We are currently working on three different  aspects of the
integron system :  

- The distribution of super-integrons and their evolutionary history,  
- The inventory of the cassette encoded functions,
- The recombination process and cassette genesis.  

Introduction

The development of antibiotic resistance has lead to the discovery of many natural mobile elements, including transposons and conjugative plasmids. Comparative sequence analysis of these elements ultimately led to the discovery of integrons, natural cloning and expression systems that incorporate open-reading frames (ORF) and convert them to functional genes  (for review see (8, 20)). The functional integron platform consists of a gene coding for an integrase (intI) of the tyrosine-recombinase family and a proximal primary recombination sequence called an attI site. The integrase mediates recombination between the attI site and a secondary target called an attC site (or 59-base elements (59be)). The attC site is generally associated with a single ORF in a structure termed a gene cassette, and the gene cassette constitutes the mobile component of the system (9, 16, 21, 22) (Figure 1). Insertion of the gene cassette at the attI site, which is located downstream of a resident promoter internal to the intI gene, drives expression of the encoded proteins (Figure 1). 

Three classes of resistant integrons (RI) have been defined based on the divergence among their integrase genes and each class appears to be able to acquire the same gene cassettes (7). The integron platform itself is defective for self-transposition but they are often found associated with insertion sequences (ISs), transposons and / or conjugative plasmids that can serve as vehicles for the intra and inter-species transmission of genetic material. The Tn21 and Tn7 families provide examples of this (11, 23). As such, they have been found in a variety of genetic contexts and among a large number of phylogenetically diverse Gram-negative and Gram-positive isolates. 

More than 70 different antibiotic resistance gene cassettes have been characterized in the three classes of RI (Mazel & Davies, 1999) and most of their attC sites are unique (Table 1). The length and sequence of the attC sites vary considerably (from 57 to 141 bp) and their similarities are primarily restricted to their boundaries, which correspond to the inverse core-site (ICS; RYYYAAC) and the core site (CS; G(TTRRRY; (, recombination point (Collis et al., 1998; Stokes, et al., 1997)). The studies of Mazel et al. (12) and Manning and colleagues (1) examining the relationship between RI gene cassette arrays and the Vibrio cholerae repeats (VCRs) cluster led to the discovery of the Vibrio cholerae super-integron. This distinct type of integron is now known to be an integral component of many ?-proteobacterial genomes (18).

The super-integrons and their distribution

The integron discovered in chromosome 2 of V. cholerae has two characteristics that distinguish it from known RIs; the large number of cassettes that are gathered and the high homology observed between the attC sites of these cassettes, the VCRs in the case of V. cholerae (12). These are the key features that define a super-integron (SI) (Figure 2). Both the V. cholerae SI and RIs possess specific and related integrases that are responsible for the insertion of coding sequences (ORFs) into a unique chromosomal attachment site, leading to the formation of tandem arrays of genes. In the case of V. cholerae, the cluster of VCR-associated ORFs represents at least 216 unidentified genes in an array of 179 cassettes which starts from the VchintIA gene and occupies about 3% of the genome (10). The extent to which this system has impacted genome evolution has emerged with the discovery of SIs in several diverse proteobacterial genera and the various species include human, animal and plant pathogens as well as non-pathogenic bacteria. SI structures have been identified among the Vibrionaceae and their close relatives, the Shewanella, the Xanthomonads (3, 18), and the Pseudomonads (18, 25) (Table 1). They share the same general characteristics (i. e. a large size and a high homology between their endogenous cassette attC sites) and clearly predate the antibiotic era, as they are present in isolates from the last century (12). Integron integrase-like genes have also been identified in the genomes of other proteobacteria including Acidithiobacillus ferroodoxans, Nitrosomonas europaea, Geobacter sulfurreducens, and Treponema denticola (Table 2), but they have not been further characterized (13, 18).

