1 Review paper.
Figure 1 Circular plasmid replication by the rolling-circle model (RCR). The light and heavy lines in each part of the plasmid diagram represent the leading and lagging strand of DNA, respectively. Key events include (1) binding of the replication initiator protein Rep (whose active form may be mono-, di-, or multimeric) to the double-strand origin (ori) produces a structural change in the DNA at ori (e.g., cruciform DNA in pT181); (2) Rep then nicks the leading strand at a specific site within ori, and an initiation complex is formed between Rep and host replication factors such as DNA polymerase III, DNA helicase, and single-stranded-DNA binding protein; (3) DNA replication from the Rep-dependent nick site proceeds with leading strand displacement until ori is regenerated; (4) Rep, which is believed to remain in close proximity to the replication fork, terminates replication via sequential nicking-closing reactions at ori. This releases a circular leading strand of DNA and an inactivated Rep protein (Rep*) and produces a regenerated doublestrand plasmid; (5) Lagging-strand replication is then initiated at the single-stranded origin (sso) exclusively by host-encoded proteins that may include RNA polymerase and the DNA polymerases I and III. Other host factors such as DNA ligase and DNA gyrase, are also likely involved in plasmid RCR. (Adapted from Khan, 1997.)
control mechanisms are incompatible, because the control systems cannot distinguish between each molecule, and so replication of either plasmid becomes random (Snyder and Champness, 1997). Incompatibility between RCR plasmids from the same family has been noted, and this phenomenon was attributed to cross recognition between each molecule's Rep proteins and ori sequences (Groh-mann et al., 1998). Plasmid incompatibility between gene-cloning vectors and native plasmids may also contribute to low transformation efficiencies in LAB (Luchansky et al., 1988; Posno et al., 1991; Van der Lelie et al., 1988).
b. Theta Replication In contrast to RCR, theta-type plasmid replication does not involve formation of large regions of ssDNA, and so theta plasmids are far less vulnerable to DNA rearrangements. The practical significance of this attribute is highlighted most effectively by studies on Bacillus subtilis that showed cloning vectors derived from theta replicons can stably accommodate very large (>300-kb) or multimeric DNA inserts (Itaya and Tanaka, 1997; Lee et al., 1998). Improved stability of large-insert DNA in vectors derived from theta replicons has also been demonstrated in Lc. lactis (Kiewiet et al., 1993).
Replication of theta plasmids involves strand separation at one or more specific loci, synthesis of an RNA primer, and then progressive uni- or bidirectional DNA replication with simultaneous synthesis of leading and lagging strands. Theta-type replicons are very common in gram-negative (gram") bacteria but, as noted above, they appear to occur less frequently than RCR plasmids in LAB and other gram+ bacteria. Nonetheless, theta replicons have been identified on small, intermediate-sized, and large plasmids from Lb. helveticus, Lb. sake, Lc. lactis, P. pentosaceous, T. halophilus, and from several enterococci and pathogenic streptococci (Benachour et al., 1997; Bruand et al., 1993; Kantor et al., 1997; Kearney et al., 2000).
Differences in genetic structure and the requirement for host encoded DNA polymerase I during replication can be used to separate eubacterial theta replicons into six distinct classes, designated A-F (Bruand et al., 1993; Del Solar et al., 1998; Meijer et al., 1995; Tanaka and Ogura, 1998). Class A replicons encode a replication initiation protein, Rep, and have an origin of replication, oriA, composed of an AT-rich region and a series of short, directly repeated sequences called iterons (which also play an important role in regulation of plasmid copy number). These plasmids do not require host DNA polymerase I for replication. Class B, C, E, and F replicons are distinguished by the absence of a typical oriA sequence, the presence of a plasmid-coded Rep protein (class C and F replicons), and a requirement for DNA polymerase I (classes B and C). Class D replicons encode Rep and have an oriA-like sequence, but it is not required for replication. They are also similar in structure and in their requirement for DNA polymerase I to class C replicons, but the replicative regions of class D and C plasmids lack any significant DNA sequence homology (Bruand et al., 1993). Like RCR
plasmids, theta-type replicons may also encode a rep repressor protein or antisense RNAs that serve to regulate plasmid copy number (Actis et al., 1999).
Many theta plasmids that have been identified in LAB appear to possess a class A replicon, and most of these have been isolated from Lc. lactis (Kearney et al., 2000). However, class D replicons have been found in enterococci and pathogenic streptococci, and it now looks as though several LAB species may possess class F theta replicons (Kearney et al., 2000). As with basic studies of RCR replicons, research into theta plasmid replication in LAB will continue to provide new insight into basic mechanisms for plasmid replication, copy control, and segregation in gram+ bacteria (Bruand et al., 1993; Gravesen et al., 1997; Kearney et al., 2000). Because these factors are directly related to plasmid incompatibility (Actis et al., 1999), studies in this area will also facilitate strain improvement efforts that involve introduction of extrachromosomal vectors into LAB hosts that contain native plasmids. On this note, it is important to point out that although many Lc. lactis class A theta replicons share regions of high sequence homology, these plasmids are often compatible with one another (Gravesen et al., 1995). Nonetheless, incompatibility groups and determinants have been identified for some theta plasmids in LAB (Gravesen et al., 1997; Seegers et al., 1994), and additional research is needed to define plasmid incompatibility groups within and among (for broad host range plasmids) different species of LAB.
Transposable elements are discrete sequences that have the ability to move from one site to another in DNA. Three types of mobile genetic elements have been found in LAB: insertion sequences (IS), transposons, and introns. By virtue of their mobility, these elements promote genetic rearrangements that can affect the organization, expression, and regulation of existing genes. In addition to inser-tional inactivation of target or adjacent genes, transpositional elements can also induce expression of flanking genes. The latter activity is thought to result from creation of new promoters that comprise an out-directed -35 promoter consensus sequence that is present in terminal inverted repeats of some elements, and an appropriately spaced -10 hexamer in DNA that flanks the insertion site (Mahil-lon and Chandler, 1998).
