Mutualism Between Self‑Splicing Introns and Their Hosts David R Edgell1*, Venkata R Chalamcharla2,3 and Marlene Belfort2*
Total Page:16
File Type:pdf, Size:1020Kb
Edgell DR et al. BMC Biology 2011, 9:22 http://www.biomedcentral.com/1741-7007/9/22 review Open Access Learning to live together: mutualism between self‑splicing introns and their hosts David R Edgell1*, Venkata R Chalamcharla2,3 and Marlene Belfort2* Abstract data argue that the relationship is more elaborate than previously appreciated. Group I and II introns can be considered as molecular parasites that interrupt protein-coding and structural Mobile introns: ribozymes with baggage RNA genes in all domains of life. They function as self- One group of mobile genetic elements comprises the group I splicing ribozymes and thereby limit the phenotypic and II introns. These sequences interrupt protein-coding and costs associated with disruption of a host gene while structural RNA genes in all domains of life and can be con they act as mobile DNA elements to promote their sidered as molecular parasites. When the gene is transcribed spread within and between genomes. Once considered into RNA, the intron sequence acts as a ribozyme (an RNA purely selfish DNA elements, they now seem, in the with enzymatic activity), which removes the intron sequence light of recent work on the molecular mechanisms from the primary RNA transcript, thus limiting the regulating bacterial and phage group I and II intron phenotypic cost associated with insertion of the element into dynamics, to show evidence of co-evolution with their a host gene and promoting their maintenance in the genome. hosts. These previously underappreciated relationships In the case of group I and II introns, the host-parasite serve the co-evolving entities particularly well in times relationship is enriched by the fact that the introns of environmental stress. themselves have been invaded by smaller parasitic elements – genes that encode mobility-promoting activities that enable the DNA element to move within and between One of the most intricate relationships in biology is that genomes [10]. Thus, at least two levels of parasitism exist for between a host and a parasite. Almost all organisms mobile introns: the intron in the host gene it interrupts, and studied so far harbor mobile genetic elements and/or the invading gene in the intron. Collectively, the intron and their derivatives. At the genomic level, the traditional its encoded mobility protein (often termed an intron- view of mobile elements is that they provide seemingly encoded protein, IEP) collaborate to form a composite little or no benefit to the host while parasitizing the host’s mobile element that utilizes host DNA replication, recombi cellular machinery to promote element mobility through nation and repair pathways to spread [11], while the ribozyme complex molecular pathways [1,2]. The host’s response to activity ensures that it does not disrupt the function of genes these elements is primarily defensive, as evidenced by the into which it is inserted. Accordingly, it has become evident many forms of negative regulation that downregulate the that there is an extraordinary degree of co-evolution among activity of mobile elements [3-8]. The persistence of a IEPs, the introns that house them, and the host organism. mobile element in a given population is thus the result of This review highlights several recent studies probing the a delicate balance between an excessive mutational interplay among self-splicing introns in bacterial and phage burden on the host caused by the element’s unrestricted genomes, their genes, and their bacterial and phage hosts. activity, and excessive negative regulation imposed by the host on the element to limit mobility. While the Group I introns relationship between host and mobile element is often Group I introns commonly inhabit bacterial, organellar, viewed as a molecular arms race [9], recent experimental bacteriophage and viral genomes, and the ribosomal RNA genes (rDNA) of eukaryotes, and produce a self- splicing RNA [12]. Group II introns have a similar *Correspondence: [email protected], [email protected] 1Department of Biochemistry, Schulich School of Medicine and Dentistry, distribution, except that they are not found in eukaryotic The University of Western Ontario, London, Ontario, Canada N6A 5C1 nuclear genes. Group I and group II introns show little Full list of author information is available at the end of the article primary sequence conservation, yet their RNAs each © 2010 Author et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Edgell DR et al. BMC Biology 2011, 9:22 Page 2 of 9 http://www.biomedcentral.com/1741‑7007/9/22 adopt characteristic secondary and tertiary structures Many group I introns move by a DNA-based necessary for ribozyme activity [13,14] (Figure 1a,b). transposition