Proof of In Vivo Fusion of Murine Tumor Cells With Host Cells by Universal Fusers: Brief Communication 1,2 Rosalie Ber, 3,
4, 5
Francis Wiener, 3 and Eva-Maria Fenyo 3
ABSTRACT-Two mouse cell lines (A9HT CI.3C and 501-1), each carrying both a recessive and a dominant mutation (and therefore designated "universal fusers"), were utilized for selection of in vivo hybrids from tumors produced intraperitoneally and subcutaneously in appropriate hosts. The double selective medium eliminated both parental (tumor and host) cells but allowed the survival and proliferation of the fused product. This proved that hybridization between tumor and host cells occurred in vivo.-J Natl Cancer Inst 60: 931-933, 1978.
MATERIALS AND METHODS Tumor cell iines.-A9HT clone 3C was developed by mutagenesis of a highly tumorigenic variant of the A9 subline of C3H mouse L-cells with ethyl-methanesulfonate. A9HT lacks the genetic information for the enzyme HGPRT and therefore cannot grow in HAT medium (11, 12). The m utagenized cells were selected stepwise for resistance to ouabain - a highly toxic membrane Na+K+ ATPase poison (13). The cloned and recloned line, designated CI.3C, is resistant to 6 mM ouabain and 50 ILg TG/ml. The chromosome constitution of CI.3C is similar to that of A9HT as described (14). The modal number of chromosomes for CI.3C is 53 with 23-24 biarmed chromosomes that serve as the marker chromosomes of the line. Cell line 501-1 (a gift from Dr. J. M. Eisenstadt, Yale University School of Medicine, New Haven, Conn.) is also a subline of A9 cells and carries a "dominant" cytoplasmic genetic marker - CAP resistance (15). From the original 501-1 cell line, which is resistant to 50 ILg CAP/ml, we selected, stepwise, cells resistant to 500 ILg CAP/ml. Without further mutagenization, 501-1 cells VOL. 60, NO.4, APRIL 1978
RESULTS Growth of Parental Cells on Selective Media A9HT CI.3C cells grown in the presence or absence ABBREVIATIONS USED: HGPRT = hypoxanthine-guanine-phosphoribosyl transferase; HAT = hypoxanthine, amethopterin, and thymidine; TG = thioguanine; CAP = chloramphenicol.
Received October 20,1977; accepted November 7,1977. Supported by Public Health Service research grant 5ROI CA140554-04 from the National Cancer Institute and by funds from the Swedish Cancer Society. 3 Department of Tumor Biology, Karolinska Institutet, S-104 01 Stockholm 60, Sweden. 4 Recipient of a long-term EMBO fellowship and a Yad Avi HaYishuv stipend. 5 Address reprint requests to Dr. Ber at Technion -Abba Khoushi, School of Medicine, Haifa, Israel. 1
2
931
J NATL CANCER INST
Downloaded from http://jnci.oxfordjournals.org/ at Yale University on May 18, 2015
Fusion of mouse tumor cells with host cells has been shown to be a common event (1-4). With appropriate selective measures, in vivo hybrids can be obtained from various tumors-both ascites and solid (5-7). Experiments designed to determine the type of host cell implicated in the fusion revealed that bone marrowderived cells (presumably macrophages) are the most common host cells involved in vivo fusion (8, 9). C3 production of hybrids selected from ascites tumors growing in chimeric mice also lends support to the macrophage presumption (10). Attempts to clarify the "in vitro" versus "in vivo" hybridization possibilities involved high-dilution plating (2), but the final formal proof was still lacking. Using highly tumorigenic cell lines carrying both a dominant and a recessive marker, we demonstrated that fusion of host cells with tumor cells occurred in vivo.
