Znt. J. Cancer: 16, 24-32 ( 1 975)
SPLEEN-CELL CYTOTOXICZTY FOR CYTOMEGALOVIRUSTRANSFORMED CELLS. 11. INHIBITION BY CYTOMEGALOVTRUS ANTISERUM
Donna M. MURASKO and Robert N. LAUSCH Department of Microbiology, College of Medicine, Milton S. Hershey Medical Center, Pennsylvania State University, Hershey, Pennsylvania 17033, USA
Serum from hamsters immunized repeatedly with cytomegalovirrrs ( C M V ) was able to block spleen-cell cytotoxicity for C M V-infected and C M V-transformed cells. Inhibition was observed regardless of whether spleen cells were obtained froni hamsters sensitized to the virus or to isografts of the transfbrmed cells (designated C X - ~ O - ~T-2). B, Such serum did not significantly block eflector cell response against transformed cells expressing herpes simplex virus or S V40-associated nienibrane antigens. The tnrchanism of blocking was investigated. It was found that the blocking activity could be absorbed out with Cx-90-3B, T-2 cells but not by untransjormed hamster embryo fibroblasts. Cx-90-3B, T-2 target cells treated with serum, arid then washed, remained resistant to effector cell attack, On the other hand, if serum-treated attacker cells were washed their cytotoxic activity was not significantly inipaired. These results suggest that the blocking factor in the serum is antibody-directed against cytomegalovirus-related membrane antigen. This conclusion is supported by the finding that the serum contained antibody specifically reactive with the transformed cell surface in isotopic antiglobulin tests.
Serologic tests have established that hamster cells transformed in vitro by cytomegalovirus express membrane antigen(s) also found on CMV-infected human cells (Lausch et a[., 1974a). Further studies showed that spleen cells from hamsters bearing isografts of the transformed cells were specifically cytotoxic for both homologous tumor cells and CMVinfected human embryonic lung cells (Murasko and Lausch, 1974). Spleen cells from CMVimmunized animals displayed similar activity in the microcytotoxicity assay. However, neither virus-immunized nor tumor-bearing hosts possessed detectable complement-dependent cytotoxic antibody in their sera. It was subsequently observed that CMVReceived: April 1 , 1975.
24
immunized animals were not resistant to challenge with small numbers of tumor cells (Lausch et al., 1974b). Preliminary in vitro studies indicated that serum from such hosts could block spleen-cell cytotoxicity for the tumor cells. Numerous studies have established that serum from tumor-bearing animals can block cellmediated immunity in vitro (Hellstrom and Hellstrom, 1974). However, the possibility that virus immunization may result in the production of a humoral factor(s) which interferes with tumor cell killing has received little attention. Therefore, further investigations regarding the specificity and mechanism of blocking were carried out. This report describes our observations.
BLOCKING OF CELLULAR IMMUNITY B Y CMV ANTISERUM MATERIAL A N D METHODS
Cells The preparation of hamster embryo fibroblasts (HEF) and the transformation of these cells by CMV, herpes simplex virus type 1 (HSV-1) and PARA-(defective SV40)-adenovirus 7 have been described previously. Briefly, the Cx-90-3B line was obtained from a transformed focus of HEF arising after exposure of the cells to ultraviolet (UV)-irradiated CMV strain C-87 (Albrecht and Rapp, 1973). 