proc. Nat. Acad. Sci. USA Vol. 72, No. 12, pp. 4971-4975, December 1975
Cell Biology
Failure of human cells transformed by simian virus 40 to form tumors in athymic nude mice (growth regulation/oncogenesis/humoral antibodies/rejection)
CHARLES DEAN STILES, WALTER DESMOND, JR., GORDON SATO, AND MILTON H. SAIER, JR. Department of Biology, John Muir College, University of California at San Diego, La Jolla, Calif. 92093
Communicated by Nathan 0. Kaplan, August 29, 1975
ral antibodies or inability of mouse serum to support growth. In this communication we report that we have been unable to induce tumor formation in nude mice from inoculations of SV40-transformed human cells.
Four individual lines and one subline of ABSTRACT human cells, permanently established in tissue culture after infection with simian virus 40, failed to form tumors- when inoculated into athymic nude mice. Under identical conditions, three established human cell lines of neoplastic origin and a spontaneously established human lymphocyte line formed tumors. Nude mice that failed to grow tumors from inocula of simian virus 40-transformed human cells, grew tumors from subsequent injections of authentic human cancer cells. Further efforts to demonstrate an immunologic basis for the growth suppression of human simian virus 40 transformants were also negative. The data suggest that the changes in morphology and in vitro growth behavior induced by the viral information are not sufficient for, or are only coincidentally related to, the neoplastic state.
MATERIALS AND METHODS Nude Mice. A colony of athymic nude mice with the Balb/c genetic background was established with breeding stock obtained from the Pathological Anatomical Institute, Kommunehospitalet, Copenhagen, Denmark. Cell Cultures. All cell lines were routinely cultured in Dulbecco-Vogt modified Eagle's medium supplemented with 12.5% horse serum and 2.5% fetal bovine serum. The history of the cell lines used in this study is summarized in Table 1. More than 95% of the cells in each of the SV40transformed human lines contained intranuclear SV40 tumor (T) antigen, as assayed by immunofluorescence (7). All cell lines were screened for mycoplasma by autoradiography and were negative for infection by this test. Tumorigenicity Test. Confluent cell cultures were exposed for 5-10 min to 0.01% trypsin in phosphate-buffered saline and then resuspended in growth medium. The cell suspensions were inoculated subcutaneously into the scapular region. All cell lines were tested in both male and female animals. Mice were scored as tumor positive when they carried progressively growing nodules greater than 0.5 cm in diameter that had grown at the site of cell inoculation. Nude mice injected with cells of neoplastic origin were tumor positive by 5 days to 2 months after injection. Test animals were observed for a minimum of 4 months before being declared tumor negative. Tumors induced by inoculation of human cancer cell lines were never observed to regress. Antilymphocyte Serum Treatment. Antilymphocyte serum, lot no. 13120, was purchased from Microbiological Assoc. Inc. and administered according to the protocol of Stanbridge and Perkins (8).
The relationship between the process of transformation, as defined by the growth properties of animal cells in tissue culture, and oncogenesis is poorly defined. The alterations in morphology and growth behavior resulting from exposure of animal cells in vitro to oncogenic viruses or chemicals frequently correlate with acquisition of tumorigenic potential although a causal relationship has not been demonstrated. Several DNA and RNA viruses, including simian virus 40 (SV40), induce morphologic transformation of human cells in culture (1, 2). Whether morphologic transformation of human cells by SV40 correlates -with the acquisition of malignant potential has been difficult to determine due to lack of a suitable animal host. Jensen et al. (3) found that subcutaneous inocula of 1.5 to 3.0 107 SV40-transformed W-18 cells produced nodules in terminal cancer patients, but the nodules did not grow progressively and always regressed within 10 days of implantation accompanied by an inflammatory reaction. Stanbridge et al. (4) reported that SV40transformed human buccal mucosa cells formed "palpable tumor nodules" in mice treated with antilymphocyte serum. Croce et al. have stated that SV4O-transformed LeschNyhan syndrome fibroblasts are tumorigenic in athymic nude mice (5). The congenitally athymic nude mouse (6) appears to be the best host for assessing the malignant potential of animal and human cells. In recent studies* workers in our laboratory found: (i) that all human and animal cell lines derived from neoplasms were tumorigenic in nude mice; (ii) that secondary cultures of animal and human embryos were not tumorigenic in nude mice; (iii) that most animal cell lines established from normal tissue explants and never intentionally exposed to carcinogens in vitro were not tumorigenic; and (iv) that those cell lines that failed to grow tumors in nude mice did not fail for reasons such as rejection by humoX
RESULTS Tumorigenicity of human cells in athymic nude mice Cell lines were tested for tumorigenicity with an inoculum of one to two million cells. Table 2 demonstrates that with an inoculum of this magnitude each of the three human cell lines derived from neoplasms generated tumors in nude mice. Additionally, a spontaneously established human lymphocyte line derived from a patient free of malignant disease but bearing antibodies to an Epstein-Barr type virus (9) was tumorigenic. The human origin of tumors induced by injection of RPMI-2650 cells was verified by karyotypic analysis of tumor cells that had been replated into culture (data not shown). Under the same conditions none of the SV40-transformed human cell lines produced tumors. Mice
Abbreviation: SV40, simian virus 40. * C. D. Stiles, W. Desmond, L. M. Chuman, G. Sato, and M. H. Saier, manuscript in preparation.