        Using PCR primers directed against conserved regions of the integron-integrase genes and attC sites, Stokes and colleagues were able to identify three new classes of integrons from four markedly different environmental DNA samples that had no known previous exposure to antibiotics (13). The protocol they used to retrieve the integron-integrase loci allowed recovery of the majority of the intI gene, the sequence covering the attI site and the first cassette up to the ICS of its associated attC site. Unfortunately, such a short sequence does permit determination of the source of these integrons, be it the endogenous SI of a soil bacterium or an integron located in a mobile structure. However, these findings support the hypothesis developed by the discovery of SIs; that integrons are widespread among bacterial populations either as components of mobile DNA elements or the chromosome and that they are not confined to pathogenic or multidrug-resistant bacteria. 

Integrons are ancient evolutionary apparatuses.

All characterized integron-integrases clearly group together and form a specific clade within the tyrosine recombinase family (Figure 3). Furthermore, it has been noticed that all contained a specific stretch of 16 amino acids (13) located between conserved patchs II and III of the tyrosine recombinase family (14). The role of this integron-integrase specific sequence is unknown. The integron platform is undoubtedly ancient as attested to by the species-specific clustering of the respective SI integrase genes in a pattern that adheres to the line of descent among the bacterial species in which they are found (Figure 3) (18). Thus, the establishment of SIs likely predates speciation within the respective genera, indicating that integrons are ancient structures that have been steering the evolution of bacterial genomes for hundreds of millions of years. It is however possible that transfer of either a part or all of a SI occured (from an Hfr-type strain, for example) during such a long period of evolution. This could be the origin of the discrepency observed in the SI-integrase and 16S rRNA gene trees for the V. fischeri branching points among the other bacterial species (18). However, the phylogenetic analysis performed on the rplT genes from the same species, which encode the conserved ribosomal L20 protein, produced a branching order identical to the one found for the intI dendogram. This supports co-evolution of the V. fischeri SI integrase gene with the rest of the V. fischeri genome.

Cassette functions

The SIs identified to date collectively equal a small genome in size, suggesting that the process of cassette genesis is constant and efficient. The majority of the cassettes examined thus far appeared to be unique to the host species. Furthermore, most of their encoded genes have no counterparts in the database or the sole homologues are unassigned ORFs of viral, bacterial or eukaryotic origins, indicating their recruitment from all kingdoms of life (10, 17). A precise inventory of the functions encoded by the cassettes remains to be established, however, a preliminary study indicates that many of the SI cassettes encode adaptive functions, sensu lato, beyond pathogenicity and antibiotic resistance. In V. cholerae, two pathogencity genes, the heat-stable toxin gene (sto) (15) and the mannose-fucose-resistant haemagglutinin gene, mrhA (26), and a lipoprotein gene have been found to be cassette encoded (2). We have determined the metabolic function of three SI cassettes: a sulfate binding protein in a V. cholerae SI cassette, a psychrophilic lipase in a Moritella marina SI cassette and a restriction enzyme (Xba I) and its cognate methylase, in a Xanthomonas campestris pathovar badrii SI cassette (18). Genes with homology to DNA methylases, immunity proteins, restriction endonucleases, dNTP triphophohydrolases, periplasmic sulphate binding proteins, lipases and 8-oxoguanine triphosphatases (MutT), among others, have been found (3, 10, 17, 18). Although a known antibiotic-resistance gene cassette has not yet been identified within a SI, several potential progenitor cassettes with significant homology to aminoglycoside, phosphinotricin, fosfomycin, streptothricin and chloramphenicol resistance genes are present. The determination of the metabolic activities of several SI cassettes, whose activities are not related to antibiotic resistance or virulence, confirms that integrons operate as a general gene capture system in bacterial adaptation (2, 18). If each bacterial species harboring a SI has its own cassette pool, the resource in terms of gene cassette availability will be immense and the functions of the encoded genes have fantastic potential from both genetic and biotechnological standpoints.

