Stability without a centromere

Abstract
Every chromosome is precious. Each one carries vital genetic information that must not be lost. To ensure chromosome retention and segregation when the cell divides, the spindle apparatus attaches to the chromosomes at the centromere. In higher eukaryotes, the centromere is a big region, often occupying a megabase or more in mammalian chromosomes. Dedicating large amounts of DNA to centromere function is not a problem for mammalian cells. They carry thousands of megabases of DNA in their genome, the great majority of which does not code for proteins. But what about a virus? Compactness is the viral strategy. If a virus adopts the attractions of latency, remaining in the cell until conditions are right for lytic growth, how can it stably remain in the nucleus? A centromere seems out of the question. A different strategy is outlined in the paper by Lehman and Botchan in this issue of Proceedings (1). Once in the nucleus, a virus bound for latency must take action to ensure its survival against loss through dissolution of the nuclear membrane upon mitosis or through nuclear pores. An association with the chromosomes would ensure viral retention by taking advantage of chromosomal segregation and stability. Some viruses, like retroviruses and adeno-associated viruses, integrate into the chromosomes, forming covalent bonds that ensure that the viral genome will travel with the chromosomes. A less sure, but adequate, method to colocate with the chromosomes is to rely on a noncovalent association with chromosomes. This is the route of bovine papilloma virus (BPV) (Fig. 1). Two ways in which viral genomes can take advantage of the chromosomal centromere to maintain themselves long term in mammalian cells. (A) Integration is a covalent interaction with the chromosomes. It is the maintenance mode of viral genomes such as retroviruses. (B) Genomes such as BPV …