Download Role of Cytotoxic T Lymphocytes in Murine Cytomegalovirus Infection

J. gen. Virol. (I98O), 47, 503-508
503
Printed in Great Britain
Role of Cytotoxic T Lymphocytes in Murine
Cytomegalovirus Infection
(Accepted 8 November I979)
SUMMARY
Cell-mediated immunity is important in host control of CMV infection.
A chromium release microcytotoxicity assay was used to evaluate the role of
cytotoxic T lymphocytes (CTL) in murine CMV infection. Within a few days
after intranasal inoculation virus was detected in cultures of buffy-coat, spleens,
anterior cervical lymph nodes and salivary glands. CTL were first detected on
day 5 post-infection in spleen and peripheral blood, and on day 6 in anterior
cervical nodes. The course of the CTL response approximated to that of virus
titres during the acute phase of infection in the spleen and blood. The findings
indicate that CTL are distributed to infected tissues and appear to be important
during the acute, viraemic phase of infection.
Cell-mediated immunity is an important part of the host defence against cytomegalovirus
(CMV) infection. Human patients with suppressed cellular immunity, such as those with
leukaemia, and recipients of renal, cardiac or bone marrow transplants experience frequent, often severe, infection despite their capacity to produce antibody to the virus (Betts &
Hanshaw, I977). The special importance of T lymphocytes is suggested by studies of CMV
infection in nude mice (Starr & Allison, 1977) and of mice treated with anti-lymphocyte
serum (Jordan et al. 1977). We have previously demonstrated a cytotoxic T lymphocyte
(CTL) response after intraperitoneal (i.p.) infection with murine cytomegalovirus (MCMV;
Quinnan et al. T978). Recent work by Jordan provided evidence that natural infection of
mice probably occurs by the upper respiratory route (Jordan, I978). In order to clarify the
role of CTLs in MCMV infection we have studied their distribution in infected tissues after
intranasal infection with MCMV.
The Smith strain (Kim et al. I974) of MCMV was initially given to our laboratory by
Dr D. Henson. Infection of mice was accomplished with an isolate that had been maintained by continuous in vivo passage, while target cells were prepared using an isolate
adapted to tissue culture by ~3 in vitro passages in primary BALB/c mouse embryo cells
(MEC).
Weanling male BALB/c mice (Animal Production Unit, NIH, Bethesda) were inoculated
intranasally under light ether anaesthesia, with 0"o5 ml of a suspension containing I x ro 4
p.f.u, of MCMV. Lymphocyte suspensions were prepared from spleens, anterior cervical
lymph nodes and peripheral blood lymphocytes. Specimens taken for virus culture included
anterior cervical node and spleen cell suspensions, salivary gland homogenates and samples
of buffy-coats from heparinized peripheral blood. All virus cultures were performed by
inoculation of specimens on to monolayer cultures of primary MEC prepared from preterm embryos. After an adsorption period of I h at 37 °C, monolayer cultures inoculated
with burly coat cells were fed with minimum essential medium (MEM) supplemented with
2 % foetal bovine serum (FBS) and observed for 2 weeks for c.p.e. Other virus cultures
were overlaid for plaque assay with MEM supplemented with 5 % FBS and 0"5 % agarose.
After incubation at 37 °C for 72 h in a 5 % CO2 in air atmosphere, a grid was placed on the
bottom of each flask and plaques were counted at 35 × magnification using an inverted
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 20:08:08
504
Short communications
microscope. Target cells for cytotoxicity assays were also prepared from primary BALB/c
MEC, or from passage 8 to Io cultures of MEC from the congeneric mouse strains, BIo. D2
and BIo. BR (Jackson Laboratories, Bar Harbor, Maine).
The capsules of spleens and anterior cervical lymph nodes were disrupted and lymphocytes teased into suspension. Erythrocytes were removed from spleen cell suspensions by
lysis with ammonium chloride buffer (Roos & Loos, I97O). Lymphocytes were separated
from heparinized peripheral blood specimens by purification on Ficoll-hypaque gradients.
After centrifugation, the cells were routinely resuspended in MEM with Io ~ FBS. For some
studies, adherent cells were removed by passage through nylon wool columns (Julius et aL
I973). In other experiments T lymphocytes were removed from cell suspension by lysis with
anti-theta serum (Bionetics, Kensington, Md.) and rabbit complement (Grand Island
Biological Co., Grand Island, N.Y.) as previously described (Quinnan et al. I978).