Intra-species cassette content variations

The activity of integron cassettes offers a fast-track to bacterial innovation. The size of SIs and the ancient and dynamic nature of the system is a reminder that the cassettes that currently occupy the SI represent only a fraction of those that may have participated in the evolution of the host, since the cassettes will presumably be subject to episodic selection. The more than 165 different O serotypes of V. cholerae are represented by species of ecological, geographical and temporal diversity. Thus, comparison of SI organization from recent and earlier isolates as well as between recent isolates from different geographical locations and ecological niches may yield valuable information. Clark et al. examined the global SI organization of 65 different V. cholerae O serotypes by PCR and Southern hybridization (3). Extensive restriction polymorphism was observed even among closely related isolates, suggesting an appreciable plasticity for that these structures and their microevolution through integrase-mediated gene acquisition and gene loss (24) as well as via cassette rearrangement events (5, 6, 22). Concerning this latter occurrence, an important question is can cassettes be mobilized in clusters? The SI organization of two V. cholerae strains suggests they can; the cassettes in positions 1-4 of the SI of strain 569B are found in the same order in the SI of strain N16961 (10) but they occupy positions 79 to 82 (a displacement of more than 40 Kb) (17). This has also been seen for other cassettes (3). These observations could represent a true group mobilization event or simply temporal differences in cassette acquisition.

There is evidence that not all repeated sequences are equally functional; some are known to contain mutations or deletions within the CS that could render them non-functional (Rowe-Magnus, Guerout and Mazel unpublished results). Therefore, their movement would have to be co-ordinated with those of other cassettes. Collis and Hall demonstrated that integron gene cassettes are excised as covalently closed circles and observed differences in the resulting recombination products (4). Some cassettes could be mobilized as individual units while others were only excised in tandem with another cassette. Whether such cassette hitch-hiking is by design to ensure simultaneous transmission of genes, or is just a matter of happenstance is not known.

The integron system is remarkably versatile with its ability to recognize highly variable target recombination sequences and apparently limitless capacity to exchange and stockpile cassettes. Such flexibility permits rapid adaptation to the unpredictable flux of environmental niches by allowing bacteria to scavenge foreign genes that may ultimately endow increased fitness to the host. Likewise, genes that fail to provide a meaningful function can be readily eliminated. It is also quite likely that many of the cassettes that presently occupy the SI are not expressed, but they may persist in the absence of selective pressure nevertheless and provide a genetic basis for the evolution and subsequent retention of novel derived activities. In addition to the plethora of antibiotic resistance genes, two virulence genes of V. cholerae are also structured as gene cassettes, underscoring the potential of this system to participate in the establishment of pathogenicity islands. It is conceivable, that any ORF can be structured as a gene cassette and it is vital to decipher the mechanism governing cassette genesis. According to the GC % and codon usage differences observed in the ORFs found in the cassettes, they must have many different origins, while their associated recombination elements, are highly homologous (e.g. 74 % identity for the VCRs). This last characteristic suggests that the VCRs were added to the ORFs inside the Vibrio cell. Therefore, it is very likely that the capture process occurs in vivo, but the nature of this process remains unknown.

As such, integron-driven gene capture is likely to be an important factor in the more general process of horizontal gene transfer in the evolution of bacterial genomes. It appears that multi-resistant integrons have evolved from super-integrons, through entrapment of intI genes and their cognate attI sites into highly mobile structures like transposons. The combination of this mobility and the selection pressure exerted by antibiotic use may have driven the specific capture of resistance cassettes from the many different kinds of super-integron cassette pools through multiple lateral transfers. To support this hypothesis, we showed that MRIs can randomly recruit genes directly from the cache of V. cholerae SI cassettes. By applying a selective constraint for the development of antibiotic resistance, we demonstrated bacterial resistance evolution through the recruitment a novel, but phenotypically silent, chloramphenicol acetyltransferase gene from the V. cholerae SI and its precise insertion into the MRI(19). The resulting resistance profile could then be disseminated by conjugation to other clinically relevant pathogens at high frequency.

