Vol. 41, No.1 INFECTION AND IMMUNITY, JUlY 1983, p. 197-204 0019-9567/83/070197-08$02.00/0 Copyright ( 1983, American Society for Microbiology Recovery from Lethal Herpes Simplex Virus Type 1 Infection Is Mediated by Cytotoxic T Lymphocytes HAL S. LARSEN,'* ROBERT G. RUSSELL,2 AND BARRY T. ROUSE1 Departments of Microbiology1 and Pathobiology,2 College of Veterinary Medicine, University of Tennessee, Knoxville, Tennessee 37996-0845 Received 17 January 1983/Accepted 18 April 1983 The ability of herpes simplex virus (HSV)-specific, cytotoxic T lymphocytes to mediate recovery of mice lethally infected with HSV was examined. Adoptive transfer of splenocytes from mice that had been primed in vivo with HSV and restimulated with HSV in vitro protected lethally infected normal and cyclophos- phamide-immunosuppressed mice from death. In contrast, equal numbers of normal splenocytes or immune splenocytes cultured without antigen failed to mediate recovery. Recovery was also transferred by noncultured, primary immune splenocytes, although the protective efficacy of these cells was 10-fold less than when immune splenocytes restimulated in vitro were used. Treatment of the cells with anti-Thy 1 or anti-Lyt 2.1 plus complement before adoptive transfer abrogated recovery. No decrease in protection was seen when in vitro-restimulat- ed splenocytes were treated with anti-Lyt 1.1 plus complement. Splenocytes expression natural killer activity also failed to effect recovery.The administration of hyperimmune anti-HSV antibody to normal, immunocompetent mice resulted in recovery, whereas no significant protection of immunosuppressed mice by anti- HSV was observed. When antibody was given concurrently with cultured immune splenocytes, the percentage of mice that recovered from infection was greater than that seen with either antibody or cultured immune splenocytes alone. These experiments demonstrate that Lyt 2-positive cells with cytotoxic activity generat- ed by in vitro immunization can mediate recovery from lethal HSV infection, whereas, under the conditions chosen, Lyt 1-positive cells were unable to mediate recovery. The immunological nature of in vivo resist- ance mechanisms to herpes simplex virus (HSV) remains to be clarified as to the respective importance of the many components of adaptive immunity that can be measured and shown spe- cifically reactive to HSV antigens in vitro. Prob- ably multiple mechanisms can operate, this ac- counting in part for opposing viewpoints on the relative protective role of one or another compo- nent. Furthermore, in discussing the protective function of various immune mechanisms, it is important to specify which stage of virus-host interaction is involved. Thus, there seems little doubt that antibody, prepared against either the whole virus (13, 22, 23, 26) or several of the glycoproteins of the envelope (2, 5), can protect against primary infection providing that the anti- bodies are present in sufficient concentrations at the site of challenge. Such antibody probably prevents attachment of virus to target cells, although the exact mechanism in vivo has not been elucidated. Usually protection against pri- mary infections is accomplished by natural resistance mechanisms rather than by adaptive immunity. Such natural resistance mechanisms against HSV include natural killer cells (1, 7, 27), interferon (7, 9, 27), and macrophages (8, 10, 14, 15, 25, 27, 28). Worthington et al. (26) supported the role of antibody in recovery from HSV infection and suggested that any protection mediated by adop- tive transfer of immune cells could be attributed to the production of antibody by the transferred cell population. This is probably an oversimpli- fication, since several observations support a role for T cells in protection and recovery. Nash et al. (20) have clearly shown protection is mediated by T cells, but they favor a principal role for the I region-restricted Lyt 1-positive, Lyt 2-negative (Lyt 1+2-) delayed-type hyper- sensitivity-mediating T cell, rather than T cell subsets that are cytotoxic. A principal role for Lyt 1+2- T cells was also observed by Nagafu- chi et al. (18), but this was in a model of protection rather than recovery. The purpose of our research was to investigate the role of cyto- toxic T cells in mediating recovery as measured by survival of lethally infected mice. The results 197 on May 29, 2018 by guest http://iai.asm.org/ Downloaded from
Transcript
Vol. 41, No.1INFECTION AND IMMUNITY, JUlY 1983, p. 197-2040019-9567/83/070197-08$02.00/0Copyright ( 1983, American Society for Microbiology
Recovery from Lethal Herpes Simplex Virus Type 1 InfectionIs Mediated by Cytotoxic T LymphocytesHAL S. LARSEN,'* ROBERT G. RUSSELL,2 AND BARRY T. ROUSE1
Departments of Microbiology1 and Pathobiology,2 College of Veterinary Medicine, University of Tennessee,Knoxville, Tennessee 37996-0845
Received 17 January 1983/Accepted 18 April 1983
The ability of herpes simplex virus (HSV)-specific, cytotoxic T lymphocytes tomediate recovery of mice lethally infected with HSV was examined. Adoptivetransfer of splenocytes from mice that had been primed in vivo with HSV andrestimulated with HSV in vitro protected lethally infected normal and cyclophos-phamide-immunosuppressed mice from death. In contrast, equal numbers ofnormal splenocytes or immune splenocytes cultured without antigen failed tomediate recovery. Recovery was also transferred by noncultured, primaryimmune splenocytes, although the protective efficacy of these cells was 10-foldless than when immune splenocytes restimulated in vitro were used. Treatment ofthe cells with anti-Thy 1 or anti-Lyt 2.1 plus complement before adoptive transferabrogated recovery. No decrease in protection was seen when in vitro-restimulat-ed splenocytes were treated with anti-Lyt 1.1 plus complement. Splenocytesexpression natural killer activity also failed to effect recovery.The administrationof hyperimmune anti-HSV antibody to normal, immunocompetent mice resultedin recovery, whereas no significant protection of immunosuppressed mice by anti-HSV was observed. When antibody was given concurrently with cultured immunesplenocytes, the percentage of mice that recovered from infection was greaterthan that seen with either antibody or cultured immune splenocytes alone. Theseexperiments demonstrate that Lyt 2-positive cells with cytotoxic activity generat-ed by in vitro immunization can mediate recovery from lethal HSV infection,whereas, under the conditions chosen, Lyt 1-positive cells were unable to mediaterecovery.
