Chromosomal alterations cause the high rates and wide ranges of drug resistance in cancer cells Ruhong Li a , Ruediger Hehlman b , Rainer Sachs c , Peter Duesberg a, * a Department of Molecular and Cell Biology, Donner Laboratory, University of California, Berkeley, Berkeley, CA 94720 b III. Medizinische Klinik Mannheim of the University of Heidelberg at Mannheim, Wiesbadener Str. 7-11, 68305 Mannheim, Germany c Departments of Mathematics and Physics, Evans Hall, University of California, Berkeley, Berkeley, CA 94720 Received 16 December 2004; received in revised form 30 April 2005; accepted 2 May 2005 Abstract Conventional mutation-selection theories have failed to explain (i) how cancer cells become spon- taneously resistant against cytotoxic drugs at rates of up to 10 23 per cell generation, orders higher than gene mutation, even in cancer cells; (ii) why resistance far exceeds a challenging drugda state termed multidrug resistance; (iii) why resistance is associated with chromosomal alterations and proportional to their numbers; and (iv) why resistance is totally dependent on aneuploidy. We pro- pose here that cancer-specific aneuploidy generates drug resistance via chromosomal alterations. According to this mechanism, aneuploidy varies the numbers and structures of chromosomes auto- matically, because it corrupts the many teams of proteins that segregate, synthesize, and repair chro- mosomes. Aneuploidy is thus a steady source of chromosomal variation from which, in classical Darwinian terms, resistance-specific aneusomies are selected in the presence of chemotherapeutic drugs. Some of the thousands of unselected genes that hitchhike with resistance-specific aneuso- mies can thus generate multidrug resistance. To test this hypothesis, we determined the rates of chromosomal alterations in clonal cultures of human breast and colon cancer lines by dividing the fraction of nonclonal karyotypes by the number of generations of the clone. These rates were about 10 22 per cell generation, orders higher than mutation. Chromosome numbers and structures were determined in metaphases hybridized with color-coded chromosome-specific DNA probes. Further, we tested puromycin-resistant subclones of these lines for resistance-specific aneusomies. Resistant subclones differed from parental lines in four to seven specific aneusomies, of which dif- ferent subclones shared some. The degree of resistance was roughly proportional to the number of these aneusomies. Thus, aneuploidy is the primary cause of the high rates and wide ranges of drug resistance in cancer cells. Ó 2005 Elsevier Inc. All rights reserved. 1. Introduction Conventional mutation-selection theories have failed to answer four critical sets of questions about drug resistance in cancer cells. First: How do cancer cells become spontaneously resis- tant against cytotoxic drugs at the exceedingly high rates of up to 10 23 per cell generation [1–6], whereas for muta- tion the rate per cell generation is restricted to 10 27 for dom- inant and to 10 214 for recessive genes [6–10], even in most cancer cells [9–18]? How can an initial low resistance be enhanced stepwise by more than 1,000-fold [1,3–5,19–28], although multiple mutations of specific genes are extremely unlikely in view of the low rates of conventional gene mu- tation? And how does resistance acquired by many, but not all, cancer cells revert in the absence of selective drugs at about the same high rates at which it was acquired [3–5,23,26,27,29–32]? And why are many cancer cells even intrinsically, or a priori, resistant to cytotoxic drugs, which kill normal cells by definition [24,33,34]? Second: How does the range of drug resistance of cancer cells far exceed a challenging drug, including drugs that are entirely unrelated to the one that induced it [5,28,35–41]? (That state, termed multidrug resistance, is the nemesis of chemotherapy [5,35].) Why is multidrug resistance so com- mon, although simultaneous mutation of multiple genes is extremely unlikely [5,28,33,35–40]? And why are many cancers intrinsically multidrug resistant, although spontane- ous mutation is very rare [24,34]? And how do cancer cells * Corresponding author. Tel.: (510) 642-6549; fax: (510) 643-6455. E-mail address: duesberg@berkeley.edu (P. Duesberg). 0165-4608/05/$ – see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.cancergencyto.2005.05.003 Cancer Genetics and Cytogenetics 163 (2005) 44–56