Tumor and serum tamoxifen concentrations in the athymic nude mouse

Summary The athymic nude mouse has been used as an in vivo model for pharmacologic studies of the antiestrogen, tamoxifen. Serum and tumor tamoxifen and metabolite concentrations were examined during and after 100 and 1000 g/day doses injected s.c. Tamoxifen and tamoxifen metabolites were quantitated by high-performance liquid chromatography. Tamoxifen was detected in tumors after a dose of 100 g/day, although serum concentrations were not detected. At a dose of 1000 g/day, tumor tamoxifen and its active metabolites were detected in high concentrations ranging up to >6 mmol/g tissue. Serum tamoxifen metabolites were not detected at either dose. In summary, high doses of tamoxifen were required in the nude mouse to obtain clinically relevant serum concentrations, and significant tumor levels were achieved at doses that resulted in undetectable serum levels. The relationship between serum tamoxifen concentrations, tumor tamoxifen levels, and the biologic activity of the drug requires further study.This work was supported in part by NIH Grant CA 30251 from the National Cancer Institute, the Louis R. Lurie Foundation and the Children's Cancer Research Institute, San Francisco, CA, USA     

Induction of active melanocytes in mouse skin by carcinogens: a new method for detection of skin carcinogens

Application of potent skin carcinogens, such as 7, 12-dimethylbenz[a]anthracene,3-methylcholanthrene, benzo[a]pyrene and 4-nitroquinoline-1-oxide,induced numerous dihydroxyphenylalanine (dopa)-positive cellsin the interfollicular epidermis of C57BL/6 mice in a dose-and time-dependent fashion. Chrysene, a weak skin carcinogen,and croton oil, a tumor promoter, also induced 3–4 timesmore dopa-positive cells than acetone. Liver carcinogens, suchas 3'-methyl-4-di-methylaminoazobenzene and N-2-acetylaminofluorene,and non-carcinogenic aromatic hydrocarbons, such as anthracene,fluoranthene, fluorene and pyrene, did not induce increase inthese cells. These results indicate that increase in the numberof dopa-positive cells after application of chemicals is wellcorrelated with the abilities of these compounds to induce skincarcino-genesis and suppress sebaceous glands.     

The induction of adenomata in mouse lung homografts by chemical carcinogens

Implants of lung from 18-day-old embryo BALB/c mice of an inbred strain were exposed to 3,4-benz(a)pyrene or 1,2,5,6-dibenzanthracene and introduced subcutaneously into 6-week-old mice of the same strain. Lung adenomata developed within 16 weeks.     

Induction of different genetic changes by different classes of chemical carcinogens during progression of mouse skin tumors

By analysis of skin tumors from F₁ hybrid mice we demonstrated that the genetic events that occur during tumor progression depend on the type of chemical carcinogenesis protocol used to induce tumor growth. More than 95% of tumors induced by initiation with 7, 12-dimethylbenz[a]anthracene (DMBA) and promotion with 12-O-tetradecanoyl-phorbol-13-acetate (TPA) exhibited mutations in Ha-ras and trisomy of chromosome 7. Carcinomas induced with multiple DMBA treatments had a lower frequency of alterations on chromosome 7 (50%), but only in tumors with Ha-ras mutations, and had a much wider spectrum of alterations, including trisomy, mitotic recombination, deletion, and gene duplication. Carcinomas induced with multiple N-methyl-N-nitro-N-nitrosoguanidine treatments only rarely exhibited alterations on chromosome 7 (8%), even if they contained mutant Ha-ras. More frequent numerical alterations of chromosome 11 were also seen in TPA-promoted tumors (23%) than in tumors induced by multiple carcinogen treatments (8%). These results show that postinitiation events are nonrandom and fit a model in which promoting agents induce numerical chromosomal alterations but in which mutagens cause more directed mutational events. ©1994 Wiley-Liss, Inc.     

Vascularity and perfusion of human gliomas xenografted in the athymic nude mouse.