were exposed to increasing concentrations of ouabain, and a line resistant to 3 mM ouabain, thus carrying three genetic markers, was selected. The chromosome constitution of this line is similar to that of the A9 line (16); i.e., it has a modal number of 55 chromosomes with 25 biarmed chromosomes, which, as in the case of A9HT CI.3C, serve as the marker chromosomes of the line. Chromosome preparations.-Metaphase spreads of cultured cells were prepared by the air-drying method (17). Colcemid (GIBCO, Grand Island, N. Y.) was added to the culture in a final concentration of 0.04 ILg/ ml, 2-3 hours before the cells were harvested. After hypotonic treatment (0.075 M KCI solution for 5 min), the pellet was fixed in acetic acid-methanol (1:3), spread on chilled wet slides, dried, and stained with alkaline Giemsa solution. We calculated the total number of chromosomes by counting chromosomes directly under the microscope. Isoantisera and adsorption tests.-Groups of 10-12 adult inbred C3H mice and hybrids of the appropriate genetic constitution (table 1) were immunized with pooled cell suspensions prepared from spleen, kidney, and liver of mice carrying the requisite H-2 complex. The mice were inoculated sc once every 2 weeks for 12-14 weeks. The activity and specificity of all isoantisera were checked by cytotoxicity tests against lymph node cells of the appropriate genotype as described (18). To measure the concentration of H-2 antigens on the surface of the cell lines, quantitative absorption tests were performed as described previously (18). The ratio of bound to free antibody was calculated by the method of Reif (19).
BER, WIENER, AND FENVO
932 TABLE l.~enotype,
chromosome constitution, and H-2 isoantigen expression of parental and hybrid cells
Biarmed chromosomes
Total chromosomes b Genotype of host
Tumor type a
C3H C3H (C3H x (C3H x (CBA x (CBA x (C3H x (C3H x (C3H x (C3H x (C3H x (C3H x (C3H x
A9HToub 501-1 A9HToub-H7A ascites A9HToub-H7B ascites A9HToub- B7 A ascites A9HToub-B7B ascites A9HToub-HSA solid tumor A9HToub-HCA solid tumor 501-1-HCA solid tumor 501-1-HCB solid tumor 501-1-HCC ascites 501-1-H7B ascites 501-1-H7C solid tumor a b
C57BLlF, C57BLlF, C57BLlF, C57BLlF, A.SWlF, A.CAlF, A.CA)F, A.CA)F, A.CA)F, C57BLlF, C57BLlF,
Range
Modal No.
Range
Modal No.
48-57 5Q-54 72-97 88-99 82-108 82-101 71-96 86-92 88-96 90-98 91-96 90-94 86-99
53 53 93 89 91 95 78 90 92 93 91 93 90
18-24 20-25 12-21 20-27 12-23 15-24 11-18 15-25 20-27 21-27 19-28 20-28 22-26
22 23 17 23 20 20 16 22 24 25 24 22 24
H-2 isoantigen complex H_2k H_2k H_2kH_2b H_2kH_2b H_2 kH_2b H_2kH_2h H_2kH_2s H_2kH_2f H_2kH_2f H-2kH-2' H-2kH-2' H_2kH_2b H_2kH_2b
A9HToub indicates ouabain-resistant A9HT cells. For each hybrid 20 metaphase plates were examined.
-6
H-2 f
H-2 s
H-2 b
4
2
5 2 . I- IlI- Ix X zX z zz en z i-~en en en en en 0l0l0l
ct ~ctctct ~ctct ~ct ct ct
ID 0 ID ct ID lDu ID -ct -·ct ct III III ct III (.J U "ao 0 I'I'- U 'au UU I'0 I'~UUUI'-UI'-
, ,. ..
:r:zz :r:zz:r: g Z:r:Zz c zz z:r: z z:r: II
II
III
II
II
I•I
.Q .Q ;- TT .a Jc: ;.a:0 Jc: .c.T"T~.&l :0 :0 Q.,,:0 Q :0:0 0
00 0 00 EOOOO EO 0 EO 11)11) :...11) >. It) 11)11) It) It) :...11)
ct ct
!it
10 II) It)
00 00 OO II- IIlI-
X z zen ZX en en 0l0l en ct ct ct ct
I.-Absorptive capacities of cell lines are shown for the following antisera: C3H anti-C57BL X DBA/2 anti-C3H (H-2k), C3H antiC57BL (H_2b), A.CA X C3H anti-A.SW (H-2s). and A X CBA anti-A.CA (H-2f). Log llG indicates the amount of antibody bound per cell when 50% of cytotoxic activity of antiserum had been bound. Numbers at top of columns denote number of absorption tests.
