research papers\(\def\hfill{\hskip 5em}\def\hfil{\hskip 3em}\def\eqno#1{\hfil {#1}}\)

Journal logoSTRUCTURAL
BIOLOGY
ISSN: 2059-7983

Structural and functional investigation of the human snRNP assembly factor AAR2 in complex with the RNase H-like domain of PRPF8

crossmark logo

aLaboratory of RNA Biochemistry, Freie Universität Berlin, Takustrasse 6, 14195 Berlin, Germany, bLaboratory of Structural Biochemistry, Freie Universität Berlin, Takustrasse 6, 14195 Berlin, Germany, and cMacromolecular Crystallography, Helmholtz-Zentrum Berlin für Materialien und Energie, Albert-Einstein-Strasse 15, 12489 Berlin, Germany
*Correspondence e-mail: mpreussner@zedat.fu-berlin.de, gert.weber@helmholtz-berlin.de

Edited by C. S. Bond, University of Western Australia, Crawley, Australia (Received 14 August 2022; accepted 5 October 2022; online 27 October 2022)

Small nuclear ribonucleoprotein complexes (snRNPs) represent the main subunits of the spliceosome. While the assembly of the snRNP core particles has been well characterized, comparably little is known of the incorporation of snRNP-specific proteins and the mechanisms of snRNP recycling. U5 snRNP assembly in yeast requires binding of the the Aar2 protein to Prp8p as a placeholder to preclude premature assembly of the SNRNP200 helicase, but the role of the human AAR2 homolog has not yet been investigated in detail. Here, a crystal structure of human AAR2 in complex with the RNase H-like domain of the U5-specific PRPF8 (PRP8F RH) is reported, revealing a significantly different interaction between the two proteins compared with that in yeast. Based on the structure of the AAR2–PRPF8 RH complex, the importance of the interacting regions and residues was probed and AAR2 variants were designed that failed to stably bind PRPF8 in vitro. Protein-interaction studies of AAR2 with U5 proteins using size-exclusion chromatography reveal similarities and marked differences in the interaction patterns compared with yeast Aar2p and imply phosphorylation-dependent regulation of AAR2 reminiscent of that in yeast. It is found that in vitro AAR2 seems to lock PRPF8 RH in a conformation that is only compatible with the first transesterification step of the splicing reaction and blocks a conformational switch to the step 2-like, Mg2+-coordinated conformation that is likely during U5 snRNP biogenesis. These findings extend the picture of AAR2 PRP8 interaction from yeast to humans and indicate a function for AAR2 in the spliceosomal assembly process beyond its role as an SNRNP200 placeholder in yeast.

1. Introduction

Precursor messenger RNA (pre-mRNA) splicing is catalyzed by a highly dynamic, multi-megadalton ribonucleoprotein (RNP) machinery, the spliceosome (Will & Lührmann, 2011[Will, C. L. & Lührmann, R. (2011). Cold Spring Harb. Perspect. Biol. 3, a003707.]; Wahl et al., 2009[Wahl, M. C., Will, C. L. & Lührmann, R. (2009). Cell, 136, 701-718.]). The small nuclear RNPs (snRNPs) U1, U2, U4, U5 and U6 are the main subunits of the major, U2-type spliceosome. Each of these snRNPs contains a particle-specific snRNA, seven common Sm proteins [or like Sm (LSm) proteins in the case of U6] and a set of particle-specific proteins (Will & Lührmann, 2001[Will, C. L. & Lührmann, R. (2001). Curr. Opin. Cell Biol. 13, 290-301.]). The U5 snRNP is the only snRNP subunit that is also employed by the minor spliceosome (Will & Lührmann, 2005[Will, C. L. & Lührmann, R. (2005). Biol. Chem. 386, 713-724.]). Apart from the snRNPs, a multitude of proteins and protein complexes that are not stably associated with an snRNP also join the spliceosome to facilitate and regulate pre-mRNA splicing (Agafonov et al., 2011[Agafonov, D. E., Deckert, J., Wolf, E., Odenwälder, P., Bessonov, S., Will, C. L., Urlaub, H. & Lührmann, R. (2011). Mol. Cell. Biol. 31, 2667-2682.]). For each round of splicing, the spliceosome assembles anew on a pre-mRNA by the stepwise association of snRNPs and non-snRNP proteins. Its catalytic center is not preformed but only emerges during assembly by repeated, extensive remodeling of the specific RNA–protein interaction network of each assembly stage to eventually elicit intron excision and exon ligation via two transesterification reactions, referred to as step 1 and step 2 (Wahl et al., 2009[Wahl, M. C., Will, C. L. & Lührmann, R. (2009). Cell, 136, 701-718.]; Will & Lührmann, 2011[Will, C. L. & Lührmann, R. (2011). Cold Spring Harb. Perspect. Biol. 3, a003707.]).

SnRNPs themselves are assembled via complex pathways, which in the cases of U1, U2, U4 and U5 include cytoplasmic and nuclear phases (Will & Lührmann, 2001[Will, C. L. & Lührmann, R. (2001). Curr. Opin. Cell Biol. 13, 290-301.]; Matera & Wang, 2014[Matera, A. G. & Wang, Z. (2014). Nat. Rev. Mol. Cell Biol. 15, 108-121.]; Gruss et al., 2017[Gruss, O. J., Meduri, R., Schilling, M. & Fischer, U. (2017). Chromosoma, 126, 577-593.]). The corresponding snRNAs are synthesized by RNA polymerase II (Pol II), modified with an m7G cap, processed by the integrator complex and exported to the cytoplasm. Here, the Sm proteins are assembled stepwise to form a ring-like structure around a U-rich Sm site in the snRNAs via the protein arginine methyltransferase 5 complex and the survival motor neuron (SMN) complex. Trimethylguanosine synthase 1 then catalyzes hypermethylation of the m7G cap to generate an m2,2,7G cap. The hypermethylated cap and the assembled Sm core domain act as a composite nuclear localization signal that facilitates re-entry of the Sm core RNPs into the nucleus.

The integration of particle-specific proteins into snRNPs also requires specific assembly factors and chaperones. For example, in human cells, the adaptor protein, nuclear FMR1-interacting protein 1, and the heat-shock protein 90 (HSP90)/Rvb1–Rvb2–Tah1–Pih1 (RT2P) chaperone machinery in collaboration with a nuclear-localized SMN complex facilitate integration of the U4/U6-specific proteins NHP2-like protein 1 and pre-mRNA processing factor (PRPF) 31 into the U4/U6 di-snRNP (Bizarro et al., 2015[Bizarro, J., Dodré, M., Huttin, A., Charpentier, B., Schlotter, F., Branlant, C., Verheggen, C., Massenet, S. & Bertrand, E. (2015). Nucleic Acids Res. 43, 8973-8989.]). The HSP90/RT2P chaperone machinery also supports the assembly of a U5 snRNP module composed of the PRPF8 protein (Prp8p in yeast), the 116 kDa U5 small nuclear ribonucleoprotein component (EFTUD2; Snu114p in yeast), the U5 small nuclear ribonucleoprotein 200 kDa helicase (SNRNP200; Brr2p in yeast) and the SNRNP40 protein in the cytoplasm, thereby promoting formation of the mature U5 snRNP (Malinová et al., 2017[Malinová, A., Cvačková, Z., Materějů, D., Hořejší, Z., Abéza, C., Vandermoere, F., Bertrand, E., Staněk, D. & Verheggen, C. (2017). J. Cell Biol. 216, 1579-1596.]). Several additional proteins have been implicated in U5 snRNP or U4/U6-U5 tri-snRNP assembly in metazoans (Malinová et al., 2017[Malinová, A., Cvačková, Z., Materějů, D., Hořejší, Z., Abéza, C., Vandermoere, F., Bertrand, E., Staněk, D. & Verheggen, C. (2017). J. Cell Biol. 216, 1579-1596.]; Klimešová et al., 2021[Klimešová, K., Vojáčková, J., Radivojević, N., Vandermoere, F., Bertrand, E., Verheggen, C. & Staněk, D. (2021). Nat. Commun. 12, 3646.]; Erkelenz et al., 2021[Erkelenz, S., Stanković, D., Mundorf, J., Bresser, T., Claudius, A. K., Boehm, V., Gehring, N. H. & Uhlirova, M. (2021). Nucleic Acids Res. 49, 1688-1707.]; Bergfort et al., 2022[Bergfort, A., Hilal, T., Kuropka, B., Ilik, İ. A., Weber, G., Aktaş, T., Freund, C. & Wahl, M. C. (2022). Nucleic Acids Res. 50, 2938-2958.]).