4971
Cell Biology: Stiles et al.
4972
Proc. Nat. Acad. Sci. USA 72 (1975) Table 1. Description and history of cell lines
Cell line
Tissue of origin
Morphology
Comments or literature refs.
SVT2 VA2-8-aza-Gr
Embryonic mouse Human buccal mucosa
Fibroblast-like Fibroblast-like
RBSV3
Human epidermis
Fibroblast-like
LNSV
Human epidermis
Fibroblast-like
Embryonic mouse cells transformed with SV40. SV40-transformed W18 cells selected for resistance to 8-aza-guanine (21). The line is termed simply "VA2" in other communications. We use the longer notation in this report to avoid confusion with the ancestral W18VA2 line. SV40-transformed skin fibroblasts from patient with inborn metabolic disorder, "cystinosis." SV40-transformed skin fibroblast from patient with
RBSV1A
Human epidermis
Fibroblast-like
A-9
Mouse muscle
Fibroblast-like
Lesch-Nyhan syndrome (22). SV40-transformed skin fibroblasts from patient with inborn metabolic disorder, "cystinosis." Derived from "L-cells" by selection for resistance to
HeLa
Human cervical carcinoma Human nasopharyngeal carcinoma Human melanoma
Epithelioid
(24)
Epithelioid
Cells have quasi-diploid human karyotype (25).
Epithelioid
Human spleen
Lymphoid
Human buccal mucosa Human skin
Fibroblast-like Fibroblast-like
Cell line donated by Dr. Grace Cannon, Litton.Bionetics Research Laboratories, Bethesda, Md. Cells were derived from a patient free from malignant disease but bearing antibodies to an Epstein-Barr type virus (9). (26) Cells from patient with Fanconi's anemia.
8-aza-guanine (23). RPMI-2650 BRL-4130 Hiene WI-L2 W18VA2 SV80
infected with lines LNSV, W18VA2, and SV80 exhibited swellings of 0.3-0.4 cm diameter, which were apparent 1 day after injection and disappeared by 12 days after injecTable 2. Tumorigenicity of heterologous cells in nude mice
No. of mice Cell line Human lines RPMI2650 BRL-4143 HeLa WI-L2 Human SV40transformed lines VA2-8-aza-Gr RBSV3
No. of mice
No. of cells
to
injected
injected
per mouse
form tumors
10 10 10 5
9 5 10 2
106 0.5 x 106 106 2 x 106
9 5
106 106 (precrisis
4
4 x 106 (precrisis
0 0
cells) RBSV1A
0
cells) RBSV1A LNSV W18VA2 SV80 VA2 8-aza-Gr LNSV W18VA2 Mouse lines SVT2 A-9
6
6 5 6 5 3 1 10 10
2x 2 x 2x 6x
106 (postcrisis cells) 106 106 106 106 107 107 106 106
0 0 0 0 0 0 0
10 10
tion. Figure LA illustrates a tumor induced by 106 human lymphocytes of the WI-L2 line. The transient swelling formed by inoculation of 2 X 106 SV4O-transformed human fibroblasts (SV80 line) is depicted in Fig. 1B. Two of the SV40-transformed human lines were retested for tumorigenicity in nude mice with inocula 10- to 20-fold in excess of those used with human cancer cell lines. Table 2 shows that the larger inocula were similarly ineffective in provoking tumor formation. When 107 LNSV cells were injected, swellings to a diameter of 0.5-0.6 cm were apparent after 1 day and disappeared within 7 days. Two lines, RBSV3 and TBSV1A, were injected at early passages after SV40 infection, and these cells subsequently entered into an apparent "senescent crisis" (10) in culture. The RBSV1A line was tested for tumorigenicity at a later passage when the cells had recovered from the crisis. The "postcrisis" RBSV1A cultures were similarily nontumorigenic. Lines VA2-8-aza-Gr and LNSV are deficient in hypoxanthine phosphoribosylpyrophosphorylase activity (Table 1). To determine whether this enzyme is crucial to tumor formation in nude mice, we injected several animals with the A-9 substrain of mouse L cells which is deficient in the enzyme activity. A-9 cells were tumorigenic in 100% of the nude mice tested (Table 2); this result corroborated the findings of Freedman and Shin (11), who demonstrated, in addition, that cells derived from A-9 induced tumors are still deficient in hypoxanthine phosphoribosylpyrophosphorylase. SV40-transformed mouse cells (the SVT2 line) are tumorigenic in nude mice. Growth of human tumors in nude mice that have failed to grow tumors from SV40-transformed human
cells
Nude mice do not possess the cell-mediated immune response and do not reject heterologous skin grafts (12). However, these animals do form humoral antibodies to heterolo-