14-012-8-1 cells were obtained by infection of HEF with UV-inactivated HSV-1 strain 14-012 (Duff and Rapp, 1973). Most of the experiments in this report were performed with cell lines derived from tumors induced by the above parent cells. The cell line designated PARA-7 was developed by the in vitro transformation of HEF with adenovirus-7-SV40 hybrid virus (Lausch and Rapp, 1971). Primary HEL cultures were obtained from HEM Research, Inc., Rockland, Maryland, USA. All cells were maintained on medium 199 or Eagle's medium supplemented with 10-20% fetal calf serum, 10% tryptose phosphate broth, 0.075 % sodium bicarbonate, 100 I U of penicillin per ml, and 1OOpg of streptomycin per ml. CMV-infected cells were used as target cells fdr certain tests. Suspensions of HEL cells infected with CMV at about 1 PFU/cell were plated into 35 mm plastic Petri dishes. Approximately 72 h later, the cells were harvested by trypsinization and utilized for the assays. Virus CMV-strain C-87 was propagated and titrated by the plaque technique in HEL cells (Albrecht and Rapp, 1973). Rabbit kidney cells were utilized for the propagation and titration of HSV-1 strain 35 and HSV-2 strain 333. Simian virus 40 (SV40) was grown and titrated in green monkey kidney cells. Immunization of hamsters Inbred hamsters (LSH strain) were obtained from Lakeview Hamster Colony, New Field, N. J., USA. Hamsters were immunized against CMV, HSV-1, HSV-2 or SV40 by intraperitoneal or subcutaneous inoculations of the respective virus given at weekly intervals. Hamsters immunized against CMV were given 2-5 inoculations
of 1 x lo6 plaque-forming units (PFU) of virus. Three inoculations of 1 x lo8 PFU of SV40 or 5 x lo7 PFU of HSV-1 were given. Immunization against HSV-2 consisted of successive inoculations of 1 x lo5, 1 x los and 1 x lo7 PFU of virus. Seven to 14 days after the final inoculation, the hamsters were bled and the sera collected. Tumors were induced by intradermal or subcutaneous inoculation of lo4-lo5 of the desired cells. Method of serum absorption Cells were harvested for absorption by trypsinization. After the cells had been washed three times, small aliquots (0.5 ml or 1.0ml) of appropriately diluted heat-inactivated (56" C for 30min) sera were absorbed with an equal volume of packed cells or with a known number of the desired cells. All absorptions were performed at 37" C for 1 h with continuous mixing. In vitro assays of cell-mediated immunity and serum blocking activity The details of the cellular cytotoxicity assays have been described previously (Murasko and Lausch, 1974). Briefly, spleen cells from virussensitized, tumor-bearing or normal hamsters were obtained by means of a Ficoll-Hypaque gradient. These were added at a ratio of 500:l to target cells attached to micro-titer test plates. After incubation overnight, the cells were fixed, stained, and the number of attached target cells counted utilizing an inverted microscope. The cytotoxic effect of the spleen cells was determined by calculating the percentage difference between the number of cells remaining in eight wells with normal spleen cells (SC,) versus that in eight wells with sensitized spleen cells (SC,) according to the equation:
% cytotoxicity
=
sc, -sc, x 100. SCn
The blocking assay was performed in a manner similar to the microcytotoxicity assay. However, normal or test serum were added to the wells prior to the addition of the spleen cells. After incubation for 1 h at 37" C, the plate was inverted, shaken once, and then the spleen cells were added. The assay was then completed as described above.
25
MURASKO A N D LAUSCH
The percentage blocking was calculated using the difference between the percentage cytotoxicity of the sensitized spleen cells in the presence of normal hamster serum (SC,i-N) and the percentage cytotoxicity of the sensitized spleen cells in the presence of test serum (SC,t A) according to the equation: %,. cytotoxicity . . SC,-tN -
blocking
=
-% cytotoxicity SC, I A cytotoxicity SC,+N
x
loo.