Some snRNPs are profoundly remodeled during pre-mRNA splicing, necessitating specific recycling mechanisms to reassemble the particles for further rounds of splicing. For example, during spliceosome activation the U4 and U6 snRNAs, which are extensively base-paired in the U4/U6 di-snRNP, are unwound by the SNRNP200 helicase, and U4 snRNA and all U4/U6-associated proteins are displaced (Laggerbauer et al., 1998[Laggerbauer, B., Achsel, T. & Lührmann, R. (1998). Proc. Natl Acad. Sci. USA, 95, 4188-4192.]; Raghunathan & Guthrie, 1998[Raghunathan, P. L. & Guthrie, C. (1998). Curr. Biol. 8, 847-855.]; Agafonov et al., 2011[Agafonov, D. E., Deckert, J., Wolf, E., Odenwälder, P., Bessonov, S., Will, C. L., Urlaub, H. & Lührmann, R. (2011). Mol. Cell. Biol. 31, 2667-2682.]). Furthermore, in human cells the U5 snRNP enters the spliceosome as a 20S particle but is released as a 35S particle after splicing due to incorporation of the PRPF19 complex and additional factors (Makarov et al., 2002[Makarov, E. M., Makarova, O. V., Urlaub, H., Gentzel, M., Will, C. L., Wilm, M. & Lührmann, R. (2002). Science, 298, 2205-2208.]). Late de novo snRNP biogenesis steps and recycling of snRNPs after pre-mRNA splicing take place in nuclear Cajal bodies (Staněk & Neugebauer, 2004[Staněk, D. & Neugebauer, K. M. (2004). J. Cell Biol. 166, 1015-1025.], 2006[Staněk, D. & Neugebauer, K. M. (2006). Chromosoma, 115, 343-354.]; Sleeman et al., 2001[Sleeman, J. E., Ajuh, P. & Lamond, A. I. (2001). J. Cell Sci. 114, 4407-4419.]; Sleeman & Lamond, 1999[Sleeman, J. E. & Lamond, A. I. (1999). Curr. Biol. 9, 1065-1074.]).

The U5-specific PRPF8 protein is one of the most conserved nuclear proteins and coordinates proteins, snRNAs and the pre-mRNA at the catalytic center of the spliceosome (Grainger & Beggs, 2005[Grainger, R. J. & Beggs, J. D. (2005). RNA, 11, 533-557.]). PRPF8 harbors two regulatory pseudo-enzyme domains at its very C-terminus, comprising an RNase H-like (RH) and a Jab1/MPN-like (JM) fold (Pena et al., 2007[Pena, V., Liu, S., Bujnicki, J. M., Lührmann, R. & Wahl, M. C. (2007). Mol. Cell, 25, 615-624.], 2008[Mozaffari-Jovin, S., Wandersleben, T., Santos, K. F., Will, C. L., Lührmann, R. & Wahl, M. C. (2013). Science, 341, 80-84.]; Zhang et al., 2007[Zhang, L., Shen, J., Guarnieri, M. T., Heroux, A., Yang, K. & Zhao, R. (2007). Protein Sci. 16, 1024-1031.]; Yang et al., 2008[Yang, K., Zhang, L., Xu, T., Heroux, A. & Zhao, R. (2008). Proc. Natl Acad. Sci. USA, 105, 13817-13822.]; Ritchie et al., 2008[Ritchie, D. B., Schellenberg, M. J., Gesner, E. M., Raithatha, S. A., Stuart, D. T. & MacMillan, A. M. (2008). Nat. Struct. Mol. Biol. 15, 1199-1205.]). In yeast, the A1 cistron-splicing factor (Aar2p) has been characterized as a U5 snRNP assembly and recycling factor that mediates the formation of pre-U5 snRNPs lacking the Brr2p helicase in the cytoplasm (Gottschalk et al., 2001[Gottschalk, A., Kastner, B., Lührmann, R. & Fabrizio, P. (2001). RNA, 7, 1554-1565.]; Boon et al., 2007[Boon, K.-L., Grainger, R. J., Ehsani, P., Barrass, J. D., Auchynnikava, T., Inglehearn, C. F. & Beggs, J. D. (2007). Nat. Struct. Mol. Biol. 14, 1077-1083.]). Aar2p concomitantly binds the RH and JM domains of Prp8p (Weber et al., 2011[Weber, G., Cristão, V. F., Alves, F. L., Santos, K. F., Holton, N., Rappsilber, J., Beggs, J. D. & Wahl, M. C. (2011). Genes Dev. 25, 1601-1612.], 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]; Galej et al., 2013[Galej, W. P., Oubridge, C., Newman, A. J. & Nagai, K. (2013). Nature, 493, 638-643.]). By sequestering the Prp8p JM domain, which is a major binding site of the Brr2p RNA helicase, Aar2p initially prevents the integration of Brr2p into U5 snRNP (Weber et al., 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]; Galej et al., 2013[Galej, W. P., Oubridge, C., Newman, A. J. & Nagai, K. (2013). Nature, 493, 638-643.]). Aar2p accompanies the Brr2p-deficient pre-U5 snRNP particle into the nucleus, where phosphorylation of its Ser253 residue triggers refolding and release of Aar2p, allowing Brr2p entry to complete U5 snRNP biogenesis (Boon et al., 2007[Boon, K.-L., Grainger, R. J., Ehsani, P., Barrass, J. D., Auchynnikava, T., Inglehearn, C. F. & Beggs, J. D. (2007). Nat. Struct. Mol. Biol. 14, 1077-1083.]; Weber et al., 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]).

In a previous study, we demonstrated that the human Aar2p homolog AAR2 is produced from the c20orf4 gene in HeLa cells and that it stably binds the PRPF8 RH domain in vitro (Santos et al., 2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]). However, human AAR2 and yeast Aar2p exhibit only 24% sequence identity, questioning the extent to which their structures and molecular mechanisms are conserved. While proteomics studies and pull-down experiments have suggested that human AAR2 also participates in U5 snRNP assembly (Malinová et al., 2017[Malinová, A., Cvačková, Z., Materějů, D., Hořejší, Z., Abéza, C., Vandermoere, F., Bertrand, E., Staněk, D. & Verheggen, C. (2017). J. Cell Biol. 216, 1579-1596.]; Klimešová et al., 2021[Klimešová, K., Vojáčková, J., Radivojević, N., Vandermoere, F., Bertrand, E., Verheggen, C. & Staněk, D. (2021). Nat. Commun. 12, 3646.]), human AAR2 has been identified in a complex with PRPF8, EFTUD2, SNRNP200 and SNRNP40 (Malinová et al., 2017[Malinová, A., Cvačková, Z., Materějů, D., Hořejší, Z., Abéza, C., Vandermoere, F., Bertrand, E., Staněk, D. & Verheggen, C. (2017). J. Cell Biol. 216, 1579-1596.]), indeed indicating potential differences in the Aar2p/AAR2-mediated U5 snRNP assembly steps in yeast and humans.

Here, we report a co-crystal structure of human AAR2 in complex with the PRPF8 RH domain (PRPF8RH) and present further interaction studies of AAR2 with PRPF8 fragments and the SNRNP200 helicase in vitro. In contrast to the situation in yeast, we find that a human AAR2–PRPF8RH complex does not bind the PRPF8 JM domain and thus permits the formation of a trimeric AAR2–PRPF8–SNRNP200 complex. As in yeast, the human AAR2–PRPF8RH interaction is abrogated in vitro by a phosphomimetic S284E (S253E in yeast) mutation, indicating highly conserved regulation of AAR2 by phosphorylation. Furthermore, AAR2 seems to lock PRPF8RH in its first-step conformation and block the conformational switch to a step 2-like, Mg2+-coordinated conformation during U5 snRNP biogenesis. Our results shed the first light on the human AAR2–PRPF8RH interface and imply a different role of AAR2 in spliceosomal assembly than in yeast.

2. Materials and methods

2.1. Cloning, expression, protein purification and reconstitution of protein complexes

We employed a modified pFL vector encoding a truncated version of human SNRNP200 lacking the first 394 residues (residues 395–2136) and containing an N-terminal, TEV-cleavable His10 tag (Santos et al., 2012[Santos, K. F., Jovin, S. M., Weber, G., Pena, V., Lührmann, R. & Wahl, M. C. (2012). Proc. Natl Acad. Sci. USA, 109, 17418-17423.]), a pET-M11 plasmid encoding human PRPF8RH (residues 1747–2016) and containing an N-terminal, TEV-cleavable His6 tag (Pena et al., 2008[Pena, V., Rozov, A., Fabrizio, P., Lührmann, R. & Wahl, M. C. (2008). EMBO J. 27, 2929-2940.]; Weber et al., 2011[Weber, G., Cristão, V. F., Alves, F. L., Santos, K. F., Holton, N., Rappsilber, J., Beggs, J. D. & Wahl, M. C. (2011). Genes Dev. 25, 1601-1612.]), a pET-M11 plasmid encoding human PRPF8JM (residues 2064–2335) and containing an N-terminal TEV-cleavable GST tag (Mozaffari-Jovin et al., 2013[Mozaffari-Jovin, S., Wandersleben, T., Santos, K. F., Will, C. L., Lührmann, R. & Wahl, M. C. (2013). Science, 341, 80-84.]), and pFL vectors encoding human AAR2 or AAR2Δloop (AAR2 with residues 170–200 replaced by three serines) with both AAR2 constructs containing an N-terminal, TEV-cleavable His10 tag (Santos et al., 2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]), which have been described previously. The inserts were derived from codon-optimized synthetic genes. A DNA fragment encoding the C-terminal fragment of PRPF8 (encompassing the RH and JM domains; PRPF8RH-JM; residues 1760–2335) was PCR-amplified from a codon-optimized prpf8 synthetic gene and cloned into the pIDK donor vector by restriction-enzyme cloning using XhoI and NdeI (NEB). Snrnp200395–2136-pFL and prpf81760–2335-pIDK were fused by Cre-Lox recombination and used for bacmid preparation. A codon-optimized DNA fragment encoding human AAR2 residues 1–364, lacking the C-terminal 20 residues of AAR2, was cloned into a modified pFL vector by restriction-enzyme cloning using EcoRI and HindIII (NEB) to guide the production of AAR21–364 containing a C-terminal His6 tag. A codon-optimized DNA fragment encoding human AAR2 was cloned into pcDNA3.1(+) using BamHI and XhoI restriction enzymes (NEB) to guide the production of AAR2 containing an N-terminal FLAG tag. Mutations were introduced with the QuikChange site-directed mutagenesis kit (Stratagene) according to the manufacturer's instructions. All constructs were verified by dye-terminator sequencing (Seqlab).

