Late abstracts 186–187

s 3 tumors and lymphomas. Under these conditions most animals develop measurable growing tumors by day 7 and show multiple lung and lymph node metastases when autopsied on day 21. However, the production of metastases depends on the number of cells injected and on the ATS dose used. Using melanoma cell lines, we showed that this model is highly sensitive, since large numbers of metastases were still obtained following the s.c. implant of as low as 10 s cells, and since tumors and metastases were regularly obtained irrespective of the tumorigenicity of cells in athymic nude mice. From a tumorigenic, but poorly metastatic, melanoma cell line (M4Be), we selected in vivo variants with increased tumorigenicity and metastatic ability, but no variant with organ specificity or high tumorigenicity and low metastatic ability. In vitro cloning of this cell line showed heterogeneity and yielded clones with different metastatic abilities. Some of these clones have been characterized and show specific karyotype features. Thus, the experimental model described allows the quantitative and qualitative study of the metastatic ability of human tumors. 5A I n c r e a s e d ce l lu lar di f ferent iat ion and d e c r e a s e d m e t a s t a t i c potcnt ia l o f FEMX h u m a n m e l a n o m a ce l l s s e l e c t e d through repet i t ive in travenous in jec t ions of m e t a s t a t i c ce l l s in nude m i c e O. FODSTAD, I. KJONNIKSEN, A. BROGGER, V. A. FLORENES and A. PIHL Institute for Cancer Research, The Norwegian Radium Hospital, 0310 Oslo 3, Norway After i.v. injection of F E M X human melanoma cells in athymic, nude mice, extrapulmonary metastases develop within approximately 50 days in tissues such as s.c. lymph nodes and brown fat. In attempts to select a possible subpopulation of F E M X cells with increased metastatic potential, single cell suspensions prepared directly from s.c. metastases were injected i.v. into recipient nude mice, and the procedure was repeated each time a new generation of metastasis developed. Cells from each generation were also injected s.c. into other animals and seeded out in tissue culture flasks, and the growth of the resulting s.c. tumors and monolayers was followed. Contrary to expectations, the lag time before the appearance of new metastases increased with increasing number of passages, and when cells from the fifth generation were used, no metastases developed in the recipient mice. Concurrently, the growth rate of the corresponding in vivo and in vitro cultivated cells decreased. Moreover, increased tyrosinase activity and loss of marker chromosomes indicated increased cellular differentiation as a result of the repetitive in vivo selection. The changes in metastatic capacity and cellular characteristics were found to be reversible, as cells from the non-metastatic fifth generation of variants regained the initial metastatic ability of the F E M X cells when grown for 3 months as s.c. xenografts before i.v. injection. Almost identical results were obtained when a new round of selection experiments were performed. The possible implication of the present findings for the understanding of mechanisms involved in metastasis formation is discussed. 6A Extrapulmonary , t i s sue-spec i f i c m e t a s t a s i s f ormat ion in nude m i e e in jec ted in travenous ly wi th FEMX-I h u m a n m e l a n o m a ce l l s O. FODSTAD, I. KJONNIKSEN, S. AAMDAL, J. M. NESLAND and A. PIHL Institute for Cancer Research, The Norwegian Radium Hospital, 0310 Oslo 3, Norway Extrapulmonary metastases are formed in adult, nude mice after intravenous injection of cells prepared from F E M X I human malignant melanoma xenografts. Tumors develop in subcutaneous, mesenteral, and mediastinal lymph nodes, in brown fat and in the adrenal medulla, whereas lung colony formation is not observed. The s.c. metastases become apparent after a lag time of approximately 50 days. Cells obtained directly from an established in vitro cell line showed a dramatically reduced ability for metastasis formation in vivo. That the