Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Structural basis for the assembly of the Sxl–Unr translation regulatory complex


Genetic equality between males and females is established by chromosome-wide dosage-compensation mechanisms. In the fruitfly Drosophila melanogaster, the dosage-compensation complex promotes twofold hypertranscription of the single male X-chromosome and is silenced in females by inhibition of the translation of msl2, which codes for the limiting component of the dosage-compensation complex1,2. The female-specific protein Sex-lethal (Sxl) recruits Upstream-of-N-ras (Unr) to the 3′ untranslated region of msl2 messenger RNA, preventing the engagement of the small ribosomal subunit3. Here we report the 2.8 Å crystal structure, NMR and small-angle X-ray and neutron scattering data of the ternary Sxl–Unr–msl2 ribonucleoprotein complex featuring unprecedented intertwined interactions of two Sxl RNA recognition motifs, a Unr cold-shock domain and RNA. Cooperative complex formation is associated with a 1,000-fold increase of RNA binding affinity for the Unr cold-shock domain and involves novel ternary interactions, as well as non-canonical RNA contacts by the α1 helix of Sxl RNA recognition motif 1. Our results suggest that repression of dosage compensation, necessary for female viability, is triggered by specific, cooperative molecular interactions that lock a ribonucleoprotein switch to regulate translation. The structure serves as a paradigm for how a combination of general and widespread RNA binding domains expands the code for specific single-stranded RNA recognition in the regulation of gene expression.

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Rent or buy this article

Get just this article for as long as you need it


Prices may be subject to local taxes which are calculated during checkout

Figure 1: Analysis of Unr and Sxl binding to msl2 mRNA.
Figure 2: Crystal structure of the ternary dRBD3–CSD1–msl2 18-mer complex at 2.8 Å resolution.
Figure 3: Ternary complex formation and functional analysis of molecular interfaces in the Sxl–Unr–msl2-RNA complex.

Accession codes

Primary accessions

Biological Magnetic Resonance Data Bank

Protein Data Bank

Data deposits

The atomic coordinates have been deposited in the Protein Data Bank under accession number 4QQB. The NMR chemical shifts are deposited in the Biological Magnetic Resonance Data Bank, entries 25059, 25060 and 25078 for chemical shifts of CSD1 in the free, RNA-bound and SXL–RNA-bound state, respectively, and 25072 for SXL chemical shifts in the ternary complex.


  1. Conrad, T. & Akhtar, A. Dosage compensation in Drosophila melanogaster: epigenetic fine-tuning of chromosome-wide transcription. Nature Rev. Genet. 13, 123–134 (2012)

    CAS  PubMed  Google Scholar 

  2. Gelbart, M. E. & Kuroda, M. I. Drosophila dosage compensation: a complex voyage to the X chromosome. Development 136, 1399–1410 (2009)

    CAS  PubMed  PubMed Central  Google Scholar 

  3. Graindorge, A., Militti, C. & Gebauer, F. Posttranscriptional control of X-chromosome dosage compensation. Wiley Interdiscip. Rev. RNA 2, 534–545 (2011)

    CAS  PubMed  Google Scholar 

  4. Beckmann, K., Grskovic, M., Gebauer, F. & Hentze, M. W. A dual inhibitory mechanism restricts msl-2 mRNA translation for dosage compensation in Drosophila. Cell 122, 529–540 (2005)

    CAS  PubMed  Google Scholar 

  5. Abaza, I., Coll, O., Patalano, S. & Gebauer, F. Drosophila UNR is required for translational repression of male-specific lethal 2 mRNA during regulation of X-chromosome dosage compensation. Genes Dev. 20, 380–389 (2006)

    CAS  PubMed  PubMed Central  Google Scholar 

  6. Duncan, K. et al. Sex-lethal imparts a sex-specific function to UNR by recruiting it to the msl-2 mRNA 3′ UTR: translational repression for dosage compensation. Genes Dev. 20, 368–379 (2006)