TEXT-FIGURE
of 3 mM ouabain had the same doubling time. This held true for 501-1 cells in the presence or absence of 500 ILg CAP/ml. However, no living multiplying cells of either mutant line could be detected upon transfer to HAT medium. Mouse embryo fibroblasts exposed to 3 mM ouabain were incapable of multiplying. After being plated for 24 hours in medium containing 3 mM ouabain, the fibroblasts became trypan blue-positive. Although a small percentage of these cells remained attached to the
J
NATL CANCER INST
surface up to 72 hours after plating, they lacked the refraction of living cells, as judged by phase microscopy, became granulated, and at 96 hours nothing but debris could be detected. Mouse embryo fibroblasts grown in the presence of 50 ILg CAP/ml were capable of undergoing approximately four divisions, after which the culture became static. After 1 week no living cells could be detected. However, grown in the presence of 500 ILg CAP/ml, normal mouse fibroblasts disintegrated within 48 hours. VOL. 60, NO.4, APRIL 1978
Downloaded from http://jnci.oxfordjournals.org/ at Yale University on May 18, 2015
H-2 k
IN VIVO FUSION: UNIVERSAL FUSER CELLS
Selection of Chromosome Constitution and Antigen (H-2) Characteristics of Hybrid Cells
DISCUSSION "Universal fuser" is the proposed designation for a cell carrying both a dominant and a recessive genetic marker. Cells, A9HT Cl.3C and 501-1 qualify for this designation. The cells, which are/resistant to 50 p.,g TG/ ml, expire in HAT medium; no revertants have been detected even when plated at 10 7 cells per flask. HAT medium therefore selects completely against the tumor cells deficient in HGPRT. However, A9HT Cl.3C cells can grow and multiply "normally" in the presence of concentrations of ouabain that are extremely toxic to nonmutagenized mouse cell lines and/or mouse fibroblasts. Similarly, 501-1 cells grow as well in the presence of 500 p.,g CAP/ml as without the drug, whereas normal mouse cells are completely destroyed in the presence of 500 p.,g CAP/ml. The major arguments of the initial in vivo hybrid reports were related to the method of selection. The dilution plating experiments (2), which showed hybrids even when as few as 102 cells were plated, essentially ruled out the probability of fusion having occurred in vitro. However, since no selective pressure was put on the host cells, the argument arose as to the possibility of the fusion having occurred during the selection process in vitro. Our data clearly show that tumor cells fuse with host cells in vivo. On the double-selective medium of HAT and either 3 mM ouabain or 500 p.,g CAP/ml, the only
VOL. 60, NO.4, APRIL 1978
cell that can survive and multiply is a hybrid cell containing genetic information for resistance to both HAT and either ouabain or CAP. Such a cell could arise only by fusion in vivo between the mutant tumor cells and host cells. Using universal fusers has enabled us to embark on kinetic studies of in vivo hybridization. This proves to be an early event that occurs within the first 24 hours of ip tumor inoculation. The kinetic studies of in ViVO hybridization will be the subject of a future report.