The immunological nature of in vivo resist-ance mechanisms to herpes simplex virus (HSV)remains to be clarified as to the respectiveimportance of the many components of adaptiveimmunity that can be measured and shown spe-cifically reactive to HSV antigens in vitro. Prob-ably multiple mechanisms can operate, this ac-counting in part for opposing viewpoints on therelative protective role of one or another compo-nent. Furthermore, in discussing the protectivefunction of various immune mechanisms, it isimportant to specify which stage of virus-hostinteraction is involved. Thus, there seems littledoubt that antibody, prepared against either thewhole virus (13, 22, 23, 26) or several of theglycoproteins of the envelope (2, 5), can protectagainst primary infection providing that the anti-bodies are present in sufficient concentrations atthe site of challenge. Such antibody probablyprevents attachment of virus to target cells,although the exact mechanism in vivo has notbeen elucidated. Usually protection against pri-mary infections is accomplished by naturalresistance mechanisms rather than by adaptive
immunity. Such natural resistance mechanismsagainst HSV include natural killer cells (1, 7,27), interferon (7, 9, 27), and macrophages (8,10, 14, 15, 25, 27, 28).Worthington et al. (26) supported the role of
antibody in recovery from HSV infection andsuggested that any protection mediated by adop-tive transfer of immune cells could be attributedto the production of antibody by the transferredcell population. This is probably an oversimpli-fication, since several observations support arole for T cells in protection and recovery. Nashet al. (20) have clearly shown protection ismediated by T cells, but they favor a principalrole for the I region-restricted Lyt 1-positive,Lyt 2-negative (Lyt 1+2-) delayed-type hyper-sensitivity-mediating T cell, rather than T cellsubsets that are cytotoxic. A principal role forLyt 1+2- T cells was also observed by Nagafu-chi et al. (18), but this was in a model ofprotection rather than recovery. The purpose ofour research was to investigate the role of cyto-toxic T cells in mediating recovery as measuredby survival of lethally infected mice. The results
demonstrate that Lyt 2+ cells with cytotoxicactivity generated by in vitro immunization canmediate recovery from infection. Under the ex-perimental conditions chosen, Lyt 1+ cells wereunable to mediate recovery. Our studies serve toemphasize that there appear to be many alterna-tive mechanisms of host resistance to HSVinfection.
MATERIALS AND METHODSVirus. An oral isolate of HSV type 1 (HSV-1) (BK
strain) was used to infect mice in adoptive transferexperiments. HSV-1 KOS was used to immunize micefor the production of immune spleen cells. Each strainwas grown in HEp-2 cells by infecting at low multiplic-ity and harvesting cell-associated virus as previouslydescribed (3).Mouse immunization and preparation of splenocytes.
C3H/HeJ mice were obtained from the University ofTennessee Medical Research Center. Mice 6 to 9weeks of age were used as recipients in the adoptivetransfer studies. Animals serving as cell donors wereinfected intraperitoneally (i.p.) with 0.1-ml inoculacontaining 106 PFU of infectious HSV-1 KOS. At least4 weeks after infection, the mice were killed bycervical dislocation, and their spleens were removedaseptically for preparation of single-cell suspensionsas described elsewhere (12). In those cases wheretransfer of restimulated cells was desired, spleen cellswere adjusted to 2 x 106/ml in RPMI 1640 containing10% fetal calf serum, 2 mM glutamine, penicillin (100U/mi), streptomycin (100,ug/ml), and 5 x 10-5 M 2-mercaptoethanol and added to plastic tissue cultureplates at a density of 106/cm2 of surface area. UV light-inactivated virus (107 PFU) then was added, and thespleen cell cultures were incubated in a humidifiedatmosphere of 5% CO2 at 37°C. After 5 days, the cellswere harvested and washed in medium, and the celldensity was adjusted to the desired concentration.
