Rhs elements of Escherichia coli K-12: complex composites of shared and unique components that have different evolutionary histories

Abstract

The complete sequences of the RhsB and RhsC elements of Escherichia coli K-12 have been determined. These sequence data reveal a new repeated sequence, called H-rpt (Hinc repeat), which is distinct from the Rhs core repetition that is found in all five Rhs elements. H-rpt is found in RhsB, RhsC, and RhsE. Characterization of H-rpt supports the view that the Rhs elements are composite structures assembled from components with very different evolutionary histories and that their incorporation into the E. coli genome is relatively recent. In each case, H-rpt is found downstream from the Rhs core and is separated from the core by a segment of DNA that is unique to the individual element. The H-rpt's of RhsB and RhsE are very similar, diverging by only 2.1%. They are 1,291 bp in length, and each contains an 1,134-bp open reading frame (ORF). RhsC has three tandem copies of H-rpt, all of which appear defective in that they are large deletions and/or have the reading frame interrupted. Features of H-rpt are analogous to features typical of insertion sequences; however, no associated transposition activity has been detected. A 291-bp fragment of H-rpt is found near min 5 of the E. coli K-12 map and is not associated with any Rhs core homology. The complete core sequences of RhsB and RhsC have been compared with that of RhsA. As anticipated, the three core sequences are closely related, all having identical lengths of 3,714 bp each. Like RhsA, the RhsB and RhsC cores constitute single ORFs that begin with the first core base. In each case, the core ORF extends beyond the core into the unique sequence. Of the three cores, RhsB and RhsA are the most similar, showing only 0.9% sequence divergence, while RhsB and RhsC are the least similar, diverging by 2.9%. All three cores conserve the 28 repetitions of a peptide motif noted originally for RhsA. A secondary structure is proposed for this motif, and the possibility of its having an extracellular binding function is discussed. RhsB contains one additional unique ORF, and RhsC contains two additional unique ORFs. One of these ORFs includes a signal peptide that is functional when fused to TnphoA.

The dispersal of five group II introns among natural populations of Escherichia coli

“…Full sequences of E.c.I1 and E.c.I3 were obtained by PCR amplification of the introns based on predicted flanking exon sequences and host strains (Ferat et al+, 1994), followed by cloning and sequencing (see Material and Methods)+ Based on the completed sequences, all five introns can be described+ E.c.I1, E.c.I2, and E.c.I3 are related to each other and belong to the previously defined subclass "bacterial class D+" (Intron subclasses are based on phylogenetic groupings of the intron-encoded ORFs, but each subclass also has a distinct intron secondary structure; Toor et al+, 2001;Zimmerly et al+, 2001+) Typical of bacterial class D, the introns are IIB-like in RNA structure (Fig+ 1A,B), and the ORFs lack a Zn domain+ Although the Zn domain is required for mobility of Lactococcus Ll+ltrB (Cousineau et al+, 1998), many bacterial introns do not contain Zn domains and at least one of these is efficiently mobile nevertheless (Martinez-Abarca & Toro, 2000a)+ E.c.I1 and E.c.I2 are 67% identical in DNA sequence, whereas E.c.I3 is much less related, with 44% identity to E.c.I2 over its ORF+ E.c.I1 sequence from ECOR43 has a stop codon in RT domain 6 and a frame shift in domain X, suggesting loss of mobility and splicing functions+ E.c.I1 and E.c.I2 are inserted into an ISEc1 element, which is itself contained within an Recombination hot spot (Rhs) element (Ferat et al+, 1994)+ ISEc1 was previously called an H-repeat, but has been renamed ISEc1 because of its resemblance to IS elements (Mahillon & Chandler, 1998)+ The Rhs element is an ;6-kb DNA found in five copies in the sequenced K-12 genome+ There are at least eight varieties of Rhs elements (A-H), whose core sequences are .70% identical, but which differ in organization and lengths of spacer elements (Zhao et al+, 1993;Bachellier et al+, 1996;Wang et al+, 1998)+ At the 39 terminus of most Rhs elements is an ISEc1 copy, which is flanked by 11 bp inverted repeats+ E.c.I1 is inserted 4 bp after the upstream inverted repeat, whereas E.c.I2 is inserted near the 39 end of the ISEc1 ORF+ E.c.I3 is also located in an IS element, IS679+ E.c.I4 belongs to the subclass "bacterial class A," and its only other relative is the essentially identical intron in the closely related bacterium Shigella (99+4% identity)+ A possible secondary structure is shown in Figure 1C+ According to information in the databases, the intron has two homing sites in IS629 and IS911 elements, which share only 60% identity (Fig+ 2D)+ (We use the term "homing site" to denote the insertion site for all five introns in this study+ Although homing has not been experimentally demonstrated, it is suggested because the introns are repeatedly found within the same flanking sequences+)…”

Rhs elements of Escherichia coli: a family of genetic composites each encoding a large mosaic protein

“…It seems likely that peptide chains possessing such a high degree of motif repetition would fold into a very regular array. Based on an algorithm for predicting secondary structure (Gamier et ai, 1978), we have proposed that each of the Rhs motif units folds within itself into two p-sheet-p-tum portions, each p-sheet segment containing seven amino acids (Zhao et al, 1993). Another feature relevant to considerations of the core protein as a cell surface protein is the presence of a potential membrane-spanning helix beginning 28 residues from the AZ-terminus.…”

Identification of a second endogenous Porphyromonas gingivalis insertion element

“…However, IS1358 from V. cholerae (15) exhibited a 54% nucleotide residue identity with PGIS2. Protein sequence database searches also revealed significant homology between the putative polypeptide encoded by PGIS2 and the putative transposase of IS1358 from V. cholerae (15), the protein encoded by A. salmonicida IS element ISAS1 (5), the predicted product of H-rpt from E. coli K-12 (16), and a hypothetical protein encoded by a long ORF downstream of the dhlA gene of X. autotrophicus GJ10 (9). The amino acid alignment indicated that these proteins are more conserved in the central region (Fig.…”

“…In this study we identified a second endogenous IS element in P. gingivalis. The predicted protein encoded by PGIS2 exhibits a high degree of homology to the putative transposases of IS1358 and ISAS1 from V. cholerae and A. salmonicida, respectively, the predicted product of H-rpt from E. coli K-12 (16), and a hypothetical protein encoded by a long ORF downstream of the dhlA gene of X. autotrophicus GJ10 (9). The additional copy of PGIS2 in P. gingivalis Tn4351 transconjugant MSM-1 suggests that either transposition of the endogenous PGIS2 element occurred following transposition of Tn4351 or transposition of PGIS2 spontaneously occurred during laboratory passage.…”

“…In this study, we have identified a second endogenous P. gingivalis IS element and have confirmed the transposition of this element within P. gingivalis. The new IS element, designated PGIS2, is predicted to code for a polypeptide with strong homology to the putative transposases of IS1358 and ISAS1 from Vibrio cholerae and Aeromonas salmonicida, respectively (5, 15), the predicted product of H-rpt from Escherichia coli K-12 (16), and a hypothetical protein encoded by a long open reading frame (ORF) downstream of the dhlA gene of Xanthobacter autotrophicus GJ10 (9).…”