REFERENCES (1) GOLDENBERG DM, BHAN RD, PAVIA RA: In vivo human-hamster somatic cell fusion indicated by glucose-6-phosphate dehydrogenase and lactate de hydrogenase profiles. Cancer Res 31:1148-1152,1971 (2) WIENER F, FENYO EM, KLEIN G, et al: Fusion of tumour cells with host cells. Nature [New Bioi] 238: 155-159, 1972 (3) Hu F, PASZTOR LM.: In vivo hybridization of cultured melanoma cells and isogenic normal mouse cells. Differentiation 4:93-97, 1975 (4) AVILES D, JAMI J. ROUSSET JP, et al: Tumor x host cell hybrids in the mouse: Chromosomes from the normal cell parent maintained in malignant hybrid tumors. J Nat! Cancer Inst 58: 1391-1399, 1977 (5) JANZEN WH, MILLMAN AP. THURSTON OG: Hybrid cells in solid tumors. Cancer 27:445-449, 1971 (6) FENYO EM, WIENER F, KLEIN G, et al: Selection of tumor-host cell hybrids from polyoma virus- and methylcholanthreneinduced sarcomas. J Nat! Cancer Inst 51:1865-1877,1973 (7) GOLDENBERG DM, PAVIA RA, TSAO MC: In vivo hybridization of human tumour and normal hamster cells. Nature 250:649-651, 1974 (8) WIENER F, FEN YO EM, KLEIN G: Tumor-host cell hybrids in radio-chimeras. Proc Nat! Acad Sci USA 71:148-152,1974 (9) WIENER F, FENYO EM, KLEIN G, et al: Tumor-host cell hybrids in radiochimeras reconstituted with bone marrow and thymus grafts. Somatic Cell Genet 2:81-92, 1976 (10) SZPIRER C, SZPIRER J, WIENER F: The expression of differentiated functions in somatic cell hybrids: Retention and activation of C3 production. Cell Differ 5:139-149,1976 (11) SZYBALSKA E, SZYBALSKI W: Genetics of human cell lines. IV. DNA-mediated heritable transformation of biochemical trait. Proc Nat! Acad Sci USA 48:2026-2034,1962 (12) LITTLEFIELD JW: Three degrees of guanylic acid-inosinic acid pyrophosphorylase deficiency in mouse fibroblasts. Nature 203: 1142-1144, 1964 (13) KUCHERLAPATI RS, BAKER RM, RUDDLE FH: Ouabain as a selective agent in the isolation of somatic cell hybrids. Cytogenet Cell Genet 14:362-363, 1975 (14) WIENER F, KLEIN G, HARRIS H: The analysis of malignancy by cell fusion. VI. Hybrids between different tumour cells. J Cell Sci 16: 189-198, 1974 (15) BUNN CL, WALLACE DC, EISENSTADT JM: Cytoplasmic inheritance of chloramphenicol resistance in mouse tissue culture cells. Proc Nat! Acad Sci USA 71:1681-1685,1974 (16) KLEIN G, BREGULA U, WIENER F, et al: The analysis of malignancy by cell fusion. I. Hybrids between tumor cells and L cell derivatives. J Cell Sci 8:659-672,1971 (17) ROTHFELS KH, SIMINOVITCH L: An air-drying technique for flattening chromosomes in mammalian cells grown in vitro. Stain TechnoI33:73-77, 1958 (18) KLEIN G, GARS U: Isoantigen expression in hybrid mouse cells. Exp Cell Res 62:149-160, 1970 (19) REIF AE: An experimental test of two general relationships to describe the adsorption of antibodies by cells and tissues. Immunochemistry 3:267-278, 1966
J
NATL CANCER INST
Downloaded from http://jnci.oxfordjournals.org/ at Yale University on May 18, 2015
Six tumor-host cell hybrids were isolated from A9HT Cl.3C and five from 501-1 tumors growing intraperitoneally or subcutaneously. The selection of hybrid cells from enzyme-deficient tumors was detailed previously (2, 6). In the present work, A9HT tumors were explanted into HAT medium containing 3 mM ouabain, whereas 501-1 tumors were ex planted in HAT medium containing 500 p.,g CAP/ml. Both A9HT and 501-1 tumors are HGPRT-deficient and cannot survive in HA T medium. As shown above, normal host cells are killed either by ouabain or by CAP. After 8-14 days, a few colonies appeared that were resistant to both ouabain and HAT or to both HAT and CAP. The cells in these colonies proved to be hybrids. Table 1 summarizes the chromosome constitution and the antigen characteristics of the original tumor lines and the tumor host-cell hybrids derived from them. The modal chromosome number of the selected hybrids was close to what would be expected from the sum of the chromosomes of the tumor cells (53-55) and diploid mouse cells (40). Text-figure 1 shows the absorptive capacities of tumor and hybrid cells from various antisera. The results of both chromosome analyses and absorption tests proved the hybridity of the cells growing in the double-selective medium.