The procedure used for cytotoxicity assays has been previously described (Quinnan
et al. 1978). For target cells, MEC were infected at i to 4 p.f.u./cell for 24 h, harvested by
trypsinization, labelled with 51Cr-sodium chromate, washed twice, suspended in MEM with
I0~o FBS to a concentration of I x IOs cells/ml and dispersed in volumes ofo.I ml in wells
of round-bottomed microtitre plates (Linbro, Hamden, Conn.). Equal volumes containing
I × Io 6 immune lymphocytes (Ioo: I effector to target ratio) were added for test lysis, control
lymphocytes for spontaneous lysis, or IO~'o Brij-35 solution for maximum lysis determinations. After incubation for I8 h at 37 °C in a 5% CO2 atmosphere, the medium was
harvested using a Titertek Supernatant Collection Apparatus (Flow Laboratories, Rockville,
Md.) and the released chromium measured on a gamma counter. Percentage specific
immune lysis was calculated from the formula:
Test c t / m i n - control ct/min
×
Maximum ct/min-control ct/min
1OO.
Lymphocyte suspensions were prepared from tissues from two to 3° mice. Assays were
performed in replicates of four to eight and results of test and spontaneous lysis were
compared using the Student t-test.
The results of infectivity titrations of spleen cell and lymph node cell suspensions and
salivary gland homogenates are presented in Fig. I. Virus titres in spleen cell suspensions
reached a maximum of 8o p.f.u./o.2 ml of cell suspension (approx. 50 × Io6 spleen cells/ml)
by 6 days after initiation of infection, and no infectious virus was detected by I6 days. This
period corresponded approximately to the duration of viraemia detected by culturing of
buffy-coat cells, which extended from day 4 to day 2I p.i. Salivary gland virus titres did not
reach maximum values until 16 days p.i. and began to decline between days 2i and 3I. Virus
titres in anterior cervical lymph nodes were lower than in salivary glands, but continued to be
elevated until salivary gland titres began to decline. These results thus indicate a viraemic
phase lasting about 3 weeks, during which virus is also cultured from the spleen, and a
chronic phase during which virus was detected locally in the salivary gland and draining
lymph nodes, but not in the blood or spleen.
The results of cytotoxic lymphocyte assays are also shown in Fig. I. Cytotoxic lymphocytes were first detected in spleens on days 4 to 5, produced highest immune lysis values
between days 7 to I o and declined to undetectable levels by day 16. The course of the response
measured in peripheral blood was similar. The CTL response in cervical lymph nodes was
lower in magnitude and was usually detectable for only I or 2 days, between days 6 and 7
p.i. No significant cytotoxic lymphocyte activity was detected in cervical lymph nodes after
that time. In general, the highest specific immune lysis was produced by spleen lymphocytes,
but significant lysis was also produced by lymphocytes from peripheral blood and regional
lymph nodes draining infected
salivary
glands.
Downloaded
from www.microbiologyresearch.org
by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 20:08:08
505
Short c o m m u n i c a t i o n s
1000000
(a)
.....Q...
~.
100000
I
I
e?
10000
\
\\
ii I
1000
\\\
A,
t
/
,
i
a00 ~
35 -
\\
iI
I,~,,./
/ /~'X
\
\
,,
/
\
/
"
\
'o
,,
(b)
30"~
25
20
15-
~.
3
i
I
0
l
6
9
\
z ....
12 15 18 21 24
Time p.i. (days)
.
27 30 33
Fig. I. Virus infectivity titrations expressed in (a) p.f.u./o.2 ml of lymphocyte suspension or salivary
gland homogenate and (b) cytotoxic lymphocyte responses during MCMV infection. (a) • - - - - • ,
Spleen cell suspensions; A - - - ,~, cervical node cell suspensions; [] - - [Z], salivary gland
homogenates. (b) •
0 , Spleen lymphocytes; A - . - A ,
peripheral blood lymphocytes;
D - - - F 1 , cervical node lymphocytes.
Several characteristics of the cell type responsible for the immune lysis were examined
and the results are presented in Table I. The cytotoxic activity of lymphocytes from all three
sources was preserved in the non-adherent cell fraction obtained after passage of cells
through nylon wool columns and eliminated from cell suspensions depleted of T cells by
treatment with anti-theta serum and complement. Lymphocytes from all three sources were
tested for their capacity to lyse MCMV-infected target cells prepared from syngeneic mice
and from the congeneric mouse strains Bio. D2 and Bto. BR, which differ only in their H-2
haplotype (Klein, 1975). Effector lymphocytes produced lysis only of target cells sharing the
same H-2 haplotype. Lysis of syngeneic target cells was generally greater than tysis of target
cells from Bro.D2 mice, but significant lysis of both cell types was found consistently.
The results presented in this study describe lymphocytes which are cytotoxic to MCMVinfected ceils and which are non-adherent, possess theta antigen and are H-2 restricted.
