BALB/3T3-Derived Clones by Murine Leukemia Virus

Transcription

1 JouRNAL OF VIROLOGY, Jan. 1976, p Copyright ( 1976 American Society for Microbiology Vol. 17, No. 1 Printed in U.S.A. Studies on Morphological Transformation of BALB/3T3-Derived Clones by Murine Leukemia Virus HELENE S. SMITH,* ADELINE J. HACKETT, E. LOUISE SPRINGER, AND JOHN RIGGS Cell Culture Laboratory, School of Public Health, University of California-Berkeley, Berkeley, California 94720,* and California State Department of Public Health Viral and Rickettsial Disease Laboratory, Berkeley, California Received for publication 18 August 1975 We have described a cell line, UC1-B, derived spontaneously from BALB/3T3 mouse embryo cells, which, unlike the standard BALB/3T3, are morphologically transformed and produce bizarre viral forms in response to murine leukemia virus. Although UC1-B and BALB/3T3 are morphologically similar, and both form contact-inhibited monolayers at confluence, the UC1-B cells are partially transformed because: they grow to a slightly higher saturation density than 3T3 cells, they grow in medium lacking serum growth factors, and they produce tumors in mice. Another clone, 12A-3, derived from BALB/3T3, also transforms and produces bizarre viral forms after infection with murine leukemia virus. Unlike UC1-B cells, the 12A3-8 cells are identical in growth properties to BALB/3T3; therefore, a partially altered morphology is not required for the induction of transformation by murine leukemia virus. Standard mouse BALB/3T3 cells replicate murine leukemia virus (MuLV) without alteration of cellular growth or morphology. Murine sarcoma virus, transformed nonproducer lines derived from BALB/3T3, yield sarcoma virus in response to MuLV infection. We have described a cell line, UC1-B, derived spontaneously from BALB/3T3 cells, which, unlike the standard BALB/3T3, transforms in response to MuLV infection (6). Unlike the transformed nonproducer cells, sarcoma virus is not detectable in the yield. The UC1-B cells yield MuLV after infection, but in addition to the typical type C virus, bizarre forms are also observed by electron microscopy. Although UC1-B and BALB/3T3 are morphologically similar and both form contact-inhibited monolayers at confluence, the UC1-B cells are partially transformed (6) since they grow to a slightly higher saturation density than 3T3 cells and they are able to grow in medium lacking serum growth factors (factor-free medium) (7, 14). In this report we also show that UC1-B cells produce tumors in mice. It was hypothesized that UC1-B cells represent an intermediate state in which leukemia virus replication can result in morphological transformation. We now describe a second clone, 12A3-8 derived from BALB/3T3, which, like UC1-B, both transforms and produces bizarre viral forms after infection with MuLV. Unlike UC1-B cells, the 12A3-8 cells are identical in growth 269 properties to the BALB/3T3 (12, 15) cells; therefore, a partially altered morphology is not required for the induction of transformation by MuLV. Futhermore, since 12A3-8 was isolated at a different time and in a different laboratory from UC1-B, it is unlikely that UC1-B cells contain an inadvertent laboratory infection with sarcoma virus which requires helper virus to transform the cells and to replicate. MATERIALS AND METHODS Cell lines and culture. All clones were grown in Dulbeccos modification of Eagle medium supplemented with 10% calf serum (Colorado Serum Co.). BALB/3T3 clone A31 (2) was obtained from G. Todaro (National Cancer Institute), FLSV, a flat revertant of simian virus 101 (SV101), was obtained from R. Pollack (Cold Spring Harbor Laboratory, N.Y.). Factor-free medium was prepared by depletion on confluent BALB/3T3 cells as described (14). Virus Maloney leukemia virus (MLV). The stock pool was derived from a tumor homogenate (obtained from J. Moloney, lot no. M6-2400) and passaged three times at limit dilution in a mouse embryo cell line, HPME (4), and subsequently passaged three times at 1:100 dilution in HPME cells. The virus was titered at 5 x 106 plaque forming units/ml on BALB/3T3 clone A31 cells by the XC method (10). Electron microscopy. Cell cultures were fixed in 2.5% glutaraldehyde and postfixed in 1% Dalton's chrome-osmium, followed by 0.5% uranylacetate. Cell pellets were washed, dehydrated, and embedded in Epon 812. Thin sections were examined in a Siemens Elmiskop 101.

