Matthew J. Schellenberg

Recognition and maturation of effector RNAs in a CRISPR interference pathway

In bacteria and archaea, small RNAs derived from clustered, regularly interspaced, short palindromic repeat (CRISPR) loci are involved in an adaptable and heritable gene-silencing pathway. Resistance to phage infection is conferred by the incorporation of short invading DNA sequences into the genome as CRISPR spacer elements separated by short repeat sequences. Processing of long primary transcripts (pre-crRNAs) containing these repeats by an RNA endonuclease generates the mature effector RNAs that interfere with phage gene expression. Here we describe structural and functional analyses of the Thermus thermophilus CRISPR Cse3 endonuclease. High-resolution X-ray structures of Cse3 bound to repeat RNAs model both the pre- and post-cleavage complexes associated with processing the pre-crRNA. These structures establish the molecular basis of a specific CRISPR RNA recognition and suggest the mechanism for generation of effector RNAs responsible for gene silencing.

Structural elucidation of a PRP8 core domain from the heart of the spliceosome

The spliceosome is a complex ribonucleoprotein (RNP) particle containing five RNAs and more than 100 associated proteins. One of these proteins, PRP8, has been shown to interact directly with the splice sites and branch region of precursor-mRNAs (pre-mRNAs) and spliceosomal RNAs associated with catalysis of the two steps of splicing. The 1.85-Å X-ray structure of the core of PRP8 domain IV, implicated in key spliceosomal interactions, reveals a bipartite structure that includes the presence of an RNase H fold linked to a five-helix assembly. Analysis of mutant yeast alleles and cross-linking results in the context of this structure, coupled with RNA binding studies, suggests that domain IV forms a surface that interacts directly with the RNA structures at the catalytic core of the spliceosome.

FRET Analysis of in Vivo Dimerization by RNA-editing Enzymes

Members of the ADAR (adenosine deaminase that acts on RNA) enzyme family catalyze the hydrolytic deamination of adenosine to inosine within double-stranded RNAs, a poorly understood process that is critical to mammalian development. We have performed fluorescence resonance energy transfer experiments in mammalian cells transfected with fluorophore-bearing ADAR1 and ADAR2 fusion proteins to investigate the relationship between these proteins. These studies conclusively demonstrate the homodimerization of ADAR1 and ADAR2 and also show that ADAR1 and ADAR2 form heterodimers in human cells. RNase treatment of cells expressing these fusion proteins changes their localization but does not affect dimerization. Taken together these results suggest that homo- and heterodimerization are important for the activity of ADAR family members in vivo and that these associations are RNA independent.

Cas5d processes pre-crRNA and is a member of a larger family of CRISPR RNA endonucleases

Small RNAs derived from clustered, regularly interspaced, short palindromic repeat (CRISPR) loci in bacteria and archaea are involved in an adaptable and heritable gene-silencing pathway. Resistance to invasive genetic material is conferred by the incorporation of short DNA sequences derived from this material into the genome as CRISPR spacer elements separated by short repeat sequences. Processing of long primary transcripts (pre-crRNAs) containing these repeats by a CRISPR-associated (Cas) RNA endonuclease generates the mature effector RNAs that target foreign nucleic acid for degradation. Here we describe functional studies of a Cas5d ortholog, and high-resolution structural studies of a second Cas5d family member, demonstrating that Cas5d is a sequence-specific RNA endonuclease that cleaves CRISPR repeats and is thus responsible for processing of pre-crRNA. Analysis of the structural homology of Cas5d with the previously characterized Cse3 protein allows us to model the interaction of Cas5d with its RNA substrate and conclude that it is a member of a larger family of CRISPR RNA endonucleases.

Spliceosome structure: piece by piece.

Processing of pre-mRNAs by RNA splicing is an essential step in the maturation of protein coding RNAs in eukaryotes. Structural studies of the cellular splicing machinery, the spliceosome, are a major challenge in structural biology due to the size and complexity of the splicing ensemble. Specifically, the structural details of splice site recognition and the architecture of the spliceosome active site are poorly understood. X-ray and NMR techniques have been successfully used to address these questions defining the structure of individual domains, isolated splicing proteins, spliceosomal RNA fragments and recently the U1 snRNP multiprotein.RNA complex. These results combined with extant biochemical and genetic data have yielded important insights as well as posing fresh questions with respect to the regulation and mechanism of this critical gene regulatory process.