    CAS  PubMed  PubMed Central  Google Scholar 

  7. Grskovic, M., Hentze, M. W. & Gebauer, F. A co-repressor assembly nucleated by Sex-lethal in the 3′UTR mediates translational control of Drosophila msl-2 mRNA. EMBO J. 22, 5571–5581 (2003)

    CAS  PubMed  PubMed Central  Google Scholar 

  8. Schindelin, H., Marahiel, M. A. & Heinemann, U. Universal nucleic acid-binding domain revealed by crystal structure of the B. subtilis major cold-shock protein. Nature 364, 164–168 (1993)

    ADS  CAS  PubMed  Google Scholar 

  9. Abaza, I. & Gebauer, F. Functional domains of Drosophila UNR in translational control. RNA 14, 482–490 (2008)

    CAS  PubMed  PubMed Central  Google Scholar 

  10. Gebauer, F., Grskovic, M. & Hentze, M. W. Drosophila sex-lethal inhibits the stable association of the 40S ribosomal subunit with msl-2 mRNA. Mol. Cell 11, 1397–1404 (2003)

    CAS  PubMed  Google Scholar 

  11. Handa, N. et al. Structural basis for recognition of the tra mRNA precursor by the Sex-lethal protein. Nature 398, 579–585 (1999)

    ADS  CAS  PubMed  Google Scholar 

  12. Sachs, R., Max, K. E., Heinemann, U. & Balbach, J. RNA single strands bind to a conserved surface of the major cold shock protein in crystals and solution. RNA 18, 65–76 (2012)

    CAS  PubMed  PubMed Central  Google Scholar 

  13. Mayr, F., Schutz, A., Doge, N. & Heinemann, U. The Lin28 cold-shock domain remodels pre-let-7 microRNA. Nucleic Acids Res. 40, 7492–7506 (2012)

    CAS  PubMed  PubMed Central  Google Scholar 

  14. Nam, Y., Chen, C., Gregory, R. I., Chou, J. J. & Sliz, P. Molecular basis for interaction of let-7 microRNAs with Lin28. Cell 147, 1080–1091 (2011)

    CAS  PubMed  PubMed Central  Google Scholar 

  15. McLaughlin, K. J., Jenkins, J. L. & Kielkopf, C. L. Large favorable enthalpy changes drive specific RNA recognition by RNA recognition motif proteins. Biochemistry 50, 1429–1431 (2011)

    CAS  PubMed  Google Scholar 

  16. Cline, T. W. Autoregulatory functioning of a Drosophila gene product that establishes and maintains the sexually determined state. Genetics 107, 231–277 (1984)

    CAS  PubMed  PubMed Central  Google Scholar 

  17. Bernstein, M., Lersch, R. A., Subrahmanyan, L. & Cline, T. W. Transposon insertions causing constitutive Sex-lethal activity in Drosophila melanogaster affect Sxl sex-specific transcript splicing. Genetics 139, 631–648 (1995)

    CAS  PubMed  PubMed Central  Google Scholar 

  18. Maine, E. M., Salz, H. K., Cline, T. W. & Schedl, P. The Sex-lethal gene of Drosophila: DNA alterations associated with sex-specific lethal mutations. Cell 43, 521–529 (1985)

    CAS  PubMed  Google Scholar 

  19. Valcárcel, J., Singh, R., Zamore, P. D. & Green, M. R. The protein Sex-lethal antagonizes the splicing factor U2AF to regulate alternative splicing of transformer pre-mRNA. Nature 362, 171–175 (1993)

    ADS  PubMed  Google Scholar 

  20. Triqueneaux, G., Velten, M., Franzon, P., Dautry, F. & Jacquemin-Sablon, H. RNA binding specificity of Unr, a protein with five cold shock domains. Nucleic Acids Res. 27, 1926–1934 (1999)

    CAS  PubMed  PubMed Central  Google Scholar 

  21. Drosophila 12 Genomes Consortium Evolution of genes and genomes on the Drosophila phylogeny. Nature 450, 203–218 (2007)

    Google Scholar 

  22. Park, S. W. et al. An evolutionarily conserved domain of roX2 RNA is sufficient for induction of H4-Lys16 acetylation on the Drosophila X chromosome. Genetics 177, 1429–1437 (2007)