Purified human SNRNP200395–2136 (8 mg ml−1; Santos et al., 2012[Santos, K. F., Jovin, S. M., Weber, G., Pena, V., Lührmann, R. & Wahl, M. C. (2012). Proc. Natl Acad. Sci. USA, 109, 17418-17423.]), PRPF8RH (30 mg ml−1; Pena et al., 2008[Pena, V., Rozov, A., Fabrizio, P., Lührmann, R. & Wahl, M. C. (2008). EMBO J. 27, 2929-2940.]), PRPF8JM (10 mg ml−1; Mozaffari-Jovin et al., 2013[Mozaffari-Jovin, S., Wandersleben, T., Santos, K. F., Will, C. L., Lührmann, R. & Wahl, M. C. (2013). Science, 341, 80-84.]), AAR2 (12 mg ml−1; Santos et al., 2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]), AAR2Δloop (12 mg ml−1; Santos et al., 2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]), an SNRNP200395–2136–PRPF8JM complex (Mozaffari-Jovin et al., 2013[Mozaffari-Jovin, S., Wandersleben, T., Santos, K. F., Will, C. L., Lührmann, R. & Wahl, M. C. (2013). Science, 341, 80-84.]; 8 mg ml−1) and an AAR2–PRPF8RH complex (20 mg ml−1; Santos et al., 2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]) were obtained as described previously. AAR21–364 (11 mg ml−1) was produced and purified as described for AAR2 but with omission of the TEV protease cleavage step, leaving the His6 tag intact. SNRNP200395–2136 and PRPF8RH-JM were co-produced based on a recombinant baculovirus derived from the recombined bacmid in Sf9 cells. For purification of the SNRNP200395–2136–PRPF8RH-JM complex, cells were resuspended in resuspension buffer [50 mM Tris–HCl pH 8.0, 300 mM NaCl, 5%(v/v) glycerol, 1 mM DTT, 0.05%(v/v) NP40] supplemented with EDTA-free protease inhibitors (Roche) and DNase I (NEB) and lysed via sonification. After centrifugation, the lysate of about 50 column volumes was filtered and passed through Ni–NTA beads (Qiagen). The beads were washed twice with ten column volumes of resuspension buffer containing 15 mM imidazole. The captured complex was eluted with two column volumes of resuspension buffer containing 500 mM imidazole, TEV protease (0.5 mg per millitre of protein solution) was added and the mixture was dialyzed against 20 mM HEPES–NaOH pH 7.5, 200 mM NaCl, 1 mM DTT overnight at 4°C. Five column volumes of the buffer-exchanged sample were again passed through Ni–NTA beads and the flowthrough was collected. The complex was concentrated to a final concentration of 7 mg ml−1 and further purified by size-exclusion chromatography (SEC) on a Superdex S200 10/300 column (GE Healthcare) in 20 mM HEPES–NaOH pH 7.5, 200 mM NaCl, 1 mM DTT.

2.2. Analytical size-exclusion chromatography

Individual proteins and protein mixtures were analyzed by analytical SEC on a Superdex 200 Increase PC3.2/30 column (GE Healthcare) in 20 mM Tris–HCl pH 7.5, 250 mM NaCl, 0.5 mM DTT at a flow rate of 50–70 µl min−1. For analysis of complex formation, proteins (at the concentrations stated in Section 2.1[link]) were mixed in equimolar ratios in 60 µl size-exclusion buffer and incubated for 30 min on ice. Elution fractions were supplemented with SDS–PAGE loading buffer and analyzed by SDS–PAGE.

2.3. Crystallographic analyses

Crystallization of the human AAR2Δloop–PRPF8RH complex has been described previously (Santos et al., 2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]). Briefly, 1 µl purified human AAR2Δloop–PRPF8RH complex concentrated to 14 mg ml−1 in 20 mM Tris–HCl pH 7.5, 150 mM NaCl, 0.5 mM DTT was crystallized employing an equal volume of reservoir solution consisting of 0.1 M HEPES pH 7.0, 10%(w/v) PEG 6000, 5%(v/v) 2-methyl-2,4-pentanediol. Crystals were transferred to reservoir solution supplemented with 10%(v/v) 2-methyl-2,4-pentanediol and flash-cooled in liquid nitrogen. Diffraction data were collected on beamlines BL14.1, BL14.2 and BL14.3 of the BESSY II storage ring, Berlin, Germany and on beamline P14 of the PETRA III storage ring, Hamburg, Germany at 100 K and were processed with XDS (Kabsch, 2010[Kabsch, W. (2010). Acta Cryst. D66, 125-132.]). The structure was solved by molecular replacement with Phaser (McCoy et al., 2007[McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. & Read, R. J. (2007). J. Appl. Cryst. 40, 658-674.]) using chain A of the structural coordinates of human PRPF8RH (PDB entry 3e9l; Pena et al., 2008[Pena, V., Rozov, A., Fabrizio, P., Lührmann, R. & Wahl, M. C. (2008). EMBO J. 27, 2929-2940.]), omitting the water molecules. An initial model of the AAR2Δloop subunit was obtained by automated model building with phenix.autobuild (Adams et al., 2010[Adams, P. D., Afonine, P. V., Bunkóczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J. J., Hung, L.-W., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C. & Zwart, P. H. (2010). Acta Cryst. D66, 213-221.]). The model was completed through alternating rounds of automated refinement using phenix.refine (Afonine et al., 2012[Afonine, P. V., Grosse-Kunstleve, R. W., Echols, N., Headd, J. J., Moriarty, N. W., Mustyakimov, M., Terwilliger, T. C., Urzhumtsev, A., Zwart, P. H. & Adams, P. D. (2012). Acta Cryst. D68, 352-367.]) and manual model building using Coot (Emsley et al., 2010[Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486-501.]). The quality of the final model was assessed with MolProbity (Chen et al., 2010[Chen, V. B., Arendall, W. B., Headd, J. J., Keedy, D. A., Immormino, R. M., Kapral, G. J., Murray, L. W., Richardson, J. S. & Richardson, D. C. (2010). Acta Cryst. D66, 12-21.]). Of note, a relatively large number of difference density peaks were observed in the Fo − Fc map. Structural figures were prepared with PyMOL (Schrödinger).

3. Results

3.1. Crystal structure of a human AAR2Δloop–PRPF8RH complex

Structures of yeast Aar2p in complex with the Prp8p RH and JM domains or with full-length Prp8p have been reported (Weber et al., 2011[Weber, G., Cristão, V. F., Alves, F. L., Santos, K. F., Holton, N., Rappsilber, J., Beggs, J. D. & Wahl, M. C. (2011). Genes Dev. 25, 1601-1612.], 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]; Galej et al., 2013[Galej, W. P., Oubridge, C., Newman, A. J. & Nagai, K. (2013). Nature, 493, 638-643.]). In contrast, the structure of human AAR2, which exhibits only 24% sequence identity to the yeast ortholog (Supplementary Fig. S1), and the structural basis for its interaction with PRPF8 remain unknown. We previously reported the crystallization of a human AAR2 variant in which an internal loop (residues 170–200) was replaced by three serine residues (AAR2Δloop) in complex with PRPF8RH (Santos et al., 2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]). In yeast Aar2p, the corresponding internal loop was shown to hinder crystallization, to be protease-cleavable and to be irrelevant for the interaction with Prp8p C-terminal domains (Weber et al., 2011[Weber, G., Cristão, V. F., Alves, F. L., Santos, K. F., Holton, N., Rappsilber, J., Beggs, J. D. & Wahl, M. C. (2011). Genes Dev. 25, 1601-1612.], 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]; Santos et al., 2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]). Here, we report the crystal structure of the human AAR2Δloop–PRPF8RH complex at 2.35 Å resolution. The structure was solved by molecular replacement with Phaser using the structural coordinates of PRPF8RH (PDB entry 3e9l; Pena et al., 2008[Pena, V., Rozov, A., Fabrizio, P., Lührmann, R. & Wahl, M. C. (2008). EMBO J. 27, 2929-2940.]) as a molecular-replacement search model (Table 1[link]).

Table 1
Crystallographic data (Santos et al., 2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]) and refinement

Values in parentheses are for the highest resolution shell.