Detection of B lymphocytes. B lymphocytes weredetected with fluorescein-conjugated anti-mouseimmunoglobulins (Cappel Laboratories, West Ches-ter, Pa.). Briefly, the 5-day restimulated immunespleen cells were pelleted, and 0.1 ml of fluoresceinat-ed antiserum (diluted 1:30) was added. After 30 min at4°C, the cells were washed twice in phosphate-buff-ered saline and suspended in a mixture of 1 drop ofphosphate-buffered saline and 2 drops of45% bufferedglycerol. The cell suspensions were then examinedwith a Nikon Photovolt fluorescent microscope.
Cytotoxicity assay. The assay for cytotoxic T lym-phocytes (CTL) was carried out as previously de-scribed (12). Splenocytes were cultured for 5 days invitro with 107 PFU of UV-irradiated HSV-1 to inducesecondary CTL. Cells were assayed by using syngene-ic and allogeneic virus-infected and uninfected targetcells at effector cell/target cell ratios of 25:1 and 5:1,and the results are reported as percent specific lysis.
Preparation and assay of hyperimmune serum. Rab-bit anti-HSV-1 hyperimmune serum was prepared byscarification of the cornea, followed by inoculation ofinfectious HSV-1 KOS. The process was repeatedafter 10 days, and the rabbits were bled by cardiacpuncture 10 days after the booster immunization. Serawere measured for neutralizing antibody by a plaque
'reduction method. Sera and virus dilutions were madein McCoy medium without fetal calf serum. Approxi-mately 100 PFU of HSV-1 was added to equal volumesof twofold dilutions of the serum. The mixtures wereincubated at 37°C for 1 h and then assayed for infec-tious virus. Each dilution was assayed in triplicate.Controls contained sera from nonimmunized animals.The antibody titer was expressed as the reciprocal ofthe highest dilution of serum that caused a 50%reduction compared with controls.
Cell fractionation. Splenocytes from immune miceor splenocytes cultured in vitro were harvested andwashed twice, and the viable cells were isolated bycentrifugation with Lympholyte-M (Accurate Chemi-cal and Scientific Corp., Westbury, N.Y.) densityseparation medium. The banded cells, or those treatedwith antisera and complement to remove selected cellpopulations, were adoptively transferred into infectedrecipients. Antibody treatment was accomplished byreacting 100 x 106 spleen cells wit 7 ml of eithermonoclonal anti-mouse Lyt 1.1, 2.1, or Thy 1.2 (Accu-rate) diluted 1:30 for 1 h at 4°C with occasional mixing.The cells were then washed, and 7 ml of a 1:8 dilutionof Low-Tox-M rabbit complement (Accurate) wasadded. The mixture was incubated for 1 h at 37°C withoccasional mixing. After complement treatment, thecells were washed and adjusted to appropriate num-bers for injection.
Induction of NK cells. Splenocytes expressing natu-ral killer (NK) activity were produced by intravenousinjection of 107 PFU of HSV-1 KOS into nonimmunemice. At 48 h after injection the animals were sacri-ficed, spleens were removed, and a single-cell suspen-sion was prepared for adoptive transfer and assay of invitro cytotoxicity.Cyclophosphamide treatment, virus inoculation, and
adoptive transfer of cells or antibody. In experimentswhere immunosuppression was desired, mice wereinitially injected i.p. with 200 mg of cyclophosphamide(Sigma Chemical Co., St. Louis, Mo.) per kg. Viruswas inoculated i.p. 24 h after cyclophosphamide treat-ment, and cells or antibody was given after an addi-tional 24 h. Syngeneic cells were injected intravenous-ly, whereas antibody was given i.p. When normal,immunocompetent mice were utilized, the experimen-tal protocol was as above, with the exception thatcyclophosphamide was not given. Mice were observeddaily for 21 days, and the number of deaths each daywas recorded. Statistical analysis was done by the chi-square method.
RESULTSRecovery of normal mice from HSV-1 infection.
