Hamster Tumor Research Articles

Adenovirus type 12 T antigen-related surface antigen detected by immunofluorescence on the membrane of transformed and infected cells

THA cells, derived from a hamster tumor induced by adenovirus type 12 (Ad12), were analyzed by immunofluorescence for the presence of virus-specific surface antigens. S antigen is a surface antigen detected on suspended living cells by a serum collected from hamsters immunized with Ad12. Besides S antigen, a new surface antigen (surface-T) was observed in formaldehyde-fixed cells treated with anti-T serum. Anti-T serum is not reactive against S antigen in living cells and anti-S serum did not detect surface-T antigen. The membrane location of surface-T antigen was confirmed with a Staphylococcus aureus binding assay. Surface-T antigen seems specific for Ad12, as it is not observed in hamster embryo cells, and is not recognized by antisera against SV40- or polyoma-T antigens nor by antiserum against hamster embryonic antigens. Surface-T antigen is also seen in infected KB cells but not in abortively infected RK13 cells. Anti-P serum, raised against an extract of infected RK13 cells, failed to detect surface-T antigen. Experiments with adsorbed anti-T and anti-S sera confirmed the specificity of surface-T antigen and its distinctiveness from S antigen.

The role of cell surface proteins in the induction of histogenetic differentiative responses by some human breast tumours and hamster tumors implanted into chick embryos.

Cells of human breast tumours and fibrocystic hyperplasia grown in culture, and three hamster tumours were implanted between the cell layers of 18-hour-old chick blastoderm. Their ability to induce histogenetic responses in the ectodermal and endodermal embryonic tissues was investigated. The surface proteins of these tumour cells were labelled by lactoperoxidase-catalysed radioiodination. It is shown that the ability to induce the histogenetic effects may be related to the expression of 265K (K = 10(3) daltons) and 233K proteins on the surface of human tumour cells and of 115K proteins on the hamster tumour cells. The antiproteinase, aprotinin, inhibits the induction of the histogenetic responses by and apparently also prevents the deletion of 115K proteins from the hamster tumour cells. It is therefore suggested that cell surface proteins are involved in the complex processes of interaction between embryonic and tumour cells and in the recognition by the embryonic cells of the tumour cells implanted into their midst.

Tumors in control hamsters, rats, and mice: literature tabulation.

(1982). Tumors in Control Hamsters, Rats, and Mice: Literature Tabulation. CRC Critical Reviews in Toxicology: Vol. 10, No. 1, pp. 49-79.

Respiratory tract tumours in hamsters exposed to acetaldehyde vapour alone or simultaneously to benzo(a)pyrene or diethylnitrosamine

Syrian golden hamsters were exposed to 0 or 2500–1650 ppm acetaldehyde vapour, 7 hr/day, 5 days/week for a period of 52 weeks. A proportion of the animals was given, simultaneously, either intratracheal instillations of benzo(a) pyrene (BP) or subcutaneous injections of diethylnitrosamine (DENA). All treatments were stopped after 52 weeks. The study was terminated after 81 weeks. Major effects attributed to acetaldehyde exposure included growth retardation, rhinitis, hyperplasia and metaplasia of the nasal, laryngeal and tracheal epithelium, nasal and laryngeal carcinomas, and a markedly increased incidence of BP-initiated tracheobronchial carcinomas. There was no evidence of acetaldehyde enhancing the development of DENA-initiated tumours of the respiratory tract. It was concluded that acetaldehyde is an irritant, as well as a carcinogen, for the respiratory tract of Syrian golden hamsters. Possible mechanisms of the carcinogenicity of this aldehyde are briefly discussed.

Enhancement of amyloidosis by chenodeoxycholic acid ingestion in the hamster.

A long-term experiment was performed to test the ability of chenodeoxycholic acid (CDCA) to induce intestinal tumors in the Syrian golden hamster. Animals were fed a CDCA-enriched diet up to 12 months, and 4 of them were sacrificed every 2 months. CDCA exhibited no carcinogenic activity. However, unexpected amyloid deposits were discovered in both treated and nontreated animals. Their incidence was significantly higher in treated (71%) than in nontreated (27%) hamsters. During CDCA administration, amyloidosis appeared as early as the second month of treatment. It involved the liver, spleen and kidneys in more than half of the animals. On the contrary, in control animals, amyloid deposits appeared later and were most frequently restricted to one organ. The intensity of the lesions also appeared to be greater in the treated group. No monoclonal components were found in either serum or urine of the amyloidotic hamsters. Such an amyloidosis-promoting effect of CDCA has never been reported.