References

1.Barker, A., C. A. Clark, and P. A. Manning. 1994. Identification of VCR, a repeated sequence associated with a locus encoding a hemagglutinin in Vibrio cholerae O1. Journal of Bacteriology. 176:5450-5458.

2.Barker, A., and P. A. Manning. 1997. VlpA of Vibrio cholerae O1: the first bacterial member of the alpha 2-microglobulin lipocalin superfamily. Microbiology. 143(Pt 6):1805-13.

3.Clark, C. A., L. Purins, P. Kaewrakon, T. Focareta, and P. A. Manning. 2000. The Vibrio cholerae O1 chromosomal integron. Microbiology. 146:2605-12.

4.Collis, C. M., and R. M. Hall. 1992. Gene cassettes from the insert region of integrons are excised as covalently closed circles. Mol Microbiol. 6(19):2875-85.

5.Collis, C. M., and R. M. Hall. 1992. Site-specific deletion and rearrangement of integron insert genes catalyzed by the integron DNA integrase. Journal of Bacteriology. 174(5):1574-85.

6.Collis, C. M., G. D. Recchia, M. J. Kim, H. W. Stokes, and R. M. Hall. 2001. Efficiency of recombination reactions catalyzed by class 1 integron integrase IntI1. J Bacteriol. 183(8):2535-42.

7.Hall, R. M., and C. M. Collis. 1998. Antibiotic resistance in gram-negative bacteria: the role of gene cassettes and integrons. Drug Resistance Updates. 1(2):109-119.

8.Hall, R. M., and C. M. Collis. 1995. Mobile gene cassettes and integrons: capture and spread of genes by site-specific recombination. Molecular Microbiology. 15(4):593-600.

9.Hall, R. M., and H. W. Stokes. 1993. Integrons: novel DNA elements which capture genes by site-specific recombination. Genetica. 90(2-3):115-32.

10.Heidelberg, J. F.,. 2000. DNA sequence of both chromosomes of the cholera pathogen Vibrio cholerae. Nature. 406:477 - 483.

11.Liebert, C. A., R. M. Hall, and A. O. Summers. 1999. Transposon Tn21, flagship of the floating genome. Microbiol Mol Biol Rev. 63(3):507-22.

12.Mazel, D., B. Dychinco, V. A. Webb, and J. Davies. 1998. A distinctive class of integron in the Vibrio cholerae genome. Science. 280(5363):605-608.

13.Nield, B. S., A. J. Holmes, M. R. Gillings, G. D. Recchia, B. C. Mabbutt, K. M. Nevalainen, and H. W. Stokes. 2001. Recovery of new integron classes from environmental DNA. FEMS Microbiol Lett. 195(1):59-65.

14.Nunes-Duby, S. E., H. J. Kwon, R. S. Tirumalai, T. Ellenberger, and A. Landy. 1998. Similarities and differences among 105 members of the Int family of site-specific recombinases. Nucleic Acids Research. 26(2):391-406.

15.Ogawa, A., and T. Takeda. 1993. The gene encoding the heat-stable enterotoxin of Vibrio cholerae is flanked by 123-base pair direct repeats. Microbiology and Immunology. 37(8):607-616.

16.Recchia, G. D., and R. M. Hall. 1995. Gene cassettes: a new class of mobile element. Microbiology. 141:3015-3027.

17.Rowe-Magnus, D. A., A.-M. Guerout, and D. Mazel. 1999. Super-Integrons. Research in Microbiology. 150:641-651.

18.Rowe-Magnus, D. A., A.-M. Guerout, P. Ploncard, B. Dychinco, J. Davies, and D. Mazel. 2001. The evolutionary history of chromosomal super-integrons provides an ancestry for multi-resitant integrons. Proceedings of the National Academy of Sciences of the United States of America. 98:652-657.

19.Rowe-Magnus, D. A., A. M. Guerout, and D. Mazel. 2002. Bacterial resistance evolution by recruitment of super-integron gene cassettes. Mol Microbiol. 43(6):1657-69.

20.Rowe-Magnus, D. A., and D. Mazel. 1999. Resistance gene capture. Current Opinion in Microbiology. 2:483-488.

21.Stokes, H. W., D. B. O'Gorman, G. D. Recchia, M. Parsekhian, and R. M. Hall. 1997. Structure and function of 59-base element recombination sites associated with mobile gene cassettes. Molecular Microbiology. 26(4):731-745.

22.Sundstrom, L. 1998. The potential of integrons and connected programmed rearrangements for mediating horizontal gene transfer. APMIS. Supplementum. 84:37-42.

23.Sundstrom, L., P. H. Roy, and O. Skold. 1991. Site-specific insertion of three structural gene cassettes in transposon Tn7. Journal of Bacteriology. 173(9):3025-8.

24.Takeda, T., Y. Peina, A. Ogawa, S. Dohi, H. Abe, G. B. Nair, and S. C. Pal. 1991. Detection of heat-stable enterotoxin in a cholera toxin gene-positive strain of Vibrio cholerae O1. FEMS Microbiol Lett. 64(1):23-7.

25.Vaisvila, R., R. Morgan, and E. Raleigh. 1999. The pacIR gene resides within a potential super-integron in the Pseudomonas alcaligenes genome. IXth international congress of bacteriology and applied microbiology, Sidney, 1999.

26.van Dongen, W. M. A. M., M. M. A. van Vlerken, and F. K. De Graaf. 1987. Nucleotide sequence of a DNA fragment encoding a Vibrio cholerae haemagglutinin. Mol. Gen. (Life Sci. Adv.). 6:85-91.

 Figure Legends

 Figure 1 :
 Schematic representation of a circularized cassette (A) and model for cassette exchange (B). A, The key features of attC sites (yellow) are the complementary inverse core-site (ICS) and core-site (CS) consensus sequences (red arrows) and the imperfect palindromic variable region (red dotted arrows). B, Outline of the process by which circular antibiotic resistance gene cassettes (antR) are repeatedly inserted at the specific attI site in a class1 integron downstream of the strong promoter Pant. The vertical line in the attC symbol (triangle) represents the recombination point in the CS sequence. intI, integrase encoding gene; Int, integrase IntI; attC, 59-be. See text for details.

 Figure 2 :  
 Structural comparison of a "classical" multi-resistant integron (A) and the V. cholerae super-integron (B). A, Schematic representation of In40 (22). The various resistance genes are associated with different 59-be (see text) whose consensus sequence is shown. B, The ORFs are separated by highly homologous sequences, the VCRs, for which the consensus sequence is shown. Recombination occurs between the G and the first T of the core-site sequence.   

 Figure 3 :
Phylogenetic relationship of the integron intI genes among the proteobacteria and chromosomal location of the Vibrio Super-integrons. Unrooted dendrogram based on known intI gene sequences. The integrases from the five classes of RI are boxed. Organism abbreviations for the SIs are as follows: Vibrio cholerae (Vch), V. metschnikovii (Vme), V. parahaemolyticus (Vpa), V. fischeri (Vfi), Listonella pelagia (Lpe), Shewanella oneidensis (Son), S. putrefaciens (Spu), Xanthomonas campestris pathovar camperstris (Xca pv ca) or badrii (Xca pv ba), X. species (Xsp), Nitrosomonas europaea (Neu). The sources of intI6-2, intI7-2, intI8-2 and the intI of plasmid pRVS1 (GB accession number AJ277063) are unknown. For the integrase and recombinase proteins: XerC/D (of E. coli). Representation of the genetic context in which each of the SIs were found relative to their phylogenetic distribution according to their IntIA genes is shown. SIs in identical locations are group within the same phylogenetic clade. The corresponding orthologues in V. cholerae are marked with VC(A) followed by the number designation for the ORF. The attI site and VXRs are also indicated. For clarity, only the first cassette within each SI is shown. rplT and rpmL, ribosomal genes; dark gray or black boxes, adjacent gene(s) which are not part of the SI structure.