Statistical analysis of the microcytotoxicity data was carried out using Student’s t-test. Serological tests
The isotopic antiglobulin test was performed as described previously (Lausch et a / . , 39746). RESULTS
Preliminary experiments had indicated that serum from animals immunized to CMV could inhibit the cytotoxic effect of Cx-90-3B, T-2 tumor-bearer spleen cells for homologous target cells. In order to investigate this phenomenon systematically, antiserum to CMV was raised in hamsters. This was done via weekly intraperitoneal inoculations of loe PFU CMV. Groups of animals were bled after the second, third, fourth and fifth immunizations, and the corresponding serum pools tested for virus neutralizing antibody. The dilution of serum producing a >75% reduction in plaque-forming units was 1 :4, 1 :40, 1 :160, and I : 160 for pools 2, 3, 4, and 5 respectively. None of the four preparations was cytotoxic for Cx-90-3B, T-2 cells in the presence of guinea-pig complement ( I :4 dilutions tested ), Blocking of effector cell cytotoxicity
The capacity of the different serum pools to abrogate spleen-cell cytotoxicity was then examined. The data in Table I show that sera (pools 4 and 5 ) from animals immunized four or five times with CMV could block the killing of Cx-903B, T-2 cells by spleen cells from animals bearing isografts of the CMV-transformed cells. Sera from animals inoculated two or three times with CMV, or sensitized to HSV-I, did not inhibit Cx-90-3B, T-2 sensitized spleen cells. Further tests with pool 5 showed that it could also block
26
spleen-cell cytotoxicity for CMV-infected cells. However, it did not inhibit the cytotoxic response of spleen cells from non-Cx-90-3B, T-2 tumorbearing hosts for their respective target cells. Spleen cells from CMV-sensitized hosts have previously been shown to be cytotoxic for CMVtransformed cells. I t was found (Table It) that serum pools 4 and 5 could also significantly inhibit this immune reaction. The specificity of the inhibition was indicated by the failure of pool 5 to abrogate the activity of spleen cells from HSV-I- or SV40-immunized hosts. Antiserum to HSV types I and 2 did not block in the CMV spleen-cell/Cx-90-3B,T-2 target-cell system but could inhibit the cytotoxic reaction of attacker cells from HSV-I -immunized hosts to HSV-I-transformed cells. Serum from a PARA-7 tumor bearer could block SV40sensitized spleen cell activity. The latter observations conform with previous reports (Laux and Lausch, 1974; Lausch et al., 1974b; Lausch ef al., 1975~).The effect of dilution on the blocking activity of pools 4 and 5 is shown in Table 111. It is evident that the serum preparations produced significant inhibition only at relatively low dilutions ( I :8-1 :l6). Efect of’ahsorption on serum blocking activity
Absorption experiments can be used to investigate the nature of the reactive factor in blocking serum. For example, if the reactivity can be specifically absorbed out with target cells, this suggests that the blocking factor is antibody, o r possibly an antigen-antibody complex. Table IV shows the results of one such experiment. I t can be seen that unabsorbed serum pools 4 and 5 strongly inhibited effector cell activity. Absorption with lo7 HEF transformed by CMV removed all blocking activity in both sera. In contrast, absorption with a five-fold greater number of untransformed HEF did not significantly alter serum blocking. These results indicated that the blocking factor in the CMV antiserum could bind specifically to Cx-90-3B, T-2 target cells. Further investigation of the natrwc’ of the serum blocking factor
In the foregoing blocking tests, the serum was added to the target cells, incubated for 1 h at 37” C and removed from the cells by inversion of the microtiter plate. Since residual serum was
TABLE I
14-012-8-1, T-10
PARA-7
14-012-8-1, T-10
PARA-7
Sera were tested at a dilution of I :4 or I :6.