2 270 SMITH ET AL. Tumorigenicity. Rapidly growing, subconfluent cell cultures were suspended with trypsin-edta, washed, and resuspended in L15 medium with 10% fetal calf serum. Newborn BALB/c mice were inoculated in the thigh with ml containing 1 x 106 to 2 x 106 cells. SV40 hybridization. The preparation of 32P_ labeled simian virus 40 (SV40) probes for the hybridization experiments has been described in detail (5, 12). The techniques of hybridization, Cjt analysis, and computation of gene copies/diploid equivalent have been described (5) with the following exception: separation of single-stranded DNA from doublestranded DNA was performed by batch elution from hydroxyapatite with 50-ml conical centrifuge tubes maintained in a 60 C water bath. Approximately a 5-ml packed volume of hydroxyapatite was used per assay point and the radioactivity was eluted with three washes of 15 ml of phosphate buffer (0.14 M) followed by three washes with 0.4 M buffer (see reference 5 for description of buffers). RESULTS Origin of the cells. Of the 13 subclones of BALB/3T3 (clone A31) tested for focus formation by MuLV, only one clone besides UC1-B morphologically transformed (Table 1). The clones tested include: abortive transformants induced by SV40 infection of BALB/3T3 in factor-free medium (12, 14), two flat SV40 transformants (11, 13, 14), one flat revertant of an SV40 transformed clone (9), and three subclones of uninfected BALB/3T3. The positive clone, 12A3-8, is a reclone of an SV40-induced abortive transformant; however, neither the original abortive transformant, 12A-3, nor other reclones of it produced foci when infected with MuLV. It has been reported that 12A-3 and all of the reclones contain varying amounts of SV40 genetic material in a TABLE 1. cryptic form (12). However, the cells have been extensively passaged since that report and they no longer contain any DNA which hybridizes to SV40 DNA probes (Table 2). By all criteria tested, 12A3-8 is indistinguishable from BALB/3T3 (Table 3). The two cell lines are similar in morphology, growth pattern, saturation density, and neither line grows in factor-free medium. Neither 12A3-8 nor BALB/3T3 produce tumors in syngeneic hosts (2, 15), whereas UC1-B innoculated under similar conditions readily produced tumors (Table 3). All of the tumors were poorly differentiated sarcomas which appeared 8 to 12 weeks postinoculation and progressively grew until the mice were sacrificed 16 weeks postinoculation. Transformation of 12A3-8 by MuLV. The morphology of the foci induced by MuLV on 12A3-8 and on UC1-B (Fig. 1) were similar to foci induced by MSV on BALB/3T3 (6). The efficiency of transformation of the UC1-B and 12A3-8 was determined by assaying simultaneously with the A31-XC system. Dilutions of MuLV were plated on subconfluent monolayers of each cell type. At the time of fluid changing TABLE 2. SV40 genome equivalents in various clones Cells SV40 copies/diploid equivalent Standard transformant Abortive transformant 12A-3 <0.1 12A-3 redone 3 <0.1 12A-3 redone 8 <0.1 12A-3 redone 10 <0.1 Control BALB/3T3 <0.1 Balb/3T3 clones tested for transformation by MuLV Cell line History of clone Reference Transformation 12A-1 Abortive SV40 transformant 12 12A2 Abortive SV40 transformant 12 12A3 Abortive SV40 transformant 12 12A-3 Reclone 3 Abortive SV40 transformant 13 Reclone 8 Abortive SV40 transformant 13 + Reclone 10 Abortive SV40 transformant 13 12A-4 Abortive SV40 transformant 12 1OA7-J4 Flat SV40 transformant 10, 12 SEA31-57 Flat SV40 transformant 10 FLSV Flat revertant of SV40 transformant 8 UCl-B Branch stock of BALB/3T3 clone A A31 clone 1 Reclones of BALB/3T3 clone A31 A31 clone 2 Reclones of BALB/3T3 clone A31 A31 clone 3 Reclones of BALB/3T3 clone A31 J. VIROL.