Tumor induction by human adenovirus type 12 in hamsters: loss of the viral genome from adenovirus type 12-induced tumor cells is compatible with tumor formation.

The patterns of integration of adenovirus type 12 (Ad12) DNA in 39 virus-induced hamster tumors were determined. Both the amount of Ad12 DNA persisting and the apparent sites of insertion differed from tumor to tumor. In 30 tumors, the intact Ad12 genome persisted in colinear arrangement and in multiple copies. In nine tumors, only the left- or the left- and right-hand parts of the Ad12 genome persisted in the tumor cells. In three other cell lines the Ad12 genomes were lost completely during continuous passage in culture. A shift from epithelioid to fibroblastic morphology correlated with loss of Adl2 genomes. The cell line H1111(1) derived from an Ad12-induced tumor had lost all viral DNA by the thirteenth subpassage, but was still oncogenic when reinjected into animals. This finding raises the question, to what extent persistence of the Ad12 genome is essential for the oncogenic phenotype. Tumor cells could be detected histologically inside local lymphatic vessels. In those experiments in which Ad12 preparations were used which contained sizeable proportions of the symmetric recombinant between Ad12 and KB cellular DNA (Deuring et al., 1981), tumors were observed in the nuchal region of the animals.

Extrachromosomal bovine papillomavirus type 1 DNA in hamster fibromas and fibrosarcomas

BPV 1-induced hamster tumors were used as model system to investigate the state of viral DNA in non-virus-producing tumors arising from papillomavirus infection. DNAs from a fibroma and three fibrosarcomas were analyzed for BPV 1 DNA by the Southern blot technique. The vast majority of viral DNA persists in a nonintegrated form, since: (i) supercoiled BPV-DNA can be purified by CsCl-ethidium bromide equilibrium centrifugation, (ii) restriction endonucleases, which do not cut viral DNA, do not change the DNA pattern in comparison to uncleaved samples, and (iii) digests of BPV-DNA with restriction enzymes HpaI, HpaII, and HhaI, respectively, lead to identical patterns with DNA from hamster tumors and from warts. Less than one genome equivalent per cell could persist in an integrated form. Virus-specific, high-molecular-weight DNA in the tumors may be interpreted as oligomeric or catenated DNA. Cleavage analysis of tumor DNA with the restriction enzymes HpaII and HhaI, both sensitive to DNA methylation, provided no evidence for methylation of BPV 1 DNA in hamster tumors.

Analysis of BK virus-transformed cells and BK virus-induced tumors by DNA-DNA reassociation kinetics.

BK virus (BKV)-transformed cells and BKV-induced tumors as well as in vitro derived tumor cell lines were all found to contain BKV DNA sequences when analysed by DNA-DNA reassociation kinetics. In transformed cells the number of viral genome equivalents per diploid cell genome (CG divided by VG) decreased with increasing generations. Likewise, single cell clones had a lower CG divided by VG than parental transformed cells. BKV-induced tumors had a high CG divided by VG. Tumor cells cultivated in vitro and their clones had a lower CG divided by VG than BKV-induced tumors from which they were derived, suggesting a multiclonal origin of tumors. Hamster tumors induced by subcutaneous inoculation of BKV-transformed cells or tumor cell lines had a higher CG divided by VG than cells producing them. Variation in CG divided by VG is discussed in terms of cell selection depending on different in vitro or in vivo conditions of cell growth. In some cases, however, the decrease in CG divided by VG most likely depends on loss on free viral sequences from transformed cells or tumors.

Ureido-ethyl-imidazolines active against autochtonous diethylnitrosamine-induced epidermoid, papillary and adenocarcinomatous tumors of the respiratory tract of Syrian hamsters and against human bronchogenic carcinomas in nu/nu mice.

The detection of a new class of tumor inhibiting substances is described. Employing a chemical reaction discovered several years ago, a series of imidazolinylureas were prepared. It was found that some compounds of this group were active against diethylnitrosamine (DENA)-induced tumours in hamsters. CGP 15720 A (1-[2-[2-(4-pyridyl)-2-imidazoline-1-yl]-ethyl]-3-(4-carboxy-phenyl)urea. Xb), the most active compound at present, was developed through a series of structural variations. CGP 15720 A inhibits significantly in oral or parenteral treatment with well tolerated doses (10-30 mg/kg) the progressive growth of autochthonous, DENA-induced papillary, epidermoid and adenocarcinomatous tumors of the respiratory system in Syrian hamsters and prolongs significantly the survival. The substance also inhibits significantly the growth of 2 poorly differentiated human epidermoid or anaplastic bronchogenic carcinomas in nu/nu Balb/c mice and prolongs the mean survival time. In these mice, the substance is also active against the rodent ascites tumors Ehrlich carcinoma, CrSa 180 and Yoshida Sa AH 66, although it is only marginally active or inactive against these tumors in normal mice or rats.-In the therapeutic trials, hamsters tolerated the highest dose administered for 4 weeks, 1000 mg/kg p.o., without signs or symptoms of toxicity.

