Znt. J.
Cancer: 31, 309-314 (1983)
NATURAL KILLER CELL ACTIVITY DURING MOUSE HEPATITIS
VIRUS INFECTION: RESPONSE IN THE ABSENCE OF INTERFERON
Stephen A. S T O H L M A N ' ~ ~Peter
, R. BRAYTC",~,Richard C. I-IARMON~, Douglas STEVENSON*,
Roland G. GANGES'S~ and Glenn K. MATSUSHIMA'~~
Departments of Neurology' and Microbiology2, University of Southern California, School of Medicine,
2025 Zonal Avenue, Los Angeles, California 90033, USA.
The ability of the JHM' strain of mouse hepatitis virus
(MHV) to induce natural killer (NK) cells was examined.
Infection of C57BU6 (86) mice with this virus resulted in
the augmentation of natural cytotoxicity against YAC-I
target cells in the absence of a detectable interferon re-
sponse. The cells responsible for this increased cytotoxici-
ty were sensitive to complement-mediated lysis with an
anti-Q-5 reagent but not with a Thy 1.2 antiserum, indi-
cating that they possess an NK-like surface phenotype.
Although variation in the N K response of individual B6
mice following JHM virus infection was found, even the
animal with the most responsive N K cell population had
no detectable interferon in the spleen. This finding con-
trasted with observations with an unrelated virus (lym-
phocytic choriomeningitis virus) and a serologically re-
lated strain of MHV. Infection with both of these viruses
induced augmented N K cell activity and interferon re-
sponses. In addition, we found that neither the ability to
mount an augmented N K cell response nor preferential
lysis of virus-infected targets correlated with resistance or
susceptibility to JHM virus infection.
Natural killer (NK) cells are a subpopulation of
lymphocytes, present in normal animals, which are
cytolytic for certain tumor or normal cell targets in
vitro (Herberman and Holden, 1978; Kiessling and
Wigzell, 1979; Clark and Harmon, 1980). Although
the physiological role of NK cells is not known, they
play a protective role against some malignancies in
vivo (Kasai et al., 1979; Cheever et al., 1980; Kawase
et al., 1982). Cold target inhibition (Herberman et al.,
1975; Kiessling et al., 1975; Herberman and Holden,
1978) and monolayer adsorption (Herberman and
Ortaldo, 1980) experiments have indicated that NK-
mediated target cell lysis involves the specific recogni-
tion of cell-surface determinants; however, very little
is known concerning the NK cell receptor or the deter-
minants which they recognize.
One potential class of target antigens, viral-spe-
cified cell-surface determinants, has been studied in
particular detail. An early study (Herberman et al.,
1975) indicated that effective cold target inhibition of
NK activity was associated with the expression of re-
trovirus cell surface antigens. Subsequently Lee and
Ihle (1977) showed that NK activity against AKR lym-
phoma cells could be blocked by murine leukemia
virus gp 70. In contrast, other studies have failed to
show a correlation between the expression of target
cell viral antigens and sensitivity to NK-mediated lysis
(Becker et al., 1976; Kiessling et al., 1978). Roder et
al. (1979) have characterizedpotential target molecules
from an NK-sensitive target cell line, YAC-1. These
molecules were not found to possess determinants in
common with retroviral antigens.
Viral infections induce a number of both specific
and non-specific defense systems. One of the non-
specific defense systems is the synthesis of interferon
which directly interferes with virus replication. Re-
cently, however, the role of interferon as a modulator
of the immune system has been appreciated, partially
because murine NK activity is particularly subject to
augmentation by interferon and interferon inducers
(Gidlund et al., 1978). Viruses, for example lym-
phocytic choriomeningitis virus (LCMV), are among
the most effective inducers of both interferon and
augmented NK activity (Welsh, 1981). We have been
interested in the pathogenesis of members of the
mouse hepatitis virus (MHV) group. The JHM strain
(JHMV) causes acute encephalomyelitis with both
acute and chronic demyelination (Stohlman and
Weiner, 1981). The MHV-3 strain, on the other hand,
causes fulminant hepatitis in susceptible mice and
chronic neurological disease in semi-susceptible hosts
(Virelizier et al., 1975). In this report, we have ex-
amined the ability to these serologically related strains
of MHV to induce NK-mediated cytotoxicity in a sus-
ceptible strain of mice. This is of particular interest
because MHV-3 induces NK activity and interferon,
which plays a role in pathogenesis (Schindler et al.,
1982), while JHMV does not appear to induce inter-
feron (Stohlman et al., 1978). In addition, we have
attempted to correlate the ability of a virus to elicit
NK-mediated cytolysis with the well-established gene-
tic control of host resistance to JHMV (Stohlman and
Frelinger, 1978). Finally, we attempted to correlate
such resistance with the ability to preferentially lyse
JHMV-infected targets.
MATERIAL AND METHODS
Animals
C57BL/6 (B6), C3H and SJL strains were purchased
from Jackson Laboratories, Bar Harbor, ME, at 5
weeks of age. All mice were used between 6 and 8
weeks of age, except for the 12-week-old SJL which
were purchased at 5 weeks of age and held locally.
Viruses
The Armstrong strain of lymphocytic choriomenin-
gitis virus (LCMV) was propagated and plaque assay-
ed in L-2 cell monolayers as previously described (Om
et al., 1982). The DS variant of the JHM (JHMV)
strain of mouse hepatitis virus (MHV) and the type-3
strain of MHV (MHV-3) were propagated and plaque
assayed on DBT cells as previously described (Stohl-
man and Weiner, 1978). Mice were injected with 100
PFU of LCMV and MHV-3 (approximately 100 LD,,)
3Abbreviutions: NK, natural killer; LCMV, lymphocytic
choriomeningitis virus; MHV, mouse hepatitis virus; i.c., in-
tracranial; i.p., intraperitoneal; B6, C57BL/6; JHMV, JHM
strain of MHV; MHV-3, the type-3 MHV.
Received: December 28, 1982.
310 STOHLMAN ET A L