Normal Pool 5
CMV-infected HEL
C X - ~ O - ~T-2 B,
’
Normal Pool 4 Pool 5 Anti-HSV-1
C X - ~ O - ~T-2 B,
C X - ~ O - ~T-2 B,
Normal PARA-7 Pool 5
Normal 14-012-8-1, T-10 Pool 5
Pool 4
Normal Pool 2 Pool 3
Serum
C X - ~ O - ~T-2 B,
Target cells
C X - ~ O - ~T-2 B,
from hamsters bearing indicated tumor
Spleen cells
48i2
78 k 4
4211
8 2 11
95rt3
Normal spleen cells
3512 4612 39%2
6013 80&1 55 i 4
27i2 48kl
48 1t2 86i3 87i2 4811
88 1 3
76+3 74h2 7713
Immune spleen cells
Number of surviving target cells ) treatment with per well ( ~ s Eafter
21 4 19
23 -3 29
36 - 14
0.02
0.001
0.02
0.05
0.001 0.01
0.001
-6 41
5
0.001
0.001 0.001 0.001
P
SPLEEN-CELL CYTOTOXICZTY FOR CYTOMEGALOVIRUSTRANSFORMED CELLS. 11. INHIBITION BY CYTOMEGALOVTRUS ANTISERUM
Donna M. MURASKO and Robert N. LAUSCH Department of Microbiology, College of Medicine, Milton S. Hershey Medical Center, Pennsylvania State University, Hershey, Pennsylvania 17033, USA
Serum from hamsters immunized repeatedly with cytomegalovirrrs ( C M V ) was able to block spleen-cell cytotoxicity for C M V-infected and C M V-transformed cells. Inhibition was observed regardless of whether spleen cells were obtained froni hamsters sensitized to the virus or to isografts of the transfbrmed cells (designated C X - ~ O - ~T-2). B, Such serum did not significantly block eflector cell response against transformed cells expressing herpes simplex virus or S V40-associated nienibrane antigens. The tnrchanism of blocking was investigated. It was found that the blocking activity could be absorbed out with Cx-90-3B, T-2 cells but not by untransjormed hamster embryo fibroblasts. Cx-90-3B, T-2 target cells treated with serum, arid then washed, remained resistant to effector cell attack, On the other hand, if serum-treated attacker cells were washed their cytotoxic activity was not significantly inipaired. These results suggest that the blocking factor in the serum is antibody-directed against cytomegalovirus-related membrane antigen. This conclusion is supported by the finding that the serum contained antibody specifically reactive with the transformed cell surface in isotopic antiglobulin tests.
Serologic tests have established that hamster cells transformed in vitro by cytomegalovirus express membrane antigen(s) also found on CMV-infected human cells (Lausch et a[., 1974a). Further studies showed that spleen cells from hamsters bearing isografts of the transformed cells were specifically cytotoxic for both homologous tumor cells and CMVinfected human embryonic lung cells (Murasko and Lausch, 1974). Spleen cells from CMVimmunized animals displayed similar activity in the microcytotoxicity assay. However, neither virus-immunized nor tumor-bearing hosts possessed detectable complement-dependent cytotoxic antibody in their sera. It was subsequently observed that CMVReceived: April 1 , 1975.
24
immunized animals were not resistant to challenge with small numbers of tumor cells (Lausch et al., 1974b). Preliminary in vitro studies indicated that serum from such hosts could block spleen-cell cytotoxicity for the tumor cells. Numerous studies have established that serum from tumor-bearing animals can block cellmediated immunity in vitro (Hellstrom and Hellstrom, 1974). However, the possibility that virus immunization may result in the production of a humoral factor(s) which interferes with tumor cell killing has received little attention. Therefore, further investigations regarding the specificity and mechanism of blocking were carried out. This report describes our observations.
BLOCKING OF CELLULAR IMMUNITY B Y CMV ANTISERUM MATERIAL A N D METHODS
Cells The preparation of hamster embryo fibroblasts (HEF) and the transformation of these cells by CMV, herpes simplex virus type 1 (HSV-1) and PARA-(defective SV40)-adenovirus 7 have been described previously. Briefly, the Cx-90-3B line was obtained from a transformed focus of HEF arising after exposure of the cells to ultraviolet (UV)-irradiated CMV strain C-87 (Albrecht and Rapp, 1973). 14-012-8-1 cells were obtained by infection of HEF with UV-inactivated HSV-1 strain 14-012 (Duff and Rapp, 1973). Most of the experiments in this report were performed with cell lines derived from tumors induced by the above parent cells. The cell line designated PARA-7 was developed by the in vitro transformation of HEF with adenovirus-7-SV40 hybrid virus (Lausch and Rapp, 1971). Primary HEL cultures were obtained from HEM Research, Inc., Rockland, Maryland, USA. All cells were maintained on medium 199 or Eagle's medium supplemented with 10-20% fetal calf serum, 10% tryptose phosphate broth, 0.075 % sodium bicarbonate, 100 I U of penicillin per ml, and 1OOpg of streptomycin per ml. CMV-infected cells were used as target cells fdr certain tests. Suspensions of HEL cells infected with CMV at about 1 PFU/cell were plated into 35 mm plastic Petri dishes. Approximately 72 h later, the cells were harvested by trypsinization and utilized for the assays. Virus CMV-strain C-87 was propagated and titrated by the plaque technique in HEL cells (Albrecht and Rapp, 1973). Rabbit kidney cells were utilized for the propagation and titration of HSV-1 strain 35 and HSV-2 strain 333. Simian virus 40 (SV40) was grown and titrated in green monkey kidney cells. Immunization of hamsters Inbred hamsters (LSH strain) were obtained from Lakeview Hamster Colony, New Field, N. J., USA. Hamsters were immunized against CMV, HSV-1, HSV-2 or SV40 by intraperitoneal or subcutaneous inoculations of the respective virus given at weekly intervals. Hamsters immunized against CMV were given 2-5 inoculations
of 1 x lo6 plaque-forming units (PFU) of virus. Three inoculations of 1 x lo8 PFU of SV40 or 5 x lo7 PFU of HSV-1 were given. Immunization against HSV-2 consisted of successive inoculations of 1 x lo5, 1 x los and 1 x lo7 PFU of virus. Seven to 14 days after the final inoculation, the hamsters were bled and the sera collected. Tumors were induced by intradermal or subcutaneous inoculation of lo4-lo5 of the desired cells. Method of serum absorption Cells were harvested for absorption by trypsinization. After the cells had been washed three times, small aliquots (0.5 ml or 1.0ml) of appropriately diluted heat-inactivated (56" C for 30min) sera were absorbed with an equal volume of packed cells or with a known number of the desired cells. All absorptions were performed at 37" C for 1 h with continuous mixing. In vitro assays of cell-mediated immunity and serum blocking activity The details of the cellular cytotoxicity assays have been described previously (Murasko and Lausch, 1974). Briefly, spleen cells from virussensitized, tumor-bearing or normal hamsters were obtained by means of a Ficoll-Hypaque gradient. These were added at a ratio of 500:l to target cells attached to micro-titer test plates. After incubation overnight, the cells were fixed, stained, and the number of attached target cells counted utilizing an inverted microscope. The cytotoxic effect of the spleen cells was determined by calculating the percentage difference between the number of cells remaining in eight wells with normal spleen cells (SC,) versus that in eight wells with sensitized spleen cells (SC,) according to the equation:
% cytotoxicity
=
sc, -sc, x 100. SCn
The blocking assay was performed in a manner similar to the microcytotoxicity assay. However, normal or test serum were added to the wells prior to the addition of the spleen cells. After incubation for 1 h at 37" C, the plate was inverted, shaken once, and then the spleen cells were added. The assay was then completed as described above.
25
MURASKO A N D LAUSCH
The percentage blocking was calculated using the difference between the percentage cytotoxicity of the sensitized spleen cells in the presence of normal hamster serum (SC,i-N) and the percentage cytotoxicity of the sensitized spleen cells in the presence of test serum (SC,t A) according to the equation: %,. cytotoxicity . . SC,-tN -
blocking
=
-% cytotoxicity SC, I A cytotoxicity SC,+N
x
loo.
Statistical analysis of the microcytotoxicity data was carried out using Student’s t-test. Serological tests
The isotopic antiglobulin test was performed as described previously (Lausch et a / . , 39746). RESULTS
Preliminary experiments had indicated that serum from animals immunized to CMV could inhibit the cytotoxic effect of Cx-90-3B, T-2 tumor-bearer spleen cells for homologous target cells. In order to investigate this phenomenon systematically, antiserum to CMV was raised in hamsters. This was done via weekly intraperitoneal inoculations of loe PFU CMV. Groups of animals were bled after the second, third, fourth and fifth immunizations, and the corresponding serum pools tested for virus neutralizing antibody. The dilution of serum producing a >75% reduction in plaque-forming units was 1 :4, 1 :40, 1 :160, and I : 160 for pools 2, 3, 4, and 5 respectively. None of the four preparations was cytotoxic for Cx-90-3B, T-2 cells in the presence of guinea-pig complement ( I :4 dilutions tested ), Blocking of effector cell cytotoxicity
The capacity of the different serum pools to abrogate spleen-cell cytotoxicity was then examined. The data in Table I show that sera (pools 4 and 5 ) from animals immunized four or five times with CMV could block the killing of Cx-903B, T-2 cells by spleen cells from animals bearing isografts of the CMV-transformed cells. Sera from animals inoculated two or three times with CMV, or sensitized to HSV-I, did not inhibit Cx-90-3B, T-2 sensitized spleen cells. Further tests with pool 5 showed that it could also block
26
spleen-cell cytotoxicity for CMV-infected cells. However, it did not inhibit the cytotoxic response of spleen cells from non-Cx-90-3B, T-2 tumorbearing hosts for their respective target cells. Spleen cells from CMV-sensitized hosts have previously been shown to be cytotoxic for CMVtransformed cells. I t was found (Table It) that serum pools 4 and 5 could also significantly inhibit this immune reaction. The specificity of the inhibition was indicated by the failure of pool 5 to abrogate the activity of spleen cells from HSV-I- or SV40-immunized hosts. Antiserum to HSV types I and 2 did not block in the CMV spleen-cell/Cx-90-3B,T-2 target-cell system but could inhibit the cytotoxic reaction of attacker cells from HSV-I -immunized hosts to HSV-I-transformed cells. Serum from a PARA-7 tumor bearer could block SV40sensitized spleen cell activity. The latter observations conform with previous reports (Laux and Lausch, 1974; Lausch et al., 1974b; Lausch ef al., 1975~).The effect of dilution on the blocking activity of pools 4 and 5 is shown in Table 111. It is evident that the serum preparations produced significant inhibition only at relatively low dilutions ( I :8-1 :l6). Efect of’ahsorption on serum blocking activity
Absorption experiments can be used to investigate the nature of the reactive factor in blocking serum. For example, if the reactivity can be specifically absorbed out with target cells, this suggests that the blocking factor is antibody, o r possibly an antigen-antibody complex. Table IV shows the results of one such experiment. I t can be seen that unabsorbed serum pools 4 and 5 strongly inhibited effector cell activity. Absorption with lo7 HEF transformed by CMV removed all blocking activity in both sera. In contrast, absorption with a five-fold greater number of untransformed HEF did not significantly alter serum blocking. These results indicated that the blocking factor in the CMV antiserum could bind specifically to Cx-90-3B, T-2 target cells. Further investigation of the natrwc’ of the serum blocking factor
In the foregoing blocking tests, the serum was added to the target cells, incubated for 1 h at 37” C and removed from the cells by inversion of the microtiter plate. Since residual serum was
TABLE I
14-012-8-1, T-10
PARA-7
14-012-8-1, T-10
PARA-7
Sera were tested at a dilution of I :4 or I :6.
Normal Pool 5
CMV-infected HEL
C X - ~ O - ~T-2 B,
’
Normal Pool 4 Pool 5 Anti-HSV-1
C X - ~ O - ~T-2 B,
C X - ~ O - ~T-2 B,
Normal PARA-7 Pool 5
Normal 14-012-8-1, T-10 Pool 5
Pool 4
Normal Pool 2 Pool 3
Serum
C X - ~ O - ~T-2 B,
Target cells
C X - ~ O - ~T-2 B,
from hamsters bearing indicated tumor
Spleen cells
48i2
78 k 4
4211
8 2 11
95rt3
Normal spleen cells
3512 4612 39%2
6013 80&1 55 i 4
27i2 48kl
48 1t2 86i3 87i2 4811
88 1 3
76+3 74h2 7713
Immune spleen cells
Number of surviving target cells ) treatment with per well ( ~ s Eafter
21 4 19
23 -3 29
36 - 14
0.02
0.001
0.02
0.05
0.001 0.01
0.001
-6 41
5
0.001
0.001 0.001 0.001
P