    CAS  PubMed  PubMed Central  Google Scholar 

  23. Weber, G., Trowitzsch, S., Kastner, B., Luhrmann, R. & Wahl, M. C. Functional organization of the Sm core in the crystal structure of human U1 snRNP. EMBO J. 29, 4172–4184 (2010)

    CAS  PubMed  PubMed Central  Google Scholar 

  24. Pomeranz Krummel, D. A., Oubridge, C., Leung, A. K., Li, J. & Nagai, K. Crystal structure of human spliceosomal U1 snRNP at 5.5 Å resolution. Nature 458, 475–480 (2009)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  25. Bono, F., Ebert, J., Lorentzen, E. & Conti, E. The crystal structure of the exon junction complex reveals how it maintains a stable grip on mRNA. Cell 126, 713–725 (2006)

    CAS  PubMed  Google Scholar 

  26. Mackereth, C. D. et al. Multi-domain conformational selection underlies pre-mRNA splicing regulation by U2AF. Nature 475, 408–411 (2011)

    CAS  PubMed  Google Scholar 

  27. Mackereth, C. D. & Sattler, M. Dynamics in multi-domain protein recognition of RNA. Curr. Opin. Struct. Biol. 22, 287–296 (2012)

    CAS  PubMed  Google Scholar 

  28. Baltz, A. G. et al. The mRNA-bound proteome and its global occupancy profile on protein-coding transcripts. Mol. Cell 46, 674–690 (2012)

    CAS  PubMed  Google Scholar 

  29. Castello, A. et al. Insights into RNA biology from an atlas of mammalian mRNA-binding proteins. Cell 149, 1393–1406 (2012)

    CAS  PubMed  Google Scholar 

  30. Hennig, J., Wang, I., Sonntag, M., Gabel, F. & Sattler, M. Combining NMR and small angle X-ray and neutron scattering in the structural analysis of a ternary protein-RNA complex. J. Biomol. NMR 56, 17–30 (2013)

    CAS  PubMed  Google Scholar 

  31. Wen, J., Arakawa, T. & Philo, J. S. Size-exclusion chromatography with on-line light-scattering, absorbance, and refractive index detectors for studying proteins and their interactions. Anal. Biochem. 240, 155–166 (1996)

    CAS  PubMed  Google Scholar 

  32. McCoy, A. J. et al. Phaser crystallographic software. J. Appl. Cryst. 40, 658–674 (2007)

    CAS  Google Scholar 

  33. Collaborative Computational Project, Number 4. The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D 50, 760–763 (1994)

  34. Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D 60, 2126–2132 (2004)

    Article  CAS  PubMed  Google Scholar 

  35. Winn, M. D., Murshudov, G. N. & Papiz, M. Z. Macromolecular TLS refinement in REFMAC at moderate resolutions. Methods Enzymol. 374, 300–321 (2003)

    CAS  PubMed  Google Scholar 

  36. Gebauer, F., Corona, D. F., Preiss, T., Becker, P. B. & Hentze, M. W. Translational control of dosage compensation in Drosophila by Sex-lethal: cooperative silencing via the 5′ and 3′ UTRs of msl-2 mRNA is independent of the poly(A) tail. EMBO J. 18, 6146–6154 (1999)

    CAS  PubMed  PubMed Central  Google Scholar 

  37. Sattler, M., Schleucher, J. & Griesinger, C. Heteronuclear multidimensional NMR experiments for the structure determination of proteins in solution employing pulsed field gradients. Prog. Nucl. Magn. Reson. Spectrosc. 34, 93–158 (1999)

    CAS  Google Scholar 

  38. Salzmann, M., Pervushin, K., Wider, G., Senn, H. & Wuthrich, K. [13C]-constant-time [15N,1H]-TROSY-HNCA for sequential assignments of large proteins. J. Biomol. NMR 14, 85–88 (1999)

