MMAC1/PTEN mutations in primary tumor specimens and tumor cell lines.

A candidate tumor suppressor gene, MMAC1/PTEN, located in human chromosome band 10q23, was recently identified based on sequence alterations observed in several glioma, breast, prostate, and kidney tumor specimens or cell lines. To further investigate the mutational profile of this gene in human cancers, we examined a large set of human tumor specimens and cancer cell lines of many types for 10q23 allelic losses and MMAC1 sequence alterations. Loss of heterozygosity (LOH) at the MMAC1 locus was observed in approximately one-half of the samples examined, consistent with the high frequency of 10q allelic loss reported for many cancers. Of 124 tumor specimens exhibiting LOH that have been screened for MMAC1 alterations to date, we have detected variants in 13 (approximately 10%) of these primary tumors; the highest frequency of variants was found in glioblastoma specimens (approximately 23%). Novel alterations identified in this gene include a missense variant in a melanoma sample and a splicing variant and a nonsense mutation in pediatric glioblastomas. Of 76 tumor cell lines prescreened for probable LOH, microsequence alterations of MMAC1 were detected in 12 (approximately 16%) of the lines, including those derived from astrocytoma, leukemia, and melanoma tumors, as well as bladder, breast, lung, prostate, submaxillary gland, and testis carcinomas. In addition, in this set of tumor cell lines, we detected 11 (approximately 14%) homozygous deletions that eliminated coding portions of MMAC1, a class of abnormality not detected by our methods in primary tumors. These data support the occurrence of inactivating MMAC1 alterations in multiple human cancer types. In addition, we report the discovery of a putative pseudogene of MMAC1 localized on chromosome 9.

Genomic sequences capable of committing mouse and rat fibroblasts to adipogenesis.

The mouse Swiss 3T3-F442A/3T3-C2 cell system is well suited for the isolation of genes involved in commitment to adipogenesis. 3T3-F442A cells convert to adipocytes with high efficiency in response to confluence and insulin. The sister clonal line 3T3-C2 does not respond to these signals, but can convert to adipocytes when transfected with DNA from 3T3-F442A preadipocytes or from human fat. Human fat-tissue biopsy FO46 DNA transfected into 3T3-C2 gave rise to fat foci after two rounds of transfection and selection. A cosmid library of a subclone of secondary transfectant 3T3-C2/FO46-1 was screened for the human repetitive Alu sequence. Five out of eight Alu+ recombinant clones committed 3T3-C2 cells to adipogenesis. The adipose commitment (AC) activity of one cosmid, p18A4, was found to reside in two small, non-identical, subcloned sequences 1.2kb and 2.0kb in length, each separately able to commit 3T3-C2, precrisis mouse and rat fibroblasts and the multipotential C3H10T1/2 cell line to adipogenesis. We conclude that commitment to adipogenesis can be effected in vitro with high efficiency by transfection of specific sequences into a variety of host cells.

Human major HSP70 protein complements the localization and functional defects of cytoplasmic mutant SV40 T antigen in Swiss 3T3 mouse fibroblast cells.

The CT3 cytoplasmic localization mutant of SV40 T antigen is neither properly transported to the nucleus nor is it functional in rodent cells. Human precrisis cells are able to complement this mutation, as they are fully transformed by CT3 with wild-type efficiency. The human-specific factors responsible for this species-specific difference in response to CT3 were localized to human chromosome 6 by synteny in a panel of six somatic cell hybrids. A major human HSP70 heat shock protein located on chromosome 6 is expressed constitutively in human cells. Hsp70 proteins have been reported to play a role in intracellular movement of newly synthesized proteins. To test whether human HSP70 played a role in the complementation by human cells of the defect of CT3, we constructed a series of mouse cell lines expressing human HSP70 and tested them for their ability to localize CT3 T antigen in the nucleus and for their ability to be transformed by CT3 DNA. Mouse cell lines expressing human HSP70 protein were able to translocate mutant CT3 T antigen into the nucleus and were transformed by CT3 at rates comparable with wild-type SV40. Mouse-inducible HSP70 protein was not able to translocate cytoplasmic T antigen in Swiss 3T3 mouse fibroblast cells, even after heat shock. Apparently human HSP70 is capable of complementing directly or indirectly the structural and functional alterations in SV40 T antigen introduced by the CT3 mutation.