via the intraperitoneal (i.p.) route and with 1,000PFU cells were gently resuspended in a 1:lO dilution of
(approximately 1-5 LD50)of JHMV via the intracra- rabbit serum as a source of C'. This serum was previ-
nial (i.c.) or i.p. routes. Control animals were injected ously shown to have no anti-mouse thymocyte cy-
with an equivalent volume of diluent. Spleen cell sus- totoxic activity. The plates were subsequently incu-
pensions for cytotoxicity assays were prepared from bated for another 30 min at 37°C and centrifuged at
animals 72 h post infection. 300 g for 10 min, then the supernatants were aspi-
rated. The pelleted cells were resuspended in 0.10 ml
Natural killer cell cytotoxicity of medium and an equal volume containing lo4 W r -
YAC-1 cells were grown in RPMI 1640 supple- labelled YAC-1 cells was added for an effector:target
mented with 10% fetal calf serum, 10mM glutamine, ratio of 5O:l. Following 4 h at 37°C the cells were
1 0 m HEPES buffer, and 100 pg/ml kanamycin (com- centrifuged, and the percentage YAC-1 cytolysis was
plete RPMI). Before use, the cells were washed twice determined as described above.
in complete RPMI, resuspended to l o7 cells/ml and
labelled with 100 @i of N a W r (1 mCi/ml; New En- Virus-infected targets
gland Nuclear, Boston, MA) for 1h at 37°C. The cells The effect of NK cells on JHMV-infected targets
were washed twice, allowed to rest on ice for 1 h, was examined using DBT cells and the F-10 passage
washed twice more and resuspended to 5 X 105/mlfor of B16 melanoma cells (Stohlman and Weiner, 1978;
use. Single-cell suspensions were pooled from three to Fidler, 1975). Cells were propagated in suspension
five spleens, washed thrice, and resuspended in com- culture, prepared for use as targets, and assayed as
plete RPMI. Spleen cells and 5 X lo4 51Cr-labelled described above except that one portion of the cells
YAC-1 cells were mixed in triplicate or quadruplicate
in 96-well "V" bottomed plates (Linbro ScientificCo.,
Hamden, CT) at various effector to target ratios in a TABLE I - NATURAL KILLER ACMVlTY OF INFECTED
AND UNINFECI'ED MICE
total volume of 0.2 ml and incubated for 4 h at 37°C.
The plates were centrifuged at 300 g for 1min and 100 Percentage Interferon
,uJ of each supernatant removed. Total releasable Strain Infection E T ratio' cytotoxicity U/ml
counts were determined by the addition of Triton
XlOO to a final concentration of 0.05%. Percentage C3H None 100 63.551.7
cytotoxicity was determined using the following for- 50 56.5k2.4
mula: 25 42.7k1.5
12.5 23.0k1.4
Percentage of cytotoxicity =
B6 None 100 25.8k0.5 524
sample cpm - spontaneous release cpm 50 26.8k1.0
x 100 25 26.0k2.2
TXlOO lysate - spontaneous release 12.5 11.5k1.7
Results are presented as the mean 51Cr release k LCM 100 82.3k2.6 256
SD. 50 78.8k3.3
25 76.5k1.7
Interferon assay 12.5 58.823.2
To determine the interferon level, spleens were JHMV3 100 47.7 k1.6 12
placed in complete RPMI at the ratio of 1.0 ml of 50 44.2k4.2
medium per spleen. The spleens were teased apart 25 28.052.1
and a cell-free supernatant was prepared by centrifu- SJL-6 weeks None 100 1.4k0.5 12
gation at 1,000 g for 10 min. These supernatants were 50 1.6k0.2
frozen at -70°C until assayed while the cell pellet was 25 1.4k0.3
resuspended and processed as described above for NK 12.5 1.2k0.6
effectors. Interferon preparations were assayed on LCM 100 48.6k4.2 256
mouse L929 cells in 0.2-ml volumes in Falcon microt- 50 35.0k1.9
est I1 plates. The cells were challenged with 100 25 23.2k0.6
TCID,,, of vesicular stomatitis virus and the interferon 12.5 14.220.2
titer determined as described by Sakaguchi et al. JHMV3 100 1.150.7 12
(1982). In this system, 1 unit of N1 AID reference 50 1.4k0.2
mouse interferon has a titer of 1-3 U. 25 0.7k0.4
SJL-12 weeks None 100 1.2k0.4 52
Antibody cytolysis of spleen cells 50 0.6k0.4
A two-step complement ((2)-dependent lysis of 25 0.7k0.3
spleen cells was used. Equal volumes (0.01 ml) of anti- 12.5 0.3k0.3
serum and 5 X lo5 spleen cells were mixed in wells of LCM2 100 30.0k1.4 256
a 96-well plate and incubated at 37°C for 30\min. 50 20.8k 1.3
Monoclonal anti-Thy 1.2 (F7D5), anti-Qa-4, and anti- 25 12.5k0.6
Qa-5 were generously supplied by Dr. Edward Clark, 12.5 7.7k0.5
Genetic Systems Corporation, Seattle, WA. The anti- JHMV 100 1.5k1.0 1 2
Qa-4 and Qa-5 reagents have been described previ- 50 0.8k0.4
ously (Koo et al., 1980). The anti-Thy 1.2 serum was 25 0.950.7
supplied by Dr. Jeffery Frelinger of this institution.
Following incubation, the plate was centrifuged at 300 'Effector to target ratio. - '100 LDwinoculated i.p. 72 h prior to assay. -
'1,ooO PFU inoculated i.c. 72 b prior to assay. No detectable interferon. -
g for 10 min at 4°C and the supernatant aspirated. The 'No detectable interferon.
