Archaea and bacteria contain CRISPR-Cas (clustered regularly interspaced short palindromic repeat-CRISPR-associated) systems that protect themselves against invasion by viruses and plasmids. There are three major types of CRISPR-Cas systems, type I, II and III, that are further divided into at least 11 subtypes. I employed Sulfolobus islandicus Rey15A as the model to study CRISPR mechanisms. The model archaeon encodes one subtype I-A (Cascade) and two subtype III-B (Cmr-α and Cmr-β) interference systems with no apparent redundancy in cas genes or in CRISPR systems, which is ideal for genetic analysis of cas gene function. Furthermore, a range of genetic tools have been developed for S. islandicus Rey15A in our laboratory and a plasmid interference assay has been successfully developed for testing CRISPR-directed DNA targeting activity, which have provided a solid basis for studying the functions of each cas gene in CRISPRbased immunity and to demonstrate the Cmr-mediated in vivo RNA interference in this model archaeon.
This thesis includes two parts of results obtained from the PhD subjects. In Part 1, it describes the results yielded from the genetic analyses of functions of each type I-A Cas protein in pre-crRNA processing and in DNA interference which depends on detection of a DNA sequence stretch called protospacer adjacent motif (PAM). In Part 2 the results obtained from the in vivo demonstration of Cmr-directed RNA interference are presented.
Knockouts of individual cas gene or each of other cas gene cassettes of S. islandicus Rey15A were constructed for analyzing the functions of cas genes of the type I-A system. Plasmid interference assay was performed in each knockout and the results showed that Cas7, Cas5, Cas6 and Cas3 proteins are essential for the type I-A DNA interference function, whereas Csa5, along with other cas gene cassettes, is dispensable for this function. Northern hybridizations of CRISPR RNAs revealed that Cas6 is the only enzyme responsible for pre-crRNA processing while other Cas proteins, such as Cas7 and Cas5 also contribute to this process. Cas7 and Cas5 are involved in stabilizing the processing intermediates and the mature crRNAs, respectively. Additionally, crRNA maturation is independent of the DNA interference activity.
An RNA interference assay was established in which the lacS gene encoding a β- glycosidase was employed as a reporter gene for identifying the in vivo RNA targeting activity of the Cmr systems of S. islandicus Rey15A. This RNA interference assay includes a plasmid that expresses an artificial CRISPR (pAC) containing a spacer derived from the lacS gene and a β-glycosidase assay. Targeting of the mRNAs of lacS by the Cmr complexes was guided by the crRNAs produced from the plasmid-borne AC and the targeting effect on the mRNA of lacS was evaluated by testing the residual β-glycosidase activity in each pAC transformant. Testing RNA interference activity in the knockouts of each entire Cmr cassette and both Cmr cassettes showed that only the deletion mutant of both Cmr cassettes lost the ability in RNA interference, indicating that each Cmr could form ribonucleoprotein (RNP) complex with a crRNA, mediating silencing of lacS gene expression in S. islandicus. This study has, for the first time, revealed that a type III-B CRISPR system (Cmr-α), which was previously shown to be implicated in DNA targeting in a PAM-independent manner, possesses dual DNA and RNA targeting activities. Cleavage sites in the mRNA of lacS were mapped within the protospacer by rapid amplification of cDNA ends (RACE), showing that the cleavage occurs either at fixed positions (17/18 or 23/24 referring to crRNA) by Cmr-α or at UA-like sites by Cmr-β. In addition, spacer mutagenesis assay identified a trinucleotide motif located at the 3’ end of a crRNA that is important for the RNA interference.