The LAGLIDADG protein family includes the first identified and biochemically characterized intron-encoded proteins (Dujon 1980; Lazowska et al. 1980; Jacquier and Dujon 1985), as described in this volume by Dujon. It has been variously termed the 'DOD', 'dodecapeptide', 'dodecamer', and 'decapeptide' endonuclease family, based on the conservation of a ten-residue sequence motif (Dujon 1989; Dujon et al. 1989; Belfort et al. 1995; Belfort and Roberts 1997; Dalgaard et al. 1997; Chevalier and Stoddard 2001). The LAGLIDADG endonu-cleases are the most diverse of the homing endonuclease families. Their host range includes the genomes of plant and algal chloroplasts, fungal and protozoan mitochondria, bacteria and Archaea (Dalgaard et al. 1997). One reason for the wide phylogenetic distribution of LAGLIDADG genes appears to be their remarkable ability to invade unrelated types of intervening sequences, including group I introns, archaeal introns and inteins (Belfort and Roberts 1997; Chevalier and Stoddard 2001). Descendents of LAGLIDADG homing endonucleases also include the yeast HO mating type switch endonuclease (Jin et al. 1997), which is encoded by an independent reading frame rather than within an intron, but does carry remnants of an inactive intein domain (Haber and Wolfe, this Vol.), and maturases that assist in RNA splicing (Dela-hodde et al. 1989; Lazowska et al. 1989; Schafer et al. 1994; Geese and Waring 2001; Caprara and Waring, this Vol.).

B. Chevalier, B.L. Stoddard (e-mail: [email protected])

Division of Basic Sciences, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N. A3-025, Seattle, Washington 98109, USA

Departments of Pathology and Genome Sciences, Box 357470, University of Washington, Seattle, Washington 98195, USA

Nucleic Acids and Molecular Biology, Vol. 16 Marlene Belfort et al. (Eds.) Homing Endonucleases and Inteins © Springer-Verlag Berlin Heidelberg 2005

Members of the LAGLIDADG family are segregated into groups that possess either one or two copies of the conserved LAGLIDADG motif. Enzymes that contain a single copy of this motif, such as I-Crel (Thompson et al. 1992; Wang et al. 1997) and I-Ceul (Turmel et al. 1997), act as homodimers and recognize consensus DNA target sites that are constrained to palindromic or near-palindromic symmetry. Enzymes that have two copies of the LAGLIDADG motif (such as I-Scel, the first LAGLIDADG enzyme to be discovered) act as monomers, possess a pair of structurally similar nuclease domains on a single peptide chain, and are not constrained to symmetric DNA targets (Agaard et al. 1997; Dalgaard et al. 1997; Lucas et al. 2001). In both subfamilies, the LAGLIDADG motif residues play both structural and catalytic roles (see below).

Free-standing LAGLIDADG endonucleases (i.e., those that are not cov-alently associated with intein domains) recognize DNA sites that typically range from 18 to 22 base pairs. They cleave both DNA strands across the minor groove, to generate mutually cohesive four base 3' overhangs (Chevalier and Stoddard 2001). Like most, if not all nucleases, LAGLIDADG homing endonucleases require divalent cations for activity.

Upon invasion of a novel biological target site, homing endonucleases and their associated mobile introns or inteins can persist, diversify and spread to similar sites of related hosts. The evolution of related homing endonucleases subsequent to a founding intron invasion event has been elegantly described for at least one LAGLIDADG endonuclease branch, which contains the I-Crel enzyme (Lemieux et al. 1988; Turmel et al. 1995; Chevalier et al. 2003). (A similar study has more recently been reported for the intein-associated Pl-Scel lineage; Posey et al. 2004.) I-Crel is encoded within a group I intron present in the chloroplast large subunit (LSU) rDNA of the green alga Chlamydomonas reinhardtir, the insertion site of this intron corresponds to position 2593 in the Escherichia coli 23S rDNA (Turmel et al. 1995). Sequence analysis of chloroplast and mitochondrial LSU rDNAs from numerous other green algae have disclosed 15 similar open-reading frames (ORFs) within identically positioned introns. Three of these genes were shown to encode active endonucleases that are isoschizomers of I-Crel, including I-Msol from Monomastix (Lucas et al. 2001). Although the native target sites of I-Crel and I-Msol differ at 2 out of 22 base pair positions, each endonuclease efficiently cleaves both target sites. Threading the I-Msol sequence onto the I-Crel structure suggests significant protein sequence divergence, especially at residues involved in DNA binding (Lucas et al. 2001); this observation has been confirmed by an X-ray crystal structure of I-Msol (Chevalier et al. 2003). The structure also implies that the I-Msol enzyme might recognize a larger number of sites (i.e. is more promiscuous in its DNA recognition profile) than does I-Crel.

0 0

Post a comment