Experiments were performed to assess the effectof adoptively transferred cell populations oranti-HSV-1 antibody on lethal infection in non-immunosuppressed mice (Table 1). Immunespleen cells cultured in vitro for 5 days with UV-inactivated HSV-1 provided significant protec-tion from death when 5 x 106 or more cells wereadministered to mice 24 h after HSV-1 infection.This protection was dose dependent and was notseen when fewer than 5 x 106 cultured immunecells were given. In contrast, no protection was
TABLE 1. Recovery of mice from lethal HSV-1 infection by inoculation of HSV-1-primed spleen cellsrestimulated in vitroa
No. of miceCells transferred % Survival
Total Dead
10 10 05 x 10' cultured immune spleen cells' 10 0 100 (P < 0.01)2 x 107 cultured immune spleen cellsb 10 0 100 (P < 0.01)1 x 107 cultured immune spleen cells' 10 2 80 (P < 0.01)5 x 106 cultured immune spleen cellsb 10 3 70 (P < 0.01)5 x 106 cultured immune spleen cellsb 10 7 30 (NSC)2 x 107 cultured immune spleen cells (no antigen) 10 10 0 (NS)5 x 107 cultured normal spleen cells 13 12 8 (NS)0.5 ml of anti-HSV serum' 10 1 90 (P < 0.01)0.3 ml of anti-HSV serumi 10 4 60 (P < 0.05)0.1 ml of anti-HSV serurm" 10 7 30 (NS)0.5 ml of normal rabbit serum 10 10 0 (NS)
a All mice were injected i.p. with 15 50% lethal doses (45,000 PFU) of HSV-1 1 day before cell or antibodytransfer."Spleen cells from C3H/HeJ mice primed 4 to 6 weeks previously with HSV-1 were cultured for 5 days with
UV-inactivated HSV-1.' NS, Not significant.d Rabbit anti-HSV-1 injected i.p.
afforded by 5 x 107 normal spleen cells culturedwith virus or by immune cells cultured withoutantigen.The role of antibody was compared to that of
immune lymphocytes by i.p. inoculation of rab-bit anti-HSV-1 (neutralization titer, 1:128) 24 hafter infection with virus. Inoculation of anti-HSV-1 serum resulted in recovery from infec-tion when compared with the uninoculated, in-fected control mice. The percent survival wasnot affected by normal rabbit serum.
It has been suggested previously that recoveryfrom lethal HSV infection by adoptively trans-ferred immune spleen cells was due to antibodyproduced by the transferred cells (25). It wastherefore of interest to determine the number ofB lymphocytes in the adoptively transferredpopulation of cells that had been restimulated invitro by 5-days culture with UV-inactivated vi-rus. The number of B lymphocytes in the trans-ferred cell population, as measured by mem-brane-bound immunoglobulin, was found to beless than 1% of the total cell number. Additional-ly, 2 x 107 5-days cultured immune spleen cellswere injected into naive normal or cyclophos-phamide treated mice (no virus given). Serawere tested at day 10 for anti-HSV antibody. Noanti-HSV antibody, as measured by radio-immunoassay, was detected (data not shown).We conclude that the observed recovery and
protection from death was not a result of anti-body production by the transferred cells.Recovery of cyclophosphamide-treated mice
from HSV-1 infection. To assess the effect ofvarious cellular and humoral components inrecovery, mice were immunosuppressed by an
i.p. injection of 200 mg of cyclophosphamide perkg 24 h before infection with HSV-1, followedby the transfer of cells or antibody after a further24 h.Immunosuppressed mice which were subse-
quently infected with HSV-1 were given spleencells obtained from mice immunized 6 weekspreviously with infectious HSV-1 (Table 2). Re-covery and subsequent protection from deathwere achieved with the transfer of 5 x 107noncultured immune spleen cells, whereas 5 x107 nonimmune cells or less than 5 x 107 noncul-tured immune spleen cells failed to transfersignificant protection.Table 3 shows that, in contrast to that seen
with imunocompetent mice, no significant pro-
TABLE 2. Effect of noncultured immune spleencells on primary HSV-1 infection incyclophosphamide-treated mice'
No. of miceNo. of cells transferred' % Survival
Total Dead
13 13 05 x 107 14 4 71 (P< 0.01)2 x 107 14 12 14 (NSC)1 X 107 14 13 7 (NS)5 x 106 14 13 7 (NS)5 x 107 (nonimmune) 14 13 7 (NS)
a Mice injected with 200 mg of cyclophosphamideper kg i.p. 1 day before virus injection and 15 50olethal doses (540 PFU) of HSV-1 1 day before celltransfer.
b Spleen cells from C3H mice immunized 6 weekspreviously with HSV-1.
a All mice were injected with 200 mg of cyclophos-phamide per kg i.p. 1 day before receiving virus and 1550%o lethal doses (540 PFU) of HSV-1 1 day before celltransfers.
b Rabbit anti-HSV-1 was injected i.p. 1 day aftervirus infection.
NS, Not significant.
tection was realized by the administration ofhyperimmune anti-HSV-1 antibody to cyclo-phosphamide-treated, HSV-1-infected mice.
