How replication of a chromosome is initiated has been a major subject of molecular genetics, because the regulation of chromosomal replication occurs primarily at the stage of initiation (see DNA Replication). Molecules that might be involved in the initiation of replication of the bacterial circular chromosome were proposed by Jacob et al. in 1963 (1). They assumed that each replicon consists of two elements participating in positive regulation of the initiation: (1) a cis-acting DNA signal (the replicator, which is now called the replication origin), and (2) a trans-acting protein factor (the initiator), which would interact directly with the replicator. The positive regulation for the initiation of chromosomal replication arose from the observation that protein biosynthesis is required for the initiation of each round of replication in Escherichia coli. Later, it was found that initiation occurs when each individual cell reaches a fixed amount of cell mass relative to the number of the replication origins, regardless of the growth rate (2). To explain these phenomena, the initiator accumulation theory was proposed (3). In this model, an initiator is synthesized and accumulates in the cell at the same rate as the cell mass increases, and initiation takes place every time a certain number of initiators accumulate for each replication origin. The model also proposed that the initiator should act as a repressor of its own gene expression to maintain the cellular content of the protein (autogenous regulation) as constant and so it is accumulated in concert with an increase in cell mass. Although Jacob and his colleagues merely stated that the model was a simplified one, their far-sighted hypothesis has now proven to be correct concerning the prokaryotic replicons, those of bacteria, bacteriophage, and plasmids (see DNA Replication).
1. Initiation of Bacterial Plasmid Replication
The replication of bacterial plasmids and phage has been shown to start from a unique site on the replicon (the origin, see Replication Origin), and most plasmids and phage encode a gene, often designated rep, for a replicon-specific protein required for the initiation of DNA replication. The replicon-specific Rep proteins have been shown to recognize specific sequences at the origin and to act positively to recruit host-encoded replication enzymes (4). The initiation frequency of the replication of plasmids is regulated so as to maintain a steady-state copy number characteristic of the particular plasmid in a given host bacterium (5). The majority of plasmids in Gram-negative bacteria contain direct repeats of the Rep binding sequence (iterons) within the origin. Many of these plasmids also contain indirect repeats of sequences related to the iterons within the promoter region of the rep gene, and the expression of the Rep protein is autoregulated, as proposed in the initiator accumulation theory. Direct repeats within the functional origin of an iteron-containing plasmid, when inserted into a plasmid normally compatible with the iteron-containing plasmid, show strong incompatibility with the plasmid. In addition, insertion of additional homologous iterons into an iteron-containing plasmid results in a reduction of the copy number of the plasmid. Based on these observations, it was proposed initially that replication begins when a sufficient amount of Rep protein is accumulated to cover the iterons at the origin, and additional Rep binding sequences reduce the effective amount of the initiator (6). However, extensive studies conducted on the regulation of initiation of the iteron-containing plasmids have revealed properties that argue against the simple titration model (5). For example, increasing the concentration of the Rep protein of several plasmids in vivo did not result in a proportional increase in the copy number. Several Rep proteins have been shown to exist as two forms; dimers that bind to the iterons in the origin and monomers that regulate its own expression. Although new models that modify the simple titration model to explain the regulation of the initiation of plasmid replication have been proposed, much remains to be learned to understand it completely (5).
2. Initiation of Replication of Bacterial Chromosome
A number of genes have been identified as being involved in the initiation of chromosomal replication in bacteria. The functions of these genes during the initiation process have now been well defined through genetic and biochemical studies in E. coli (4, 7). DnaA protein was found to function at the first step in the initiation process by binding to specific DNA repeats (the DnaA box) in the replication origin of the chromosome (oriC), and then opening the double strand for formation of the primosome, the machinery for the first primer synthesis (see DNA Replication). The combination of a DnaA box and DnaA protein functioning as cis and trans regulatory elements, respectively, in the initiation of chromosomal replication was subsequently found to be common in eubacteria (7, 8).
Expression of the E. coli dnaA gene is negatively regulated by the DnaA protein itself, by the protein binding to DnaA boxes located in the promoter region of the gene, and the concentration of the DnaA protein is kept proportional to the origin concentration. Overproduction of the DnaA protein stimulated the initiation of chromosomal replication up to two-fold, and that stimulation was proportional to the increase in the DnaA protein concentration within this limited range. DnaA overproduction also resulted in a two-fold increase in minichromosome (oriC plasmid) copy number. Thus, E. coli DnaA protein is considered to fulfill the qualifications required for an initiator that controls the timing of initiation of replication during the cell cycle, and the time of initiation is assumed to be set by the DnaA protein accumulating to a threshold level (9). However, it should be noted that other factors affecting initiation control have been identified. The characteristic of oriC of enterobacteria is the presence of many GATC sequences, which are sites for the Dam restriction-modification methylase. It was found that newly replicated oriC DNA strands are not methylated during up to one-third of the cell cycle after the replication. Such hemi-methylated DNA tends to bind to the cell membrane and does not act as a template for activation by DnaA protein. Furthermore, initiation of replication occurs more or less at random in relation to the cell cycle in a mutant lacking Dam methylase (7). The DnaA protein can be modified in different ways by binding ATP or ADP. Phospholipids were found to reactivate the inactive form (ADP-DnaA) in vitro by exchanging ADP for ATP. Initiation from oriC seemed to be inhibited in vivo in a mutant in which the synthesis of phospholipids was limited (10).
The dnaA gene is also autoregulated in Bacillus subtilis. However, expression of the B. subtilis dnaA gene is coupled to initiation of replication of the chromosome. It is assumed that the autorepression by interaction of DnaA protein with many DnaA boxes near the promoter is so strong in B. subtilis that the gene is expressed only when the chromosome undergoes vigorous conformational changes during the replication of the gene itself (11). In B. subtilis, fewer than one oriC plasmid/cell is allowed to coexist with the chromosome. Introduction of extra copies of DnaA boxes would be incompatible with chromosomal replication. This stringent control of the initiation observed may be ascribed to the strong negative control of dnaA expression. Accumulation of the DnaA protein to a threshold level was shown to be prerequisite for the initiation of replication in B. subtilis. However, control of the timing of the initiation may require other factors, because the DnaA protein seems to be synthesized only in the early stage of the replication cycle in this cell. Although the role of DnaA as a key factor for regulation of initiation is clear in both E. coli and B. subtilis, the overall view of the regulation of the timing of initiation is not still clear. And, different devices to control the timing seem to have been evolved in two bacteria.
3. Initiation in Eukaryotes
A unique initiator protein (T Antigen) has been identified to regulate the initiation of replication of the eukaryotic SV40 virus. However, no single initiator protein like bacterial DnaA has been found for eukaryotic chromosomes. Instead, multiprotein complexes have been identified to initiate replication of the multiple replicons of eukaryotes. Furthermore, a complex network of protein-protein interactions (complex formation) and protein modifications ( phosphorylation) regulates the initiation (see DNA Replication).