In 1980, Ronald W. Davis and coworkers provided significant new insights into eukaryotic DNA replication when they discovered yeast DNA sequences, called autonomously replicating sequence elements(ARS elements), that allow plasmids containing selectable genes to replicate autonomously in yeast cells. The experiment illustrates one method for demonstrating the existence of ARS elements. Two sets of plasmids, one containing an ARS element and LEU and the other just containing LEU, are introduced into leu- yeast cells, which are then spread on agar growth medium lacking leucine. A large number of colonies form when cells transformed with ARS-plasmids are spread on the agar medium because the plasmids can replicate autonomously. In contrast, very few colonies appear when plasmids lacking ARS are spread on the agar medium and the colonies that do form contain cells in which LEU is integrated into the yeast genome.
Yeast cells have approximately 400 ARS elements, while cells from higher animals probably have thousands. Most yeast ARS elements are named by using a number that designates their chromosome and position on the chromosome. Although the ARS elements were initially named because of their ability to support autonomous plasmid replication in yeast, they function as replicators, tors in yeast and overlap sites of replication initiation. There are two main reasons why each yeast chromosome requires several ARS elements while the E. coli chromosome requires just one replicator. First, yeast forks migrate about 30 times more slowly than those in E. coli. Second, yeast chromosomes are much longer than the E. coli chromosome.
ARS plasmids that lack centromeres are preferentially retained by the mother cell, leading to low copy numbers in the daughter cells and high copy numbers in the mother cells. When ARS plasmids also contain centromeres, more equal plasmid segregation is observed between mother and daughter cells. Attempts to isolate ARS units from multicellular animals by constructing plasmids have not yet been successful. One possible reason for the lack of success may be that nuclear membranes remain intact throughout the entire yeast life cycle, whereas nuclear membranes disassemble to form membrane vesicles during prophase in animal cells. Therefore, even if plasmid replication occurred in animal cells, the plasmids would escape when the nuclear membrane disassembles. A possible solution to the problem is to incorporate a centromere into the plasmid, but animal cell centromeres are too large to be included in simple plasmids.
Several different observations suggest that ARS elements direct the initiation of yeast DNA replication. For instance, ARS plasmids and yeast chromosomes both are replicated only during the S phase of the cell growth cycle and both require the same gene products for their replication. One may therefore hope to obtain considerable information about the function of ARS elements by studying them in ARS plasmids, where they are easily manipulated. Two ARS elements that have been studied in greatest detail, ARS1 and ARS307. Sequence analysis revealed that ARS elements are more A/T rich than surrounding chromosomal DNA. Furthermore, all contain an 11-bp element [5'-(A/T) TTTA (T/C) (A/G) TTT (A/G)-3'] known as the A-element or ARS consensus sequence (ACS).