3 VOL. 17, 1976 TABLE 3. growth properties of clones TRANSFORMATION BY MuLV 271 Characteristics Clone A31 UCl-B 12A3-8 Saturation density 3 x 10' to 5 x 104 cells/cm2 10 x 10' cells/cm2 3 x 10' to 5 x 10' cells/cm Growth in FFMa - + Morphology Epithelial flat Epithelial flat Epithelial flat Tumorigenicity 0/12 9/18 0/22 a FFM, Factor-free medium as described by Smith et al. (12). (day 3), XC cells were added to half of the BALB/3T3 cultures. BALB/3T3 cells alone showed no change in morphology indicating that no contaminating sarcoma virus was present in the dilution of stock MuLV. Typically, the number of plaques formed on BALB/3T3 clone A 31 plus XC monolayers was equal to the number of foci formed on both UC1-B and 12A3-8. The data shown in Table 4 was taken from a recent experiment as representative of the type of comparative counts obtained. The twofold greater yield on A31 may be due to either secondary plaque formations or technical error, but such a difference is not observed reproducibly. Thus, 12A3-8 is as sensitive to leukemia virus infection as are UC1-B and BALB/3T3 clone A31. Virus production by 12A3-8. As previously described for UC1-B (6), 12A3-8 produced only infectious leukemia virus after transformation; no focus-forming virus similar to murine sarcoma virus has been found in the yield after transformation of either 12A3-8 or UC1-B. Electron microscopy of MLV-infected 12A3-8 cells revealed aberrant viral maturation. Typical budding type C virus was found at the plasma membrane. (Fig. 2). Mature particles (not shown here) were present in the extracellular spaces as well. In the same cell, atypical budding formations were found at the periphery of an intracytoplasmic vacuole (Fig. 2). Morphologically complete particles released into the lumen of the vacuoles were rare; the membrane of these vacuoles consisted of linear or rodshaped tubular formations of unit membrane with election dense material typical of the inner membrane and nucleoid of type C virus particles. A single vacuole contained either bizarre forms or typical type C virus, not both, suggesting that the two forms mature at separate sites. In all the preparations observed, at least 95% of the cells contained both types of virus particles. The bizarre forms were not found in A31 cells infected with MLV, and no virus particles were observed in the noninfected UC1-B or 12A3-8 cells. The aberrant virus matured slightly differently in UC1-B than in 12A3-8. Intracysternal particles, also atypical, which were frequently found maturing at the endoplasmic reticulum of UC1-B cells, were rarely found in 12A3-8. The ultrastructural features of viral maturation are summarized in Table 5. DISCUSSION Leukemia virus infection results in morphological transformation of two independently isolated cell lines derived from BALB/3T3. One line, UC1-B, was a spontaneously appearing branch culture of BALB/3T3 to which no selective pressure was applied. The second line was one of three reclones subcultured from an abortive transformant induced by SV40 infection. After infection with MuLV, both lines produced bizarre viral forms as well as standard C type viruses. These bizarre viral forms have not been seen after MuLV infection of any other murine cell line that we have observed, nor have they been seen after MuLV infection of XC cells (3). One explanation for our observations is that the two lines contain an endogenous sarcoma genome which can be activated by MuLV infection. Since endogenous sarcoma virus has never been activated in BALB/3T3 cells, we hypothesize that an altered regulation of gene expression must have occurred in both clones. The relationship between SV40 infection and the alteration of gene regulation responsible for transformation by MuLV is unclear. Although prior infection with SV40 is neither necessary nor sufficient to induce the altered gene regulation (UC1-B was not infected with SV40 and most SV40-infected clones were not transformed by MuLV), SV40 infection may increase the probability that the regulatory change occurs. Since the clones were independently isolated in different laboratories, it is unlikely that both lines contain an inadvertent laboratory infection with sarcoma virus which requires helper virus to transform the cells and to replicate. UC1-B cells show some alterations of growth properties characteristic of tumor cells. Although morphologically similar to BALB/3T3,

4 272 SMITH ET AL. J. VIROL. 7 V"..,,P.Z. ~ ~ ~ S * 4,,, 4'. 4. -j Cj ka. -s * 44k' ;r, 4'4- ;t t- W 4A."%nft% ri# 4. '¼'- $ V4'o*X4'st* * 'ea. S 8 E 'I,.4 qh` 4' A.z.. A r.id 1I I Cl Downloaded from "IP i' on June 30, 2018 by guest i"i FIG. 1. Focus of transformed cells induced by MuLV on two cell lines derived from BALB/3T3 cells. (A) UCJ-B cells with focus of transformed cells; (B) UCJ-B cells, uninfected; (C) 12A3-8 cells with focus of transformed cells; (D) 12A3-8 cells, uninfected. the UC1-B cells grow to a slightly higher saturation density, grow in factor-free medium, and produce tumors when inoculated into syngeneic mice. Because 12A3-8, which has identical growth properties to BALB/3T3, is also transformed by MuLV, an intermediate state of growth control is not necessary to induce the morphological transformation.

