Metastatic variants are generated spontaneously at a high rate in mouse KHT tumor.

Abstract

Using the Luria-Delbruck fluctuation analysis, we have examined the lung tumor-forming ability of a series of parallel clones derived from the KHT tumor, grown to small defined sizes. From these studies, we conclude that metastatic variants arise spontaneously in the clonal lines during their growth, at an apparent rate of 10-5 per cell per generation. This rapid rate has implications for our understanding of tumor heterogeneity and the process of tumor progression. Previous (1)(2)(3)(4)(5)(6)(7)(8)(9)(10). Such results are consistent with the current concept of tumor progression-that mutation-like events spontaneously occur in tumors, generating heterogeneity that, in combination with the selectivity of the host environment, results in the emergence of subpopulations with increased malignancy and metastatic potential (11-15). In our previous study of clonal heterogeneity in KHT mouse sarcoma (10), we observed that the apparent frequency of metastatic cells (i.e., cells with the ability to form lung tumors when injected intravenously into mice) increased significantly with time in culture in certain clonal lines. This raised the possibility that we were observing a de novo generation of metastatic variants. In the present report, we examine the validity of this hypothesis by formally measuring the rate of generation of such presumptive metastatic variants in KHT clonal lines by using the Luria-Delbruck fluctuation analysis. MATERIALS AND METHODSThe cells were derived originally from the mouse KHT sarcoma line (16) and were grown and handled as reported (10). Clonal populations were prepared in vitro and in vivo from KHT tumor cells, and stocks were maintained frozen at the population size indicated. Cells were recovered from the frozen stocks in plastic tissue culture flasks in a-minimal essential medium (17) with 10% fetal calf serum and then cloned in vitro by plating at limiting dilution in 24-well Linbro trays. Subclones ofclones 24 and 35 were transferred by trypsinization to 75-cm2 tissue culture flasks after -10 days and grown to-m5 x 10O cells per clone before injection into mice. Subclones from clones 3 and 13 were dispersed with trypsin on days 11-13 and injected into mice when they were counted in situ to have (1.0-1.7) X 105 cells per well. To allow the testing of essentially the whole of these cell populations in mice, the cell number in Linbro monolayers was determined by counting the number of cells in three microscope fields (x320) chosen at random. The number of cells recovered after trypsinization was found to be linearly related to the number of cells counted by this procedure and the standard curves so produced were used to determine the total number of cells per Linbro well.The metastatic ability ofthe KHT cells was assessed by counting lung tumors from cells injected intravenously as described (10). Briefly, cells from each clonal population were trypsinized and prepared at 5 x 104 cells per ml in a-minimal essential medium with 10% fetal calf serum. They were maintained on ice ...

Origin of metastases: Subspecies of cancers generated by intrinsic karyotypic variations

“…Consistent with a valid theory, the saltational multi-chromosomal theory of the origin of metastases would directly explain the peculiar kinetics of metastasis, namely, that they are unexpectedly "rapid" 43 or arise at "high rates" 44 and at "early and late stages of genetic divergence," 40 that they unpredictably change via "nonlinear progressions" 8,15 and follow a "nongenetic mechanism" 38 and "abrupt" 36 or "punctuated clonal expansions" 24 and generate "unpredictable" phenotypes 36,37 as originally described by Foulds and then in other studies listed in the Introduction.…”

“…41 (8) Studying breast cancers, Navin et al advanced a phylogenetic theory that "tumors grow by punctuated clonal expansions-borrowing a term from species evolution to explain gaps in the fossil record," because metastasis could not be reconciled with the low rates predicted by "a linear progression model." 24 Animal studies have also shown "evolutions" of metastases 42 at "rapid" 43 or "high rates" 44 that exceed those of conventional mutations by orders of magnitude. [44][45][46] All these observations, including Foulds' rules, are hard to reconcile with the conventional view that metastases are end products of multiple steps of sequential mutations.…”

“…24 Animal studies have also shown "evolutions" of metastases 42 at "rapid" 43 or "high rates" 44 that exceed those of conventional mutations by orders of magnitude. [44][45][46] All these observations, including Foulds' rules, are hard to reconcile with the conventional view that metastases are end products of multiple steps of sequential mutations. 19,23,28 Thus conventional mutation theories cannot explain (1) why metastases are karyotypically related to parental cancers but not to those of other cancers and (2) why cancers metastasize at rates that often exceed those of conventional mutations by several orders.…”

Aneuploidy and Cancer: From Correlation to Causation

“…2) and change phenotypes at rates that far exceed those at which genotypes and phenotypes are changed by conventional mutation [53][54][55]. For example, highly aneuploid cancer cells become drug resistant at rates of up to 10 Ϫ3 per cell generation [53,54,[56][57][58] or become metastatic at 'high rates' [59,60]. As a result of this inherent chromosomal instability most cancers are enormously heterogeneous populations of nonclonal and partially clonal, or sub-clonal cells [13,61].…”

“…The mutation hypothesis has to assume mutation rates of up to 10 Ϫ3 per cell generation to explain the frequent, spontaneous variation of phenotypes in highly aneuploid cancer cells. Examples are the 'high rates', compared to mutation, at which some cancers generate metastatic cells [59,60], or generate drug-resistant variants [53,54,56,58]. But the mutation rates of most cancers are not higher than those of normal cells [6,19,20,47,66,[70][71][72][73][74].…”