mechanism known as ‘homing’. Such Moreover, the introns can tolerate the insertion of large introns harbor genes encoding so-called homing amounts of sequence into terminal loops of the ribozyme endonucleases, site-specific but sequence-tolerant DNA secondary structure with little or no effect on splicing, endonucleases that introduce double-strand breaks providing convenient ‘hiding’ spots for parasitic genes. (DSBs) in cognate alleles that lack the intron, initiating intron mobility via a DSB-repair process [11] (Figure 1c). (a) Group I intron (b) Group II intron The outcome of a homing event is the unidirectional movement of the intron and endonuclease open reading P9 DIII frame (ORF) to an unoccupied allele, leaving a copy of P5 * 3′ Exon the intron in its original location (Figure 1c). Group I P1 DII P7 introns can also harbor other ‘baggage’. Many group I Exon Exon ′ P4 ′ 3′ Exon 5 5 introns in organellar genomes encode maturases – P3 DI DV DVI proteins that help promote intron splicing by a variety of DI DI mechanisms [15,16]. Some maturases also function in P6 P2 P8 trans to promote splicing of other group I introns in the same genome [17,18]. Interestingly, many maturases (c) Group I homing (d) Group II retrohoming characterized so far are degenerate or bifunctional D homing endonucleases of the LAGLIDADG class – so 5′ 3′ D 3′ 5′ named for their conserved sequence motif – that have dsDNA ssDNA 5′ acquired an RNA chaperone activity independent of their 1 5′ 3′ 3′R 3′ 5′ R 5′ DNA endonuclease activity [19,20]. Group I introns can 1 3′ 1 also harbor ORFs unrelated to mobility or splicing R [21,22], as exemplified by the astonishing case of an 2 approximately 18-kilobase-long intron inserted in the 2 2 mitochondrial ND5 gene of the mushroom coral cDNA Discosoma that encodes 15 mitochondrial genes in the P8 3 cDNA loop of the intron [23,24]. Interestingly, these 15 genes 3 3 include both the small and large subunit rRNA genes and the cox1 gene, which is interrupted by another self- splicing group I intron. Figure 1. Models of group I and group II introns and their Some bacterial group I introns have been invaded by ‘homing’ mechanisms. (a,b) Schematic representations of (a) group I and (b) group II intron secondary structures [13,37]. In both cases, mobile elements other than those that encode homing secondary structures are represented by solid lines indicating endonucleases. Notable among these are the chimeric conserved stem-loop structures, named P1 to P10 for group I introns, intron/insertion sequence (IS) elements (IStrons) of and DI to DVI for group II introns. The positions of ORFs and other Clostridium that contain an IS605-like element inserted insertions are depicted by solid red lines. The asterisk (*) next to at the 3’ end of the intron [25]. It is not known, however, domain II of group II introns indicates bioinformatic predictions of the ORF start sites, but these remain uncharacterized. Dashed gray whether the chimeric intron/IS element is mobilized by lines indicate joining regions of unpaired nucelotides, with arrows the IS605 machinery. Intriguingly, another unusual indicating a 5’-3’ orientation. The 5’ and 3’ exons are indicated by grey clostridial group I intron arrangement was recently found rectangles. (c) Homing of a group I intron. In this DNA-based mobility by a bioinformatic search for riboswitches [26], RNA pathway, the intron donor (D) expresses the intron endonuclease (red structural elements that control gene expression through enzyme symbol) (step 1). After cleavage of the allelic intron recipient sequence (R) at the homing site (step 2) the donor and recipient alternative secondary structures in response to binding of engage in double-strand break (DSB) repair to generate two intron- secondary metabolites. In this case, the tandem containing alleles. (d) Group II intron retrohoming by means of an riboswitch/intron lies in the upstream region of a RNA intermediate. The intron donor (D) in this case is the spliced putative virulence factor gene, and sensing of cyclic di- intron lariat RNA (dashed red line), whereas the recipient (R) can be guanosyl-5’-monophosphate by the riboswitch controls either double-stranded DNA (dsDNA) or single-stranded DNA (ssDNA), as at a replication fork. A ribonucleoprotein complex between the choice of the 3’ splice junction by the intron to modulate RNA and the IEP catalyzes a reverse splicing (step 1). In the dsDNA expression of the virulence factor. pathway the IEP then cleaves the second strand to generate the While many ORFs embedded within group I introns are primer for cDNA synthesis by the IEP, whereas in the ssDNA pathway entirely located in loop regions, a surprising number of an Okazaki fragment at the replication fork (solid gray line) acts as ORFs extend beyond peripheral loops to contribute a primer (step 2).