 NK CELLS IN MOUSE HEPATITIS
was infected with JHMV at a multiplicity of infection
of 2-5 during the Wr-labelling period. Both cell lines
express viral cell surface antigens at 5 h post infection
and destruction of a monolayer prepared in this man-
ner is complete at 12-13 h post infection.
RESULTS
Natural killer response during J H W infection
The endogenous NK activity of splenic lymphocytes
from three inbred strains of mice was examined. Table
I shows that, using the YAC-I cell line as target, the
activity oflymphocytesfrom these strains comesponds
to the designations “high” (C3H), “intermediate”
@@, and ‘‘10W’’ (SJL) (Herberman et a1.p 1977).
Table I also shows that both strains tested, €36 and
SJL, exhibit significantly increased NK activity 3 days
after LCMV infection and that this increase COlTeS-
Ponds to the Presence of in~e.rfemnin the spleen.
LYmPhocfies from K2-Week-old SJL mice were also
augmented following LCMV infection; however, the
level of the increased response was lower than that
found in identically treated 6-week-old SJL mice.
TABLE n - NATURALC~TOTOXICIT~ OF B6 FOLLOWING
DEPLETION BY SPECIFIC ANTlSERA
Treatment
Percentage cytotoxicity’
Infected mice2 Ufifectd mice
None 54.4
Complement (C‘) only 55.6
Anti-Thy 1.2 sera C‘ + 57.8
Anti-Thy 1.2 monoclonal + C‘ 33.7
Anti-Qa-4 c‘ + 44.7
Anti-Qa-5 C‘ + 17.5
lEffector to target ratio = 50:1, - 2~~ inoculated with 1,000 pFU
JHMV i.p. 72 h prior to assay. - ’ND = not tested.
We have previouS’y reported that interferon is not
induced by JHMV9a neurotropic strain Of
(Stohlman et a1*7 1978)* We therefore examined fhe
ability of this virus to elicit an NK response. NK activi-
ty was examined after infection of both JHMV-sus-
ceptible mice (B6 and 6-week-old sJL) and JHMV-
resistant mice (12-week-old SJL) (Stohlman et al.,
1980; Stohlman and Weiner, 1981). Table I shows a
significant increase in NK activity in the pooled B6
spleen cell population from mice inoculated i.c. with
JHMV when compared to uninfected controls and
that there was not detectable interferon associated
with the increased response. In addition, no interferon
311
was detected following treatment at pH 2.0 to inacti-
vate any residual virus. This is similar to the in vitro
results we previously reported (Stohlman et al., 1978).
This indicates that residual JHMV is not inhibiting
cellular RNA and protein synthesis during the assay
(data not shown). No response was detectable in SJL
mice infected with JHMV regardless of age. In addi-
tion, no NK activity was found in SJL mice tested 2,4,
5 and 6 days post infection (data not shown). Thus,
following JHMV infection, elevated NK activity did
not correlate with either detectable interferon produc-
tion or JHMV susceptibility.
To ensure that the cytotoxic cells elicited in B6 mice
by JHMv infection in the absence of detectable in-
terferon possessed a cell surface phenotype charac-
teristic of NK cells, splenic effector cells were de-
pleted using antisera or monoclonal antibodies di-
rected against Thy 1.2, Qa-4, and Qa-5 as described
by Clark et ul. (1981). Table I1 shows the results of
such an experiment. The anti-Thy 1.2 antiserum had
no effect on the cytotoxic activity while depletion us-
ing the anti-my 1.2 monoclonal F7D5 plus C‘ reduced
the cytotoxicity by approximately 39 %. Treatment
with anti-Qa-4 monoclonal antibody plus C‘ slightly
decreased the cytotoxicity while treatment with an
anti-Qa-5 monoclonal antibody plus C‘ eliminated ap-
proximately 70 % of the JHMV-induced response.
This pattern is consistent with the previously de-
scribed reactivity for these antisera (Clark et d., 1981)
and indicates that the cytotoxic effectors are indeed
14.6 NK cells.
13.1
17.6 Interferon response to M H V infection
ND To ensure that JHMV did not induce a highly vari-
ND
ND able response and that a low but detectable amount of
interferon was not masked by pooling, the NK re-
sponse and interferon synthesis in individual B6 mice
infected with JHMV were tested. At an E:T ratio of
50: 1, the cytotoxicity for the five infected mice ranged
from 28.5 to 68.8 %, with an average of 48.5 % (Table
111). The values of cytotoxicity for the sham-infected
mice ranged from 11.8% to 34.3% with an average
cflotoxicity of 23.4%. No interferon was detected in
the spleens of any of the individual mice, which indi-
cates that pooling the spleen-cell suspensions had not
diluted a minimally detectable interferon response to
an response’
Our inability to detect interferon in the spleens of
B6 mice concomitant with enhanced NK activity fol-
lowing JHMV infection prompted tests with MHV-3,
a strain that induces both interferon and NK cells
(Schindler et al., 1982). Table IV shows that MHV-3,