Forssman antigen in BK virus-induced tumor cell lines.

A complement-mediated microcytotoxicity assay was used to detect Forssman antigen in BK virus (BKV)-induced tumor cells. BKT-1B, a transplantable hamster tumor cell line, contained Forssman antigen as did some other hamster cell lines tested for comparison. A rat embryo cell line transformed by BKV in vitro was negative for Forssman antigen. The presence of forssman antigen in hamster cell lines was confirmed by absorption experiments of rabbit forssman antiserum with the tumor cells, although in quantitative terms the results obtained by the two techniques were somewhat different. Immunization of rabbits with tumor cells also induced production of Forssman antibody; a much weaker response was observed in hamsters and rats.

The carcinogenicity of N-nitrosodiethanolamine, an environmental pollutant, in Syrian hamsters

Weekly subcutaneous injections of N-nitrosodiethanolamine (NDELA) at doses of 1000, 500 and 250 mg/kg body wt for life induced tumors in Syrian hamsters which primarily affected the upper respiratory tract. The incidence of these malignant neoplasms arising exclusively from the olfactory region was between 73% (highest dose) and 35% (lowest dose). Lower numbers of neoplasms were found with decreasing frequency in the trachea, larynx and lungs. The results indicate that doses of NDELA lower than 250 mg/kg body wt may also be carcinogenic. Hence, NDELA and its precursors should be regarded as hazardous to human health.

Site of linkage between adenovirus type 12 and cell DNAs in hamster tumour line CLAC3.

The patterns of integration of persisting viral DNA have been examined in 30 to 40 different adenovirus type 2 (Ad2)- and type 12 (Ad12)-transformed hamster cell lines, Ad12-induced hamster and rat tumours and tumour cell lines1–5. These patterns have proved to be quite complex, and so far no evidence has been found for specific sites of virus integration. However, the methods of analysis used so far, are not sufficiently sensitive to decide definitively the issue of specificity. Moreover, specific signals directing the interactions between viral and cellular DNA may reside in DNA sequences distant from the site of junction between the two genomes and may be hidden in hitherto undetected structural peculiarities of DNA sequences. From the DNA of an Ad12-induced hamster tumour line (CLAC3) carrying five copies of the Ad12 genome1, the site of junction between the left terminus of Ad12 DNA and cellular DNA has now been recloned in a λ gtWES·λB′ vector. Determination of the nucleotide sequence at this site of junction reveals that the Ad12 DNA sequence is preserved starting with base pair (bp) 46 of the authentic left end of Ad12 DNA. The first 82 bp of the recloned fragment are not viral and contain scattered, patch-like octa- to undecanucleotide pair homologies to distant sequences within the first 2,722 left terminal base pairs of Ad12 DNA. Close to and encompassing the site of junction a stemmed loop can be constructed based on extensive regions of dyad symmetry. The stability of this loop and its biological function are still uncertain.

Restriction endonuclease analysis of mitochondrial DNA from virus-transformed, tumor and control cells of human, hamster and avian origin. Sequence conservation and intraspecific variation

This study compares over 70 recognition sites for restriction endonucleases on mtDNAs from various control versus malignant cells, derived from Syrian hamster, chick embryo, viper and human cells, exhibiting a wide spectrum of cellular transformation and tumor histories. Agents for transformation in vitro and in vivo include Rous sarcoma viruses, simian virus 40, polyoma virus and adenovirus. The results show a striking intraspecific sequence homogeneity of different mtDNAs regardless of tissue origin and oncogenic history. mtDNA from human biopsy specimens of tumor versus pathologically normal areas yielded indistinguishable restriction cleavage patterns reflecting either the ‘wild-type’ form (with seven restriction endonucleases) or, in one individual, a variant pattern detected with HpaI. The precise position of the HpaI variant site was determined on the physical map of human mtDNA. Additional cleavage sites in the previously reported restriction map of Syrian hamster mtDNA are also presented. It is concluded that (1) mtDNA sequences in higher animal cells are highly conserved in malignant transformation; (2) no evidence for integration of viral sequences in mtDNA is apparent; (3) variant patterns in mtDNA are likely to be intraspecific polymorphisms that pre-exist neoplastic transformation. The possibility is discussed that altered regulatory interaction with the mitochondrial genome, rather than evident changes in mtDNA primary structure, determine anomalous mitochondrial functions in malignant transformation.