Data collection
 Wavelength (Å) 1.24
 Temperature (K) 100
 Space group C2
a, b, c (Å) 145.28, 57.26, 111.23
α, β, γ (°) 90, 112.74, 90
 Resolution (Å) 99.90–2.35 (2.41–2.35)
 Unique reflections 34687 (2502)
 Completeness (%) 97.6 (95.6)
 Multiplicity 3.8 (3.9)
 〈I/σ(I)〉 15.6 (1.7)
Rmerge(I) 0.05 (0.92)
 CC1/2 99.9 (76.5)
Refinement
 Resolution (Å) 48.21–2.35 (2.42–2.35)
 Total No. of reflections 34678 (2667)
 Completeness (%) 97.7 (95.7)
 Reflections in test set (%) 4.96 (4.87)
Rwork§ 0.189 (0.347)
Rfree 0.235 (0.418)
 ESU†† (Å) 0.35
 Contents of asymmetric unit
  Protein molecules/residues 2/574
  Water oxygens 147
 Mean B factors (Å2)
  Wilson 55.39
  Protein 77.81
  Solvent 65.02
 Ramachandran plot‡‡ (%)
  Favored 95.58
  Allowed 4.42
  Outliers 0
 R.m.s.d.§§ from target geometry
  Bond lengths (Å) 0.009
  Bond angles (°) 0.990
  Dihedral angles (°) 115.920
 PDB code 7pjh
Rmerge(I) = [\textstyle \sum_{hkl}\sum_{i}|I_{i}(hkl)- \langle I(hkl)\rangle|/][\textstyle \sum_{hkl}\sum_{i}I_{i}(hkl)] for i observations of a given reflection hkl; 〈I(hkl)〉 is the average intensity of the i observations.
‡CC1/2 = (〈I2〉 − 〈I2)/(〈I2〉 − 〈I2) + [\sigma^{2}_{\varepsilon}], where [\sigma^{2}_{\varepsilon}] is the mean error within a half data set (Karplus & Diederichs, 2012[Karplus, P. A. & Diederichs, K. (2012). Science, 336, 1030-1033.]).
§Rwork = [\textstyle \sum_{hkl}\big ||F_{\rm obs}|-|F_{\rm calc}|\big |/][\textstyle \sum_{hkl}|F_{\rm obs}|] (for the working set; no σ cutoff was applied).
Rfree is the same as Rwork but calculated on the test set of reflections that were excluded from refinement.
††Estimated overall coordinate error based on maximum likelihood.
‡‡Calculated with phenix.refine (Afonine et al., 2012[Afonine, P. V., Grosse-Kunstleve, R. W., Echols, N., Headd, J. J., Moriarty, N. W., Mustyakimov, M., Terwilliger, T. C., Urzhumtsev, A., Zwart, P. H. & Adams, P. D. (2012). Acta Cryst. D68, 352-367.]).
§§Root-mean-square deviation.

Apart from ten N-terminal residues, the region spanning residues 159–201 including the three serines connecting residues 170–200 of the internal loop, a loop connecting helices α9 and α10 (residues 313–321) and six C-terminal residues of AAR2Δloop (Supplementary Fig. S1), all residues of AAR2Δloop and PRPF8RH could be reliably modeled into well defined electron density (Supplementary Fig. S2). Residues 65–71 of AAR2 and residues 2001–2008 of PRPF8RH were modeled with low confidence due to weaker electron density.

Despite the low sequence identity, the overall structure of human AAR2Δloop in the AAR2Δloop–PRPF8RH complex is very similar to that of yeast Aar2pΔloop in complex with the Prp8p RH and JM domains (Weber et al., 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]; Galej et al., 2013[Galej, W. P., Oubridge, C., Newman, A. J. & Nagai, K. (2013). Nature, 493, 638-643.]; root-mean-square deviation of 2.13 Å for 236 pairs from 330 AAR2 and 342 Aar2p Cα atoms; Supplementary Fig. S3). As previously observed for Aar2p (Weber et al., 2011[Weber, G., Cristão, V. F., Alves, F. L., Santos, K. F., Holton, N., Rappsilber, J., Beggs, J. D. & Wahl, M. C. (2011). Genes Dev. 25, 1601-1612.], 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]; Galej et al., 2013[Galej, W. P., Oubridge, C., Newman, A. J. & Nagai, K. (2013). Nature, 493, 638-643.]), human AAR2Δloop exhibits an N-terminal domain (NTD; residues 10–158) mainly composed of β-strands, an α-helical C-terminal domain (CTD; residues 202–364) and a C-terminal, irregularly structured tail (residues 365–384) (Fig. 1[link]a).

[Figure 1]
Figure 1
(a) Crystal structure of the human AAR2Δloop–PRPF8RH complex. Colour scheme for this and the following figures: AAR2Δloop, orange; PRPF8RH, light blue; dashed orange lines indicate a flexible loop (labeled Ser3) in AAR2 connecting its two domains, which is replaced by three serine residues in AAR2Δloop (Santos et al., 2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]), and another smaller flexible loop between residues 313 and 321 (labeled flexible loop). The N- and C-termini as well as the β-finger module of PRPF8RH are labeled. (b) Superposition of the RH domains of human PRPF8 and yeast Prp8p in complex with human AAR2Δloop and yeast Aar2p/PRPF8JM (PDB entry 4i43; Galej et al., 2013[Galej, W. P., Oubridge, C., Newman, A. J. & Nagai, K. (2013). Nature, 493, 638-643.]), respectively, to illustrate the human AAR2 in a larger PRPF8 context. Colour scheme for this and the following figures: Aar2p, maroon; Prp8pRH, dark blue; Prp8pJM, cyan. (c) Comparison of the human AAR2Δloop–PRPF8RH complex and the yeast Aar2pΔloop–Prp8pRH–Prp8pJM complex (PDB entry 4ilg; Weber et al., 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]) by superposition of the RH domains. Dashed maroon line, flexible linker preceding the C-terminal tail of Aar2p (Weber et al., 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]). (d, e) Close-up views of interface regions I of the yeast Aar2pΔloop–Prp8pRH–Prp8pJM complex (d) and the human AAR2Δloop–PRPF8RH complex (e). (f, g) Close-up views of interface regions II of the yeast Aar2pΔloop–Prp8pRH–Prp8pJM complex (f) and the human AAR2Δloop–PRPF8RH complex (g). In (df) and the following figures interacting residues are shown as sticks colored by atom type, with carbon colored as the respective protein, nitrogen in blue, oxygen in red and sulfur in yellow; green spheres are water molecules, dashed black lines represent hydrogen bonds or salt bridges and rotation symbols represent orientations relative to (a).

In the full-length yeast Aar2p–Prp8p structure (PDB entry 4i43; Galej et al., 2013[Galej, W. P., Oubridge, C., Newman, A. J. & Nagai, K. (2013). Nature, 493, 638-643.]), the C-terminal peptide of Aar2p is fully structured and contacts several other Prp8p domains. Superposition with the present human AAR2–RH complex structure suggests that due to the shorter AAR2 C-terminal peptide in humans, contacts with other PRPF8 domains may be limited. Hence, the differences in the functionally important C-terminal peptide of yeast and human Aar2p/AAR2 may hint at a somewhat different mode of action of AAR2 in U5 snRNP or U4/U6-U5 tri-snRNP assembly in humans.

Despite the overall structural similarity of both individual components, the protein interfaces between yeast and human Prp8pRH/PRPF8RH and Aar2p/AAR2 are markedly different. As in yeast, the NTD lacks direct interactions with PRPF8RH, while the CTD and the C-terminal tail of AAR2 establish two interfaces with PRPF8RH (interfaces I and II, respectively; Figs. 1[link]a–1[link]c). In interface I, an edge of the AAR2Δloop CTD laterally contacts PRPF8RH (Figs. 1[link]d and 1[link]e). Interface II is built by the C-terminal residues 366–377 of AAR2 extending across the PRPF8 RH domain below the protruding β-finger module (Figs. 1[link]f and 1[link]g). Both interfaces bury a comparable surface area in the yeast and human systems (interface I, 399 and 412 Å2, respectively; interface II, 733 and 511 Å2, respectively).

Interface I is dominated by hydrophobic contacts, with only four of 12 PRPF8RH-interacting residues conserved between yeast Aar2p and human AAR2, underlining the different organization of the interactions. The conserved core of interface I includes interactions between Ile225 of AAR2 (Ile189 in Aar2p) and Val1874 of PRPF8RH (Val1946 in Aar2p) as well as between Met230 of AAR2 (Met195 in Aar2p) and Trp1839 of PRPF8RH (Trp1911 in Aar2p) (Figs. 1[link]d and 1[link]e; Supplementary Fig. S1). Compared with yeast Aar2p, the AAR2 CTD harbors two extended helices (α11 and α12; Supplementary Fig. S3). In addition, Ile225 and Met230 are shifted by four residues (about one helical turn) along the α5 helix compared with the equivalent residues in yeast Aar2p (Figs. 1[link]d and 1[link]e), giving rise to a markedly different angle with which human AAR2Δloop contacts the PRPF8 RH domain compared with yeast Aar2pΔloop in the Aar2pΔloop–Prp8pRH–Prp8pJM complex (Fig. 1[link]c). Also, the AAR2Δloop residues participating in interface II are only partially conserved between yeast and humans (two of eight residues; Supplementary Fig. S1).

3.2. Similarities and differences in AAR2–PRPF8–SNRNP200 and Aar2p–Prp8p–Brr2p interactions in humans and yeast

The low degree of conservation of AAR2 and observed marked differences in the interface with PRPF8RH have apparent consequences for AAR2 function and likely for interactions within the spliceosome. To test the importance of the specific contacts between AAR2Δloop and PRPF8RH that are observed in our co-crystal structure, we conducted analytical SEC runs with wild-type (WT) proteins and variants. To this end, we investigated the binding of WT AAR2 to WT PRPF8RH in previous work, which is only shown here for comparison (Figs. 2[link]a–2[link]c; Santos et al., 2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]). In yeast, the C-terminal tail of Aar2p is dispensable for Prp8RH binding (Weber et al., 2011[Weber, G., Cristão, V. F., Alves, F. L., Santos, K. F., Holton, N., Rappsilber, J., Beggs, J. D. & Wahl, M. C. (2011). Genes Dev. 25, 1601-1612.]). Conversely, in the human system, AAR21–364, which lacks the C-terminal tail, no longer binds stably to PRPF8RH (Figs. 2[link]a–2[link]d). Likewise, converting Trp1839 of PRPF8RH or Met230 of AAR2, which are part of the conserved core of interface I, individually to alanine residues abrogated complex formation (Fig. 2[link]e and 2[link]f). Again, the situation differs in yeast, where only Trp1911 of Prp8pRH (equivalent to Trp1839 in human PRPF8RH), but not Met195 of Aar2p (equivalent to Met230 in human AAR2), is essential for the interaction (Weber et al., 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]).