TABLE m - NATURAL KILLER ACTIVITY IN INDIVIDUAL MICE FOLLOWING JHMV INFECTION

~~ ~

Percentage cytotoxicity
Infected mice ’ Uninfected mice
E:T ratio2 1 2 3 4 5 Average 1 2 3 4 Average
~ ~ ~~~ ~

50:l 32.3 67.8 45.3 68.8 28.5 48.5 11.8 26.3 21.5 34.3 23.4
25: 1 9.8 37.8 20.7 60.5 8.5 27.5 6.8 0.0 11.5 11.8 7.5
12.5:l 3.8 25.8 12.7 46.3 5.8 18.8 1.0 9.5 6.0 2.5 4.8
Interferon 523 52 52 52 52 52 52 52 12
‘Mice inoculated with 1,ooO PFU i.c. 72 h prior to assay. - 2Effector to target ratio equal to 5 0 1 . - ’No interferon detectable.
312 S T O H L W ET AL.

TABLE N - NATURAL KILLER A(JTIVlTY FOLLOWING INFECTION WITH MHV-3

Route of Expriment I Experiment I1 Experiment 111

Mouse strain V i S
% cvtotoxicitv' Intederon % cvtotoxicitv Interferon % cvtotoxicitv Interferon

B6 None 21.2k1.1 S24 13.8f1.3 52' 30.3k2.1 52'

MHV-3' i.p. 67.5 f 1.3 16 47.8k2.5 16 70.352.1 32
JHMV3 1.c. 47.8f1.6 52 42.5k2.4 52 61.854.6 52
i.p. ND ND 33.0f1.6 52 65.3f2.2 52
SJL None 0.6f0.3 12 1.0k0.2 52 ND ND
MHV-3 ' i.p. 2.5f0.8 52 3.150.4 52 ND ND
JHMV3 1.p. ND ND 2.720.2 12 ND ND
IYAC-1 targets, effector to target ratio = 501. - 2Mice infected with 100 PFU 72 h prior to assay. -'Mice infected with 1,ooO PFU 72 h prior to
assay. -'No interferon detectable. - 5ND= Not detennined.

an hepatotropic strain of MI-IV, induces both NK ac-
tivity and detectable interferon in B6 mice. M W - 3
induced little or no detectable NK response or inter-
feron in SJL following i.p. infection, and this resem-
bles the results following infection with JHMV (Table
I). Table IV also shows that i.p. infection of JHMV
elicits a vigorous NK response without detectable in-
and the minimal cytotoxicity was only demonstrable
during a short-term assay. Lymphocytes from both
susceptible 6-week-old and resistant 12-week-old SJL
mice failed to lyse either target, irrespective of
whether the target was infected or not. The increased
spontaneous release from the infected cell lines paral-
lels the appearance of virally induced cytopathology.

TABLE V - NATURAL KILLER ACTIVTlT AGAINST JHMV-INFECED AND UNINFECTED TARGETS'

DBT targets2 F-10 targets2

Hours post infection Mouse strain
Control Infected Control Infected

4 B6 27.7k3.2 27.0f1.0 3.7f 0.6 3.0kO.O

C3H 20.3f1.2 15.7+ 1.2 1.0+ 1.0 2.7k1.5
SJL-6 weeks 03 1.7f0.6 2.7k 0.6 2.0f2.0
SJL-12 weeks 0 0 2.0k 2.0 3.3f2.1
7 B6 31.750.6 31.7k2.3 0 0
C3H 23.3f2.1 24.0k1.7 0 0
SJG6 1.320.6 1.750.6 0 0
SJL-12 weeks 1.3f1.5 0 0 0
10 86 21.7k3.2 24.0k1.7 0 0
C3H 5.3f2.3 6.3k2.3 0 0
SJL-6 weeks 0 0 0 0
SJL-12 weeks 0 0 0 0
% Spontaneous release 4 h 16.0f1.7 13.0kO.O 12.7k 0.6 11.350.6
7h 18.3f4.2 19.7f0.6 31.7k12.9 39.0k3.6
10 h 25.5 f 3 . 5 41.0k3.0 30.3k15.3 62.0f8.9
IEffector to target ratio = 1OO:l.- 2Percentagecytotoxicity. - 'Release of 5'Cr less than or equal to spontaneous release.

terferon, indicating that the route of viral infection is DISCUSSION

not critical in eliciting an augmented NK response in
the absence of a detectable interferon response.
N K activity against infected targets
Pooled spleen cells from C3H, B6, and both 6- and
12-week-old SJL mice were tested for their ability to
lyse two target cells infected with JHMV. Viral cell-
surface antigen is expressed at approximately 5 h post
infection in both these cell lines and complete destruc-
tion of the cell sheet occurs by 12 h post infection.
Table V shows that splenic lymphocytes from both
C3H and B6 mice lyse DBT target cells and that this
cell line was moderately susceptible to spontaneous
cytotoxicity. However, there was no differential kill-
ing of JHMV-infected over uninfected targets. Lym-
phocytes from C3H and B6 mice showed only mini-
mally detectable levels of cytotoxicity against the F-10
cell line. Again there was no difference between their
ability to lyse JHMV-infected versus uninfected cells,
In the past few years, it has become apparent that
NK activity is greatly augmented during viral infection
(Welsh, 1981), including augmentation during MHV
infection (Herberman et al., 1977; Schindler et al.,
1982). Non-specific cellular cytotoxicity following vi-
ral infection was first ascribed to non-specific T-cell
activation or cytolytic macrophages (Blanden and
Gardner, 1976; Rodda and White, 1976); it is now
clear, however, that this activity is due to NK cells
(Welsh, 1981). It appears prior to the virus-specific T-
cell response and is generally believed to be due to the
ability of virally induced interferon to increase NK
activity. It has been reported that NK-cell activity was
not increased over control values during Newcastle
disease virus infection in mice treated with anti-inter-
feron antibody (Gidlund et al., 1978) and that mice
chronically infected with LCMV show no augmenta-
tion of NK cell activity or interferon (Hansson et al.,
 NK CELLS IN MOUSE HEPATITIS
1980). However, it has recently been demonstrated
that NK activity can be augmented by lectins in the
absence of a detectable interferon response (Brunda et
al., 1982). To gain a clearer understanding of the role
of NK cells during viral infection, we have examined
the role of NK cell augmentation in the pathogenesis
of JHMV infection, a virus that does not induce in-
terferon in vitro (Stohlman et al., 1978). We first con-
firmed that there is an augmentation of NK activity
during LCMV infection that can be demonstrated us-
ing the YAC-I target and that this augmentation is
concomitant with a detectable amount of interferon in
the spleens of infected mice. Augmentation occurs in
SJL mice even though they express very low to unde-
tectable levels of endogenous NK activity at E:T ratios
of less than 1OO:l. This probably results from the pre-
sence of suppressor cells (Riccardi et al., 1981).
The data presented show that the ability to exhibit
augmented NK cytotoxicity does not correspond with
resistance to viral infection. Both B6 and 6-week-old
SJL are susceptible to JHMV, while 12-week-old SJL
are resistant (Stohlman et al., 1980; Stohlman and
Weiner, 1981). Although B6 mice respond with aug-
mented NK activity, neither age group of SJL mice
was responsive following JHMV infection. Recently,
it has been shown that there was an inverse correlation
between host resistance and both the interferon and
NK responses of mice to MHV-3 infection (Schindler
et al., 1982). Both the data of Schindler et al. (1982)
and the data presented here are in contrast to reports
that resistance to herpes viruses correlates with NK
cell activity (Armerding and Rossiter, 1981; Bancroft
et al., 1981).
Infection with JHMV resulted in the augmentation
of NK activity in B6 mice in the absence of a detect-
able interferon response. Although our method of de-
tecting interferon has excellent sensitivity, it was pos-
sible that individual animals responded with a mini-
mally detectable interferon response which was
masked by dilution in the pooled spleen preparations
due to unresponsive individuals. To exclude this possi-
bility, we examined the NK activity and interferon
levels in individual mice. No interferon response was