Oncogenicity of heat-inactivated simian adenovirus SA7.

Simian adenovirus SA7 heated for 30 minutes at 70 degrees C retained part of its ability to induce tumors in newborn hamsters. Tumors were also induced by DNA extracted from the heated virus. However, neither residual replicating, i.e. cytopathogenic, nor transforming and T-antigen-inducing activities could be detected in vitro in CV-1 and hamster embryo cells inoculated with the heated virus. The results are discussed in terms of a possible release of viral nucleic acid at the high temperature. They also show that current tests for effective virus inactivation based exclusively on the recognition of replicating virus should be reevaluated.

Unintegrated viral DNA sequences in a hamster tumor induced by bovine papilloma virus.

A fibrosarcoma was induced in a hamster by bovine papilloma virus type 2 (BPV2). The content of BPV2 DNA sequences was measured by DNA-DNA and cRNA-DNA hybridizations. The tumor contained approximately 300 BPV2 genome equivalents per cell. Southern blot hybridization indicated that the viral DNA was in free form, the entire genome most likely being present. In situ hybridization with BPV2 cRNA showed that multiple genome copies were present in each cell. Neither virus particles nor virus coat antigens could be detected in the tumor. A cell line was established from the fibrosarcoma, and the cells contained multiple copies of the BPV2 genome. The latter was in free form, and all of the DNA sequences appeared to be present in multiple copies and in all cells. An extensive search failed to reveal the presence of virus or viral antigens.

Hormonal induction of kidney tumours in male hamsters. Inhibition of growth by a phenothiazine derivative (perphenazine)

Experiments have shown that long-term treatment of male hamsters with oestrogen will cause renal carcinomas, neoplastic changes in the pars intermedia of the pituitary and highly elevated levels of melanocyte-stimulating hormone (MSH). To test if MSH was fundamentally involved in renal carcinogenesis, a group of hamsters was treated for 317 days with perphenazine which is known to elevate MSH levels. A second group was treated with oestrogen while a third received a combination of oestrogen and perphenazine for a similar period of time. Serum MSH levels were significantly elevated in the perphenazine group but kidney tumours were absent. All oestrogen-treated animals developed kidney tumours but only 19% of those given the combined treatment developed kidney tumours despite highly elevated levels of MSH. The findings suggest that MSH does not play a significant role in renal carcinogenesis in the hamster. Perphenazine may inhibit the induction of oestrogen-induced tumours by competing with oestrogen for receptors in the kidney or may protect cells from toxic injury by stabilising cell and microsomal membranes. The phenothiazines are known to induce, and increase the activity of, microsomal enzymes and that may lead to enhanced detoxification of the carcinogen. Other effects of phenothiazines include elevation of prolactin levels. The fact that the body weight of animals treated with DES and perphenazine was significantly lower than that of the other groups may indicate that the reduction in tumour incidence was the result of a certain degree of undernutrition. Further work is required to elucidate the mechanism of tumour inhibition by phenothiazine derivatives.

Identification of polypeptide components of adenovirus type 12 tumor antigen from productively and abortively infected cells and from tumor cells

Adenovirus type 12 (Ad12) tumor antigen (T antigen) induced in productively infected KB cells, abortively infected rabbit RK13 cells, and hamster tumor cells (THA cells) was studied by immunoprecipitation and polyacrylamide gel electrophoresis. [ 35S]Methionine-labeled cell extracts were precipitated with anti-T serum, obtained from hamsters bearing tumors induced by Ad12, and with anti-P serum, obtained from rabbits immunized against Ad12 early proteins induced in RK13 cells. Two polypeptides with 31,000 (31K) and 19K molecular weights are precipitated from all cells. Polypeptides with 16K, 15K, 13K, and 10.5K molecular weights are observed in extracts from KB and THA cells only. T antigen from tumor cells also contains 57K, 50K, 45K, 41K, and 34K polypeptides. Polypeptides with 39K, 28K, and 25K molecular weights are synthesized in KB and RK13 cells, mainly when the cells are pretreated with cycloheximide, but are never seen in extracts from tumor cells. Anti-P serum precipitates the 39K, 31K, 28K, 25K, and 19K polypeptides and also a 60K polypeptide from KB and RK13 cells, which probably corresponds to the DNA-binding protein.