In vitro stimulation of immune spleen cellswith UV-inactivated virus results in a populationthat contains potent HSV-specific cytotoxic Tlymphocytes (12). The effect upon lethal HSV-1infection resulting from the adoptive transfer ofthese cells was studied (Table 4). The transfer of2 x 107, 1 x 107, or 5 x 106 restimulated immunespleen cells protected mice from death, whereas1 x 106 restimulated immune spleen cells failedto protect. In comparison, 5 x 107 nonculturedimmune cells were required for protection,which is a 10-fold greater number of cells than isneeded when restimulated immune cells aretransferred. No protection resulted from thetransfer of immune cells cultured without UV-inactivated virus (2 x 107 cells), naive spleencells cultured with virus (2 x 1O7 cells), or
INFECT. IMMUN.
noncultured normal spleen cells (5 x 107 orfewer cells).To characterize the cell type(s) responsible for
recovery, cultured immune spleen cells weretreated with monoclonal antibodies against vari-ous mouse lymphocyte surface antigens pluscomplement before adoptive transfer (Table 5).Recovery from infection was conferred by thetransfer 2 x 107 cultured immune cells, culturedcells treated with anti-Lyt 1.1 plus complement,and cells treated with normal mouse serum pluscomplement. Treatment of cultured immunecells with anti-Thy 1.2 or anti-Lyt 2.1 pluscomplement abrogated recovery. The in vitroassay for HSV-1-specific CTL with 5'Cr releasecorrelated with elimination of recovery. That is,CTL activity in vitro was abolished by treatmentwith anti-Thy 1.2 or anti-Lyt 2.1 plus comple-ment. No decrease in cytotoxicity was observedby cells treated with anti-Lyt 1.1 or normalmouse serum plus complement (Table 5).NK cells have been proposed as being of
importance in the resistance of mice to HSV (1).We asessed the role of these cells in our systemby adoptively transferring a splenocyte popula-tion that expressed NK activity, but little or notCTL (Table 6). Transfer of 2 x 107 splenocytes,which demonstrated NK activity in vitro, failedto mediate recovery of infected mice. However,transfer of 2 x 107 cultured immune spleno-cytes, which demonstrated CTL but not NKactivity, resulted in significant recovery.The possibility that antibody and thymus-
derived lymphocytes act in concert in the recov-ery of mice from lethal HSV infection wassuggested by Oakes wt al. (21). We investigatedthis possiblity in our system by adoptively trans-
TABLE 4. Recovery of cyclophosphamide-treated mice from lethal HSV-1 infection by HSV-1-primedspleen cells restimulated in vitro'
No. of miceCells transferred % Survival
Total Dead
25 25 02 x 107 cultured immune spleen cellsb 11 2 82 (P < 0.01)1 x 107 cultured immune spleen cells" 21 3 86 (P < 0.01)5 x 106 cultured immune spleen cells" 19 11 42 (P < 0.01)1 x 106 cultured immune spleen cellsb 19 16 16 (NS')5 x 107 noncultured immune spleen cells 12 2 83 (P < 0.01)2 x 107 noncultured immune spleen cells 12 10 17 (NS)1 x 107 noncultured immune spleen cells 12 10 17 (NS)5 x 107 noncultured normal spleen cells 12 10 17 (NS)2 x 10' cultured immune spleen cells (no antigen) 5 5 0 (NS)2 x 107 cultured normal spleen cellsd 10 9 10 (NS)aAll mice were injected with 200 mg of cyclophosphamide per kg i.p. 1 day before receiving virus and 15 50%o
lethal doses (540 PFU) of HSV-1 1 day before cell transfers." Spleen cells from C3H/HeJ mice primed 4 to 6 wks previously with HSV-1 were cultured for 5 days withUV-inactivated HSV-1.
I NS, Not significant.d Nonimmune spleen cells cultured for 5 days with UV-inactivated HSV-1.
plus complement) (38.7)ca All mice were injected with 200 mg/of cyclophosphamide per kg i.p. 1 day before receiving virus and 15 50%
lethal doses (540 PFU) of HSV-1 1 day before cell transfers.b Spleen cells from C3H/HeJ mice primed 4 to 6 weeks previously with HSV-1 were cultured for 5 days with
UV-inactivated HSV-1.C Percent specific lysis.d Cultured (5 days) immune spleen cells treated with antiserum plus complement before intravenous transfer
into infected recipients.e NS, Not significant.
ferring cultured immune spleen cells and anti-HSV antibody to cyclophosphamide-treated,HSV-infected mice. When antibody was givenconcurrently with 2 x 106 cultured immune
TABLE 6. Effect of spleen cells expressing NKactivity on primary HSV-1 infection in
cyclophosphamide-treated micea
Cells transferred % % Specific lysis'(2X 107) Survi-(2 x 107) valb L-HSV L A31-HSV
cellsea All mice were injected with 200 mg of cyclophos-
phamide per kg i.p. 1 day before receiving virus and 1550%o lethal doses (540 PFU) of HSV-1 1 day before celltransfers.
