Cell Research (2016) 26:511-512. npg © 2016 IBCB, SIBS, CAS All rights reserved 1001-0602/16 $ 32.00 RESEARCH HIGHLIGHT www.nature.com/cr

Drosha and : Slicers cut from the same cloth Cell Research (2016) 26:511-512. doi:10.1038/cr.2016.19; published online 29 April 2016

DROSHA and its partner DGCR8 domains (RIIIDa and RIIIDb) and one to cleavage specificity and efficiency form a heterotrimeric complex named dsRNA-binding domain (dsRBD). The [5], providing additional constraints on Microprocessor, which is essential for pair of RIIIDs dimerize intramolecu- Microprocessor recognition. microRNA biogenesis. A recent study larly to form a composite processing During the last decade, structural by Kwon et al. in Cell reveals the center with the ability to cut the 3′ and 5′ information on key miRNA processing structure of a DROSHA construct in strands of the pri-miRNA stem, thereby DICER [6] and [7, complex with the C-terminal region producing characteristic staggered 8] have added to our understanding of of DGCR8, thereby unveiling the ends containing 2-nt 3′-overhangs. The the later steps of miRNA biogenesis. Al- topology and interactions between dsRBDs of DROSHA have weak RNA- though DROSHA was only discovered components of the Microprocessor binding capacity, which is augmented over a decade ago [9], it was recently and insights into its ‘ruler’-based by DGCR8 (domain architecture in that Kwon et al. [10] have reported the cleavage activity and function. Figure 1A, bottom panel), such that 3.2 Å crystal structure of a DROSHA (miRNAs) are a class DROSHA and DGCR8 together form construct (aa 390-1 365) in complex of ~22-nt endogenous non-coding a complex termed Microprocessor [1, with the C-terminal helix (aa 728-750) with important roles in the post- 2]. The binding stoichiometry of Micro- of DGCR8 (Figure 1C). The overall transcriptional regulation of in eu- processor has been elucidated from bio- structure adopts an elongated shape karyotes. The biogenesis of miRNAs is chemical assays to consist of a hetero- with the two RIIIDs located on one under strict control and their dysregula- trimer composed of one DROSHA and side and the N-terminal segment of the tion leads to the onset of diverse disease two DGCR8 molecules [3]. The central CED domain located on the other side. states. In humans, miRNA maturation RNA-binding heme domain (Rhed) of One small 23-aa DGCR8 helix (named involves two consecutive cleavage steps DGCR8 contributes to dimerization as G1) interacts with each DROSHA mediated by RNaseIII enzymes DRO- and Microprocessor processing fidelity RIIID, with the interaction between SHA and DICER. In the nucleus, DRO- [4], while the dsRBDs of DGCR8 have G1s and RIIIDs playing an important SHA together with its partner DiGeorge higher binding affinity to RNA and the role in stabilizing DROSHA. The syndrome critical region 8 (DGCR8) C-terminal tail region (CTT) is known superposition of RNaseIII domains of site-specifically crops a long primary to stabilize DROSHA. Drosha with Aquifex aeolicus RNaseIII transcript (pri-miRNA) to release a Pri-miRNAs, which are transcribed (AaRNaseIII) [11] indicates that DRO- hairpin RNA (pre-miRNA) of ~65-nt in by RNA Polymerase II (Pol II), contain SHA RIIIDs utilize a canonical catalytic length. Pre-miRNA is subsequently ex- a ~33-35-bp stem containing helical mechanism involving two metal ions ported into the and undergoes imperfections, a hairpin loop at the at the catalytic sites, thereby confirm- a second cleavage by DICER, liberating apical junction and single-strand RNA ing previous predictions that a pair of a ~22-nt mature miRNA duplex. This (ssRNA) overhangs at the basal junction RIIIDs generate a composite processing duplex is subsequently loaded onto (Figure 1B). Based on previous bio- center. However, there is an unusual Argonaute, with one strand termed the chemical data, it has been proposed that long conserved insertion (aa 898-964) guide retained, to form a RNA-induced Microprocessor cleaves at a distance of positioned within the DROSHA RIIIDa silencing complex (RISC). ~11-bp from the basal junction and at a predicted to involve two helical seg- DROSHA belongs to the RNaseIII distance of ~22-bp from the apical junc- ments. One of these, designated as the endonuclease family with the domain tion. Recent studies have identified four Bump helix, is buried into the CED architecture shown in Figure 1A (top primary sequence determinants located region, while the other one is disordered panel). The critical domains of DRO- in the pri-miRNA scaffold (UGUG ele- in the structure due to flexibility. RIIIDa SHA include a highly conserved central ment in apical loop, GHG (H = A, U or is linked to the bottom of the CED by domain (CED), which is essential for its C) elements in the stem, and UG and a long helix (named Connector) that is cleavage activity, and a C-terminal seg- CNNC elements in the basal ssRNA surrounded by the N-terminal part of ment containing two tandem RNaseIII overhangs (Figure 1B)) that contribute CED (named Platform) and the long npg 512