    CAS  PubMed  Google Scholar 

  39. Delaglio, F. et al. NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J. Biomol. NMR 6, 277–293 (1995)

    CAS  PubMed  Google Scholar 

  40. Goddard, T. D. & Kneller, D. G. SPARKY 3. (Univ. California).

  41. Cordier, F., Dingley, A. J. & Grzesiek, S. A doublet-separated sensitivity-enhanced HSQC for the determination of scalar and dipolar one-bond J-couplings. J. Biomol. NMR 13, 175–180 (1999)

    CAS  PubMed  Google Scholar 

  42. Yang, D. & Kay, L. E. Improved 1HN-detected triple resonance TROSY-based experiments. J. Biomol. NMR 13, 3–10 (1999)

    CAS  PubMed  Google Scholar 

  43. Otting, G., Ruckert, M., Levitt, M. H. & Moshref, A. NMR experiments for the sign determination of homonuclear scalar and residual dipolar couplings. J. Biomol. NMR 16, 343–346 (2000)

    CAS  PubMed  Google Scholar 

  44. Dosset, P., Hus, J. C., Marion, D. & Blackledge, M. A novel interactive tool for rigid-body modeling of multi-domain macromolecules using residual dipolar couplings. J. Biomol. NMR 20, 223–231 (2001)

    CAS  PubMed  Google Scholar 

  45. Gosh, R. E. et al. A computing guide for small-angle scattering. ILL Technical Report ILL06GH05T. (2006)

  46. Konarev, P. V., Volkov, V. V., Sokolova, A. V., Koch, M. H. J. & Svergun, D. I. PRIMUS: a Windows-PC based system for small-angle scattering data analysis. J. Appl. Cryst. 36, 1277–1282 (2003)

    CAS  Google Scholar 

  47. Svergun, D. I., Barberato, C. & Koch, M. H. J. CRYSOL - a program to evaluate x-ray solution scattering of biological macromolecules from atomic coordinates. J. Appl. Cryst. 28, 768–773 (1995)

    CAS  Google Scholar 

  48. Svergun, D. I. et al. Protein hydration in solution: experimental observation by x-ray and neutron scattering. Proc. Natl Acad. Sci. USA 95, 2267–2272 (1998)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  49. Svergun, D. I. Restoring low resolution structure of biological macromolecules from solution scattering using simulated annealing. Biophys. J. 76, 2879–2886 (1999)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  50. Svergun, D. I. Determination of the regularization parameter in indirect-transform methods using preceptual criteria. J. Appl. Cryst. 25, 495–503 (1992)

    CAS  Google Scholar 

  51. Volkov, V. V. & Svergun, D. I. Uniqueness of ab initio shape determination in small-angle scattering. J. Appl. Cryst. 36, 860–864 (2003)

    CAS  Google Scholar 

  52. Kozin, M. & Svergun, D. I. Automated matching of high- and low-resolution structural models. J. Appl. Cryst. 34, 33–41 (2001)

    CAS  Google Scholar 

  53. Jacrot, B. Study of biological structures by neutron-scattering from solution. Rep. Prog. Phys. 39, 911–953 (1976)

    ADS  CAS  Google Scholar 

  54. Voss, N. R. & Gerstein, M. Calculation of standard atomic volumes for RNA and comparison with proteins: RNA is packed more tightly. J. Mol. Biol. 346, 477–492 (2005)

    CAS  PubMed  Google Scholar 

  55. Cornilescu, G., Marquardt, J. L., Ottiger, M. & Bax, A. Validation of protein structure from anisotropic carbonyl chemical shifts in a dilute liquid crystalline phase. J. Am. Chem. Soc. 120, 6836–6837 (1998)

    CAS  Google Scholar 

  56. Gouet, P., Courcelle, E., Stuart, D. I. & Metoz, F. ESPript: analysis of multiple sequence alignments in PostScript. Bioinformatics 15, 305–308 (1999)

    CAS  PubMed  Google Scholar 

  57. St Pierre, S. E., Ponting, L., Stefancsik, R., McQuilton, P. & the FlyBase Consortium FlyBase 102–advanced approaches to interrogating FlyBase. Nucleic Acids Res. 42, D780–D788 (2014)