[Figure 2]
Figure 2
Probing interacting regions and residues in AAR2Δloop–PRPF8RH. (ah) SDS–PAGE analyses (left) and UV elution profiles (right) of analytical size-exclusion chromatography runs monitoring the interactions among AAR2 variants, PRPF8RH variants and PRPF8JM. (a)–(c) were adapted from Santos et al. (2015[Santos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421-1428.]) and are shown for comparison. Lane M, molecular-mass standard (kDa); lane I, input samples. Protein bands are identified on the right. Elution fractions are indicated at the top of the gels and profiles; elution volumes are indicated at the bottom of the profiles. Icons are explained at the bottom. Variants are indicated below the respective icons. Peaks labeled by transparent icons represent an excess of the respective protein.

The low sequence conservation and the resulting structural differences in the AAR2–PRP8RH interface might also have consequences for the wider protein interaction network around AAR2. Concomitant binding of the Prp8p RH and JM domains by Aar2p in yeast sequesters the JM domain, preventing binding of the Brr2p RNA helicase to Aar2p–pre-U5 snRNP (Weber et al., 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]; Galej et al., 2013[Galej, W. P., Oubridge, C., Newman, A. J. & Nagai, K. (2013). Nature, 493, 638-643.]). In yeast Aar2p–Prp8p complexes (Weber et al., 2011[Weber, G., Cristão, V. F., Alves, F. L., Santos, K. F., Holton, N., Rappsilber, J., Beggs, J. D. & Wahl, M. C. (2011). Genes Dev. 25, 1601-1612.], 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]; Galej et al., 2013[Galej, W. P., Oubridge, C., Newman, A. J. & Nagai, K. (2013). Nature, 493, 638-643.]), the C-terminal tail of Aar2p runs along the protruding Prp8pRH β-finger module, stringing the β-finger and the central Prp8pJM β-sheet into an extended, intermolecular β-structure (Figs. 1[link]b and 1[link]c).

While the beginning of the C-terminal tail in human AAR2Δloop maintains similar interactions with PRPF8RH as in yeast, for example employing Val373–Val375 to form a short β-sheet of three hydrogen bonds to PRPF8RH, distal parts of the C-terminal tail (beyond Val374) deviate from the direction of the Aar2p C-terminal tail (Figs. 1[link]b and 1[link]c). In yeast, the formation of the penultimate β-strand of Aar2p and the concomitant sequestration of JM from Brr2p is mediated exclusively by a series of hydrophobic residues at the very C-terminus of Aar2p, which are complementary to hydrophobic residues of the neighboring β-strands of RH and JM (Weber et al., 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]; Galej et al., 2013[Galej, W. P., Oubridge, C., Newman, A. J. & Nagai, K. (2013). Nature, 493, 638-643.]). A structure-based alignment revealed that the respective very C-terminal residues of AAR2, Pro378, Glu379, Gly380 and Glu382, are unlikely to support β-sheet formation with the corresponding highly conserved residues of the PRPF8RH β-finger and PRPF8JM due to their steric or polar properties (Supplementary Fig. S1; compare Figs. 1[link]f and 1[link]g). However, we cannot exclude that in the context of the full-length proteins the very C-terminus of hAAR2 may engage in a yeast-like interaction with the PRPF8 JM domain.

The C-terminal tail of human AAR2Δloop in the observed conformation would not be able to concomitantly bind the PRPF8JM domain as observed in yeast. Indeed, also confirming a prior study (Malinová et al., 2017[Malinová, A., Cvačková, Z., Materějů, D., Hořejší, Z., Abéza, C., Vandermoere, F., Bertrand, E., Staněk, D. & Verheggen, C. (2017). J. Cell Biol. 216, 1579-1596.]), AAR2–PRPF8RH did not stably bind PRPF8JM in analytical SEC (Figs. 2[link]g and 2[link]h) and failed to sequester PRPF8JM from a pre-formed SNRNP200395–2136–PRPF8JM complex (Fig. 3[link]a and 3[link]b).

[Figure 3]
Figure 3
Probing AAR2Δloop–PRPF8–SNRNP200 interactions and AAR2 phosphorylation. (ad) SDS–PAGE analyses (left) and UV elution profiles (right) of analytical size-exclusion chromatography runs monitoring the interactions among AAR2, PRPF8RH-JM and SNRNP200395–2136 (a, b) and among AAR2, PRPF8RH, PRPF8JM and SNRNP200395–2136 (c, d). (e) Close-up view of the region in AAR2Δloop–PRPF8RH surrounding Ser284 of AAR2Δloop. The corresponding region in yeast Aar2p is profoundly restructured upon replacement of the equivalent Ser253 by a phosphomimetic glutamate residue (Weber et al., 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]). (f) SDS–PAGE analysis (top) and UV elution profile (bottom) of an analytical size-exclusion chromatography run monitoring the interaction between AAR2S284E and PRPF8RH. In panels showing SDS–PAGE gels and elution profiles lane M contains molecular-mass standard (kDa) and lane I contains input samples. Protein bands are identified on the right. Elution fractions are indicated at the top of the gels and profiles; elution volumes are indicated at the bottom of the profiles. Icons are explained at the bottom. Variants are indicated below the respective icons. Peaks labeled by transparent icons represent an excess of the respective protein.

AAR2 alone or in complex with PRPF8RH did not bind stably to SNRNP200395–2136 or to a SNRNP200395–2136–PRPF8JM complex (Figs. 3[link]c and 3[link]d). Instead, a stable AAR2–PRPF8RM-JM–SNRNP200395–2136 ternary complex was formed upon mixing the components (Fig. 3[link]b).

3.3. Conserved Aar2 phosphorylation between humans and yeast

Aar2p can be phosphorylated at five positions in vivo (Ser253, Thr274, Tyr328, Ser331 and Thr345) and phosphomimetic S253D or S253E variants of Aar2p interfered with Aar2p–Prp8p interaction in yeast extracts (Weber et al., 2011[Weber, G., Cristão, V. F., Alves, F. L., Santos, K. F., Holton, N., Rappsilber, J., Beggs, J. D. & Wahl, M. C. (2011). Genes Dev. 25, 1601-1612.]). Structural analysis of a phosphomimetic Aar2pS253E variant suggested that phosphorylation leads to a local conformational rearrangement of the Aar2p CTD and thereby to disruption of the Prp8pRH binding site (Weber et al., 2011[Weber, G., Cristão, V. F., Alves, F. L., Santos, K. F., Holton, N., Rappsilber, J., Beggs, J. D. & Wahl, M. C. (2011). Genes Dev. 25, 1601-1612.]). A structure-based sequence alignment revealed that Ser284 in human AAR2 corresponds to Ser253 in yeast Aar2p (Supplementary Fig. S1; Fig. 3[link]e), and AAR2 has been found to be phosphorylated at Ser284 in human liver cancer cells (Hornbeck et al., 2012[Hornbeck, P. V., Kornhauser, J. M., Tkachev, S., Zhang, B., Skrzypek, E., Murray, B., Latham, V. & Sullivan, M. (2012). Nucleic Acids Res. 40, D261-D270.]). Recapitulating the situation in yeast, an AAR2S284E phosphomimetic variant failed to stably bind PRPF8RH in analytical SEC (Fig. 3[link]f). Taken together, our interaction studies reveal differences in the relative importance of AAR2/Aar2p regions in maintaining a stable interaction with the PRPF8/Prp8p RH domain in the human and yeast systems. Furthermore, AAR2 does not sequester the PRPF8 JM domain to intermittently prevent SNRNP200 association with the U5 snRNP. AAR2 displacement from PRPF8 may involve reversible phosphorylation of AAR2 at Ser284. Thus, the U5 snRNP assembly steps apparently differ in detail in yeast and humans.

3.4. Human AAR2 counteracts a step 2-like conformation in the PRPF8 RH domain

Mutations in the prpf8 gene can lead to retinitis pigmentosa (RP; Růžičková & Staněk, 2017[Růžičková, Š. & Staněk, D. (2017). RNA Biol. 14, 544-552.]), a disease that causes blindness in humans, and the corresponding PRPF8/Prp8p variants cause defects in U5 snRNP assembly (Malinová et al., 2017[Malinová, A., Cvačková, Z., Materějů, D., Hořejší, Z., Abéza, C., Vandermoere, F., Bertrand, E., Staněk, D. & Verheggen, C. (2017). J. Cell Biol. 216, 1579-1596.]) and splicing (Mayerle & Guthrie, 2016[Mayerle, M. & Guthrie, C. (2016). RNA, 22, 793-809.]; Mozaffari-Jovin et al., 2013[Mozaffari-Jovin, S., Wandersleben, T., Santos, K. F., Will, C. L., Lührmann, R. & Wahl, M. C. (2013). Science, 341, 80-84.]) in humans and yeast. In baker's yeast, two sets of prp8 mutant alleles, corresponding to RP-related mutations in humans that disrupt either the first or the second step of splicing, cluster in the Prp8p RH domain (Grainger & Beggs, 2005[Grainger, R. J. & Beggs, J. D. (2005). RNA, 11, 533-557.]).

