.
MacPherson, I. & Stoker, M. G. (1962). Polyoma transformation of
hamster cell lines: an investigation of genetic factors affecting cell
competence. Virology 16, 147–151.
Mannick, J. B., Asano, K., Izumi, K., Kieff, E. & Stamler, J. S. (1994).
Nitric oxide produced by human B lymphocytes inhibits apoptosis and
Epstein–Barr virus reactivation. Cell 79, 1137–1146.
Mason, D. & Fowell, D. (1992). T-cell subsets in autoimmunity. Current
Opinion in Immunology 4, 728–732.
Moncada, S. & Higgs, A. (1993). The -arginine–nitric oxide pathway.
New England Journal of Medicine 329, 2002–2012.
Nash, A. A. & Wildy, P. (1983). Immunity inrelation tothe pathogenesis
of herpes simplex virus. In Human Immunity to Viruses, pp. 179–192.
Edited by F. A. Ennis. New York: Academic Press.
Nathan, C. & Xie, S. Q. (1994). Regulation of biosynthesis of nitric
oxide. Journal of Biological Chemistry 269, 13725–13728.
Reed, L. J. & Muench, H. (1938). A simple method of estimating fifty
percent endpoints. American Journal of Hygiene 27, 493–449.
Robertson, L. M., MacLean, A. R. & Brown, S. M. (1992). Peripheral
replication and latency reactivation kinetics of the non-neurovirulent
herpes simplex virus type 1 variant 1716. Journal of General Virology 73,
Rolph, M. S., Ramshaw, I. A., Rockett, K. A., Ruby, J. & Cowden, W. B.
infection, but may not be essential for virus clearance. Virology 217,
Scott, P. (1993). IL-12: the initiation cytokine for cell-mediated
immunity. Science 260, 696.
Sher, A. & Coffman, R. L. (1992). Regulation of immunity to parasites
by T cells and T-cell derived cytokines. Annual Review of Immunology 10,
Taylor-Robinson, A. W., Liew, F. Y., Severn, A., Xu, D., McSorley, S. J.,
Garside, P., Padron, J. & Phillips, R. S. (1994). Regulation of the
immuneresponsebynitric oxidedifferentiallyproduced byThelpertype
1 and T helper type 2 cells. European Journal of Immunology 24, 980–984.
Trinchieri, G. (1993). Interleukin-12and itsrole inthe generationof Th1
cells. Immunology Today 14, 335–338.
Tucker, P. C., Griffin, D. E., Hchoi, S., Bui, N. & Wesselingh, S. (1996).
Inhibition of nitric oxide synthesis increases mortality in sindbis virus
encephalitis. Journal of Virology 70, 3972–3977.
Wei, X.-Q., Charles, I. G., Smith, A., Ure, J., Feng, G.-J., Huang, F.-P.,
Xu, D. M., Muller, W., Moncada, S. & Liew, F. Y. (1995). Altered
immune response in mice lacking inducible nitric oxide synthase. Nature
Received 4 September 1997; Accepted 11 November 1997