    CAS  PubMed  Google Scholar 

  58. Sievers, F. et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol. (11 November 2011)

  59. Dereeper, A. et al. robust phylogenetic analysis for the non-specialist. Nucleic Acids Res. 36, W465–W469 (2008)

    CAS  PubMed  PubMed Central  Google Scholar 

  60. Jacques, D. A., Guss, J. M., Svergun, D. I. & Trewhella, J. Publication guidelines for structural modelling of small-angle scattering data from biomolecules in solution. Acta Crystallogr. D 68, 620–626 (2012)

    CAS  PubMed  Google Scholar 

Download references


We thank H.-S. Kang, M. Pabis and L. Warner for discussions and reading the manuscript. We acknowledge the ILL, Grenoble for BAG SANS beamtime on D22 (local contact A. Martel), the ESRF for beamtime on BM29 (local contact L. Zerrad), the Swiss Light Source for beamtime, the crystallization facility at the Max-Planck-Institute for Biochemistry, Martinsried, and the Bavarian NMR Center for NMR measurement time. We thank T. Madl and C. Göbl for assisting in SAXS measurements of the 34-mer complex. J.H. acknowledges postdoctoral fellowships from the Swedish Research Council (Vetenskapsrådet) and the European Molecular Biology Organization (EMBO, ALTF276-2010), C.M. acknowledges a La Caixa Foundation fellowship. Funding support is acknowledged by grants BFU2009-08243 and Consolider CSD2009-00080 from the Spanish Ministry of Economy and Competitiveness (to. F.Ge.) and the Deutsche Forschungsgemeinschaft, grants SFB1035 and GRK1721 (to M.Sa.).

Author information

Authors and Affiliations



J.H. produced samples, performed NMR, SAXS/SANS, biophysical measurements, crystallization and model building/refinement. C.M. produced samples, performed and analysed EMSA, transfection and translation assays. G.M.P. acquired diffraction data and performed molecular replacement and model building. I.W., M.So. and A.G. contributed to sample production. F.Ga. performed and analysed SANS measurements. J.H., F.Ge., C.M. and M.Sa. interpreted results and wrote the manuscript. All authors discussed the results and commented on the manuscript.

Corresponding authors

Correspondence to Fátima Gebauer or Michael Sattler.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Extended data figures and tables

Extended Data Figure 1 Biochemical and biophysical characterization of Sxl–Unr–msl2-RNA complex formation.

a, Sxl and Unr do not interact in the absence of RNA. The lack of interaction between Sxl dRBD3 and Unr CSD1 in the absence of RNA was tested by an NMR titration, where 15N-labelled Sxl resonances were observed in a 1H,15N-HSQC experiment (blue). Upon titration of CSD1 up to a ratio of 1:1 (red), no significant changes of the spectrum could be observed. Some line broadening is observed for Leu 127, Gly 167, Asp 210, Val 238, Arg 244 and Ser 281. However, as these residues are not involved in complex formation, this probably reflects unspecific interactions and/or aggregation. b, Mapping of the Unr binding region on msl2 mRNA using EMSA. Serial mutation of the EF region replacing the native nucleotides by CU repeats showed that the first 32 nucleotides were relevant (Fig. 1), while the last 14 nucleotides (mutant (Mut) 5 and 6) are dispensable for ternary complex formation. c, Size-exclusion chromatography combined with static light scattering for molecular weight (MW) determination of dRBD3–CSD1–msl2-RNA complexes using different lengths of RNA. The refractive index (RI, red) and the right-angle light scattering (RALS, green) profiles are shown. AU, arbitrary units. d, Structural superposition of Sxl bound to transformer pre-mRNA (Sxl, yellow; RNA, orange, PDB: 1B7F11) and bound to msl2 mRNA in the ternary complex presented in this study (Sxl, green; RNA, magenta). e, Comparison of the structure of Unr CSD1 bound to msl2 within the ternary complex (left: CSD1, blue; RNA, magenta), and the homologous CSD of CspB from Bacillus subtilis bound to U5 RNA (right: CSD, orange; RNA, yellow, PDB: 3PF511).