Figure 1 The structure of DROSHA-DGCR8 complex. (A) Schematic representation of the domain architecture of hu- man DROSHA and DGCR8. PAZ, Piwi-Argonaute-Zwille; RIIID, RNase III domain; dsRBD, double-stranded RNA- binding domain; Rhed, RNA-binding heme domain; CTT, C- terminal tail. (B) A schematic representation of pri-miRNA sequence and DROSHA cleavage site. The key invariant nucleotides are highlighted in red and the cleavage sites are marked with red arrows. (C) The crystal structure of human DROSHA in complex with DGCR8 (PDB code: 5B16). The Platform, PAZ-like, Connector, RIIIDa, RIIIDb, dsRBD, and CTT domains are colored in red, green, blue, orange, cyan, brown, and yellow, respectively. (D) Modeling of pri-miRNA onto the DROSHA complex with full-length DGCR8. Note: in B and C, the DROSHA-DGCR8 complex is aligned in the same orientation. We thank Drs V Narry Kim and Jae-Sung Woo for providing the coordinates of DROSHA-DGCR8 com- plex and DROSHA-DGCR8-pri-miRNA complex.

of the complex of DROSHA, full-length DGCR8 and bound pri-miRNA, so as to more fully decipher the rules governing pri-miRNA recognition, thereby provid- insertion of RIIIDa. Platform has two with the model guided by the structure ing a molecular understanding underly- unanticipated zinc finger (ZnF) mo- of dsRNA-bound AaRNase III [11]. ing accurate cleavage-based processing. tifs, with the ZnF1 motif stabilizing a Given the directionality of the DGCR8 1 1, 2 protruding loop from the Platform that CTT domain relative to the DROSHA Sisi Li , Dinshaw J Patel interacts with RIIIDa. The DROSHA RIIIDs, it is conceivable that full-length 1Department of Biology, South University of dsRBD domain is positioned adjacent DGCR8 can form a symmetrically Science and Technology of China, Shenzhen, to the RIIIDb site and projects away elongated complex, such that dimerized Guangdong 518055, China; 2Structural Biol- from the main scaffold in the absence Rhed and dsRBD domains are posi- ogy Program, Memorial Sloan Kettering Cancer of bound pri-miRNA. tioned to interact with the upper stem Center, New York, NY 10065, USA Correspondence: Dinshaw J Patel An unexpected and insightful ob- and apical loop of bound pri-miRNAs. E-mail: [email protected] servation is that the overall folding of The long insertion loop of RIIIDa DROSHA in the human DROSHA- (Bump and Mobile helices) together References DGCR8 complex resembles that ob- with Platform and PAZ-like domains served for Giardia Dicer [6], although could collaborate so as to recognize and 1 Denli AM, Tops BB, Plasterk RH, et al. Na- ture 2004; 432:231-235. there is no between hold the basal junction of pri-miRNA, 2 Gregory RI, Yan KP, Amuthan G, et al. Na- the two proteins, except that they con- thereby allowing the RIIIDa catalytic ture 2004; 432:235-240. tain a pair of RIIIDs. Moreover, they site to measure ~11-bp away from the 3 Nguyen TA, Jo MH, Choi YG, et al. Cell both contain a Connector helix of simi- basal junction. 2015; 161:1374-1387. 4 Barr I, Smith AT, Chen Y, et al. Proc Natl lar size and a Platform of different size This landmark contribution reveals Acad Scis USA 2012; 109:1919-1924. surrounding the Connector that adopts the structure of the catalytic core of 5 Fang W, Bartel DP. Mol Cell 2015; 60:131- the same folding topology, implying DROSHA, clarifies how DROSHA 145. that DROSHA and DICER most likely assembles with an interacting segment 6 Macrae IJ, Zhou K, Li F, et al. Science 2006; 311:195-198. evolved from a common ancestor. There of DGCR8 and confirms a 1:2 bind- 7 Wang Y, Juranek S, Li H, et al. Nature 2009; is a partially disordered region between ing stoichiometry of DROSHA and 461:754-761. Platform and Connector tentatively DGCR8. Interestingly, the similarity 8 Schirle NT, Sheu-Gruttadauria J, MacRae named PAZ-like region in DROSHA. between DROSHA and DICER impli- IJ. Science 2014; 346:608-613. 9 Lee Y, Ahn C, Han J, et al. Nature 2003; In order to illustrate how Micro- cates that class II RNaseIII 425:415-419. processor could potentially recognize (DORSHA) may have evolved from 10 Kwon SC, Nguyen TA, Choi YG, et al. Cell substrate pri-miRNAs, Kwon et al. class III RNaseIII nucleases (DICER) 2016; 164:81-90. built a model of the pri-miRNA bound early in evolution. A future challenge 11 Gan J, Tropea JE, Austin BP, et al. Cell 2006; 124:355-366. to DROSHA-DGCR8 (Figure 1D), remains the structural characterization

Cell Research | Vol 26 No 5 | May 2016