Evidence that adenovirus type 2 can infect human placenta in vitro

Of the more than 80 identified serotypes of adenovirus (Ad), 31 are known to infect humans. These viruses cause respiratory infections and may persist in lymphoid tissue without causing recurrent illness. Types 1, 2 and 5 are commonly found in adenoids in latent form, and a high proportion of the population has serum antibodies to them. Indeed, antibodies to types 1 and 2 have already developed in three-quarters of all children by the age of five years (Oswald and Frey, 1962). Thirteen of the 51 human Ad have been classified into groups according to their oncogenic potential (Green and Mackey, 1977). Group A (Ad 12, 18 and 31) are highly oncogenic and rapidly produce tumours in most inoculated baby hamsters. Group B viruses (Ad 3, 7, 11, 14, 16 and 21) induce tumours in only a small fraction of inoculated test animals. Group C viruses (Ad 1, 2, 5 and 6) are not oncogenic in newborn rodents but have been shown to transform cells in culture (Green, 1970). Because of the widespread existence of the adenoviruses in the human population and their oncogenic potential, the possibility exists that these viruses may be responsible for some human tumours. An extensive search for viral sequences in human cancer tissue has been conducted employing the rationale that, like Ad-induced hamster tumours, Ad-induced human tumours should contain viral DNA sequences. The emphasis of this work has been on group A Ad (Mackey, Rigden and Green, 1976). However, searches for group BAd sequences (Mackey et al, 1979; Wold et al, 1979) and group CAd sequences (Green and Mackey, 1977) in human turnouts have also been carried out. To date, no adenovirus-related sequences have been detected in DNA isolated from human tumours. Although none of the adenovirus genome has been found in human tumours, there is some evidence for the presence of some or all of the adenovirus genome in apparently normal mammalian tissue. Analysis of human tonsil DNA by Green et al (1979) revealed multiple copies of the entire Ad 5 genome free in the cells, as well as some Ad 5 sequences possibly integrated into the cellular genome. Furthermore, in a recent report, Frolova and Georgiev (1979) demonstrated the presence of Ad 5-related sequences in the genome of normal rat liver cells. By digesting the cellular DNA with restriction endonuclease Bam HI, separating the fragments on agarose gels and then hybridizing 32P-labelled Ad 5 restriction fragments to the cellular DNA immobilized on nitrocellulose filters, these authors were able to detect at least three regions of homology between the Ad 5 genome and cellular DNA.

Organization of viral genome in a T antigen-negative hamster tumor induced by human papovavirus BK.

We analyzed by blot hybridization the state and structure of the viral DNA in an exceptional BK virus-induced hamster tumor (choroid plexus papilloma Vn-324) that contains about one copy of the BK virus genome per cell, but no intranuclear T antigen as assayed by indirect immunofluorescence. The BK viral DNA was found to be integrated into cellular DNA at a site in the middle of the early region of the viral genome (between 0.32 and 0.41 map units). The structure of the inserted viral DNA shows that it cannot encode a full-size large T antigen, but may encode small T antigen and an N-terminal portion of large T antigen.

Biochemical and biological effects of nonionic nucleic acid methylphosphonates.

Oligodeoxyribonucleoside methylphosphonates with base sequences complementary to the anticodon loop of tRNALys and to the -ACCA-OH amino acid accepting stem of tRNA were prepared by chemical synthesis. Oligodeoxyadenosine methylphosphonates form stable, triple-stranded complexes with both poly(U) and poly(dT). These analogues selectively inhibit cell-free aminoacylation of tRNALys (E. coli) but have no effect on aminoacylation of tRNALys (rabbit). The extent of inhibition is temperature dependent and parallels the ability of the oligomer to bind to poly(U), which suggests that inhibition occurs as a result of oligomer binding to the -UUUU- anticodon loop of tRNALys (E. coli). The failure of the oligodeoxyadenosine methylphosphonates to inhibit tRNALys (rabbit) amino-acylation suggests that there may be a difference between the structure of tRNALys or its interaction with aminoacyl synthetase in the Escherichia coli and rabbit systems. The oligodeoxyadenosine analogues also effectively inhibit polyphenylalanine synthesis in cell-free translation systems derived from both E. coli and rabbit reticulocytes. The extent of inhibition parallels the Tm values of the oligo(A) phosphonate-poly(U) complexes and suggests that the inhibition is a consequence of complex formation with the poly(U) message. Tritium-labeled oligodeoxyribonucleoside methylphosphonates with a chain length of up to nine nucleotidyl units are taken up intact by mammalian cells in culture. All the oligomer analogues tested inhibited, to various extents, colony formation by bacterial, hamster, and human tumor cells in culture.