Different Facets of Copy Number Changes: Permanent, Transient, and Adaptive Sweta Mishra, Johnathan R. Whetstine Massachusetts General Hospital Cancer Center and Department of Medicine, Harvard Medical School, Charlestown, Massachusetts, USA Chromosomal copy number changes are frequently associated with harmful consequences and are thought of as an underlying mechanism for the development of diseases. However, changes in copy number are observed during development and occur dur- ing normal biological processes. In this review, we highlight the causes and consequences of copy number changes in normal physiologic processes as well as cover their associations with cancer and acquired drug resistance. We discuss the permanent and transient nature of copy number gains and relate these observations to a new mechanism driving transient site-specific copy gains (TSSGs). Finally, we discuss implications of TSSGs in generating intratumoral heterogeneity and tumor evolution and how TSSGs can influence the therapeutic response in cancer. I t was long thought that the DNA sequences of healthy individ- uals were 99.9% identical to each other (1). However, genome- wide sequencing efforts in individuals from multiple ethnicities have revealed more variations in the genetic architecture than were previously appreciated (2–4). These genomic alterations have been termed structural vari- ants, which are further classified as microscopic or submicrosco- pic, depending on the amount of DNA involved (5). The micro- scopic variations have historically been identified through chromosome banding techniques (6) and comprise at least 500 kb of DNA (7). Examples of these variants are whole-chromosome gain or loss (referred to as aneuploidy [7, 8]), translocation (change in location of a chromosomal segment [9]), deletion (de- letion of a DNA segment relative to the rest of the chromosome [10]), duplication (a chromosomal segment occurs in two or more copies per haploid genome [11]), and inversion (reversal in orientation of a DNA segment compared to the rest of the chro- mosome [12, 13]). A schematic of structural variants resulting in copy number changes is shown in Fig. 1. With the development of more sophisticated tools, such as array-based comparative genomic hybridization (GGH) arrays (14–16), smaller variants (submicroscopic alterations) in the size range of 1 to 500 kb can be detected (5). Genome sequencing has further revealed small inser- tions and deletions (indels) spanning from 1 to 10,000 bp across the human genome which could cause considerable variability in the human population (17, 18). The most common variant identified under submicroscopic alterations is copy number variation (CNV). CNV is defined as a genomic segment of more than 1 kb present at a variable copy number in comparison to a reference genome (19–22). The first studies documenting the genome-wide presence of CNVs in the normal human genome came from work in the laboratories of Lee (23) and Wigler (24). These studies described more than 200 large-scale CNVs (LCVs; about 100 kb or greater) in normal indi- viduals. These studies also paved the way for the creation of the Database of Genomic Variants (DGV) in 2004, which catalogs all the human CNVs and structural variations present in healthy in- dividuals. The sequencing efforts from the International HapMap Con- sortium (25) and 1000 Genomes Project (26) have led to the iden- tification and frequency determinations of novel CNVs in the hu- man genome. CNVs are now known to contribute to 4.8% to 9.5% of the variability in the human genome (27, 28), which is more than what is accounted for by single nucleotide polymorphisms (SNPs; accounting for 0.1% of the variations) (29). Recently, the CNV map for the human genome was constructed (28), and it documented all the small- and large-scale CNVs present in nor- mal healthy individuals. CNVs can either have no phenotypic con- sequences in individuals (4, 23, 24) or lead to adaptive benefits that have been observed in a wide range of species (5). One of the major challenges in the field is to distinguish benign CNVs (events that do not lead to phenotypic consequences) from pathogenic CNVs that underlie diseases (30). Pathogenic CNVs are often associated with deleterious consequences because of an imbalance in gene dosage (31) and/or aberrant chromosomal structure (5, 7, 32, 33). Pathogenic CNVs have been associated with several disorders, including the following: obesity (34), dia- betes (35), developmental disorders (36), psychiatric diseases (37) such as autism spectrum disorder (38), schizophrenia (39), and Alzheimer’s disease (40, 41), and cancer (42–44). In this review, we focus mainly on copy number alterations observed in cancer and their functional implications. CNVs can either be present in the germ line or can arise in phenotypically normal tissues and organs, which are referred to as somatic CNVs (45, 46). Instead of being randomly present in the genome, CNVs are preferentially found to occur in regions that are rich in low-copy-number repeats (segmental duplications) (47–50), heterochromatic areas (e.g., telomeres and centro- meres), and replication origins and palindromic regions (28). There are several proposed mechanisms that underlie the genera- tion of somatic CNVs: nonallelic homologous recombination (NAHR), nonhomologous end joining (NHEJ), defects in DNA Accepted manuscript posted online 11 January 2016 Citation Mishra S, Whetstine JR. 2016. Different facets of copy number changes: permanent, transient, and adaptive. Mol Cell Biol 36:1050 –1063. doi:10.1128/MCB.00652-15. Address correspondence to Johnathan R. Whetstine, jwhetstine@hms.harvard.edu. Copyright © 2016 Mishra and Whetstine This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license. MINIREVIEW crossmark 1050 mcb.asm.org April 2016 Volume 36 Number 7 Molecular and Cellular Biology