b Ten mice per group.c The values shown represent means of quadupli-
cate cultures run at effector cell target cell ratios of25:1. The standard errors for the 4-h assays werealways below 4%. The L targets were L929 cells whichwere infected with HSV-1 (L-HSV) or uninfected (L).The A31 targets were BALB/c 3T3 clone A31 whichwere infected with HSV-1 (A31-HSV).
d Spleen cells from C3H/HeJ mice primed 4 to 6weeks previously with HSV-1 were cultured for 5 dayswith UV-inactivated HSV-1.
e Spleen cells from nonimmune C3H/HeJ mice in-fected intravenously 2 days previously with 107 PFUof HSV-1.
cells, the number of mice that recovered frominfection was greater than that seen with eitherantibody or cultured immune spleen cells alone(Table 7). Inoculation of 2 x 10 cultured normalcells plus antibody failed to result in recovery.
DISCUSSIONThe essential aim of our studies was to define
the role of CTL in recovery from HSV-1 infec-tion in mice. To achieve this, normal or immuno-suppressed mice were infected with virus, andthe effect of adoptive transfers of various cellpopulations was recorded. We have shown thatspleen cells taken from mice previously infectedwith HSV (HSV-immune mice) and immunizedin vitro with inactivated virus express potent H-2-restricted, HSV-specific, cytotoxic activity(12). Such cells conferred protection, as judgedby survival from an otherwise lethal infection, toboth normal and cyclophosphamide-immuno-suppressed mice. Protection was not afforded bysimilar numbers of normal splenocytes, immunesplenocytes cultured in the absence of antigen,or immune splenocytes before in vitro restimula-tion. However, transfer of the last cell popula-tion resulted in recovery when given at highercell numbers. On the average, the protectiveefficacy of immune, noncultured splenocyteswas 10-fold less than that of in vitro-immunizedimmune splenocytes. These data strongly implythat CTL were responsible for the protection,although the case for CTL would have beenstronger had their effectiveness in comparison tothe other cell transfers been even greater. Sever-al factors should be considered which couldexplain why greater differences were not ob-served. First, to exert protection the immune
a All mice were injected with 200 mg of cyclophosphamide per kg i.p. 1 day before receiving virus and 15 50%lethal doses (540 PFU) of HSV-1 1 day before cell transfers.
b Mice were injected i.p. with rabbit anti-HSV-1 antiserum.c Results obtained from two separate experiments.d NS, Not significant.e Spleen cells from C3H/HeJ mice primed 4 to 6 weeks previously with HSV-1 were cultured for 5 days with
UV-inactivated HSV-1.
component in the cell transfers must gain accessto the sites of virus replication. It is possible thatcultured cells after injection are held up in theliver, spleen, and lungs, with only a few reachinginfected cells. Second, in the noncultured im-mune cell populations CTL precursors werepresent (24). These cells could rapidly expandand differentiate into CTL upon exposure toinfected cells in vivo. Third, other subsets of Tcells, particularly the Lyt 1+1- cell that medi-ates delayed-type hypersensitivity to HSV, werepresent, especially in the noncultured immunecell population. This subset of cells also acted toconfer anti-herpesvirus immunity (A. A. Nash,Abstr. 6th Cold Spring Harbor Meet. 1982, p.213). A fourth possibility is that antibody pro-duction or NK cell activity in the adoptivetransfers could confer protection, albeit to alesser degree than CTL. Thus we showed, ashave others (13, 16, 19, 22), that antibody couldprotect immunocompetent mice. In addition, wealso demonstrated that a cyclophosphamide-sensitive component of the immune system actsin concert with the adoptively transferred anti-body (see below). We know that adoptive trans-fers of cells taken freshly from animals canproduce antibody upon stimulation in vitro or invivo (Larsen and Rouse, unpublished data), butafter culture and antigen stimulation for 5 daysin vitro they lose this capacity and are almostlacking in viable B lymphocytes.Thus, taking into account the above men-
tioned possibilities, we are strongly of the con-tention that recovery from lethal infection byHSV was principally mediated by CTL. Strongsupport for this notion came from experimentsshowing that the protective effect of the in vitro-immunized cell population was ablated by treat-ment with anti-Thy 1 and anti-Lyt 2 antisera pluscomplement, but not removed by treatment withanti-Lyt 1 plus complement. This profile of