The vascularisation and perfusion of seven subcutaneously xenografted human glioma lines established from surgical specimens has been analysed using an anti-collagen type IV antibody to visualise the vascular walls in combination with a perfusion marker (Hoechst 33342). A computer-based digital image processing system was employed for quantitative analysis of the parameters. The vascular architecture of individual tumours belonging to the same tumour line showed a consistent similarity, while substantial differences occurred between the various tumour lines derived from different patients. Despite the presence of a large inter-tumour variation in vascular area as a proportion of the tumour area, this vascular parameter clearly showed tumour line-specific characteristics. The perfused fraction of the tumour vessels also showed a large inter-tumour variation for all tumour lines ranging from 20% to 85%, but the majority of tumours of all lines had perfusion fractions of more than 55%. Despite large variation, the perfused vascular area as a proportion of the tumour cross-sectional area exhibited clear tumour line-specific tendencies. These observations suggest that consistent differences in vascular parameters are present between glioma xenograft lines, although the tumour lines all originated from histologically similar human high-grade gliomas. These differences may have important consequences for treatment and clinical behaviour of this type of tumour.     

Methylation of deoxycytidine in replicating cells treated with ultraviolet radiation and chemical carcinogens

5-Methyl-2'-deoxycytidine (m⁵dC) levels were measured in DNAfrom three types of cultured cells following treatment withu.v. radiation and two chemical carcinogens, N-methyl-N-nitrosourea(MNU) and N-acetoxy-2-acetylaminofluorene (NA-AAF). Controlvalues for m⁵dC in Raji cells (a human lymphoblastoid cell line),S49 cells (a mouse thymic lymphoma cell line) and human diploidfibroblasts are 3.6%, 3.6% and 3.2%, respectively. None of thedamaging agents produced a detectable change in methylationlevels of newly replicated DNA, even at levels of damage thatinhibited replication by 95%. In contrast, treatment with 5-aza-2'-deoxycytidine,a known methyltransferase inhibitor, transiently reduced genomicmethylation by 89% and 74% in Raji and S49 cells, respectively.Although other investigators have found a marked reduction inm⁵dC in DNA replicated after carcinogen treatment, our experimentsindicate that extensive demethylation is not a necessary consequenceof DNA damage.     

Serum tamoxifen concentrations in the athymic nude mouse after three methods of administration

Summary The athymic nude mouse has been used as an in vivo model for pharmacologic studies of the antiestrogen, tamoxifen. We hav examined the steady-state serum tamoxifen concentrations achieved in mice with s. c. slow-release pellets, s. c. injections, and i. p. injections, in an attempt to identify a method that would yield serum levels similar to those observed in patients receiving tamoxifen therapy. Tamoxifen and tamoxifen metabolites were examined by a high-performance liquid chromatography assay which has a sensitivity of 8 ng/ml. Tamoxifen metabolites were not observed with any dose or schedule. After slow-release pellets containing 5 or 25 mg tamoxifen no tamoxifen was detectable, even after 2 weeks of treatment. Very low levels (0.07 M) were found with 50-mg pellets. Tamoxifen was also not detected either with daily s. c. injections of 500 g/mouse or with i. p. injections of 2.5 mg/kg. However, daily s. c. injections of 1000 g or i. p. injections of 25, 50, o4 100 mg/kg resulted in tamoxifen concentrations ranging from 0.21 to 0.51 M which are similar to those observed in patients. Thus, clinically relevant tamoxifen concentrations can be achieved in the nude mouse with either of these methods of administration.This work was supported in part by NIH Grant CA 30251 from the National Cancer Institute and the Louis R. Lurie Foundation     

Cancer-associated colonic mucin in cultured human tumor cells and athymic (nude) mouse xenografts