Furthermore, the human PRPF8 RH domain can undergo a conformational switch in a protruding β-finger module, with one conformation promoting the first step and an alternative, Mg2+-bound conformation supporting the second step of splicing (Schellenberg et al., 2013[Schellenberg, M. J., Wu, T., Ritchie, D. B., Fica, S., Staley, J. P., Atta, K., LaPointe, P. & MacMillan, A. M. (2013). Nat. Struct. Mol. Biol. 20, 728-734.]). Despite the biochemical and structural evidence reported previously, which supports this switch, a caveat of our AAR2–PRPF8RH structure may be that the RH β-finger module makes crystal contacts with a neighboring symmetry-related RH β-finger module. However, recent cryogenic electron-microscopy structures of spliceosomes also confirm this conformational switch, rationalize some of the effects of PRPF8 RH domain variants and demonstrate repeated, long-range repositioning of the PRPF8/Prp8p RH domain during the splicing reaction in yeast and humans (Wan et al., 2016[Wan, R., Yan, C., Bai, R., Huang, G. & Shi, Y. (2016). Science, 353, 895-904.]; Bertram, Agafonov, Liu et al., 2017[Bertram, K., Agafonov, D. E., Liu, W.-T., Dybkov, O., Will, C. L., Hartmuth, K., Urlaub, H., Kastner, B., Stark, H. & Lührmann, R. (2017). Nature, 542, 318-323.]; Yan et al., 2015[Yan, C., Hang, J., Wan, R., Huang, M., Wong, C. C. L. & Shi, Y. (2015). Science, 349, 1182-1191.], 2017[Yan, C., Wan, R., Bai, R., Huang, G. & Shi, Y. (2017). Science, 355, 149-155.]; Zhang et al., 2017[Zhang, X., Yan, C., Hang, J., Finci, L. I., Lei, J. & Shi, Y. (2017). Cell, 169, 918-929.]; Bertram, Agafonov, Dybkov et al., 2017[Bertram, K., Agafonov, D. E., Dybkov, O., Haselbach, D., Leelaram, M. N., Will, C. L., Urlaub, H., Kastner, B., Lührmann, R. & Stark, H. (2017). Cell, 170, 701-713.]; Rauhut et al., 2016[Rauhut, R., Fabrizio, P., Dybkov, O., Hartmuth, K., Pena, V., Chari, A., Kumar, V., Lee, C.-T., Urlaub, H., Kastner, B., Stark, H. & Lührmann, R. (2016). Science, 353, 1399-1405.]; Galej et al., 2016[Galej, W. P., Wilkinson, M. E., Fica, S. M., Oubridge, C., Newman, A. J. & Nagai, K. (2016). Nature, 537, 197-201.]; Plaschka et al., 2017[Plaschka, C., Lin, P.-C. & Nagai, K. (2017). Nature, 546, 617-621.]; Fica et al., 2017[Fica, S. M., Oubridge, C., Galej, W. P., Wilkinson, M. E., Bai, X.-C., Newman, A. J. & Nagai, K. (2017). Nature, 542, 377-380.]; Wilkinson et al., 2021[Wilkinson, M. E., Fica, S. M., Galej, W. P. & Nagai, K. (2021). Mol. Cell, 81, 1439-1452.]).

Comparison of our AAR2Δloop–PRPF8RH structure with the PRPF8RH step 1 and step 2 conformations revealed that AAR2 binding is compatible with the PRPF8RH step 1 conformation but that steric clashes ensue between the AAR2 C-terminal tail and the PRPF8 RH domain in the Mg2+-bound step 2 conformation (Figs. 4[link]a–4[link]d). To test whether AAR2 is likely to prevent a switch of PRPF8RH into the step 2 conformation, we explored whether AAR2 binds a PRPF8RH variant that is stabilized in the step 2 conformation (T1789P; Schellenberg et al., 2013[Schellenberg, M. J., Wu, T., Ritchie, D. B., Fica, S., Staley, J. P., Atta, K., LaPointe, P. & MacMillan, A. M. (2013). Nat. Struct. Mol. Biol. 20, 728-734.]). Indeed, unlike WT PRPF8RH, PRPF8RH,T1789P partly dissociated from AAR2 upon increasing the Mg2+ concentration in analytical SEC (Figs. 4[link]e–4[link]h), suggesting that a step 2 conformation in PRPF8RH is incompatible with AAR2 binding.

[Figure 4]
Figure 4
AAR2-mediated blocking of a step 2 conformation in PRPF8RH. (ad) Structure of the AAR2Δloop–PRPF8RH complex (a) and close-up views comparing the AAR2Δloop C-terminal tail (sticks) traversing the PRPF8 RH domain below the protruding β-finger module (surface views) as observed in the AAR2Δloop–PRPF8RH complex (b) or modeled onto the PRPF8RH domain in the step 1 conformation (PDB entry 4jk7; Schellenberg et al., 2013[Schellenberg, M. J., Wu, T., Ritchie, D. B., Fica, S., Staley, J. P., Atta, K., LaPointe, P. & MacMillan, A. M. (2013). Nat. Struct. Mol. Biol. 20, 728-734.]) (c) or onto the PRPF8RH domain in the step 2 conformation (PDB entry 4jk7; Schellenberg et al., 2013[Schellenberg, M. J., Wu, T., Ritchie, D. B., Fica, S., Staley, J. P., Atta, K., LaPointe, P. & MacMillan, A. M. (2013). Nat. Struct. Mol. Biol. 20, 728-734.]) (d) by superposition of the RH domains. Yellow sphere, coordinated Mg2+ ion. The AAR2 C-terminus clashes with the PRPF8RH domain in the step 2 conformation. (eh) SDS–PAGE analyses (left) and UV elution profiles (right) of analytical size-exclusion chromatography runs monitoring the interaction between AAR2 and PRPF8RH (e, g) or PRPF8RH,T1789P (f, h) at 2 mM (e, f) or 40 mM (g, h) magnesium chloride. Lane M, molecular-mass standard (kDa); lane I, input samples. Protein bands are identified on the right. Elution fractions are indicated at the top of the gel and profile in (e); elution volumes are indicated at the bottom of the profile in (h). Icons are explained at the bottom left. Variants are indicated below the respective icons. Peaks labeled with transparent icons represent an excess of the respective protein.

4. Discussion

We have elucidated similarities and differences in the structures and interaction profiles of yeast Aar2p and human AAR2 and have identified a putative, conserved phosphorylation event that is most likely to be involved in the functional cycle of AAR2 as a U5 snRNP assembly factor. Based on our findings, we conclude that the precise roles of Aar2p and AAR2 in U5 snRNP biogenesis differ. In yeast, an Aar2p–pre-U5 snRNP, from which the Brr2p RNA helicase is excluded, seems to constitute an important U5 snRNP assembly intermediate (Boon et al., 2007[Boon, K.-L., Grainger, R. J., Ehsani, P., Barrass, J. D., Auchynnikava, T., Inglehearn, C. F. & Beggs, J. D. (2007). Nat. Struct. Mol. Biol. 14, 1077-1083.]; Weber et al., 2013[Weber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525-540.]). In contrast, our observations of (i) human AAR2 failing to sequester the PRPF8 JM domain from SNRNP200395–2136 and (ii) AAR2 concomitantly binding to a PRPF8 fragment encompassing the RH and JM domains and SNRNP200395–2136 suggest that an equivalent, long-lived intermediate is not formed in the human system. Association of AAR2 with the PRPF8 RH domain as in our AAR2Δloop–PRPF8RH structure would prevent the PRPF8 RH domain from engaging with other regions of PRPF8, the N-terminal region of SNRNP200, the C-terminal region of PRPF31, PRPF6, U4/U6 di-snRNAs and U5 snRNA as observed in the human U4/U6-U5 tri-snRNP (Agafonov et al., 2016[Agafonov, D. E., Kastner, B., Dybkov, O., Hofele, R. V., Liu, W.-T., Urlaub, H., Lührmann, R. & Stark, H. (2016). Science, 351, 1416-1420.]; Charenton et al., 2019[Charenton, C., Wilkinson, M. E. & Nagai, K. (2019). Science, 364, 362-367.]). This finding suggests that prevention of the premature association of U4/U6 di-snRNP components with pre-U5 particles may be an important function of AAR2 in the human system. In addition, transient blocking of binding sites on the PRPF8 RH domain, possibly supported by allosteric effects due to the selective stabilization of a step 1-like conformation in the PRPF8 RH domain by AAR2, may help to order assembly steps during U5 snRNP biogenesis. The above findings and suggestions are in agreement with the previous observation of the interaction of human AAR2 with a PRPF8–EFTUD2–SNRNP200–SNRNP40 U5 submodule (Malinová et al., 2017[Malinová, A., Cvačková, Z., Materějů, D., Hořejší, Z., Abéza, C., Vandermoere, F., Bertrand, E., Staněk, D. & Verheggen, C. (2017). J. Cell Biol. 216, 1579-1596.]).