susceptibility indicates the essential role of a Lyt1-2+ T cell, markers expressed on CTL (andsuppressor cells) (27). Ablation of protectiveeffects by anti-T cell sera also argues stronglyagainst the suggestion by Worthington et al. (26)that any protective effect observed with adop-tive cell transfers can be attributed to antibodyproduction by the transferred cells. It also rulesout protection due to the transfer of helper Tcells that permits antibody production by thedonor or recipient B cells. It is of interest to notethat the adoptive protective effects of spleencells from animals recently activated with virusis affected by anti-Lyt 1 serum, not by Lyt 2serum, an observation taken to argue for the roleof a helper cell- or delayed-type hypersensitivi-ty-mediating T cell, both of which express theLyt 2 phenotype (4, 18). This shows that othercell types can also mediate immunity, but, inaddition, it should be noted that CTL precursorsare Lyt 1+2 cells and so also would also beremoved by the anti-Lyt 1 serum treatment (24).Our observation that CTL can confer to mice
the ability to recover from HSV infection doesnot mean that these cells mediate their protec-tive effects in vivo by expressing cytotoxicity. Astudy with a cloned cytotoxic T cell line demon-strated the production of interferon upon expo-sure to antigen (17). Such interferon could medi-ate recovery either directly by protectinguninfected cells or indirectly by NK cell activa-tion and recycling (1, 27). Precursors of the NKcells could either be in the adoptively trans-ferred population or in the recipient. We consid-er it unlikely that protection was mediated byactive NK cells in the transferred population,because, apart from the Lyt data discussedabove, we have failed to observe protection withpopulations of spleen cells taken from frecentlyinfected (2 days) mice at a time when theyexpress NK cell activity, but little or no CTL. In
some models the CTL seem able to exert theirprotective effects in vivo by direct cytotoxicity(6), but whether this is occurring in HSV immu-nity merits further investigation.The final aspect of our study deserving of
comment is to emphasize the point that in vivothere are numerous ways to effect immunity andthat additive (and presumably negative) mecha-nisms may occur between different immunecomponents. Some excellent examples of posi-tive interactions to effect protection againstHSV (not recovery) were reported by Kohl andLoa, who showed interaction between cells andantibody, an effect assumed to be the in vivoequivalent of antibody-dependent cell cytotoxic-ity (11). In our study, we observed a possiblepositive interaction in the cyclophosphamideimmunosuppressed model between anti-HSVantibody and CTL-containing cell populations inthat protection was observed in mixtures ofconcentrations at which either alone was notprotective. We do not consider this effect to beexplained by antibody-dependent cell cytotoxic-ity, since transfers of cultured, unstimulatedcells with antibody did not give similar results.This is in contrast to the studies of Rager-Zisman and Allison, who showed normal spleencells and antibody given together to cyclophos-phamide-treated mice conferred protection (23).However, these authors transferred 1 x 108noncultured spleen cells, a number 50-fold great-er than utilized in the present study (2 x 106).Further work is clearly required to describe theexact mechanism involved.
ACKNOWLEDGMENTS
The technical assistance of Linda Miller was most appreci-ated.
This work was supported by Public Health Service grant Al14981 from the National Institutes of Health.
ADDENDUM IN PROOF
Sethi et al. (J. Gen. Virol. 64:443-447, 1983) havereported protection of mice from fatal HSV-1 infectionby adoptive transfer of cloned anti-HSV cytotoxic Tlymphocytes.
LITERATURE CITED
1. Armerding, D., and H. Rossiter. 1981. Induction of naturalkiller cells by herpes-simplex virus type 2 in resistant andsensitive inbred mouse strains. Immunobiology 158:369-379.
2. Balachandran, N., S. Bacchetti, and W. E. Rawls. 1982.Protection against lethal challenge of BALB/c mice bypassive transfer of monoclonal antibodies to five glyco-proteins of herpes simplex virus type 2. Infect. Immun.
37:1132-1137.3. Bone, D. R., and R. J. Courtney. 1974. A temperature
sensitive mutant of herpes simplex virus type 1 defectivein the synthesis of the major capsid polypeptide. J. Gen.Virol. 24:17-27.
4. Cantor, H., and E. A. Boyse. 1977. Regulation of theimmune response by T-cell subclasses. Contemp. Top.Immunobiol. 7:47-67.
5. Dix, R. D., L. Pereira, and J. R. Baringer. 1981. Use of
monoclonal antibody directed against herpes simplex vi-rus glycoproteins to protect mice against acute virus-induced neurological disease. Infect. Immun. 34:192-199.
6. Engers, H. D., G. D. Sorenson, G. Terres, A. L. Glase-brook, C. Horvath, and K. T. Brunner. 1981. Functionalactivity in vivo of cytolytic T lymphocytes generated invitro, p. 632-635. In K. Resch and H. Kirchner (ed.),Mechanisms of lyphocyte activation. Elsevier/North-Hol-land, Amsterdam.
7. Engler, H., R. Zawatzky, A. Goldbach, C. H. Shroder, C.Weyand, G. J. Hammerling, and H. Kirchner. 1981.Experimental infection of inbred mice with herpes sim-plex virus. II. Interferon production and activation ofnatural killer cells in the peritoneal exudate. J. Gen. Virol.55:25-30.