933
4, 5
Francis Wiener, 3 and Eva-Maria Fenyo 3
ABSTRACT-Two mouse cell lines (A9HT CI.3C and 501-1), each carrying both a recessive and a dominant mutation (and therefore designated "universal fusers"), were utilized for selection of in vivo hybrids from tumors produced intraperitoneally and subcutaneously in appropriate hosts. The double selective medium eliminated both parental (tumor and host) cells but allowed the survival and proliferation of the fused product. This proved that hybridization between tumor and host cells occurred in vivo.-J Natl Cancer Inst 60: 931-933, 1978.
MATERIALS AND METHODS Tumor cell iines.-A9HT clone 3C was developed by mutagenesis of a highly tumorigenic variant of the A9 subline of C3H mouse L-cells with ethyl-methanesulfonate. A9HT lacks the genetic information for the enzyme HGPRT and therefore cannot grow in HAT medium (11, 12). The m utagenized cells were selected stepwise for resistance to ouabain - a highly toxic membrane Na+K+ ATPase poison (13). The cloned and recloned line, designated CI.3C, is resistant to 6 mM ouabain and 50 ILg TG/ml. The chromosome constitution of CI.3C is similar to that of A9HT as described (14). The modal number of chromosomes for CI.3C is 53 with 23-24 biarmed chromosomes that serve as the marker chromosomes of the line. Cell line 501-1 (a gift from Dr. J. M. Eisenstadt, Yale University School of Medicine, New Haven, Conn.) is also a subline of A9 cells and carries a "dominant" cytoplasmic genetic marker - CAP resistance (15). From the original 501-1 cell line, which is resistant to 50 ILg CAP/ml, we selected, stepwise, cells resistant to 500 ILg CAP/ml. Without further mutagenization, 501-1 cells VOL. 60, NO.4, APRIL 1978
RESULTS Growth of Parental Cells on Selective Media A9HT CI.3C cells grown in the presence or absence ABBREVIATIONS USED: HGPRT = hypoxanthine-guanine-phosphoribosyl transferase; HAT = hypoxanthine, amethopterin, and thymidine; TG = thioguanine; CAP = chloramphenicol.
Received October 20,1977; accepted November 7,1977. Supported by Public Health Service research grant 5ROI CA140554-04 from the National Cancer Institute and by funds from the Swedish Cancer Society. 3 Department of Tumor Biology, Karolinska Institutet, S-104 01 Stockholm 60, Sweden. 4 Recipient of a long-term EMBO fellowship and a Yad Avi HaYishuv stipend. 5 Address reprint requests to Dr. Ber at Technion -Abba Khoushi, School of Medicine, Haifa, Israel. 1
2
931
J NATL CANCER INST
Downloaded from http://jnci.oxfordjournals.org/ at Yale University on May 18, 2015
Fusion of mouse tumor cells with host cells has been shown to be a common event (1-4). With appropriate selective measures, in vivo hybrids can be obtained from various tumors-both ascites and solid (5-7). Experiments designed to determine the type of host cell implicated in the fusion revealed that bone marrowderived cells (presumably macrophages) are the most common host cells involved in vivo fusion (8, 9). C3 production of hybrids selected from ascites tumors growing in chimeric mice also lends support to the macrophage presumption (10). Attempts to clarify the "in vitro" versus "in vivo" hybridization possibilities involved high-dilution plating (2), but the final formal proof was still lacking. Using highly tumorigenic cell lines carrying both a dominant and a recessive marker, we demonstrated that fusion of host cells with tumor cells occurred in vivo.