5 VOL. 17, 1976 TABLE 4. Efficiency of ML V transformation of UCJ-B and 12A3-8 Cl 8 cells Cell line No. of focia No. of plaques' MLV MLV UC1-B 80,81 NDc 12A3-8 70,100 ND A31 0,0 232,240 a XC cell indicator method for MLV replication (9). b Focus assay was done as described previously (5). c ND, Not done. TRANSFORMATION BY MuLV 273 UC1-B cells readily produced tumors despite a relatively low saturation density. This observation is contrary to earlier work comparing 3T12 and 3T3 cells which showed that the ability to produce tumors in vivo correlated best with in vitro growth to high saturation density (1). Furthermore, cells with in vitro growth properties similar to UC1-B, such as flat revertants of SV40 transformed clones, rarely produce tumors (8, 9, 15). The tumorigenicity of UC1-B cells suggests that loss of density-dependent inhibition is not an intrinsic attribute of tumor e....,.i Downloaded from 2 FIG. 2. Elec tron x 100,000. micrograph of 12A3-8 cells d wit- M x Ie d t virus. infectedd with MLV. x55,000. Inset: Budding type C virus. on June 30, 2018 by guest TABLE 5. Ultrastructural features of viral maturation Vesicles with Vesicles with Cell Cell line granular electron trans- Intracisternal membrane matrix parent matrix maturation maturation UC1-B A Rare + BALB/3T3 (clone A31)

6 274 SMITH ET AL. cells, but rather the ability to form tumors is a dynamic balance between cell growth and host rejection. SV40 transformants and 3T12 cells (which produce complete C type viruses) may be so antigenic that only clones capable of growth to high densities can produce tumors. It is tempting to speculate that the tumorigenic potential of UC1-B results from its partial expression of the endogenous sarcoma virus genome. The UC1-B cells have been used as substrate for assaying MuLV which is as accurate, and even more rapid than, the XC-plaque assay. One disadvantage to using UC1-B is that the cells often spontaneously transform, possibly because they have already lost some growth control. Since 12A3-8 does not have this problem, it is particularly convenient to use it for the assay. ACKNOWLEDGMENTS We gratefully acknowledge the excellent technical assistance of Steven Sylvester and Nancy Robertson. This work was supported by contract E NO1- CP and NO1-CP from the Special Virus Cancer Program. LITERATURE CITED 1. Aaronson, S. A., and G. J. Todaro Basis for the acquisition of malignant potential by mouse cells cultivated in vitro. Science 162: Aaronson, S. A., and G. J. Todaro Development of 3T3-like lines from BALB/c mouse embryo cultures: transformation susceptibility to SV40. J. Cell Physiol. 72: Chan, J. C., N. Vera, J. L. East, S. Hiraki, and L. Dmochwski Lack of sycytium formation by a type C virus-producing XC cell line in the mixed J. VIROL. culture cytophogenicity test. Cancer Res. 34: Field, A. K., and J. Fong Mouse salivary gland virus plaque assay on a stable line of mouse embryo cells. J. Bacteriol. 87: Gelb, L. D., D. E. Kohne, and M. A. Martin The detection and quantitation of viral genomes within normal and transformed mammalian cells. J. Mol. Biol. 57: Hackett, A. J., and S. S. Sylvester Cell line derived from BALB/3T3 that is transformed by murine leukemia virus: a focus assay for leukaemia virus. Nature (London) New Biol. 239: Jainchill, J., and G. J. Todaro Stimulation of cell growth in vitro by serum with and without growth factor. Exp. Cell Res. 59: Pollack, R. E., and G. W. Teebor Relationship of contact inhibition to tumor transplantability, morphology, and growth rate. Cancer Res. 29: Pollack, R. E., H. Green, and G. J. Todaro Growth control in cultured cells: selection of sublines with increased sensitivity to contact inhibition and decreased tumor-producing ability. Proc. Natl. Acad. Sci. U.S.A. 60: Rowe, W. P., W. E. Pugh, and J. W. Hartley Plaque assay techniques for murine leukemia viruses. Virology 42: Scher, C. D., and W. A. Nelson-Rees Direct isolation and characterization of flat SV40-transformed cells. Nature (London) New Biol. 233: Smith, H. S., L. D. Gelb, and M. A. Martin Detection and quantitation of simian virus 40 genetic material in abortively transformed Balb/3T3 clones. Proc. Natl. Acad. Sci U.S.A. 69: Smith, H. S., A. J. Hiller, E. W. Kingsbury, and C. Roberts-Dory Cell surface properties and the expression of SV40-induced transformation. Nature (London) New Biol. 245: Smith, H. S., C. D. Scher, and G. J. Todaro Induction of cell division in medium lacking serum growth factor by SV40. Virology 44: Wright, P. W., H. S. Smith, and J. McCoy Tumorigenicity and antigenicity of mouse cells infected with Simian Virus 40. I. Relationship of growth in vitro and in vivo in immunosuppressed and immunocompetent recipients. J. Natl. Can. Inst. 51: