Mechanisms of RNAi

 The introduction of exogenous double-stranded RNA (dsRNA) into a cell can trigger the gene silencing process called RNA interference or RNAi. An RNAse III family enzyme, Dicer, initiates silencing by releasing ~20 base duplexes, with 2-nucleotide 3’-overhangs called siRNAs. The RNAi pathway also mediates the function of endogenous, non-coding regulatory RNAs called miRNAs. Both miRNAs and siRNAs guide substrate selection by similar effector complexes called RISC. These contain single-stranded versions of the small RNA and additional protein components. Of those, the signature element, which virtually defines a RISC, is a member of the Argonaute family of proteins. Argonaute proteins are defined by the presence of PAZ and PIWI domains. In a central RNAi pathway, guided by the siRNA, RISC directs the cleavage of mRNA substrates.

Although there has been remarkable progress in unraveling the components of the RNAi machinery, in order to get a true mechanistic understanding of this process, how these components fit together, and how they functioned, we must understand how they work at a molecular level. Therefore, we embarked on structural and biochemical studies of key proteins in the RNAi pathway, the protein-RNA and protein-protein interactions involved.

The PAZ domain resembles a right-handed baseball glove. Aromatic residues lining the intersubdomain cleft are involved in RNA binding.
We have initially determined the crystal structure of the PAZ domain of D. melanogaster Argonaute 2 (Ago2-PAZ). The PAZ domain forms a deviant OB fold containing a central cleft lined with conserved aromatic residues. Together with Greg Hannon’s lab we showed that the PAZ domain binds specifically to single-stranded 3’ ends. This was confirmed by subsequent structural studies of PAZ complexed with nucleic acids. Based upon these studies, we first proposed a model in which the PAZ domain interacts with the 3’ ends of siRNAs in the two proteins containing this domain, Dicer and Argonaute. In RISC, the Argonaute PAZ domain would hold the 3’ end of the single-stranded siRNA, perhaps orienting recognition and cleavage of mRNA substrates. However, the nuclease responsible for cleavage, dubbed “Slicer” has so far escaped identification.

In an effort to further our understanding of the role of Argonaute in RNAi, we determined the crystal structure of the full-length Argonaute protein from P. furiosus at 2.25 Å resolution. The structure reveals a crescent-shaped base made up of the N-terminal, middle and PIWI domains. The PAZ domain is held above the base by a “stalk”-like region. The PIWI domain is similar to RNase H, with conserved active site aspartate residues, strongly implicating Argonaute as “Slicer”. Several known characteristics of mRNA cleavage by RISC are consistent with an RNase H like enzyme. The architecture of the molecule and placement of the PAZ and PIWI domains define a groove for substrate binding. Based on the structure of PAZ with a “mini-siRNA”, we could model the siRNA binding with its 3’-end in the PAZ cleft and the siRNA-mRNA double helix extending into the groove and suggest a mechanism for siRNA guided mRNA cleavage. The notion that RISC “Slicer” activity, resides in Argonaute itself was tested in a mammalian system, by mutational analysis of hAgo2 in collaboration with Greg Hannon’s laboratory. Conserved active site aspartates in hAgo2 were altered and the mutants lost nuclease activity while retaining siRNA binding. This supports the model in which Argonaute itself functions as the Slicer enzyme in the RNAi pathway.

DNA-binding, assembly and helicase activity of the papillomavirus initiator protein E1
Structure of the E1-DBD dimer bound to sites 2 and 4 determined from the [(E1-DBD159-307)2(DNAd)] crystal structure.

In the past few years we set out to understand replication initiation in the papillomavirus system at an atomic level. This is a wonderful system for structural biology and biochemistry because the many varying activities of E1, the central protein in this process, are associated with different oligomeric states of the protein on DNA. It provides the opportunity to examine different types of protein-nucleic acid interactions: sequence-specific and non-sequence-specific, double-stranded and single-stranded, and a transition between these states that is controlled by protein-protein interactions. It is also a great system to investigate mechanisms of strand separation and DNA translocation by a replicative helicase. Moreover, there is a longstanding interest in using viral DNA replication as a model for understanding eukaryotic replication at CSHL. The high relevance to cervical cancer (see below) is another major incentive.
Structure of the E1-DBD tetramer bound to E1 binding sites 1-4 determined from the [(E1-DBD159-303)4(DNAt)] crystal structure. The DNA-binding loops are shown in yellow.

