Non-cell-autonomous mechanisms of tumor initiation and relapse by chromosomal instability

Abstract

Chromosomal instability (CIN) is a hallmark of cancer, and a primary cause of genetic heterogeneity in tumors. Depending on the degree of CIN and the affected tissue, CIN can promote or suppress tumor formation, and high CIN induction has been proposed as a therapeutic strategy. How CIN achieves these effects is unclear. Here we use a conditional mouse model of graded CIN in combination with longitudinal monitoring of DMBA/TPA-initiated skin tumors to show that low CIN increases the frequency of skin tumor initiation, while higher CIN accelerates onset and growth rates. Strikingly, gene recombination analysis of the tumors reveals that upon high CIN induction the fast-growing tumors originate from rare low CIN cells, suggesting a strong non-cell-autonomous effect of high CIN. Gene expression analysis and immunohistochemistry show that high CIN causes epidermal hyperplasia, immune evasion and a regenerative response that stimulates low CIN tumor growth beyond what is achieved by induction of low CIN alone. Such cell-extrinsic effects may be a common mechanism of tumor formation by CIN, as we observe it also in CIN-induced tumors of the intestine, breast and mesentery. When high CIN is induced in established skin tumors, mimicking CIN-based therapy, tumors regress but relapse quickly. Relapsed tumors, too, arose from rare low CIN cells. Our findings have implications for our understanding of the contributions of CIN to cancer initiation and progression and give caution to the rationale for CIN therapies. Significance statementChromosomal instability (CIN) is a source of genetic diversity in cancer. Different degrees of CIN can be highly oncogenic or tumor suppressive, the latter being a rationale for CIN-based therapies. Using a mouse model for precise control of CIN levels, we show that low CIN promotes tumor initiation, while moderate CIN accelerates both initiation and tumor growth. Surprisingly, moderate CIN acts through its effects on the surrounding tissue rather than directly within tumor cells, by triggering a regenerative response accompanied by immune evasion. When very high CIN is induced in existing tumors, tumors initially regress but rapidly relapse. These findings inform on the contributions of CIN to cancer initiation and progression and give caution to the rationale for CIN therapies.