Cell Biology: Stiles et al.
Proc. Nat. Acad. Sc. USA 72 (1975)
4973
*
::%.'.C 11
f
A.
il
..
B.
FIG. 1. Appearance of human tumors and abortive nodules in nude mice. (A) Nude mouse with tumor formed from injection of WI-L2 cells photographed 6 weeks after injection. (B) Nude mice with abortive nodules induced by human SV40 transformants (the SV80 line). The animals were photographed one week after injection of 2 x 106 cells subcutaneously. The upper mouse was untreated and the lower mouse was treated with antilymphocyte serum as described in Materials and Methods. Arrows indicate the transient swellings (abortive nodules) which appeared at the site of injection. gous
skin grafts (13), and
some
evidence suggests that nude
mice have a residual capacity to reject heterologous cells (14). To examine the possibility that SV40-transformed human cells might be more efficient inducers of residual immune
capabilities than authentic cancer cells,
we
tested the
capacity of the human viral transformants to "vaccinate" nude mice against authentic human tumor cells. Table 3 shows that nude mice that did not develop tumors from injections of SV40-transformed human fibroblasts formed tu-
when subsequently injected with authentic human cells on the contralateral flank. Additionally, SV40transformed mouse cells (the SVT2 line) formed tumors in nude mice that had failed to form tumors from injections of SV40-transformed human cells. Growth behavior of SV40 human transformants in "supersuppressed" nude mice Nude mice form humoral antibodies to heterologous skin grafts, and it has been shown that these humoral antibodies can induce necrosis in skin allografts when the mice are injected with rabbit complement (13). To examine the possibility that human SV40-transformed cells were for some reason more vulnerable than other types of human tissue to the action of humoral antibodies and endogenous mouse complement, we tested for tumorigenicity in nude mice that had been treated to suppress humoral immune responses. Table 4 shows that human SV40-transformed cells do not form tumors in neonatal nude mice, in nude mice treated with 700 rads of whole body irradiation, or in nude mice treated with antilymphocyte serum. Antilymphocyte serum had no effect mors cancer
on the size or duration of the transient swelling induced by inoculation of SV80 cells into nude mice (Fig. 1B). In an attempt to reproduce findings of Stanbridge et al. that human SV40 transformants were tumorigenic in mice
treated with antilymphocyte serum (4), the tumorigenicity of W18VA2 cells and also of SV80 cells was tested in Balb/c mice treated with antilymphocyte serum. Palpable nodules arose in the treated Balb/c mice that were similar in size and duration to those produced in either control mice or nude mice treated with antilymphocyte serum. The potency of the antilymphocyte serum used in these experiments was confirmed by the ability of treated Balb/c mice to form tumors from an inoculation of 106 HeLa cells; untreated Balb/ c mice did not form tumors from HeLa cells. Comparative morphology and growth behavior of mouse cells and human cells infected with SV40 SV40 infection of 3T3 cells in culture generates characteristic changes in growth behavior and morphology which frequently correlate with enhanced tumorigenic potential (15, 16). Relative to normal 3T3 cells, SV40-transformed cultures show a decrease in serum growth requirement, an increased saturation density and enhanced ability to plate in agar suspension, and/or loss of density dependent inhibition of growth. Morphologically, SV40-transformed 3T3 cells tend to overlap each other in tissue culture dishes and grow in multilayers, whereas control cultures do not overlap (17). Table 4. Behavior of SV40-transformed cells in baby nude mice and "supersuppressed" nude mice
Table 3. Growth of tumors in nude mice subsequent to challenge with SV40-transformed human cells
First inoculation
Second inoculation
RBSV3 RBSV1A
RPMI-2650 HeLa
VA2-8-aza-Gr
SVT2
No. of tumors formed/no. of