Infection with the human papillomavirus (HPV) is the most common sexually transmitted disease, with an infection rate of above 40% in the college-aged population. Women with persistent infections from certain types of HPV are at risk for cervical cancer, as 99% of cervical cancers around the world are associated with HPV infection. The papillomaviral E1 protein belongs to a family of multifunctional viral proteins whose main function is viral DNA replication. These proteins bind to the origin of DNA replication (ori), melt the DNA duplex, posses DNA helicase activity and recruit other cellular replication proteins, with various oligomeric forms of the same protein controlling its different activities. A sequential assembly of these complexes ensures an ordered transition between these different activities. Ultimately, E1 forms a hexameric ring helicase on each strand that serve as a replicative DNA helicases that unwinds the DNA in front of the replication fork. We are pursuing structural and biochemical studies of E1, its DNA binding activity and its assembly on DNA. This system is providing general insights into the biochemical events that are involved in viral DNA replication in particular and replication initiation and DNA helicase activity in general. Our studies can also provide a basis for the development of clinical intervention strategies.
A cartoon depicting a model for the assembly of two hexameric helicases around single strands at the ori. The two DBLs (in yellow) of the upstream monomers of each dimer bind to one strand, while the DBLs of the downstream monomers of each dimer bind to the other strand. One hexameric helicase assembles from the upstream monomers around the top strand and the other hexameric helicase assembles from the downstream monomers around the other strand.

Based on structures of the DNA-binding domain of E1 bound to DNA that we determined a few years ago in collaboration with Arne Stenlund’s lab, we suggested a mechanism for DNA strand separation. We captured structural snapshots of two sequential steps in the assembly process with structures of both the dimeric and tetrameric forms of the E1-DBD bound to the origin DNA. We found that the mode of DNA-binding employed by E1 partitions the two individual DNA strands onto distinct binding surfaces of the protein. The organization of these surfaces on the origin suggests how E1 ultimately progresses from a double-stranded origin recognition complex to a hexameric helicase where the two strands are fully separated and each strand is encircled by a hexameric ring.

However, the mechanism that couples the ATP cycle to DNA translocation has been unclear. The E1 hexameric helicase adopts a ring shape with a prominent central channel that has been presumed to encircle substrate DNA during the unwinding process, but the atomic details of this binding have been uncertain, including whether the ring encircles one or both strands of DNA during unwinding.

Our crystal structure of the E1 hexameric helicase bound to single-stranded DNA demonstrates that only one strand of DNA passes through the central channel and reveals the details of single-stranded DNA binding. The b-hairpins (DNA-binding hairpins) of each subunit sequentially track the sugar-phosphate backbone of the DNA in a one nucleotide per subunit increment. This configuration resembles a spiral staircase.
Views of the E1 hexamer parallel and perpendicular to the central channel with individual subunits are color-coded. Single-stranded DNA is bound discretely within the channel, and nucleotides are present at the subunit interfaces.



ATP-binding (and hydrolysis) sites are located at the subunit interfaces, and multiple configurations are observed within the hexamer. These have been assigned as ATP-type, ADP-type, and apo-type. The configuration of the site for a given subunit correlates with the relative height of its DNA-binding hairpin in the staircase arrangement. The subunits that adopt an ATP-type configuration place their hairpins at the top of the staircase while the hairpins of apo-type subunits occupy the bottom positions of the staircase. The hairpins of the ADP-type subunits are placed at intermediate positions.



Details of DNA coordination viewed parallel (left) and perpendicular (right) to the hexamer channel

A straightforward “coordinated escort” DNA-translocation mechanism is inferred from the staircased DNA-binding and its correlation with the configuration at the ATP-binding sites. Each DNA-binding hairpin maintains continuous contact with one unique nucleotide of ssDNA and migrates downward via ATP-hydrolysis and subsequent ADP-release at the subunit interfaces. ATP-hydrolysis occurs between subunits located toward the top of the staircase, while ADP-release occurs between subunits located toward the bottom of the staircase. The hairpin at the bottom of the staircase releases its associated ssDNA phosphate to conclude its voyage through the hexameric channel. Upon binding a new ATP molecule, this subunit moves to the top of the staircase to pick up the next available ssDNA phosphate, initiating its escorted journey through the channel and repeating the process. For one full cycle of the hexamer, each subunit hydrolyzes one ATP molecule, releases one ADP molecule, and translocates one nucleotide of DNA through the interior channel. A full cycle, therefore, translocates 6 nucleotides with associated hydrolysis of 6 ATPs and release of 6 ADPs.
A cartoon depicting the “coordinated escort” mechanism for DNA translocation.

The crystal structure of Argonaute from Pyrococcus furiosus. Ribbon representation of Argonaute showing the N-terminal domain (blue), the “stalk (light blue), the PAZ domain (red), the middle domain (green), the PIWI domain (purple), and the interdomain connector (yellow).