Pre-mRNA splicing: a complex picture in higher definition

Intron excision from pre-mRNAs of higher eukaryotes requires a transition from splice-site recognition across short exons to organization of the spliceosome across long introns. Recently, insight into this transition has been provided and, in addition, it has been shown that an alternative splicing factor, the polypyrimidine-tract-binding protein, can exert its control on splice-site choice by blocking this key step in the assembly of the splicing machinery.

A conformational switch in PRP8 mediates metal ion coordination that promotes pre-mRNA exon ligation.

Splicing of pre-mRNAs in eukaryotes is catalyzed by the spliceosome, a large RNA-protein metalloenzyme. The catalytic center of the spliceosome involves a structure comprising the U2 and U6 snRNAs and includes a metal bound by U6 snRNA. The precise architecture of the splicesome active site, however, and the question of whether it includes protein components, remains unresolved. A wealth of evidence places the protein PRP8 at the heart of the spliceosome through assembly and catalysis. Here we provide evidence that the RNase H domain of PRP8 undergoes a conformational switch between the two steps of splicing, rationalizing yeast prp8 alleles that promote either the first or second step. We also show that this switch unmasks a metal-binding site involved in the second step. Together, these data establish that PRP8 is a metalloprotein that promotes exon ligation within the spliceosome.

Structural model of the p14/SF3b155 · branch duplex complex.

Human p14 (SF3b14), a component of the spliceosomal U2 snRNP, interacts directly with the pre-mRNA branch adenosine within the context of the bulged duplex formed between the pre-mRNA branch region and U2 snRNA. This association occurs early in spliceosome assembly and persists within the fully assembled spliceosome. Analysis of the crystal structure of a complex containing p14 and a peptide derived from p14-associated SF3b155 combined with the results of cross-linking studies has suggested that the branch nucleotide interacts with a pocket on a non-canonical RNA binding surface formed by the complex. Here we report a structural model of the p14 · bulged duplex interaction based on a combination of X-ray crystallography of an adenine p14/SF3b155 peptide complex, biochemical comparison of a panel of disulfide cross-linked protein-RNA complexes, and small-angle X-ray scattering (SAXS). These studies reveal specific recognition of the branch adenosine within the p14 pocket and establish the orientation of the bulged duplex RNA bound on the protein surface. The intimate association of one surface of the bulged duplex with the p14/SF3b155 peptide complex described by this model buries the branch nucleotide at the interface and suggests that p14 · duplex interaction must be disrupted before the first step of splicing.

Context-Dependent Remodeling of Structure in Two Large Protein Fragments

Protein folding involves the formation of secondary structural elements from the primary sequence and their association with tertiary assemblies. The relation of this primary sequence to a specific folded protein structure remains a central question in structural biology. An increasing body of evidence suggests that variations in homologous sequence ranging from point mutations to substantial insertions or deletions can yield stable proteins with markedly different folds. Here we report the structural characterization of domain IV (D4) and ΔD4 (polypeptides with 222 and 160 amino acids, respectively) that differ by virtue of an N-terminal deletion of 62 amino acids (28% of the overall D4 sequence). The high-resolution crystal structures of the monomeric D4 and the dimeric ΔD4 reveal substantially different folds despite an overall conservation of secondary structure. These structures show that the formation of tertiary structures, even in extended polypeptide sequences, can be highly context dependent, and they serve as a model for structural plasticity in protein isoforms.

Selectable high-yield recombinant protein production in human cells using a GFP/YFP nanobody affinity support: GFP-Tag Protein Expression in Human Cells

Recombinant protein expression systems that produce high yields of pure proteins and multi‐protein complexes are essential to meet the needs of biologists, biochemists, and structural biologists using X‐ray crystallography and cryo‐electron microscopy. An ideal expression system for recombinant human proteins is cultured human cells where the correct translation and chaperone machinery are present. However, compared to bacterial expression systems, human cell cultures present several technical challenges to their use as an expression system. We developed a method that utilizes a YFP fusion‐tag to generate recombinant proteins using suspension‐cultured HEK293F cells. YFP is a dual‐function tag that enables direct visualization and fluorescence‐based selection of high expressing clones for and rapid purification using a high‐stringency, high‐affinity anti‐GFP/YFP nanobody support. We demonstrate the utility of this system by expressing two large human proteins, TOP2α (340 KDa dimer) and a TOP2β catalytic core (260 KDa dimer). This robustly and reproducibly yields >10 mg/L liter of cell culture using transient expression or 2.5 mg/L using stable expression.