detected in the individual animals tested even though
the average NK activity of both the infected and sham-
infected groups approximated the data obtained from
the pooled samples.
The augmentation of NK activity by two strains of
MHV, one that has been reported to induce interferon
and NK activity (MHV-3), and one that does not in-
duce interferon in vitro (JHMV), was compared. In
three separate experiments, we found that MHV-3 in-
REFERENCES
ARMERDING, D., and R O S S ~ RH.,
, Induction of natural killer
cells by herpes-simplex virus type 2 in resistant and sensitive in-
bred mouse strains. Immunobiology, 158, 369-379 (1981).
B A N C R OG.,~ , SHELLAM, G., and CHALMER, J . , Genetic influ-
ences on the augmentation of natural killer (NK) cells during
murine-cytomegalovirus infection: correlation with patterns of
resistance. J. Immunol., 126,988-994 (1981).
BECKER, S., FENYO,E.M., and KLEIN,E., The “natural killer”
cell in the mouse does not require H-2 homology and is not
directed against type or group-specificantigens of murine C viral
proteins. Europ. J. Immunol., 6, 882-885 (1976).
313
duced interferon and augmented NK activity. JHMV
infection was less efficient at augmenting NK activity,
but still resulted in clearly enhanced activity in the
absence of interferon. These results confirm that there
is augmented NK activity in mice infected with a virus
that elicits an interferon response (Welsh, 1981).
However, it is clear from the data that NK cell activity
can also be augmented during infection that does not
elicit a detectable interferon response and that this
result is not dependent upon the route of infection.
One mechanism whereby NK cells could play a role
in resistance to viral infection would be the recogni-
tion of virus-infected cells in deference to uninfected
cells. In general, virus-infected cells are not more sen-
sitive to NK cell lysis than uninfected cells (Santoli et
al., 1978) and, in some cases, have been found to be
more resistant to cytolysis than uninfected cells (San-
toli et al., 1978; Welsh and Hallenbeck, 1980). It has
been suggested, however, that MHV-infected targets
have increased sensitivity to NK-cell-mediated cytoly-
sis and, therefore, that NK cells may have a direct
effector function during MHV infections (Welsh,
1981). To test this hypothesis we compared high,
medium, and low NK responder mouse strains for
their ability to lyse infected and uninfected cells. In
addition, we tested lymphocytes from SJL mice at two
ages, of which one is susceptible to JHMV, while the
other is resistant (Stohlman et al., 1980). Although
only two target cell lines were tested, there was no
preferential lysis of infected cells over control cells by
splenic lymphocytes.
The data in this paper contain two interesting obser-
vations. First, it is clear that normal splenic lympho-
cytes do not have the capacity to preferentially lyse
JHMV-infected targets when compared to uninfected
targets. This finding held true irrespective of whether
the NK cells were derived from susceptible or from
resistant mice. Second, it is clear that JHMV is able to
augment NK activity in the absence of a detectable
interferon response, independent of the route of infec-
tion. This may indicate that the augmentation of NK
activity following viral infection can proceed without
obligate interferon induction. However, these studies
indicates that NK activity does not play a major role in
host survival following JHMV infection.
ACKNOWLEDGEMENTS
We wish to thank Mr. Raymond L. Mitchell I1 for
his excellent assistance with the manuscript. This work
was supported in part by grants NS 07149, NS 12967,
NS 15079 and A1 3874 from the National Institutes of
Health.
BLANDEN, R.V., and GARDNER, J., The cell-mediated immune
response to ectromelia virus infection. I. Kinetic and characteri-
zation of the primary effector T-cell response in vivo. Cell. Im-
munol., 22, 271-282 (1976).
BRUNDA, M., VARESIO, L., HERFIERMAN, R., and HOLDEN, H.,
Interferon-independent, lectin-induced augmentation of murine
natural killer cell activity. In?. J. Cancer, 29. 299-307 (1982).
CHEEVER, M.A., GREENBERG,P.D., and FEFER,A , , Therapy of
leukemia by nonimmune syngeneic spleen cells. J . Immunol.,
124, 2137-2142 (1980).
314 STOHLMAN ET AL.