Excess wild-type p53 blocks initiation and maintenance of simian virus 40 transformation.

Wild-type (wt) murine p53 has been tested for its ability to block and reverse the transforming effects of simian virus 40 (SV40) large T antigen. Established and precrisis mouse cells overexpressing exogenously introduced wt p53 became resistant to SV40 transformation. The introduction of excess wt p53 into SV40-transformed precrisis cells reverted their transformed phenotype. However, the phenotype of SV40-transformed established cells was not reverted by excess wt p53. We conclude that an antioncogenic action of wt p53 is exerted during SV40 transformation and that in precrisis cells, the antitransforming action of wt p53 can be exerted both at initiation and during the maintenance of transformation.

Commitment of mouse fibroblasts to adipocyte differentiation by DNA transfection.

Cells of the mouse cell line 3T3-F442A can be induced by various hormones to differentiate into adipocytes, whereas cells of 3T3-C2, a subclone of 3T3, cannot. However, transfection of DNA from uninduced 3T3-F422A cells into 3T3-C2 cells permits recovery of 3T3-C2 transfectants that differentiate into adipocytes in the presence of insulin. DNA isolated from human fat tissue, when transfected into 3T3-C2 mouse cells, also gives rise to mouse transfectants that are induced to differentiate into adipocytes by the addition of insulin. Apparently, transfection of a trans-regulatory gene (or genes) from 3T3-F442A or human fat cells into 3T3-C2 cells is sufficient to commit 3T3-C2 cells to adipocyte differentiation.

Skin fibroblasts from individuals hemizygous for the familial adenopolyposis susceptibility gene show delayed crisis in vitro.

Normal human fibroblast cells have not been reported to escape crisis--that is, they die after about 24 doublings in culture. We have been studying the growth properties of skin fibroblast cells from persons in families with familial adenopolyposis of the colon (FAP). An individual hemizygous at the FAP locus will develop hyperplasia of the colonic epithelium followed by colonic polyps, both at an early age. Polyps themselves still retain a single functional FAP allele. A mutation or deletion in this allele in a polyp is hypothesized to lead to further loss of growth control; thus, a tumor is formed. We found that the in vitro life-span of skin fibroblast cells from FAP individuals and from some asymptomatic children were markedly extended when compared with normal individuals.

Transformation of precrisis human cells by the simian virus 40 cytoplasmic-localization mutant pSVCT3 is accompanied by nuclear T antigen.

pSVCT3 is a cytoplasmic-localization mutant of simian virus 40 (SV40) isolated from the SV40 adenovirus 7 hybrid virus (PARA) and cloned into plasmid PBR. The large T antigen of pSVCT3 accumulates in the cytoplasm of infected monkey cells instead of being transported to the nucleus. The sole change in CT3 large T antigen is amino acid residue 128 (Lys----Asn). Transformation of precrisis rodent cells by pSVCT3 is negligible, whereas the frequency of transformation of established rodent cell lines by pSVCT3 is comparable to that of wild-type SV40. According to the model, in which transformation of precrisis cells involves the combined oncogenic action of both nuclear and cytoplasmic gene products, we predicted that pSVCT3 would localize in the cytoplasm of human cells and would therefore at most only partially and rarely transform precrisis human cells. We have found that pSVCT3 is able to transform precrisis human cells at high frequency. Furthermore, pSVCT3-transformed human precrisis cells relocalized T antigen to their nuclei. The relocalization of large T antigen was not dependent on cell growth. Wild-type and pSVCT3-transformed human cell lines both have about five copies of integrated SV40 DNA. SV40 virus-specific proteins, including the 100,000-molecular-weight super large T antigen, were expressed in pSVCT3-transformed human cells. Our results suggest that molecules in precrisis human cells, but not cells of other species, are able to complement the cytoplasmic-localization defect of the CT3 mutant large T antigen.

Two integrated partial repeats of simian virus 40 together code for a super-T antigen.