Only T lymphocytes possess all these characteristics. These same features were typical of
CTL which we have described previously in spleens of BALB/c mice infected i.p. with
MCMV (Quinnan et al.Downloaded
r978) andfrom
which
have been described in this
www.microbiologyresearch.org
by laboratory (Ennis et al.
IP: 88.99.165.207
On: Sat, 17 Jun 2017 20:08:08
506
Short communications
Table I. H-2-restricted cytotoxieity of lymphocyte suspension prepared from spleens, peripheral blood and anterior cervical lymph nodes of BALB/c mice after intranasal inoculation
of M C M V
Percent specific immune lysis of MCMV-infected
target cells prepared from:
Source of
lymphocytes*
Spleen
Spleen
Peripheral blood
Peripheral blood
Cervical lymph nodes
Cervical lymph nodes
Spleen
Lymphocyte pre-treatment
Nil
Passage through nylon
column
Nil
Passage through nylon
column
Nil
Passage through nylon
column
Cervical lymph nodes
Cervical lymph nodes
Anti-theta serum and
complement
Complement
Media
Anti-theta serum and
complement
Complement
Media
Anti-theta serum and
complement
Complement
Media
Spleen
Peripheral blood
Cervical lymph nodes
Nil
Nil
Nil
Spleen
Spleen
Peripheral blood
Peripheral blood
Peripheral blood
Cervical lymph nodes
BALB/c MEC
(H-2 a)
BIo.D2 MEC
(H-z d)
B I o . B R MEC
(H-2 ~)
20"9 (< 0.001)'~
25"6 (< o.ooi)
ND +
ND
ND
ND
9'5 (< o'o5)
18"5 (< 0'04)
ND
ND
ND
ND
I6"4 (< o.ooi)
t4'8 (o.oo6)
ND
ND
ND
ND
8'0 (> o ' 0
ND
ND
ZI'O ( < 0'OOI)
26"0 (< o.ooi)
o'0 (> o'I)
ND
ND
ND
ND
ND
ND
2I -8 ( < o.ool)
32'0 (< o'oo0
I'4 (< o'I)
ND
ND
ND
ND
ND
ND
8"4 (0"03o)
I 1.6 (o'oo5)
ND
ND
ND
ND
20-4 (o.oo2)
29"o (< o'oor)
I2.6 (o.ooz)
Ii.8 (< o.ooi)
2o'3 (o'o49)
9"9 (< o.ool)
o-o ( > o.I)
5'r (> oq)
o'o (> o.I)
* Lymphocyte suspensions used for passage through nylon yarn columns, treatment with anti-theta
serum and demonstration of H-2 restriction, respectively, were obtained from spleens at I I, 8 and 8 days,
from peripheral blood at 11, 7 and 6 days and from anterior cervical lymph nodes 7, 7 and 6 days p.i.
t Probability by Student t-test comparing ct/min of test lysis to control lysis.
~: ND = not done.
I977, I978) and by others (Doherty & Zinkernagel, I975) in a variety of murine virus
infections. These results are important since they demonstrate that a CTL response occurs
after infection with a herpesvirus administered by the natural, upper respiratory route, and
that these cells circulate in peripheral blood and are present in regional lymph nodes
draining the acutely infected salivary glands. The findings confirm the presence of CTL in
critical sites during the natural course of infection and are similar to those obtained during
murine influenza infection (Ennis et al. I978).
The results we reported previously (Quinnan et al. I978) demonstrated that the amount
of virus inoculated was proportional to the amount which replicated in the spleen and to the
magnitude of the CTL response. The data included in this report indicate, however, that far
fewer infectious foci are present in the spleen after intranasal inoculation than when the
virus is given intraperitoneally, but that for a given dose the magnitude of the splenic CTL
response is equivalent after infection by either route. In addition, the fraction of lymphocytes which are cytotoxic is similar in the spleen and the peripheral blood, but the total
number oflymphocytes in the spleen is about tenfold greater. It is suggested, therefore, that
the spleen is a source of CTL which may circulate through the peripheral blood to infected
tissues, and that the magnitude of the splenic CTL response is proportional to the amount of
virus producing the initial infection, not the amount of local virus replication.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 20:08:08
Short communications
507
The temporal relationship of the CTL response to the course of infection is also of
interest. Our findings are consistent with those of others which demortstrated that viraemia
begins about 3 to 4 days after the initiation of infection (Olding et al. I975; Wise et al.
~979)- In general, the duration of viraemia depends on the size of the inoculum and on the
strain of mice being studied. In our experiments, virus was recovered from buffy-coat
cultures during the first 3 weeks of infection, approximately the same period in which virus
was detected in spleens. CTL were measured within I to z days of the initiation of viraemia
and also persisted throughout the first 3 weeks. This association suggests that CTL have an
important role during the acute viraemic phase of infection, but that other immune
mechanisms, possibly antibody-mediated, may be more significant during the later, chronic
phase of salivary gland infection.
These experiments thus provide some basis for understanding one aspect of the cellular
immune system, CTL, in MCMV infection. An antiviral CTL response occurs during the
natural course of infection, is proportional to the virus inoculum, can be measured in
peripheral blood and in proximity to infected tissues and may be important in controlling
the acute, viraemic phase of infection. Since CTL cannot be detected after the acute phase of
infection, studies are needed to determine the role of other cellular mechanisms, such as
antibody-dependent cell-mediated cytotoxicity, in controlling chronic infection and in
protective immunity. An important role for CTL in human CMV infection can only be
speculative at present, but seems likely because of the many similarities between human and
murine CMV infection and because of the considerable clinical evidence indicating that
cellular immunity is of major importance in human CMV infection.
Department of Health, Education and Welfare
Food and Drug Administration
Bureau of Biologics, Division of Virology
88oo RockviUe Pike
Bethesda, Md. 2o2o5, U.S.A.
G. V. QUrNNAN
J. E. MANISCHEWITZ
F. A. ENNIS
REFERENCES
BETTS, R. r. & HANSHAW,J. B. (Jt977). Cytomegalovirus (CMV) in the compromised host (s). Annual Review of
Medicine 28, Io3-t IO.
OOHERTY, t'. C. & ZINKERNAGEL,R. M. 0975). H-2 compatibility is required for T-cell-mediated tysis of target
cells infected with lymphocytic cboriomeningitis virus. Journal of Experimental Medicine I4x, 5o2-5o7.
ENNIS, F. A., MARTIN, W . J . , VERBONITZ, M . W . & BUTCHKO, G . M . 0977). Specificity studies on cytotoxic
thymus-derived lymphocytes reactive with influenza virus-infected cells: evidence for dual recognition
of H-2 and viral hemagglutinin antigens. Proceedings of the National Academy of Sciences of the United
States of America 74, 3o06-3o I o.
ENNIS, F. A., VERBONITZ, M. W., BUTCHKO, G. M. & ALBRECHT, P. (I978). Evidence that cytotoxic T-cells are part
of the host response to influenza pneumonia. Journal of Experimental Medicine x48, t24t-I25 o.
JORDAN, M. C. 0978). Interstitial pneumonia and subclinical infection after intranasal inoculation of routine
cytomegalovirus. Infection and Immunity zx, 275-280.
JORDArq, M. C., SHANLEY,J. D. & STEVENS,J. G. 0977). Immunosuppression reactivates and disseminates latent
murine cytomegalovirus. Journal of General Virology 37, 419-423.
JULIUS, M., STMPSON,E. e~ HERZENBEIG, L. A. 0973). A rapid method for the isolation of functional thymusderived lymphocytes. European Journal oflmmunology 3, 645-648.
KIM, K. S., SAPIENZA,V. & CARP, R. I. (1974). Comparative studies of the Smith and Raynaud strains of murine
cytomegalovirus. Infection and Immunity xo, 672-674.
KLEIN, J. 0975). In Biology of the Mouse Histocompatibility- 2 Complex. New York: Springer Verlag.
OLDINO, L. B., JENSEN, r. C. & OLDSTONr, ~a. B. A. (I975). Pathogenesis of cytomegalovirus infection. I. Activation of virus from bone marrow-derived lymphocytes by in vitro allogenic reaction. Journal of
Experimental Medicine x4x, 56t-57z.
QUINNAN, G . V . , MANISCHEWITZ, J . E . & ENNIS, E.A. 0 9 7 8 ) . Cytotoxic T lymphocyte r e s p o n s e t o r o u t i n e
cytomegalovirus infection. Nature, London 273, 541-543.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 20:08:08
508
Short communications
ROOS, D. & LOOS, J. A. 0970). Changes in the carbohydrate metabolism of mitogenically stimulated human
peripheral lymphocytes. Bioehimica et Biophysica Acta 222, 565-582.
STARR, S.E. & ALUSON, A. ¢. 0977). Role of T lymphocytes in recovery from murine cytomegalovirus
infection. Infection and hnmunity I7, 458-462.
WISE, T, G., MANISCHEWITZ, J. ~., QU1NNAN, G. ¥., AULAKH, G. S. & ENNIS, F. A. (1979). Latent cytomegalovirus
infection of BALB/c mouse spleens detected by an explant culture technique. Journal of General
Virology 44, 551-556.
(Received 2o July I979)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 20:08:08