As most protein-coding genes in humans contain multiple introns (Lee & Rio, 2015[Lee, Y. & Rio, D. C. (2015). Annu. Rev. Biochem. 84, 291-323.]), pre-mRNA splicing is an inherent step in their expression. Moreover, pre-mRNA splicing predominantly occurs co-transcriptionally (Alpert et al., 2017[Alpert, T., Herzel, L. & Neugebauer, K. M. (2017). WIREs RNA, 8, e1401.]) and splicing is physically and functionally coupled to transcription, other pre-mRNA processing steps and mRNA export (Carrocci & Neugebauer, 2019[Carrocci, T. J. & Neugebauer, K. M. (2019). Cold Spring Harb. Symp. Quant. Biol. 84, 11-20.]; Tellier et al., 2020[Tellier, M., Maudlin, I. & Murphy, S. (2020). WIREs RNA, 11, e1593.]). Thus, efficient splicing is a prerequisite for efficient gene expression and, due to its stabiliziation of the step 1 configuration of RH, a potential role of human AAR2 in pre-mRNA splicing cannot be ruled out. AAR2 may have a moonlighting function during pre-mRNA splicing independent of its role as a U5 snRNP assembly factor. By binding the PRPF8 RH domain during a stage of splicing when it is available, for example, in the pre-catalytic B complex (PDB entry 7abg; Townsend et al., 2020[Townsend, C., Leelaram, M. N., Agafonov, D. E., Dybkov, O., Will, C. L., Bertram, K., Urlaub, H., Kastner, B., Stark, H. & Lührmann, R. (2020). Science, 370, eabc3753.]), AAR2 may hinder the transition to subsequent stages, thus impeding splicing and, as a consequence, gene expression. As in the case of U5 snRNP assembly, direct blocking of binding sites on PRPF8RH and allosteric effects due to the stabilization of a step 1 conformation in PRPF8RH may support such a splicing-inhibitory role of AAR2. Again, the observed high nuclear levels of AAR2 might ensure that sufficient AAR2 is available to serve multiple functions, as moonlighting is known for some splicing factors that are in excess over other splicing machinery. For example, U1 snRNP has additional roles in 3′-end processing of Pol II transcripts (telescripting; Di et al., 2019[Di, C., So, B. R., Cai, Z., Arai, C., Duan, J. & Dreyfuss, G. (2019). Cold Spring Harb. Symp. Quant. Biol. 84, 115-122.]). However, AAR2 has never been found to be associated with the spliceosome at any stage of splicing (Agafonov et al., 2011[Agafonov, D. E., Deckert, J., Wolf, E., Odenwälder, P., Bessonov, S., Will, C. L., Urlaub, H. & Lührmann, R. (2011). Mol. Cell. Biol. 31, 2667-2682.]), arguing against a direct effect of AAR2 on splicing. Further studies on human Aar2 in a spliceosomal context will hopefully resolve these remaining questions.

5. Data availability

Structure coordinates and diffraction data have been deposited in the Protein Data Bank (https://www.pdb.org) under accession code 7pjh. All other data supporting the findings of this study are described in the manuscript or in the supporting information or are available from the corresponding authors on request.

6. Related literature

The following references are cited in the supporting information for this article: Barton (1993[Barton, G. J. (1993). Protein Eng. Des. Sel. 6, 37-40.]), Kabsch & Sander (1983[Kabsch, W. & Sander, C. (1983). Biopolymers, 22, 2577-2637.]) and Pettersen et al. (2004[Pettersen, E. F., Goddard, T. D., Huang, C. C., Couch, G. S., Greenblatt, D. M., Meng, E. C. & Ferrin, T. E. (2004). J. Comput. Chem. 25, 1605-1612. ]).

Supporting information


Footnotes

Present address: 4TEEN4 Pharmaceuticals GmbH, Neuendorfstrasse 15A, 16761 Hennigsdorf, Germany.

§Present address: Berlin Institute for Medical Systems Biology, Hannoversche Strasse 28, 10115 Berlin, Germany.

Present address: MRC Laboratory of Molecular Biology, Francis Crick Avenue, Cambridge, United Kingdom.

Acknowledgements

We are grateful to Reinhard Lührmann, Max-Planck-Institut für Biophysikalische Chemie, Göttingen, Germany for the gift of rabbit αAAR2 polyclonal antibody. We acknowledge access to beamlines BL14.1, BL14.2 and BL14.3 of the BESSY II storage ring, Berlin, Germany via the Joint Berlin MX-Laboratory sponsored by the Helmholtz Zentrum Berlin für Materialien und Energie, the Freie Universität Berlin, the Humboldt-Universität zu Berlin, the Max-Delbrück Centrum, the Leibniz-Forschungsinstitut für Molekulare Pharma­kologie, Charité – Universitätsmedizin Berlin and the Max-Planck-Institut für Kolloid- und Grenzflächenforschung, Potsdam, Germany and to beamline P14 of the PETRA III storage ring, Deutsches Elektronen-Synchrotron, Hamburg, Germany. Author contributions are as follows. KFS, JA, CH and GW conducted molecular cloning, expression, protein purification and interaction studies. KFS, CH and GW conducted crystallographic analyses. All authors participated in data interpretation. MP, MCW, FH and GW wrote the manuscript, which was revised and approved by all authors. MP and GW initiated the study and directed the project. MCW and FH provided funding for the project. The authors declare no competing interests. Open access funding enabled and organized by Projekt DEAL.

Funding information

This work was funded by a grant from the Deutsche Forschungsgemeinschaft (TRR186/A15 to FH and MCW).