CLARK,E.A., ENGLE,D., and WINDSOR, N.T., Immune respon- Om, A., BRAYTON,P., WOODWARD, J., GOODENOW, R., HAR-
siveness of SWJ mice: hyper NK cell activity by NKl Qa 5- MON, R., FRELINGER,
+ J., and HOOD,L., Products of a transferred
cells. J. Immunol., 127, 2391-2395 (1981). H-2Ld gene act as a restriction element for LCMV-specific killer
CLARK,E.A., and HARMON,R.C., Genetic control of natural T cells. Nature (Lond.), 297, 415-417 (1982).
cytotoxicity and hybrid resistance. Advanc. Cancer Res., 31,227- RICCARDI, C., SANTONI, A., BARLOZZARI, T., CESARINI,C., and
285 (1980). HERBERMAN, R.B., Suppression of natural killer (NK) activity by
FIDLER, I., Biological behavior of malignant melanoma cells cor- splenic adherent cells of low NK-reactive mice. Int. J . Cancer, 28,
related to their survival in vivo. Cancer Res., 35,218-224 (1975). 811-818 (1981).
GIDLUND, M., O m , A., WIGZELL,H., SENIK,A., and GRESSER, RODDA,S., and WHITE,D.O., The mononuclear cell in human
I., Enhanced NK cell activity in mice injected with interferon and blood which mediates ADCC to virus infected targets. I. Identifi-
interferon inducers. Nature (Lond.), 273, 759-761 (1978). cation of the population of effector cells. J . Zmmunol., 117,2067-
HANSSON, M., KIESSLING, R., ANDERSON, B., and WELSH,R., 2072 (1976).
Effect of interferon and interferon inducers on the NK sensitivity RODER,J.C., ROSEN,A., FENYO, E.M., and TROY, F.A.,
of normal mouse thymocytes. J . Imrnunol., 125, 2225-2231 Target-effector interaction in the natural killer cell system: isola-
(1980). tion of target structures. Proc. nat. Acad. Sci. (Wash.), 76,1405-
HERBERMAN, R.B., and HOLDEN,H.T., Natural cell-mediated 1409 (1979).
immunity. Advanc. Cancer Res., 27, 305-377 (1978). SAKAGUCHI, A.Y., STEVENSON, D., and GORDON,I., Species
HERBERMAN, R.B., NU”, M.E., and LAVRIN,D.H., Natural specificity of interferon action: a functioning homospecific nu-
cytotoxic reactivity of mouse lymphoid cells against syngeneic cleus is required for induction of antiviral activity in hetero-
and allogeneic tumors. I. Distribution of reactivity and specifici- karyons. Virology, 116,441-453 (1982).
ty. Int. J . Cancer, 16, 216-229 (1975). SANTOLI,D., TRINCHIERI, G., and KOPROWSKI, H., Cell-medi-
HERBERMAN, R.B., NU”, M.E., HOLDEN, H.T., STALL,S., and ated cytotoxicity against virus-infected target cells in humans. 11.
DJEU, J.Y., Augmentation of natural cytotoxic reactivity of Interferon induction and activation of natural killer cells. J . Im-
mouse lymphoid cells against syngeneic and allogeneic target munol., 121, 532-538 (1978).
cells. Int. J . Cancer, 19, 555-564 (1977). SCHINDLER, L., ENGLER,H., and KIRCHNER, H., Activation of
HERBERMAN, R.B., and ORTALDO,J.R., Specificity of NK cells. natural killer cells and induction of interferon after injection of
In: R.B. Herberman (ed.), Natural cell-mediated immunity mouse hepatitis virus type 3 in mice. Infect. Immun., 35,869-873
against tumors, pp. 873-882, Academic Press, New York (1980). (1982)..----I

KASAI,M., LECLERC, J.C., MCVAY-BOUDREAU, L., SHEN,F.W., STOHLMAN, S.A., and FRELINGER, J.A., Resistance to fatal cen-
and CANTOR,H., Direct evidence that natural killer cells in tral nervous system disease by mouse hepatitis virus strain JHM.
nonimmune spleen cell populations prevent tumor growth in I. Genetic analysis. Imrnunogenetics, 6 , 277-281 (1978).
vivo. J. exp. Med., 149, 1260-1264 (1979).
KAWASE, I., URDAL,D.L., BROOKS, C.G., and HENNEY, C.S., STOHLMAN, S.A., FRELINGER, J.A., and WEINER,L.P., Resis-
Selective depletion of NK cell activity in vivo and its effect on the tance to fatal central nervous svstem disease bv mouse heoatitis
growth of NK sensitive and NK resistant tumor cell variants. Int. virus, strain JHM. 11. Adheren; cell mediated protection. 3. Im-
J. Cancer, 29, 567-574 (1982). munol., W, 1733-1739 (1980).
KIESSLING,R., HALLER,O., FENYO, E.M., STEINITZ,M., and STOHLMAN, S., and WEINER,L., Stability of neurotropic mouse
KLEIN,G., Mouse natural killer (NK) cell activity against human hepatitis virus (JHM strain) during chronic infection of neuro-
cell lines is not influenced by superinfection of the target cell with blastoma cells. Arch. Virol., 57, 53-61 (1978).
xenotropic murine C-type virus. Int. J . Cancer, 21, 460-465 STOHLMAN, S.A., and WEINER,L.P., Chronic central nervous
(1978). system demyelination in mice after JHM virus infection. Neurol-
KIESSLING, R., KLEIN,E., and WIGZELL,H., “Natural” killer ogy, 31, 38-44 (1981).
cells in the mouse. I. Cytotoxic cells with specificity for mouse STOHLMAN, S.A., SAKAGUCHI, A.Y., and HITI, A,, Interferon
Moloney leukemia cells. Specificity and distribution according to production and activity in mouse neuroblastoma cells. Arch.
genotype. Europ. J . Immunol., 5 , 112-117 (1975). Virol., 57, 91-96 (1978).
KIESSLING, R., and WIGZELL, H . , An analysis of murine NK cells WELSH,R., Natural cell-mediated immunity during viral infec-
as to structure, function and biological relevance. Immunol. tions. Curr. Top. Microbiol. Immunol., 92, 83-106 (1981).
Rev., 44, 165-208 (1979).
Koo, G.C., JACOBSEN, J.B., HAMMERLING,G.J., and HAMMER- WELSH,R.M., and HALLENFIECK, L.A., Effect of virus infections
LING, V., Antigenic profile of murine natural killer cells. J . Im- on target cell susceptibility to natural killer cell-mediated lysis.
munol.,125, 1003-1006 (1980). J . Immunol., 124, 2491-2497 (1980).
LEE, J.C., and IHLE,J.N., Characterization of the blastogenic VIRELIZIER, J.L., DAYAN, A.D., and ALLISON,A.C., Neuro-
and cytotoxic responses of normal mice to ecotropic C type viral pathological effects of persistent infection of the mouse hepatitis
gp 71. J . Immunol., 18, 928-934 (1977). virus (MHV-3). Infection Immun., 12, 1127-1140 (1975).