Cell line
No. of mice injected
mice from second inoculation
VA2-8-aza-Gr VA2-8-aza-Gr
5 (1-day-old mice) 4 (mice pretreated with 700 rads whole body irradiation) 2 (mice treated with
3/3 6/6 2/2
Nude mice that, after 4 months, had failed to form tumors from inocula of 106 RBSV1A (postcrisis cells), RBSV3 (precrisis cells) or VA2-8-aza-Gr were inoculated with 106 cells subcutaneously on the contralateral side.
W18VA2
No. of No. of cells mice to injected form per mouse tumors 106 6 x 106
0 0
2 x 106
0
2 x 106
0
antilymphocyte SV80
serum) 2 (mice treated with antilymphocyte serum)
4974
Proc. Nat. Acad. Sci. USA 72 (1975)
Cell Biology: Stiles et al.
e1a
Table 5. Growth behavior of tumorigenic and nontumorigenic mouse and human cells in vitro
Cell line
Saturation density (cells/cm2)
HeLa SVT2 3T3 LNSV W18 VA2
1.08 7.29 3.91 7.35 2.5
x x x x x
106 105 104 104 105
Ratio doubling time 10% FBS: 1.0% FBS*
% Cloning efficiency in Methocel
0.89 0.38 0 0.90 0
12
Cell Biology
Failure of human cells transformed by simian virus 40 to form tumors in athymic nude mice (growth regulation/oncogenesis/humoral antibodies/rejection)
CHARLES DEAN STILES, WALTER DESMOND, JR., GORDON SATO, AND MILTON H. SAIER, JR. Department of Biology, John Muir College, University of California at San Diego, La Jolla, Calif. 92093
Communicated by Nathan 0. Kaplan, August 29, 1975
ral antibodies or inability of mouse serum to support growth. In this communication we report that we have been unable to induce tumor formation in nude mice from inoculations of SV40-transformed human cells.
Four individual lines and one subline of ABSTRACT human cells, permanently established in tissue culture after infection with simian virus 40, failed to form tumors- when inoculated into athymic nude mice. Under identical conditions, three established human cell lines of neoplastic origin and a spontaneously established human lymphocyte line formed tumors. Nude mice that failed to grow tumors from inocula of simian virus 40-transformed human cells, grew tumors from subsequent injections of authentic human cancer cells. Further efforts to demonstrate an immunologic basis for the growth suppression of human simian virus 40 transformants were also negative. The data suggest that the changes in morphology and in vitro growth behavior induced by the viral information are not sufficient for, or are only coincidentally related to, the neoplastic state.
MATERIALS AND METHODS Nude Mice. A colony of athymic nude mice with the Balb/c genetic background was established with breeding stock obtained from the Pathological Anatomical Institute, Kommunehospitalet, Copenhagen, Denmark. Cell Cultures. All cell lines were routinely cultured in Dulbecco-Vogt modified Eagle's medium supplemented with 12.5% horse serum and 2.5% fetal bovine serum. The history of the cell lines used in this study is summarized in Table 1. More than 95% of the cells in each of the SV40transformed human lines contained intranuclear SV40 tumor (T) antigen, as assayed by immunofluorescence (7). All cell lines were screened for mycoplasma by autoradiography and were negative for infection by this test. Tumorigenicity Test. Confluent cell cultures were exposed for 5-10 min to 0.01% trypsin in phosphate-buffered saline and then resuspended in growth medium. The cell suspensions were inoculated subcutaneously into the scapular region. All cell lines were tested in both male and female animals. Mice were scored as tumor positive when they carried progressively growing nodules greater than 0.5 cm in diameter that had grown at the site of cell inoculation. Nude mice injected with cells of neoplastic origin were tumor positive by 5 days to 2 months after injection. Test animals were observed for a minimum of 4 months before being declared tumor negative. Tumors induced by inoculation of human cancer cell lines were never observed to regress. Antilymphocyte Serum Treatment. Antilymphocyte serum, lot no. 13120, was purchased from Microbiological Assoc. Inc. and administered according to the protocol of Stanbridge and Perkins (8).