8. Hirsch, M. S., B. Zisman, and A. C. Allison. 1970.Macrophages and age-dependent resistant to herpes sim-plex virus in mice. J. Immunol. 104:1160-1165.
9. Kirchner, H. 1982. Immunobiology of infection with her-pes simplex virus. Monogr. Virol. 13:34-37.
10. Kirchner, H., H. M. Hirt, H. Becker, and K. Munk. 1977.Production of an antiviral factor by murine spleen cellsafter treatment with Corynebacterium parvum. Cell. Im-munol. 31:172-176.
11. Kohl, S., and L. S. Loo. 1982. Protection of neonatal miceagainst herpes simplex virus infection. Probable in vivoantibody-dependent cellular cytotoxicity. J. Immunol.129:370-376.
12. Lawman, M. J. P., B. T. Rouse, R. J. Courtney, and R. D.Walker. 1980. Cell-mediated immunity against herpessimplex induction of cytotoxic T lymphocytes. Infect.Immun. 23:305-311.
13. McKendafl, R. R., T. Klassen, and J. R. Baringer. 1979.Host defenses in herpes simplex infections of the nervoussystem: effect of antibody on disease and viral spread.Infect. Immun. 23:305-311.
14. Mogenson, S. C., and H. K. Andersen. 1978. Role ofactivated macrophages in resistance of congenitally athy-mic nude mice to hepatitis induced by herpes simplexvirus type 2. Infect. Immun. 19:792-798.
15. Morahan, P. S., E. R. Kern, and L. A. Glasgow. 1977.Immunomodulator-induced resistance against herpes sim-plex virus. Proc. Soc. Exp. Biol. Med. 154:615-620.
16. Morahan, P. S., T. A. Thomson, S. Kohl, and B.K.Murray. 1981. Immune responses to labial infection ofBALB/c mice with herpes simplex type 1. Infect. Immun.32:180-187.
17. Morris, A. G., Y. L., Lin, and B. A. Askonas. 1982.Immune interferon release when a cloned cytotoxic T-cellline meets its correct influenza-infected target cell. Nature(London) 295:150-152.
18. Nagafuchi, S., I. Hayashida, K. Higa, T. Wada, and R.Mori. 1982. Role of Lyt-1 positive immune T cells inrecovery from herpes simplex virus infection in mice.Microbiol. Immunol. 26-359-362.
19. Nagafuchi, S., H. Oda, R. Mori, and T. Taniguchi. 1979.Mechanism of acquired resistance to herpes simplex virusinfection as studied in nude mice. J. Gen. Virol. 44:715-723.
20. Nash, A. A., H. J. Field, and R. Quartey-Papafio. 1980.Cell-mediated immunity in herpes simplex virus-infectedmice: induction, characterization and antiviral effects ofdelayed type hypersensitivity. J. Gen. Virol. 48:351-357.
21. Oakes, J. E., W. B. Davis, J. A. Taylor, and W. A.Weppner. 1980. Lymphocyte reactivity contributes toprotection conferred by specific antibody passively trans-ferred to herpes simplex virus-infected mice. Infect. Im-mun. 29-642-649.
22. Oakes, J. E., and H. Rosemond-Hornbeak. 1978. Anti-body-mediated recovery from subcutaneous herpes sim-plex virus type 2 infection. Infect. Immun. 21:489-495.
23. Rager-Zisman, B., and A. C. Allison. 1976. Mechanism ofimunologic resistance to herpes simplex virus 1 (HSV-1)infection. J. Immunol. 116:35-40.
24. Rouse, B. T., H. S. Larsen, and H. Wagner. 1983. Fre-
quency of cytotoxic T lymphocyte precursors to herpessimplex virus type 1 as determined by limiting dilutionanalysis. Infect. Immun. 39:785-792.
25. Starr, S. E., A. M. Visentine, M. 0. Tomek, and A. J.Nahmias. 1976. Effects of immunostimulants on resistanceof newborn mice to herpes simplex virus type 2 infection.Proc. Soc. Exp. Biol. Med. 152:57-60.
26 Worthington, M., M. A. Conliffe, and S. Baron. 1980.Mechanism of recovery from systemic herpes simplexvirus infection. II. Effectiveness of antibody reconstitution
INFECT. IMMUN.
of nude and neonatally thymectomized mice. Proc. Soc.Exp. Biol. Med. 165:462-468.
27. Zawatsky, R., H. Engler, and H. Kirchner. 1982. Experi-mental infection of inbred mice with herpes simplex virus.III. Comparison between newborn and adult C57BL/6mice. J. Gen. Virol. 60:25-29.
28. Zisman, B., M. S. Hirsch, and A. C. Allison. 1970.Selective effects of anti-macrophage serum, silica andanti-lymphocyte serum on pathogenesis of herpes virusinfection of young adult mice. J. Immunol. 104:1155-1159.