We determined that the coding sequence for a 100-kilodalton super-T antigen found in Simian virus 40 mouse transformants spanned two separate partial repeats of the viral genome. The downstream repeat contained a complete Simian virus 40 large-T-antigen gene, whereas the upstream repeat was a truncated copy of the same gene. When the repeats were separated by subcloning, the capacity to code for the super-T antigen was lost. A small insertion or deletion in the origin-control region which preceded the second repeat could also destroy the ability to code for the 100-kilodalton protein. Our data suggest that differential splicing between parts of two gene copies was responsible for the additional molecular weight of this super-T antigen.

A frameshift mutation affecting the carboxyl terminus of the simian virus 40 large tumor antigen results in a replication- and transformation-defective virus.

We have constructed a frameshift mutation in the simian virus 40 early region using a novel method of oligonucleotide-directed mutagenesis. The mutated DNA specifies an 84,000-dalton large tumor antigen that consists of approximately equal to 75,000 daltons encoded by the wild-type reading frame and 9,000 daltons, by the alternative reading frame (wild-type large tumor antigen is approximately equal to 82,000 daltons). The frameshifted carboxyl terminus of the protein bears a strong similarity to the same region of polyoma virus middle-sized tumor antigen. We have found that the mutant DNA is unable to replicate when introduced into permissive monkey cells and incapable of transforming nonpermissive mouse cells.

A functional simian virus 40 origin of replication is required for the generation of a super T antigen with a molecular weight of 100,000 in transformed mouse cells.

We used two recombinant plasmids, one containing wild-type simian virus 40 DNA (pSVR1) and the other containing a simian virus 40 genome with a defective origin of replication (pSVR1-origin-minus) to transfect NIH3T3 cells. Quantitation of T-antigen synthesis by indirect immunofluorescence at 48 h after transfection with either DNA revealed the same percentage of T-positive nuclei. The transformation frequencies observed were also similar with both plasmids. Immunoprecipitation of [35S]methionine-labeled cell extracts showed the expected 94,000-dalton (94K) T and 17K t antigens in all clones examined. In pSVR1-generated transformants, a 100K super T antigen was also detected. Transformants isolated from pSVR1-origin-minus transfection, however, never expressed this 100K super T antigen, and some of these clones originally also showed greatly reduced levels of 94K T antigen. However, after growth in culture for several generations, the levels of 94K T antigen synthesis in these underproducer clones were dramatically increased. A direct correlation between the amounts of T antigen synthesized and the ability to grow independently of anchorage was observed. The mechanism which brings about increasing levels of T-antigen synthesis in some of the clones is not clear, but it appears not to be due to changes in either the copy number or the methylation pattern of the integrated simian virus 40 DNA.

Pre-crisis mouse cells show strain-specific covariation in the amount of 54-kilodalton phosphoprotein and in susceptibility to transformation by simian virus 40.

We have used several inbred mouse strains to examine the role of the 54-kilodalton (kDa) cellular phosphoprotein in transformation by the papovavirus simian virus 40. We have measured the endogenous 54-kDa phosphoprotein in cells obtained from these inbred mouse strains. To study the effect of passage, cell cultures were measured for amount of the 54-kDa phosphoprotein at the 2nd and 12th passages. In the absence of any transforming agent, the amount of endogenous 54-kDa phosphoprotein in early pre-crisis mouse cells varied in a strain-specific way. Transformation frequency varied coordinately with endogenous 54-kDa expression. Mouse strains whose cells produced a high level of endogenous 54-kDa phosphoprotein on passage did not further increase its expression after simian virus 40 transformation.

94,000- and 100,000-molecular-weight simian virus 40 T-antigens are associated with the nuclear matrix in transformed and revertant mouse cells.

A small fraction of the 94,000-molecular-weight multifunctional large T-antigen of simian virus 40 was associated with the nuclear protein matrix derived from simian virus 40-transformed mouse cells. The interaction between this fraction of T-antigen and the matrix was largely or entirely independent of nuclear DNA. Similar amounts of T-antigen were retained by the nuclei of transformed and revertant cell lines. A 100,000-molecular-weight variant of T-antigen, which has been found to correlate specifically with anchorage-independent growth, was present in the nuclear protein matrix of a transformed cell line. A T-antigen-containing revertant selected for the reacquisition of a high serum requirement and an anchorage requirement for growth retained T-antigen in association with its matrix.