Transposons and IS elements also promote more extensive forms of intra-genomic rearrangements such as cointegrations, inversions, and deletions. Comparative genomic analysis of Lc. lactis, for example, has revealed that an inversion encompassing approximately half of the chromosome in strain ML3 is the result of homologous recombination between two copies ofIS905 (Daveran-Min-got et al., 1998). Insertion sequence-mediated plasmid cointegration is also well documented in this species (Anderson and McKay, 1984; Polzin and Shimizu-Kadota, 1987; Romero and Klaenhammer, 1990).
Finally, transposable elements can contribute to genetic variation in bacteria by facilitating horizontal gene transfer between different strains, species, and genera (Arber, 2000; Brisson et al., 1988). Among the LAB, transposons play an important role in dissemination of virulence factors among pathogenic entero-cocci and streptococci (Horaud et al., 1996; McAshen et al., 1999; Teuber et al., 1999), and recent evidence suggests IS elements were involved in horizontal transfer of genes for exopolysaccharide production between Lc. lactis and S. ther-mophilus (Bourgoin et al., 1996 and 1999). From a more practical perspective, transposable elements can be useful tools for molecular analysis of LAB genetics, physiology, and metabolism, and for development of integrative gene cloning vectors (Dinsmore et al., 1993; Israelsen et al., 1995; Le Bourgeois et al., 1992b; Maguin et al., 1996; Polzin and McKay, 1992; Ravn et al., 2000; Walker and Klaenhammer, 1994).
The IS described in LAB range in size from approximately 0.8 to 1.5 kb, with 16-40 bp inverted repeats on left and right ends (Table 2). Like other prokaryotic IS, they are compact elements that only encode transposase and cis-acting sequences required for transposition, and their location is almost always flanked by short, direct repeats (3-8 bp) that reveal the target sequence used for insertion into new sites (Mahillon and Chandler, 1998). Mechanisms involved in IS transposition are both varied and complex, and they are quite beyond the scope of this chapter. Readers interested in this topic are referred to the reviews of Haren et al. (1999) and Mizuuchi (1992).
Discovery of the first IS element in LAB arose from a series of elegant experiments to ascertain the cause of abnormal fermentations during production of a fermented skim milk beverage (Shimizu-Kadota and Sakurai, 1982; Shimizu-Kadota et al., 1983, 1985). Those studies showed that abnormal fermentations at several factories were caused by the same virulent bacteriophage, designated ^FSV, which was serologically, morphologically, and biochemically identical to a temperate phage (^FSW) harbored by the starter bacterium, Lb. casei S-1 (Shimizu-Kadota and Sakurai, 1982; Shimizu-Kadota et al., 1983). Structural analysis of the ^FSV and ^FSW genomes revealed ^FSV contained 1.3 kb of additional DNA, and nucleotide sequence analysis revealed this region contained an IS, designated ISLi. Southern hybridization showed ISLi was present on the Lb. casei S-1 chromosome, which led to the conclusion that ^FSV arose from ^FSW by ISLi transposition from the chromosome to a region of the prophage that controlled lysogeny (Shimizu-Kadota et al., 1985). With this knowledge, the Yakult company was able to isolate a prophage-cured derivative of Lb. casei S-1 and eliminate further emergence of ^FSV in their factories (Shimizu-Kadota and Sakurai, 1982).
Table 2 Insertion Sequences in Dairy Lactic Acid Bacteria
Original host and Inverted element name" Size (bp) repeat (bp)b IS Family0
ISL7 1256 40 ISi
ISL2 858 16 IS5
ISL3 1494 38 ISL5
ISLhi 962 35 IS982
1S125 1024 24 1S30
IS1163 1180 39 ISi
IS1201 1387 24 IS256
IS1223 1492 25 ISi Lactococcus lactis
IS214 809 23 IS6
IS275 (IS1077) 1448 14 ISi
15904 1241 39 ISi
15905 1313 28 IS256
IS981 1222 40 ISi
Copies per genome Host range (references)"1
1-3 Lb. casei subsp. casei. Lb. zeae (Shimizu-Kadota et al., 1985, 1988)e
4-21 Lb. helveticus (Zwahlen and Mollet, 1994)f
1-9 Lb. delbrueckii subsp. bulgaricus (Germond et al., 1995)g
NDh Lb. helveticus (Pridmore et al., 1994)
NDh Lb. plantarum (Ehrmann et al., 2000)
2 Lb. sake (Skaugen and Nes, 1994)
3-16 Lb. helveticus (Tailliez et al., 1994)
NDh Lb. johnsonii (Walker and Klaenhammer, 1994)
1-20 En. faecium. En. hirae. Lb. plantarum. Lc. lactis. In. mes-
enteroides subsp. dextranicum, S. thermophilus (Bour-goin et al., 1996; Ehrmann et al., 2000; Polzin and Shim-izu-Kadota, 1987; Polzin et al., 1993; Ward et al., 1996)' 3' En. faecium. Lc. lactis (Teuber et al., 1999)
l-7k Lc. lactis (Bolotin et al., 1999; Teuber et al., 1999)
> 16 Lc. lactis. S. thermophilus (Dodd et al., 1994; Guédon et □□
4-26 Lc. lactis. S. thermophilus (Polzin and McKay, 1991; Q_
15982 1003 18 IS982 1-20 Lc. lactis (Yu et al., 1995) O
15983 1067 25 IS30 15k Lc. lactis (Bolotin et al., 1999; A. Sorokin, personal com- 3
Was this article helpful?