Mucins derived from colonic cancers differ immunologically and chemically from those in normal colonic epithelium. It has been demonstrated that the lectin from the peanut will bind to mucins present in colonic cancers and other neoplastic lesions but not to those from the normal colon. It was hypothesized, therefore, that in transformed colonic epithelium the glycosylation of mucins occurs differently than in normal epithelium. To rule out the possibility that the differences in oligosaccharide structure were due to postsecretory degradation, studies were designed to evaluate cancer-associated colonic mucins produced under more controlled conditions. We studied nine different cancer cell lines first in monolayer culture and then as xenografts in athymic or nude mice. Eight of the nine cell lines in monolayer culture synthesized glycoconjugates that were labeled by fluorescein-conjugated lectins. After injection into nude mice, eight of the nine cell lines produced tumors typical of human colonic cancer, and six of nine secreted mucin. The mucins produced by the xenografts were labeled at fluorescence microscopy by peanut lectin and other lectins, characteristic of what had been seen in other primary human colonic cancers. One cell line, LS174T, produced large amounts of mucin in the xenograft model. Mucin was purified from these tumors and characterized biochemically. It was demonstrated that mucin purified from the xenografts bound peanut lectin. Therefore, we have concluded that cancer-associated mucins are present in cultured colorectal tumor cells. The cancer-associated mucins are also found in nude mouse xenografts, indicating that they are not the result of postsecretory degradation by colonic flora or by tumor cell necrosis. The cell culture and xenograft can therefore be useful for studying the biosynthesis of cancer-associated mucins.     

Growth control of heterologous tissue culture cells in the congenitally athymic nude mouse.

A variety of heterologous mammalian cells were inoculated into nude mice and scored for tumorigenicity. The cells tested were from primary cell cultures, established cell lines of neoplastic origin, established cell lines of nontumor origin, and primary cell cultures transformed by oncogenic viruses. Regardless of the animal species of origin, every cell line that was tumorigenic in some other animal host and every cell line of neoplastic origin was tumorigenic in nude mice. Several tissue culture cells lines capable of indefinite growth in vitro failed to form tumors in nude mice, and the basis of this growth suppression was investigated. The findings suggest that the failure of an established cell line to form tumors in nude mice is an authentic response to host-mediated growth-regulatory signals.     

A model for initiated mouse skin: suppression of papilloma but not carcinoma formation by normal epidermal cells in grafts on athymic nude mice.

The availability of a skin grafting system on nude mouse hosts and of epidermal cell lines which form papillomas when grafted has made possible the creation of a model for initiated skin in vivo from cultured cells. When grafted with 6-8 x 10(6) primary dermal fibroblasts, 10 x 10(6) primary epidermal cells form an apparently normal skin, and cell line SP-1 (0.5 x 10(6) cells) forms papillomas. Cell line SP-1 was derived from papillomas produced on SENCAR mice by initiation with 7,12-dimethylbenzaadaa]anthracene and promotion with 12-O-tetradecanoylphorbol-13-acetate. Grafting of 0.5 x 10(6) SP-1 cells along with 10 x 10(6) SENCAR newborn primary epidermal cells resulted in a 90% reduction in the average papilloma volume per mouse compared to controls without primary epidermal cells. Suppression occurred specifically with epidermal cells, either cultured or freshly prepared, and was not seen when an equivalent number of SENCAR primary dermal fibroblasts was grafted in place of epidermal cells. Nor did suppression occur when primary epidermal cells were replaced with a carcinogen-altered cell line, SCR722. SCR722 cells have a normal-skin phenotype when grafted. Furthermore, suppression of tumor formation did not occur when a malignant variant of SP-1 cells replaced benign SP-1 cells in grafts. Repeated treatment of suppressed grafts with 12-O-tetradecanoylphorbol-13-acetate resulted in an increased number of mice with papillomas and a larger mean papilloma volume per mouse compared to controls treated with solvent alone, whereas treatment of nonsuppressed grafts of papilloma cells with promoter produced no change in tumor size. These results support the concepts that normal epidermal cells suppress the growth of initiated cells and that repeated treatment with phorbol ester tumor promoters overcomes the suppression, leading to benign tumor formation.     