The relationship between the process of transformation, as defined by the growth properties of animal cells in tissue culture, and oncogenesis is poorly defined. The alterations in morphology and growth behavior resulting from exposure of animal cells in vitro to oncogenic viruses or chemicals frequently correlate with acquisition of tumorigenic potential although a causal relationship has not been demonstrated. Several DNA and RNA viruses, including simian virus 40 (SV40), induce morphologic transformation of human cells in culture (1, 2). Whether morphologic transformation of human cells by SV40 correlates -with the acquisition of malignant potential has been difficult to determine due to lack of a suitable animal host. Jensen et al. (3) found that subcutaneous inocula of 1.5 to 3.0 107 SV40-transformed W-18 cells produced nodules in terminal cancer patients, but the nodules did not grow progressively and always regressed within 10 days of implantation accompanied by an inflammatory reaction. Stanbridge et al. (4) reported that SV40transformed human buccal mucosa cells formed "palpable tumor nodules" in mice treated with antilymphocyte serum. Croce et al. have stated that SV4O-transformed LeschNyhan syndrome fibroblasts are tumorigenic in athymic nude mice (5). The congenitally athymic nude mouse (6) appears to be the best host for assessing the malignant potential of animal and human cells. In recent studies* workers in our laboratory found: (i) that all human and animal cell lines derived from neoplasms were tumorigenic in nude mice; (ii) that secondary cultures of animal and human embryos were not tumorigenic in nude mice; (iii) that most animal cell lines established from normal tissue explants and never intentionally exposed to carcinogens in vitro were not tumorigenic; and (iv) that those cell lines that failed to grow tumors in nude mice did not fail for reasons such as rejection by humoX
RESULTS Tumorigenicity of human cells in athymic nude mice Cell lines were tested for tumorigenicity with an inoculum of one to two million cells. Table 2 demonstrates that with an inoculum of this magnitude each of the three human cell lines derived from neoplasms generated tumors in nude mice. Additionally, a spontaneously established human lymphocyte line derived from a patient free of malignant disease but bearing antibodies to an Epstein-Barr type virus (9) was tumorigenic. The human origin of tumors induced by injection of RPMI-2650 cells was verified by karyotypic analysis of tumor cells that had been replated into culture (data not shown). Under the same conditions none of the SV40-transformed human cell lines produced tumors. Mice
Abbreviation: SV40, simian virus 40. * C. D. Stiles, W. Desmond, L. M. Chuman, G. Sato, and M. H. Saier, manuscript in preparation.
4971
Cell Biology: Stiles et al.
4972
Proc. Nat. Acad. Sci. USA 72 (1975) Table 1. Description and history of cell lines
Cell line
Tissue of origin
Morphology
Comments or literature refs.
SVT2 VA2-8-aza-Gr
Embryonic mouse Human buccal mucosa
Fibroblast-like Fibroblast-like
RBSV3
Human epidermis
Fibroblast-like
LNSV
Human epidermis
Fibroblast-like
Embryonic mouse cells transformed with SV40. SV40-transformed W18 cells selected for resistance to 8-aza-guanine (21). The line is termed simply "VA2" in other communications. We use the longer notation in this report to avoid confusion with the ancestral W18VA2 line. SV40-transformed skin fibroblasts from patient with inborn metabolic disorder, "cystinosis." SV40-transformed skin fibroblast from patient with
RBSV1A
Human epidermis
Fibroblast-like
A-9
Mouse muscle
Fibroblast-like
Lesch-Nyhan syndrome (22). SV40-transformed skin fibroblasts from patient with inborn metabolic disorder, "cystinosis." Derived from "L-cells" by selection for resistance to
HeLa
Human cervical carcinoma Human nasopharyngeal carcinoma Human melanoma
Epithelioid
(24)
Epithelioid
Cells have quasi-diploid human karyotype (25).
Epithelioid
Human spleen
Lymphoid
Human buccal mucosa Human skin
Fibroblast-like Fibroblast-like
Cell line donated by Dr. Grace Cannon, Litton.Bionetics Research Laboratories, Bethesda, Md. Cells were derived from a patient free from malignant disease but bearing antibodies to an Epstein-Barr type virus (9). (26) Cells from patient with Fanconi's anemia.
8-aza-guanine (23). RPMI-2650 BRL-4130 Hiene WI-L2 W18VA2 SV80
infected with lines LNSV, W18VA2, and SV80 exhibited swellings of 0.3-0.4 cm diameter, which were apparent 1 day after injection and disappeared by 12 days after injecTable 2. Tumorigenicity of heterologous cells in nude mice
No. of mice Cell line Human lines RPMI2650 BRL-4143 HeLa WI-L2 Human SV40transformed lines VA2-8-aza-Gr RBSV3
No. of mice
No. of cells
to
injected
injected
per mouse
form tumors
10 10 10 5
9 5 10 2
106 0.5 x 106 106 2 x 106
9 5
106 106 (precrisis
4
4 x 106 (precrisis
0 0
cells) RBSV1A
0
cells) RBSV1A LNSV W18VA2 SV80 VA2 8-aza-Gr LNSV W18VA2 Mouse lines SVT2 A-9
6
6 5 6 5 3 1 10 10
2x 2 x 2x 6x
106 (postcrisis cells) 106 106 106 106 107 107 106 106
0 0 0 0 0 0 0
10 10
tion. Figure LA illustrates a tumor induced by 106 human lymphocytes of the WI-L2 line. The transient swelling formed by inoculation of 2 X 106 SV4O-transformed human fibroblasts (SV80 line) is depicted in Fig. 1B. Two of the SV40-transformed human lines were retested for tumorigenicity in nude mice with inocula 10- to 20-fold in excess of those used with human cancer cell lines. Table 2 shows that the larger inocula were similarly ineffective in provoking tumor formation. When 107 LNSV cells were injected, swellings to a diameter of 0.5-0.6 cm were apparent after 1 day and disappeared within 7 days. Two lines, RBSV3 and TBSV1A, were injected at early passages after SV40 infection, and these cells subsequently entered into an apparent "senescent crisis" (10) in culture. The RBSV1A line was tested for tumorigenicity at a later passage when the cells had recovered from the crisis. The "postcrisis" RBSV1A cultures were similarily nontumorigenic. Lines VA2-8-aza-Gr and LNSV are deficient in hypoxanthine phosphoribosylpyrophosphorylase activity (Table 1). To determine whether this enzyme is crucial to tumor formation in nude mice, we injected several animals with the A-9 substrain of mouse L cells which is deficient in the enzyme activity. A-9 cells were tumorigenic in 100% of the nude mice tested (Table 2); this result corroborated the findings of Freedman and Shin (11), who demonstrated, in addition, that cells derived from A-9 induced tumors are still deficient in hypoxanthine phosphoribosylpyrophosphorylase. SV40-transformed mouse cells (the SVT2 line) are tumorigenic in nude mice. Growth of human tumors in nude mice that have failed to grow tumors from SV40-transformed human
cells
Nude mice do not possess the cell-mediated immune response and do not reject heterologous skin grafts (12). However, these animals do form humoral antibodies to heterolo-
Cell Biology: Stiles et al.
Proc. Nat. Acad. Sc. USA 72 (1975)
4973
*
::%.'.C 11
f
A.
il
..
B.
