Plasmids R1 and ColE1, which control their N-values by ctRNAs, also encode a protein as a second inhibitory element that, however, plays only an auxiliary role. In the case of R1, expression of the essential gene repA requires the translation of the leader gene tap, which is translationally coupled to repA (Blomberg et al., 1992). The main replication control element is the ctRNA, CopA, which inhibits translation of tap and, indirectly, that of repA. The second inhibitory element of R1 is the product of the copB gene, which is co-transcribed with tap and repA from promoter P1(Fig. 2). CopB protein is a transcriptional repressor of a second promoter, P2, located downstream of copB, which directs the synthesis of a tap–repA mRNA. At steady state, CopB is present at saturating concentrations, blocking transcription from P2, so that repA is expressed almost exclusively from P1. Deletion of the entire copB gene (including promoter P1) results in plasmids with an eightfold increase in N (Riise et al., 1982). The CopB regulatory loop has been suggested to serve as a rescue mechanism that prevents plasmid loss in newborn cells harbouring very few plasmids. However, computer simulation of mini-R1 plasmid replication indicated that the CopB regulatory circuit contributes little to the stability of these replicons (Rosenfeld and Grover, 1993).
The second instance of plasmids with auxiliary proteins is ColE1 (Fig. 3). In this case, replication is mediated by the synthesis of a preprimer RNA (RNA II) by the host RNA polymerase (RNAP) and the formation of a DNA–RNA hybrid between the RNA II and the template DNA strand at the origin region. This hybrid is cleaved by RNase H, generating a 3′-OH end, which is used by DNA polymerase I to initiate leading strand synthesis. The availability of the primer 3′-OH end is rate-limiting for initiation, and it is modulated by the ctRNA I control element. Interaction between ctRNA I and its complementary region in the preprimer alters the secondary structure of the latter, leading to the inhibition of stable DNA–RNA hybrid formation. This, in turn, leads to inhibition of replication. The second element of this circuit is protein Rom (Rop), which enhances the rate of formation of a stable complex between the ctRNA I and the preprimer RNA. Rom does not seem to be an essential component of the ColE1 control system. Deletion of the rom gene leads to a two- to threefold increase in N in slowly growing cells, but it has no phenotypic consequences on the N-value in fast-growing bacteria (Atlung et al., 1999). When cloned on a compatible multicopy plasmid, the rom gene is able to complement a rom− derivative, although it has no further effect upon the replication of a co-resident ColE1. The absence of incompatibility caused by extra copies of rom shows that Rom is not a primary inhibitor of ColE1 replication, as it exerts its maximum effect at the wild-type concentration (Summers, 1996). Mathematical models of the dynamic features of copy number control in ColE1 and the experimental observations about the small effect caused by variations in the dosage of the rom gene have left open the question of why there is a Rom protein (Paulson et al., 1998). At least three theoretical proposals have been made to account for an important role of Rom in the dynamics of ColE1 copy number control (Paulson et al., 1998). First, Rom concentration would be proportional to the N-value, so that the response in replication frequency to variations in the N-value would be sharper than if RNA I acted alone. This hypothesis requires that Rom is rapidly degraded, which has not been tested experimentally. Secondly, Rom would act by making the probability of plasmid replication very close to zero at high ctRNA I concentration because, in the absence of Rom, the intrinsic rate of ctRNA I–RNA II duplex formation would be too slow to ensure total inhibition of replication. Thus, Rom would ensure an efficient copy number control system. Thirdly, Rom could act as a back-up system when N (and, subsequently, Rom concentration) is greatly reduced: under normal conditions, the replication frequency would not depend on small deviations in Rom concentration but, if this concentration decreases greatly (as a result of a large reduction in N), inhibition of primer formation would decrease, thus leading to an increase in the replication frequency. However, and as far as we are aware, no experiments have been performed to clarify these hypotheses. On the other hand, experimental evidence has shown that the presence of ColE1 derivatives lacking rom reduced bacterial growth in medium impoverished in carbon sources, whereas rom+ derivatives did not show an adverse effect on cell growth (Atlung et al., 1999). This is thought to be related to the amplification of ColE1-type plasmids in slowly growing cells, which is higher in plasmids lacking rom. From these observations, a key role for Rom protein has been envisaged, that is to prevent ColE1-type plasmids from representing a metabolic burden to their hosts in natural habitats, in which cells grow slowly because of nutrient limitation.
In conclusion, CopB and Rom are inhibitory elements that limit plasmid replication, as their inactivation leads to an increase in the N-value of R1 or ColE1 respectively. However, the genes encoding these proteins do not constitute an incompatibility determinant against the wild-type plasmid, so that both proteins would be auxiliary negative elements unable to correct up-fluctuations in N. However, the influence of these proteins on the dynamics of plasmid copy number control remains to be solved.