Main Content

AG Randau - Research

Our group investigates the functionality of small non-coding RNA molecules in archaeal and bacterial model organisms. One focus of our research is placed on small RNAs that guide protein complexes to complementary target sites. The studied ribonucleoproteins include (i) antiviral defense machineries (CRISPR-Cas) and (ii) complexes that facilitate modification of target RNA molecules (C/D box sRNPs). We apply a combination of biochemical (RNA-seq), microbiological and synthetic biology approaches to obtain a mechanistic understanding of these processes and obtain insights into their evolution.

Class I CRISPR-Cas systems

Bacterial and archaeal organisms possess adaptive immune systems termed CRISPR-Cas that consist of CRISPR clusters and a large variety of CRISPR-associated (Cas) proteins. The CRISPR array is transcribed and processed into CRISPR RNAs (crRNAs), which use a so-called spacer sequence to guide Cas protein complexes towards target nucleic acids. In nature, most CRISPR-Cas systems belong to type I and utilize a CRISPR-associated complex for antiviral defense (Cascade) to identify viral DNA molecules. Cascade recruits the DNA nuclease Cas3 to degrade the invading DNA molecules. Cascade complexes are best studied in Escherichia coli (Type I-E) and contain (i) a crRNA-producing endonuclease (Cas6e), (ii) a backbone protein that protects the crRNA (Cas7), (iii) a protein that recognizes the 5'-tag of the crRNA (Cas5), (iv) a small subunit and (v) a large subunit. These two latter subunits interact with the target DNA and recognize the PAM sequence, a small DNA motif which is used in the cell to distinguish between self- and non-self DNA. Bacteriophages can produce anti-CRISPR proteins that bind to Cascade subunits and inactivate the interference mechanism. It is plausible that the evolution of anti-CRISPR proteins results in diversified Cascade variants that circumvent inactivation by anti-CRISPR proteins. Consequently, a large variety of Cascade architectures is found in nature and several subtypes have been classified.

Our group identified a minimal type I-Fv Cascade variant in Shewanella putrefaciens CN-32 that lacks large and small subunits and consists of the three proteins Cas5fv, Cas7fv and Cas6. Recombinant type I-Fv Cascade was produced in E. coli and purified. A unique Cas5fv protein was found to bind the crRNA 5'-tag and several Cas7fv subunits span the crRNA backbone. Each Cas7fv subunit can bind to a 6 nt spacer segment without any sequence specificity. The crystal structure of the type I-Fv complex was solved in collaboration with Patrick Pausch and Prof. Gert Bange (Philipps-Universität Marburg). Structures in complex with a partially unwound DNA target revealed insights into DNA target recognition and PAM identification. The Cas5fv protein was found to contain an additional alpha-helical domain that is involved in reading out the GG dinucleotide PAM by major groove interactions. Our recent work investigated the question how Cascade can induce DNA interference in the absence of a large subunit. We utilized bio-layer interferometry (BLI) and EMSA assays to follow interactions between recombinant type I-Fv Cascade and target DNA.

To date, a large variety of divergent CRISPR-Cas systems with different repeat arrangements and semi-conserved modules of cas genes have been described. Computational analyses of the Cas protein content allowed their classification into two classes and at least six types. Type IV CRISPR-Cas systems remain the only type whose functional role in bacterial cells is not known. Strikingly, an adaptation module and genes coding for known DNA nucleases are absent, which suggests that the unknown activity of these CRISPR-Cas systems might not rely on the degradation of target DNA. We investigated the type IV CRISPR-Cas system in Aromatoleum aromaticum, which contains a type I-C CRISPR-Cas system on its genome and a type IV CRISPR-Cas system on one of its megaplasmids. RNA was isolated from A. aromaticum cells and subjected to Illumina sequencing. The RNA-Seq data revealed that the type I-C CRISPR systems yields crRNAs with an 11 nt 5-terminal repeat tag, which is in agreement with previous studies on a related type I-C system. Here, a Cas5 variant is responsible for crRNA maturation. In addition, we could show that the CRISPR array located next to the type IV system of the A. aromaticum megaplasmid is transcribed and mature crRNAs were sequenced. The mature crRNAs harbor a 7 nt 5′-terminal repeat tag which deviates from the standard crRNA tag. RNA cleavage assays demonstrated that Csf5 is a crRNA endonuclease of the Cas6 family and able to generate the unusual 7 nt 5′-terminal repeat tag. A crystal structure of the Csf5 in complex with a repeat sub-strate was obtained in collaboration with Prof. Gert Bange (Philipps-Universität Marburg). Recombinant production of a Type IV crRNP revealed a complex with very similar amounts of Csf3(Cas5 variant), Csf5(Cas6 variant) and Csf1 (large subunit), while increased abundance of Csf2(Cas7 variant) was observed. Our results indicate that the Type IV Cas7 variant is represented by several subunits in a Type IV crRNP which is in agreement with its proposed role as a backbone protein that mediates crRNA spacer binding. Transmission electron microscopy indicated that the crRNP form small crescent-shaped molecules with dimensions that resemble the size of Cascade complexes. Therefore, Type IV CRISPR-Cas activity is proposed to rely on Cascade-like complexes and guidance by a specific Type IV-associated crRNA component. Future studies will investigate the functionality of type IV CRISPR-Cas systems. We have shown that they induce plasmid clearance for targeted plasmids that contain a sequence with complementarity to the crRNA and a PAM sequence.

RNA stability in Archaea

Our group has a long-standing interest in RNA processing pathways in (hyperthermophilic) archaea and we continue to investigate SRP RNA processing and rRNA modification pathways that involve BHB intron splicing events and C/D box sRNA guidance. One recent topic of interest emerged from a collaboration with the group of Andres Jäschke (Heidelberg) in which we sought to identify the presence of cap-structures in archaeal RNAs. In the last years, the redox cofactor nicotinamide adenine dinucleotide (NAD) was found to be linked to the 5′ end of bacterial and eukaryotic RNAs and different consequences on mRNA stability were observed. NAD-RNAs are stabilized in bacterial cells, while in eukaryotic cells NAD modifications serve as degradation signals. RNA polymerases are responsible for the insertion of NAD moieties at the 5′ terminus of RNAs and NUDIX domain containing proteins can remove NAD residues to regulate their relative abundance. NAD-RNA modifications have not been investigated for any member of the archaeal domain and we aimed to determine if this domain of life resembles bacterial or eukaryotic NAD-RNA processing. We utilized NAD-captureSeq methodology to identify NAD-RNAs molecules in the archaeal model organism Sulfolobus acidocaldarius. Our results indicate that enriched RNAs did not significantly overlap with the general abundance of RNAs and nucleotide frequency analysis of the +1 position exclusively revealed an adenine at this position. This data suggests that NAD-capping in Archaea also occurs during transcription initiation as previously described for Eukaryotes and Bacteria. The functional consequences of this modification remain to be established. We observed that many of these RNAs are mRNAs that contain potentially structured 5′ UTR regions. This is notable as the majority of S. acidocaldarius mRNAs does not contain leader sequences and we propose that NAD modifications might be useful for stabilizing regulatory RNA elements including riboswitches. In similar vein, modification of tRNA leader sequences could assist in tRNA precursor maturation. Enriched mRNAs are derived from a significant number of genes that encode either for transcription factors or transporters. Our future studies will be aimed (i) to investigate the half-life of mRNAs in S. acidocaldarius to correlate these values with observed levels of NAD modifications, (ii) to identify the roles of NUDIX-domain containing proteins in regulating NAD-RNA levels in archaea and (iii) to investigate the presence of regulatory RNA elements in the leader regions of the enriched RNA