FIG. 1. Appearance of human tumors and abortive nodules in nude mice. (A) Nude mouse with tumor formed from injection of WI-L2 cells photographed 6 weeks after injection. (B) Nude mice with abortive nodules induced by human SV40 transformants (the SV80 line). The animals were photographed one week after injection of 2 x 106 cells subcutaneously. The upper mouse was untreated and the lower mouse was treated with antilymphocyte serum as described in Materials and Methods. Arrows indicate the transient swellings (abortive nodules) which appeared at the site of injection. gous
skin grafts (13), and
some
evidence suggests that nude
mice have a residual capacity to reject heterologous cells (14). To examine the possibility that SV40-transformed human cells might be more efficient inducers of residual immune
capabilities than authentic cancer cells,
we
tested the
capacity of the human viral transformants to "vaccinate" nude mice against authentic human tumor cells. Table 3 shows that nude mice that did not develop tumors from injections of SV40-transformed human fibroblasts formed tu-
when subsequently injected with authentic human cells on the contralateral flank. Additionally, SV40transformed mouse cells (the SVT2 line) formed tumors in nude mice that had failed to form tumors from injections of SV40-transformed human cells. Growth behavior of SV40 human transformants in "supersuppressed" nude mice Nude mice form humoral antibodies to heterologous skin grafts, and it has been shown that these humoral antibodies can induce necrosis in skin allografts when the mice are injected with rabbit complement (13). To examine the possibility that human SV40-transformed cells were for some reason more vulnerable than other types of human tissue to the action of humoral antibodies and endogenous mouse complement, we tested for tumorigenicity in nude mice that had been treated to suppress humoral immune responses. Table 4 shows that human SV40-transformed cells do not form tumors in neonatal nude mice, in nude mice treated with 700 rads of whole body irradiation, or in nude mice treated with antilymphocyte serum. Antilymphocyte serum had no effect mors cancer
on the size or duration of the transient swelling induced by inoculation of SV80 cells into nude mice (Fig. 1B). In an attempt to reproduce findings of Stanbridge et al. that human SV40 transformants were tumorigenic in mice
treated with antilymphocyte serum (4), the tumorigenicity of W18VA2 cells and also of SV80 cells was tested in Balb/c mice treated with antilymphocyte serum. Palpable nodules arose in the treated Balb/c mice that were similar in size and duration to those produced in either control mice or nude mice treated with antilymphocyte serum. The potency of the antilymphocyte serum used in these experiments was confirmed by the ability of treated Balb/c mice to form tumors from an inoculation of 106 HeLa cells; untreated Balb/ c mice did not form tumors from HeLa cells. Comparative morphology and growth behavior of mouse cells and human cells infected with SV40 SV40 infection of 3T3 cells in culture generates characteristic changes in growth behavior and morphology which frequently correlate with enhanced tumorigenic potential (15, 16). Relative to normal 3T3 cells, SV40-transformed cultures show a decrease in serum growth requirement, an increased saturation density and enhanced ability to plate in agar suspension, and/or loss of density dependent inhibition of growth. Morphologically, SV40-transformed 3T3 cells tend to overlap each other in tissue culture dishes and grow in multilayers, whereas control cultures do not overlap (17). Table 4. Behavior of SV40-transformed cells in baby nude mice and "supersuppressed" nude mice
Table 3. Growth of tumors in nude mice subsequent to challenge with SV40-transformed human cells
First inoculation
Second inoculation
RBSV3 RBSV1A
RPMI-2650 HeLa
VA2-8-aza-Gr
SVT2
No. of tumors formed/no. of
Cell line
No. of mice injected
mice from second inoculation
VA2-8-aza-Gr VA2-8-aza-Gr
5 (1-day-old mice) 4 (mice pretreated with 700 rads whole body irradiation) 2 (mice treated with
3/3 6/6 2/2
Nude mice that, after 4 months, had failed to form tumors from inocula of 106 RBSV1A (postcrisis cells), RBSV3 (precrisis cells) or VA2-8-aza-Gr were inoculated with 106 cells subcutaneously on the contralateral side.
W18VA2
No. of No. of cells mice to injected form per mouse tumors 106 6 x 106
0 0
2 x 106
0
2 x 106
0
antilymphocyte SV80
serum) 2 (mice treated with antilymphocyte serum)
4974
Proc. Nat. Acad. Sci. USA 72 (1975)
Cell Biology: Stiles et al.
e1a
Table 5. Growth behavior of tumorigenic and nontumorigenic mouse and human cells in vitro
Cell line
Saturation density (cells/cm2)
HeLa SVT2 3T3 LNSV W18 VA2
1.08 7.29 3.91 7.35 2.5
x x x x x
106 105 104 104 105
Ratio doubling time 10% FBS: 1.0% FBS*
% Cloning efficiency in Methocel
0.89 0.38 0 0.90 0
12