The cytoskeleton in cultured cells: coordinate in vitro regulation of cell growth and shape.

Adenopolyposis of the colon and rectum (ACR) links the well-characterized phenomena of murine oncogenic virus transformation with the progression of a human cancer. The same syndrome links defects in fibroblast growth control and cytoskeletal organization to a tumor of epithelial origin. Since skin fibroblasts are involved in this colonic tumor, the syndrome is very likely to be systemic. That is, one element of normal growth regulation of epithelial cells in situ may be provided by the fibroblasts residing beneath their basement membrane. These observations have led to a novel approach to early detection of persons at risk for a tumor, via the behavior of their skin fibroblasts in culture. At present, it is rarely possible to detect persons at risk for malignancy before the appearance of a frank invasive or metastatic growth. Although many biochemical assays have been proposed as indicators of pre-malignant states or cryptic early tumors, most have eventually been shown to be related to age, sex drug treatment, or other variables linked to but not clearly indicative of, a malignancy. Two serious problems limiting attempts to detect preneoplastic states or a disposition to eventual neoplasia are the inherent low frequency of incidence of any single type of tumor and the difficulty of obtaining identical sample material from prospective or actual patients once they are located. Our approach to these problems significantly departs from the common dependence upon isolation of cells from the site of a tumor. Clearly, it is an easier task to scan skin fibroblasts for disrupted cytoskeletal patterns than it is to obtain epithelia cells from most tissues at high risk for malignancy. This line of work, buttressed by information derived from the model system of SV40 transformation, may provide a novel mode of early detection of other human malignancies, as it has for ACR.

Microfilament bundles, LETS protein and growth control in somatic cell hybrids.

Hybrid cell lines between normal rat embryo fibroblasts and TA3B mouse tumour cells, or between TA3B and BI hamster sarcoma cells, have been examined for the expression of the cell surface large external-transformation-sensitive (LETS) protein and the organization of microfilament bundles. LETS protein was detected by lactoperoxidase-catalysed radioiodination and microfilament bundles were visualized by indirect immunofluorescence with antibodies directed against actin or myosin. Hybrids which exhibited normal growth control characteristics had high levels of LETS protein and extensive microfilament bundles. Neoplastic transformation appears to be suppressed in these hybrids. Hybrids which had the growth control characteristics typical of transformed cells had reduced or zero levels of LETS protein and few microfilament bundles. These results confirm previous studies on the expression of the transformed phenotype in these hybrids and demonstrate that there is a good correlation between normal growth control and the presence of LETS protein and microfilament bundles. However, the changes in cell surface LETS protein and in the organization of microfilament bundles often appeared to be quantitative reductions rather than all-or-none effects. The magnitude of the alterations in the levels of LETS protein and in the organization of microfilaments appeared to correlate with the range of transformed characteristics exhibited by the cells. One transformed hybrid in particular, selected for growth in agar, had some surface LETS protein, some microfilament bundles and retained density-dependent inhibition of growth.

Identification of biological molecules in situ at high resolution via the fluorescence excited by a scanning electron beam.

Proteins, nucleic acids, and fluorescein-conjugated antibody are shown to be identifidable in situ via the fluorescence excited by the focused electron beam of a canning electron microscope. A molecular species is identified by its characteristic fluorescence spectrum and by a characteristic alteration of the spectrum with time under the electron beam. Primary protein fluorescence is relatively rapidly destroyed by the beam, but protein photoproduct fluorescence is more rugged and will in some cases permit detection of small numbers of protein molecules. Nucleic acid fluorescence is extremely long-lived and will permit detection of small numbers of nucleic acid residues. The theoretical resolution limit for localization of a particular molecular species -- about 20 A--is determined by the known maximum distance for molecular excitation by fast electrons. Drect extapolation from an observed resolution of 900 A in the localization of nucleic acid using a low-efficiency detector leads to an experimental resolution limit of less than 60 A. Fluorescence is strongly quenched by residual water in the specimen. Similar quenching is produced by some macromolecular associations and so may serve to localize such associations.