were exposed to increasing concentrations of ouabain, and a line resistant to 3 mM ouabain, thus carrying three genetic markers, was selected. The chromosome constitution of this line is similar to that of the A9 line (16); i.e., it has a modal number of 55 chromosomes with 25 biarmed chromosomes, which, as in the case of A9HT CI.3C, serve as the marker chromosomes of the line. Chromosome preparations.-Metaphase spreads of cultured cells were prepared by the air-drying method (17). Colcemid (GIBCO, Grand Island, N. Y.) was added to the culture in a final concentration of 0.04 ILg/ ml, 2-3 hours before the cells were harvested. After hypotonic treatment (0.075 M KCI solution for 5 min), the pellet was fixed in acetic acid-methanol (1:3), spread on chilled wet slides, dried, and stained with alkaline Giemsa solution. We calculated the total number of chromosomes by counting chromosomes directly under the microscope. Isoantisera and adsorption tests.-Groups of 10-12 adult inbred C3H mice and hybrids of the appropriate genetic constitution (table 1) were immunized with pooled cell suspensions prepared from spleen, kidney, and liver of mice carrying the requisite H-2 complex. The mice were inoculated sc once every 2 weeks for 12-14 weeks. The activity and specificity of all isoantisera were checked by cytotoxicity tests against lymph node cells of the appropriate genotype as described (18). To measure the concentration of H-2 antigens on the surface of the cell lines, quantitative absorption tests were performed as described previously (18). The ratio of bound to free antibody was calculated by the method of Reif (19).
BER, WIENER, AND FENVO
932 TABLE l.~enotype,
chromosome constitution, and H-2 isoantigen expression of parental and hybrid cells
Biarmed chromosomes
Total chromosomes b Genotype of host
Tumor type a
C3H C3H (C3H x (C3H x (CBA x (CBA x (C3H x (C3H x (C3H x (C3H x (C3H x (C3H x (C3H x
A9HToub 501-1 A9HToub-H7A ascites A9HToub-H7B ascites A9HToub- B7 A ascites A9HToub-B7B ascites A9HToub-HSA solid tumor A9HToub-HCA solid tumor 501-1-HCA solid tumor 501-1-HCB solid tumor 501-1-HCC ascites 501-1-H7B ascites 501-1-H7C solid tumor a b
C57BLlF, C57BLlF, C57BLlF, C57BLlF, A.SWlF, A.CAlF, A.CA)F, A.CA)F, A.CA)F, C57BLlF, C57BLlF,
Range
Modal No.
Range
Modal No.
48-57 5Q-54 72-97 88-99 82-108 82-101 71-96 86-92 88-96 90-98 91-96 90-94 86-99
53 53 93 89 91 95 78 90 92 93 91 93 90
18-24 20-25 12-21 20-27 12-23 15-24 11-18 15-25 20-27 21-27 19-28 20-28 22-26
22 23 17 23 20 20 16 22 24 25 24 22 24
H-2 isoantigen complex H_2k H_2k H_2kH_2b H_2kH_2b H_2 kH_2b H_2kH_2h H_2kH_2s H_2kH_2f H_2kH_2f H-2kH-2' H-2kH-2' H_2kH_2b H_2kH_2b
A9HToub indicates ouabain-resistant A9HT cells. For each hybrid 20 metaphase plates were examined.
-6
H-2 f
H-2 s
H-2 b
4
2
5 2 . I- IlI- Ix X zX z zz en z i-~en en en en en 0l0l0l
ct ~ctctct ~ctct ~ct ct ct
ID 0 ID ct ID lDu ID -ct -·ct ct III III ct III (.J U "ao 0 I'I'- U 'au UU I'0 I'~UUUI'-UI'-
, ,. ..
:r:zz :r:zz:r: g Z:r:Zz c zz z:r: z z:r: II
II
III
II
II
I•I
.Q .Q ;- TT .a Jc: ;.a:0 Jc: .c.T"T~.&l :0 :0 Q.,,:0 Q :0:0 0
00 0 00 EOOOO EO 0 EO 11)11) :...11) >. It) 11)11) It) It) :...11)
ct ct
!it
10 II) It)
00 00 OO II- IIlI-
X z zen ZX en en 0l0l en ct ct ct ct
I.-Absorptive capacities of cell lines are shown for the following antisera: C3H anti-C57BL X DBA/2 anti-C3H (H-2k), C3H antiC57BL (H_2b), A.CA X C3H anti-A.SW (H-2s). and A X CBA anti-A.CA (H-2f). Log llG indicates the amount of antibody bound per cell when 50% of cytotoxic activity of antiserum had been bound. Numbers at top of columns denote number of absorption tests.
TEXT-FIGURE
of 3 mM ouabain had the same doubling time. This held true for 501-1 cells in the presence or absence of 500 ILg CAP/ml. However, no living multiplying cells of either mutant line could be detected upon transfer to HAT medium. Mouse embryo fibroblasts exposed to 3 mM ouabain were incapable of multiplying. After being plated for 24 hours in medium containing 3 mM ouabain, the fibroblasts became trypan blue-positive. Although a small percentage of these cells remained attached to the
J
NATL CANCER INST
surface up to 72 hours after plating, they lacked the refraction of living cells, as judged by phase microscopy, became granulated, and at 96 hours nothing but debris could be detected. Mouse embryo fibroblasts grown in the presence of 50 ILg CAP/ml were capable of undergoing approximately four divisions, after which the culture became static. After 1 week no living cells could be detected. However, grown in the presence of 500 ILg CAP/ml, normal mouse fibroblasts disintegrated within 48 hours. VOL. 60, NO.4, APRIL 1978
Downloaded from http://jnci.oxfordjournals.org/ at Yale University on May 18, 2015
H-2 k
IN VIVO FUSION: UNIVERSAL FUSER CELLS
Selection of Chromosome Constitution and Antigen (H-2) Characteristics of Hybrid Cells
DISCUSSION "Universal fuser" is the proposed designation for a cell carrying both a dominant and a recessive genetic marker. Cells, A9HT Cl.3C and 501-1 qualify for this designation. The cells, which are/resistant to 50 p.,g TG/ ml, expire in HAT medium; no revertants have been detected even when plated at 10 7 cells per flask. HAT medium therefore selects completely against the tumor cells deficient in HGPRT. However, A9HT Cl.3C cells can grow and multiply "normally" in the presence of concentrations of ouabain that are extremely toxic to nonmutagenized mouse cell lines and/or mouse fibroblasts. Similarly, 501-1 cells grow as well in the presence of 500 p.,g CAP/ml as without the drug, whereas normal mouse cells are completely destroyed in the presence of 500 p.,g CAP/ml. The major arguments of the initial in vivo hybrid reports were related to the method of selection. The dilution plating experiments (2), which showed hybrids even when as few as 102 cells were plated, essentially ruled out the probability of fusion having occurred in vitro. However, since no selective pressure was put on the host cells, the argument arose as to the possibility of the fusion having occurred during the selection process in vitro. Our data clearly show that tumor cells fuse with host cells in vivo. On the double-selective medium of HAT and either 3 mM ouabain or 500 p.,g CAP/ml, the only
VOL. 60, NO.4, APRIL 1978
cell that can survive and multiply is a hybrid cell containing genetic information for resistance to both HAT and either ouabain or CAP. Such a cell could arise only by fusion in vivo between the mutant tumor cells and host cells. Using universal fusers has enabled us to embark on kinetic studies of in vivo hybridization. This proves to be an early event that occurs within the first 24 hours of ip tumor inoculation. The kinetic studies of in ViVO hybridization will be the subject of a future report.
REFERENCES (1) GOLDENBERG DM, BHAN RD, PAVIA RA: In vivo human-hamster somatic cell fusion indicated by glucose-6-phosphate dehydrogenase and lactate de hydrogenase profiles. Cancer Res 31:1148-1152,1971 (2) WIENER F, FENYO EM, KLEIN G, et al: Fusion of tumour cells with host cells. Nature [New Bioi] 238: 155-159, 1972 (3) Hu F, PASZTOR LM.: In vivo hybridization of cultured melanoma cells and isogenic normal mouse cells. Differentiation 4:93-97, 1975 (4) AVILES D, JAMI J. ROUSSET JP, et al: Tumor x host cell hybrids in the mouse: Chromosomes from the normal cell parent maintained in malignant hybrid tumors. J Nat! Cancer Inst 58: 1391-1399, 1977 (5) JANZEN WH, MILLMAN AP. THURSTON OG: Hybrid cells in solid tumors. Cancer 27:445-449, 1971 (6) FENYO EM, WIENER F, KLEIN G, et al: Selection of tumor-host cell hybrids from polyoma virus- and methylcholanthreneinduced sarcomas. J Nat! Cancer Inst 51:1865-1877,1973 (7) GOLDENBERG DM, PAVIA RA, TSAO MC: In vivo hybridization of human tumour and normal hamster cells. Nature 250:649-651, 1974 (8) WIENER F, FEN YO EM, KLEIN G: Tumor-host cell hybrids in radio-chimeras. Proc Nat! Acad Sci USA 71:148-152,1974 (9) WIENER F, FENYO EM, KLEIN G, et al: Tumor-host cell hybrids in radiochimeras reconstituted with bone marrow and thymus grafts. Somatic Cell Genet 2:81-92, 1976 (10) SZPIRER C, SZPIRER J, WIENER F: The expression of differentiated functions in somatic cell hybrids: Retention and activation of C3 production. Cell Differ 5:139-149,1976 (11) SZYBALSKA E, SZYBALSKI W: Genetics of human cell lines. IV. DNA-mediated heritable transformation of biochemical trait. Proc Nat! Acad Sci USA 48:2026-2034,1962 (12) LITTLEFIELD JW: Three degrees of guanylic acid-inosinic acid pyrophosphorylase deficiency in mouse fibroblasts. Nature 203: 1142-1144, 1964 (13) KUCHERLAPATI RS, BAKER RM, RUDDLE FH: Ouabain as a selective agent in the isolation of somatic cell hybrids. Cytogenet Cell Genet 14:362-363, 1975 (14) WIENER F, KLEIN G, HARRIS H: The analysis of malignancy by cell fusion. VI. Hybrids between different tumour cells. J Cell Sci 16: 189-198, 1974 (15) BUNN CL, WALLACE DC, EISENSTADT JM: Cytoplasmic inheritance of chloramphenicol resistance in mouse tissue culture cells. Proc Nat! Acad Sci USA 71:1681-1685,1974 (16) KLEIN G, BREGULA U, WIENER F, et al: The analysis of malignancy by cell fusion. I. Hybrids between tumor cells and L cell derivatives. J Cell Sci 8:659-672,1971 (17) ROTHFELS KH, SIMINOVITCH L: An air-drying technique for flattening chromosomes in mammalian cells grown in vitro. Stain TechnoI33:73-77, 1958 (18) KLEIN G, GARS U: Isoantigen expression in hybrid mouse cells. Exp Cell Res 62:149-160, 1970 (19) REIF AE: An experimental test of two general relationships to describe the adsorption of antibodies by cells and tissues. Immunochemistry 3:267-278, 1966
J
NATL CANCER INST
Downloaded from http://jnci.oxfordjournals.org/ at Yale University on May 18, 2015
Six tumor-host cell hybrids were isolated from A9HT Cl.3C and five from 501-1 tumors growing intraperitoneally or subcutaneously. The selection of hybrid cells from enzyme-deficient tumors was detailed previously (2, 6). In the present work, A9HT tumors were explanted into HAT medium containing 3 mM ouabain, whereas 501-1 tumors were ex planted in HAT medium containing 500 p.,g CAP/ml. Both A9HT and 501-1 tumors are HGPRT-deficient and cannot survive in HA T medium. As shown above, normal host cells are killed either by ouabain or by CAP. After 8-14 days, a few colonies appeared that were resistant to both ouabain and HAT or to both HAT and CAP. The cells in these colonies proved to be hybrids. Table 1 summarizes the chromosome constitution and the antigen characteristics of the original tumor lines and the tumor host-cell hybrids derived from them. The modal chromosome number of the selected hybrids was close to what would be expected from the sum of the chromosomes of the tumor cells (53-55) and diploid mouse cells (40). Text-figure 1 shows the absorptive capacities of tumor and hybrid cells from various antisera. The results of both chromosome analyses and absorption tests proved the hybridity of the cells growing in the double-selective medium.
933