Development of murine epidermal cell lines which contain an activated rasHa oncogene and form papillomas in skin grafts on athymic nude mouse hosts

We have developed four murine epidermal cell lines which form squamous papillomas when grafted to athymic nude mice in a reconstituted skin. Two of the lines, SP-1 and BP-4, were derived from pools of papillomas produced on SENCAR and BALB/c mouse skin, respectively, by initiation with 7,12-dimethylbenz(a)anthracene and promotion with 12-O-tetradecanoylphorbol-13-acetate. Line 308 was derived from BALB/c mouse skin initiated in vivo with 7,12-dimethylbenz(a)anthracene, culture of the epidermal cells, and selection of cells resistant to Ca2+-induced terminal differentiation. Line LC14 was derived from untreated, cultured newborn BALB/c mouse primary epidermal cells which spontaneously developed resistance to Ca2+-induced terminal differentiation. Each line has an activated rasHa gene with a mutation within codon 61. Cells from all four lines, in contrast to normal primary epidermal cells, survive in medium with Ca2+ levels greater than 0.1 mM. Clonal growth studies in culture showed a unique growth pattern for each of the four lines in medium with 1.4 mM and 0.05 mM Ca2+, with or without 12-O-tetradecanoylphorbol-13-acetate. Early passage cells of these lines should provide a valuable resource for detecting genes or genetic alterations which complement an activated ras gene to cause malignant conversion and for studying the biology of tumor promotion.     

Alterations of mouse proto-oncogenes in sarcomas induced after transplantation of human tumors in athymic nude mice

During serial subcutaneous transplantation of several types of human tumors into nude mice, the local development of malignant mouse-specific sarcomas has been observed. Although the frequency of sarcoma induction is low, this phenomenon is very important because the mouse-specific sarcomas completely replaced the human tumors during serial transplantation. The DNA of five independently induced mouse-specific sarcomas was transfected into NIH/3T3 cells in order to detect oncogenes associated with mouse-specific sarcoma induction. Two of these DNAs were found to carry activated mouse c-N-ras and c-Ki-ras genes. The sequence analysis of the molecularly cloned mouse c-N-ras oncogene showed a single nucleotide transition from G to A at the 12th codon. This results in substitution of aspartic acid for glycine at this position. The mouse c-myc gene was also found to be amplified in a sarcoma. In these mouse sarcoma DNAs, human Alu sequences were not detected. These data strongly suggest that the mouse-specific sarcomas were not induced by the transfer of human transforming sequences but by the alterations of mouse proto-oncogenes.     

Induction of polyoma tumors in athymic nude mice

We verified the oncogenic effect ofpolyoma virus in gnotobiotic nude (nu/nu) mice compared to their gnotobiotic normal counterparts (nu/+). The nude mice were found to be much more sensitive to the oncogenic activity of the virus than were their heterozygote counterparts. This increased sensitivity is illustrated by the short latent period and by the high incidence of polyoma tumors in the nude mice inoculated with the virus several weeks after birth at a time when the nu/+ mice are completely resistant to polyoma oncogenesis. Moreover, the cytolytic lesions observed in the kidney and in the salivary glands were also much more conspicuous in the nude mice. The results are discussed in relation to the role of immune surveillance in oncogenesis.     

Metastasis of human tumors in athymic nude mice

The incidence of metastasis of xenogeneic tumors transplanted to nude mice is controversial. We studied 106 malignant human tumor lines in a total of 1,045 nude mice, and observed metastasis in only 14 instances (1.3%), involving 11 different tumor lines. Three of the lines showed repeated metastasis. Breast tumor lines metastasized with significantly greater frequency than other tumor types. None of the sarcoma lines metastasized. Tumors derived from human metastases were no more prone to metastasizing in nude mice than were tumors derived from primary sites. However, deep penetration of the body wall during growth of the tumor transplant was highly correlated with metastasis (p less than 0.001). Such factors as nude mouse health, tumor size and growth rate, and age and sex of the host mouse were not correlated with metastasis. Serial passage in nude mice did not select for a more malignant tumor line, since the incidence of metastasis did not differ at various passage levels. Thus, metastasis of human malignant tumors in nude mice would appear to depend primarily upon the site of tumor growth in the nude mouse, and upon the intrinsic metastasizing capability of the tumor line employed.