References

First citationAdams, P. D., Afonine, P. V., Bunkóczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J. J., Hung, L.-W., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C. & Zwart, P. H. (2010). Acta Cryst. D66, 213–221.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationAfonine, P. V., Grosse-Kunstleve, R. W., Echols, N., Headd, J. J., Moriarty, N. W., Mustyakimov, M., Terwilliger, T. C., Urzhumtsev, A., Zwart, P. H. & Adams, P. D. (2012). Acta Cryst. D68, 352–367.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationAgafonov, D. E., Deckert, J., Wolf, E., Odenwälder, P., Bessonov, S., Will, C. L., Urlaub, H. & Lührmann, R. (2011). Mol. Cell. Biol. 31, 2667–2682.  Web of Science CrossRef CAS PubMed Google Scholar
First citationAgafonov, D. E., Kastner, B., Dybkov, O., Hofele, R. V., Liu, W.-T., Urlaub, H., Lührmann, R. & Stark, H. (2016). Science, 351, 1416–1420.  CrossRef CAS PubMed Google Scholar
First citationAlpert, T., Herzel, L. & Neugebauer, K. M. (2017). WIREs RNA, 8, e1401.  Google Scholar
First citationBarton, G. J. (1993). Protein Eng. Des. Sel. 6, 37–40.  CrossRef CAS Web of Science Google Scholar
First citationBergfort, A., Hilal, T., Kuropka, B., Ilik, İ. A., Weber, G., Aktaş, T., Freund, C. & Wahl, M. C. (2022). Nucleic Acids Res. 50, 2938–2958.  CrossRef CAS PubMed Google Scholar
First citationBertram, K., Agafonov, D. E., Dybkov, O., Haselbach, D., Leelaram, M. N., Will, C. L., Urlaub, H., Kastner, B., Lührmann, R. & Stark, H. (2017). Cell, 170, 701–713.  CrossRef CAS PubMed Google Scholar
First citationBertram, K., Agafonov, D. E., Liu, W.-T., Dybkov, O., Will, C. L., Hartmuth, K., Urlaub, H., Kastner, B., Stark, H. & Lührmann, R. (2017). Nature, 542, 318–323.  CrossRef CAS PubMed Google Scholar
First citationBizarro, J., Dodré, M., Huttin, A., Charpentier, B., Schlotter, F., Branlant, C., Verheggen, C., Massenet, S. & Bertrand, E. (2015). Nucleic Acids Res. 43, 8973–8989.  CrossRef CAS PubMed Google Scholar
First citationBoon, K.-L., Grainger, R. J., Ehsani, P., Barrass, J. D., Auchynnikava, T., Inglehearn, C. F. & Beggs, J. D. (2007). Nat. Struct. Mol. Biol. 14, 1077–1083.  CrossRef PubMed CAS Google Scholar
First citationCarrocci, T. J. & Neugebauer, K. M. (2019). Cold Spring Harb. Symp. Quant. Biol. 84, 11–20.  CrossRef PubMed Google Scholar
First citationCharenton, C., Wilkinson, M. E. & Nagai, K. (2019). Science, 364, 362–367.  Web of Science CrossRef CAS PubMed Google Scholar
First citationChen, V. B., Arendall, W. B., Headd, J. J., Keedy, D. A., Immormino, R. M., Kapral, G. J., Murray, L. W., Richardson, J. S. & Richardson, D. C. (2010). Acta Cryst. D66, 12–21.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationDi, C., So, B. R., Cai, Z., Arai, C., Duan, J. & Dreyfuss, G. (2019). Cold Spring Harb. Symp. Quant. Biol. 84, 115–122.  CrossRef PubMed Google Scholar
First citationEmsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486–501.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationErkelenz, S., Stanković, D., Mundorf, J., Bresser, T., Claudius, A. K., Boehm, V., Gehring, N. H. & Uhlirova, M. (2021). Nucleic Acids Res. 49, 1688–1707.  CrossRef CAS PubMed Google Scholar
First citationFica, S. M., Oubridge, C., Galej, W. P., Wilkinson, M. E., Bai, X.-C., Newman, A. J. & Nagai, K. (2017). Nature, 542, 377–380.  CrossRef CAS PubMed Google Scholar
First citationGalej, W. P., Oubridge, C., Newman, A. J. & Nagai, K. (2013). Nature, 493, 638–643.  CrossRef CAS PubMed Google Scholar
First citationGalej, W. P., Wilkinson, M. E., Fica, S. M., Oubridge, C., Newman, A. J. & Nagai, K. (2016). Nature, 537, 197–201.  Web of Science CrossRef CAS PubMed Google Scholar
First citationGottschalk, A., Kastner, B., Lührmann, R. & Fabrizio, P. (2001). RNA, 7, 1554–1565.  Web of Science PubMed CAS Google Scholar
First citationGrainger, R. J. & Beggs, J. D. (2005). RNA, 11, 533–557.  CrossRef PubMed CAS Google Scholar
First citationGruss, O. J., Meduri, R., Schilling, M. & Fischer, U. (2017). Chromosoma, 126, 577–593.  CrossRef CAS PubMed Google Scholar
First citationHornbeck, P. V., Kornhauser, J. M., Tkachev, S., Zhang, B., Skrzypek, E., Murray, B., Latham, V. & Sullivan, M. (2012). Nucleic Acids Res. 40, D261–D270.  Web of Science CrossRef CAS PubMed Google Scholar
First citationKabsch, W. (2010). Acta Cryst. D66, 125–132.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationKabsch, W. & Sander, C. (1983). Biopolymers, 22, 2577–2637.  CrossRef CAS PubMed Web of Science Google Scholar
First citationKarplus, P. A. & Diederichs, K. (2012). Science, 336, 1030–1033.  Web of Science CrossRef CAS PubMed Google Scholar
First citationKlimešová, K., Vojáčková, J., Radivojević, N., Vandermoere, F., Bertrand, E., Verheggen, C. & Staněk, D. (2021). Nat. Commun. 12, 3646.  PubMed Google Scholar
First citationLaggerbauer, B., Achsel, T. & Lührmann, R. (1998). Proc. Natl Acad. Sci. USA, 95, 4188–4192.  Web of Science CrossRef CAS PubMed Google Scholar
First citationLee, Y. & Rio, D. C. (2015). Annu. Rev. Biochem. 84, 291–323.  CrossRef CAS PubMed Google Scholar
First citationMakarov, E. M., Makarova, O. V., Urlaub, H., Gentzel, M., Will, C. L., Wilm, M. & Lührmann, R. (2002). Science, 298, 2205–2208.  Web of Science CrossRef PubMed CAS Google Scholar
First citationMalinová, A., Cvačková, Z., Materějů, D., Hořejší, Z., Abéza, C., Vandermoere, F., Bertrand, E., Staněk, D. & Verheggen, C. (2017). J. Cell Biol. 216, 1579–1596.  PubMed Google Scholar
First citationMatera, A. G. & Wang, Z. (2014). Nat. Rev. Mol. Cell Biol. 15, 108–121.  Web of Science CrossRef CAS PubMed Google Scholar
First citationMayerle, M. & Guthrie, C. (2016). RNA, 22, 793–809.  CrossRef CAS PubMed Google Scholar
First citationMcCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. & Read, R. J. (2007). J. Appl. Cryst. 40, 658–674.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationMozaffari-Jovin, S., Wandersleben, T., Santos, K. F., Will, C. L., Lührmann, R. & Wahl, M. C. (2013). Science, 341, 80–84.  Web of Science CAS PubMed Google Scholar
First citationPena, V., Liu, S., Bujnicki, J. M., Lührmann, R. & Wahl, M. C. (2007). Mol. Cell, 25, 615–624.  Web of Science CrossRef PubMed CAS Google Scholar
First citationPena, V., Rozov, A., Fabrizio, P., Lührmann, R. & Wahl, M. C. (2008). EMBO J. 27, 2929–2940.  Web of Science CrossRef PubMed CAS Google Scholar
First citationPettersen, E. F., Goddard, T. D., Huang, C. C., Couch, G. S., Greenblatt, D. M., Meng, E. C. & Ferrin, T. E. (2004). J. Comput. Chem. 25, 1605–1612.   Web of Science CrossRef PubMed CAS Google Scholar
First citationPlaschka, C., Lin, P.-C. & Nagai, K. (2017). Nature, 546, 617–621.  CrossRef CAS PubMed Google Scholar
First citationRaghunathan, P. L. & Guthrie, C. (1998). Curr. Biol. 8, 847–855.  Web of Science CrossRef CAS PubMed Google Scholar
First citationRauhut, R., Fabrizio, P., Dybkov, O., Hartmuth, K., Pena, V., Chari, A., Kumar, V., Lee, C.-T., Urlaub, H., Kastner, B., Stark, H. & Lührmann, R. (2016). Science, 353, 1399–1405.  Web of Science CrossRef CAS PubMed Google Scholar
First citationRitchie, D. B., Schellenberg, M. J., Gesner, E. M., Raithatha, S. A., Stuart, D. T. & MacMillan, A. M. (2008). Nat. Struct. Mol. Biol. 15, 1199–1205.  CrossRef PubMed CAS Google Scholar
First citationRůžičková, Š. & Staněk, D. (2017). RNA Biol. 14, 544–552.  PubMed Google Scholar
First citationSantos, K., Preussner, M., Heroven, A. C. & Weber, G. (2015). Acta Cryst. F71, 1421–1428.  CrossRef IUCr Journals Google Scholar
First citationSantos, K. F., Jovin, S. M., Weber, G., Pena, V., Lührmann, R. & Wahl, M. C. (2012). Proc. Natl Acad. Sci. USA, 109, 17418–17423.  Web of Science CrossRef CAS PubMed Google Scholar
First citationSchellenberg, M. J., Wu, T., Ritchie, D. B., Fica, S., Staley, J. P., Atta, K., LaPointe, P. & MacMillan, A. M. (2013). Nat. Struct. Mol. Biol. 20, 728–734.  CrossRef CAS PubMed Google Scholar
First citationSleeman, J. E., Ajuh, P. & Lamond, A. I. (2001). J. Cell Sci. 114, 4407–4419.  CrossRef PubMed CAS Google Scholar
First citationSleeman, J. E. & Lamond, A. I. (1999). Curr. Biol. 9, 1065–1074.  CrossRef PubMed CAS Google Scholar
First citationStaněk, D. & Neugebauer, K. M. (2004). J. Cell Biol. 166, 1015–1025.  PubMed Google Scholar
First citationStaněk, D. & Neugebauer, K. M. (2006). Chromosoma, 115, 343–354.  PubMed Google Scholar
First citationTellier, M., Maudlin, I. & Murphy, S. (2020). WIREs RNA, 11, e1593.  Google Scholar
First citationTownsend, C., Leelaram, M. N., Agafonov, D. E., Dybkov, O., Will, C. L., Bertram, K., Urlaub, H., Kastner, B., Stark, H. & Lührmann, R. (2020). Science, 370, eabc3753.  CrossRef PubMed Google Scholar
First citationWahl, M. C., Will, C. L. & Lührmann, R. (2009). Cell, 136, 701–718.  Web of Science CrossRef PubMed CAS Google Scholar
First citationWan, R., Yan, C., Bai, R., Huang, G. & Shi, Y. (2016). Science, 353, 895–904.  Web of Science CrossRef CAS PubMed Google Scholar
First citationWeber, G., Cristão, V. F., Alves, F. L., Santos, K. F., Holton, N., Rappsilber, J., Beggs, J. D. & Wahl, M. C. (2011). Genes Dev. 25, 1601–1612.  CrossRef CAS PubMed Google Scholar
First citationWeber, G., Cristão, V. F., Santos, K. F., Jovin, S. M., Heroven, A. C., Holton, N., Lührmann, R., Beggs, J. D. & Wahl, M. C. (2013). Genes Dev. 27, 525–540.  Web of Science CrossRef CAS PubMed Google Scholar
First citationWilkinson, M. E., Fica, S. M., Galej, W. P. & Nagai, K. (2021). Mol. Cell, 81, 1439–1452.  CrossRef CAS PubMed Google Scholar
First citationWill, C. L. & Lührmann, R. (2001). Curr. Opin. Cell Biol. 13, 290–301.  Web of Science CrossRef PubMed CAS Google Scholar
First citationWill, C. L. & Lührmann, R. (2005). Biol. Chem. 386, 713–724.  CrossRef PubMed CAS Google Scholar
First citationWill, C. L. & Lührmann, R. (2011). Cold Spring Harb. Perspect. Biol. 3, a003707.  Web of Science CrossRef PubMed Google Scholar
First citationYan, C., Hang, J., Wan, R., Huang, M., Wong, C. C. L. & Shi, Y. (2015). Science, 349, 1182–1191.  Web of Science CrossRef CAS PubMed Google Scholar
First citationYan, C., Wan, R., Bai, R., Huang, G. & Shi, Y. (2017). Science, 355, 149–155.  CrossRef CAS PubMed Google Scholar
First citationYang, K., Zhang, L., Xu, T., Heroux, A. & Zhao, R. (2008). Proc. Natl Acad. Sci. USA, 105, 13817–13822.  Web of Science CrossRef PubMed CAS Google Scholar
First citationZhang, L., Shen, J., Guarnieri, M. T., Heroux, A., Yang, K. & Zhao, R. (2007). Protein Sci. 16, 1024–1031.  Web of Science CrossRef PubMed CAS Google Scholar
First citationZhang, X., Yan, C., Hang, J., Finci, L. I., Lei, J. & Shi, Y. (2017). Cell, 169, 918–929.  CrossRef CAS PubMed Google Scholar

This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.

Journal logoSTRUCTURAL
BIOLOGY
ISSN: 2059-7983
Follow Acta Cryst. D
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds