The Plasmid-Encoded parAB-parS System
Low-copy-number linear plasmids of Streptomyces contain a pair of partitioning genes, parA and parB, which are involved in active segregation and thus stable inheritance of these plasmids. During segregation, ParB binds to a specific palindromic DNA sequence (parS) on the replicons and forms a partitioning complex, while ParA, an ATPase, forms cytoskeleton-like structures and is responsible for segregating the two daughter replicons (for review, see Gerdes et al. 2000; Bignell and Thomas 2001; Shih and Rothfield 2006). The vegetative mycelium of Streptomyces contains few septa with each compart ment containing many copies of chromosomes, and therefore precise partitioning of the genomes appears not to be critical during vegetative growth. Proper segregation and partitioning, on the other hand, is important during sporulation, when multiple septa are formed in the aerial hyphae to produce unigenome spores. Limited studies on partitioning of linear plasmids have been done on SCP1 and SLP2.
Interestingly, SCP1 contains two partitioning loci, each consisting of a parA-parB operon (Bentley et al. 2004). The protein pairs encoded by the two operons are more than 50% identical to each other (and more than 30% identical to the chromosomally encoded ParABSco), and both pairs are imperfectly related to the type Ia partitioning system (Gerdes et al. 2000). The presence of two partitioning systems is unusual, and may reflect the mosaic nature of this plasmid as a result of past recombination between heterologous plasmids.
The partitioning function encoded by SCP1 has been demonstrated by the ability of the integrated SCP1NF (Chater and Kinashi, in this volume, for details on SCP1NF) to suppress partially the chromosome partitioning defect during sporulation that is caused by a chromosomal AparABSco mutation (Bentley et al. 2004). No parSSco is present in SCP1, and presumably ParABSCP1 acts on different sequences. Therefore, one may assume that the partitioning activity restored to the AparAB host is exerted on an as yet unidentified parSSCP1 element present in the integrated SCP1NF.
SLP2 of S. lividans also contains a type Ia-like locus in the central region. In a typical type Ia locus, parS is adjacent to and downstream from the parAB operon. parSSLP2 (two short inverted repeats), however, lies within the coding sequence of parB and encodes the same amino acid sequences (C.-C. Hsu, unpublished results). The binding of ParB to its own coding sequence raises the possibility that it may regulate its own synthesis by blocking transcription elongation.
Deletion of parABSLP2 resulted in loss of SLP2 from spores of S. coelicolor, and the loss was more severe in an S. coelicolor mutant with a chromosomal AparABSco mutation, indicating that the chromosomal partitioning system may affect partitioning of SLP2, although parSSco is likewise absent from SLP2. The interesting cooperative interactions between the chromosomal and plasmid partitioning systems await further elucidation.
On linear plasmid SAP1 (94 kb) of S. avermitilis (Ikeda et al. 2003), parASAP1 exists without an accompanying parB gene. Such unpaired parA genes are also found in some linear plasmids in other actinomycete hosts, e.g., pCLP in Mycobacterium celatum (Accession number: AF312688), pRHL3 in Rhodococcus sp. RHA1 (http://www.rhodococcus.ca), pBD2 (NC_005073) in Rhodococcus erythropolis BD2, and pREL1 in R. erythropolis PR4 (Sekine et al. 2006). The biological significance of these lone parA genes on linear plasmids is not clear.
Resolution of Sister Plasmids
The apparent interaction between the TPs at the telomeres of a linear Strepto-myces replicon poses three potential complications during segregation. During and after replication, if the parental TPs persist in associating with each other, the 5' ends of the two parental DNA would also remain linked with each other, thus creating a novel Möbius strip-like situation, i.e., the two daughter DNA molecules cannot be separated from each other (Fig. 4A-D). To resolve this Möbius strip-like structure, the two parental TPs perhaps dissociate from each other and each finds a new TP partner (Fig. 4E). In this case, a mechanism must exist to allow the recognition and sorting of the old and new TPs. Alternatively, an exchange may occur between the daughter DNA molecules to resolve the structure (Fig. 4F). Such a solution is similar to the resolution of dimeric circular replicons by the XerCD recombinase (reviewed by Barre et al. 2001). Is there such an analogous site-specific recombination system in Streptomyces, or is the resolution carried out by less efficient homologous recombination? It is noteworthy that in the latter scenario, replication of the Streptomyces linear replicons would be semi-conservative with respect to DNA but conservative with respect to TPs.
A second complication may arise from homologous recombination. Between two circular sister chromosomes or plasmids, homologous recombination generates dimers that must be resolved to achieve segregation. The resolution is carried out by the site-specific recombinase XerCD on a specific sequence—dif on chromosomes and cer on plasmids. Linear DNA molecules are not expected to dimerize through homologous recombination. However, a persistent TP-TP interaction during and after replication may also result in pseudodimer formation through homologous recombination. Interestingly, the recombination products in fact would be identical to the replication products from a persistent TP-TP association (Fig. 4D), which thus may be resolved by the two alternatives outlined above: TP exchange or DNA exchange. No homologs of xerCD genes are found in the S. coelicolor genomes. Perhaps dimer resolution is solved by a different recombination system, or by TP exchange.
Another topological problem encountered during the replication of circular DNA is the formation of concatemers between two intertwined sister DNA molecules. In E. coli, an essential topoisomerase Topo IV (encoded by parC and parD) is responsible for decatenation of such concatemers (Adams et al. 1992; Zechiedrich and Cozzarelli 1995). Streptomyces genomes contain a pair of parC and parD homologs (SCO5822 and SCO5836 in S. coelicolor; SAV2422 and SAV2423 in S. avermitilis), presumably encoding Topo IV. Attempts to knock out these genes in S. coelicolor have not been successful (T.-W. Huang, unpublished results), suggesting that Topo IV is also essential for decatenation of the linear chromosomes (and presumably linear plasmids)
TP exchange DNA exchange
Fig. 4 Postreplicational Möbius strip-like problem. A A linear replicon assumes a circular configuration caused by interactions between the terminal proteins. The DNA duplex is colored in solid black, and the TPs are represented by the filled black circles. The association between them is emphasized by the connecting line. B Replication begins at an internal origin. The newly synthesized strands are colored gray. C Bidirectional replication reaches the telomeres. D When end patching is complete, if the parental TPs remain associated to each other, the two sister replicons form a Möbius strip-like structure with 5'-5' linkages. The Möbius strip-like structure may be resolved by means of two possible alternative pathways. E In one pathway, the parental TPs dissociate from each other and find proper "daughter" TPs (filled grey circles) to pair with. F In the other pathway, a recombination (at sites indicated by the vertical lines in D) resolves the pseudodimer
Fig. 4 Postreplicational Möbius strip-like problem. A A linear replicon assumes a circular configuration caused by interactions between the terminal proteins. The DNA duplex is colored in solid black, and the TPs are represented by the filled black circles. The association between them is emphasized by the connecting line. B Replication begins at an internal origin. The newly synthesized strands are colored gray. C Bidirectional replication reaches the telomeres. D When end patching is complete, if the parental TPs remain associated to each other, the two sister replicons form a Möbius strip-like structure with 5'-5' linkages. The Möbius strip-like structure may be resolved by means of two possible alternative pathways. E In one pathway, the parental TPs dissociate from each other and find proper "daughter" TPs (filled grey circles) to pair with. F In the other pathway, a recombination (at sites indicated by the vertical lines in D) resolves the pseudodimer of Streptomyces after replication. This may reflect the circular configuration formed by the linear Streptomyces replicons through persistent TP-TP interactions. Alternatively, a decatenase such as Topo IV is required to untangle the large chromosomes after replication irrespective of their topology, as is the case with eukaryotic chromosomes.
Was this article helpful?