Extended Data Figure 2 Structure validation in solution.

a, Residual dipolar couplings (RDCs) validate the domain orientations for the Sxl RRMs and Unr CSD1 seen in the crystal structure in solution. Correlations between experimental and back-calculated residual dipolar coupling values are shown. The type of dipolar coupling, the observed protein within the complex, and the number of couplings are indicated. All measurements were performed in Pf1 phages. Other alignment media tested induced aggregation of the sample and interacted with the complex as chemical shift perturbations could be observed. The RDC quality factors QRDC 55 are shown in red and indicate a good agreement between the crystal structure and the domain arrangements in solution. The alignment tensors determined are similar for dRBD3 (D = −12.4, R = 0.17; where D is the magnitude and R the rhombicity of the alignment tensor) and CSD1 (D = −14.8, R = 0.28) consistent with a fixed domain orientation in solution. Error bars indicate measurement uncertainties depending on the spectral resolution. b, NMR chemical shift perturbation (combined 1H and 15N chemical shift difference ΔδH,N) of the Sxl–RNA complex upon titration with CSD1. Major shifts occur only on residues located in RRM1, consistent with the crystal structure. Further zoom views of triple-zipper and RRM1-α1-helix residues confirm their involvement in ternary complex formation (Fig. 3a). c, Small-angle X-ray scattering (SAXS) analysis of the ternary complex showing back-calculated scattering densities (red line) and experimental scattering data (blue dots). A χ2 of 0.77 confirms that the structure of the complex in solution is the same as in the crystal structure. d, Small-angle neutron scattering (SANS) of the ternary dRBD3–CSD1–msl2-mRNA complex, where CSD1 is perdeuterated and dRBD3/msl2-mRNA is protonated, at 42% D2O buffer concentration. In this composition, scattering of dRBD3 matches the scattering density of the buffer and therefore does not contribute to the signal (indicated with a white dRBD3 component in the schematic complex used throughout this study). The back-calculated scattering curve (red line) is fitted against the experimental data (blue dots), and the resulting χ2 is shown as inset. e, as d but using 70% D2O in the buffer, where RNA matches the buffer contrast, and where the perdeuterated component (CSD1) has a positive contrast (indicated with a dark blue colour), and dRBD3 a negative contrast (light green). f, as d but here dRBD3 is perdeuterated and CSD1 is protonated. Thus, CSD1 matches the contrast of the buffer and does not contribute to the signal, whereas dRBD3 exhibits a positive contrast. g, as e but inversely labelled, thus, CSD1 having a negative contrast to the buffer and dRBD3 a positive. These data enabled us to localize each component within the overall ab initio bead models as shown in h obtained with the program MONSA50 (two views are shown; green, Sxl; blue, CSD1; magenta, RNA; see Methods for details).

Extended Data Figure 3 NMR and small-angle X-ray scattering analysis of dRBD3–CSD1 bound to different msl2 mRNA oligomers.

a, NMR chemical shift perturbations (combined 1H and 15N chemical shift difference ΔδH,N) of amide signals of CSD1 upon binding to 17-mer RNA (Site F with the 5′ flanking region) (top right histogram; conserved, predicted RNPs are indicated by yellow bars). Very similar chemical shift perturbations as for the 18-mer (Site F with 3′ flanking region) are observed (Fig. 3, main text). However, upon further titration with dRBD3 no additional shifts can be seen (bottom histogram), suggesting that there are no contacts between CSD1 and Sxl, in contrast to the 18-mer. b, Static light scattering (SLS) confirms these observations and further shows that a stable complex can be only formed between Sxl and the 17-mer RNA (RALS: right-angle light scattering; RI: refractive index). c, as in a but instead of a 17-mer, a 27-mer RNA consisting of site F and both flanking regions was used. This RNA can accommodate two CSD1 moieties as evident from NMR chemical shift perturbations (top histogram). However, upon Sxl titration, the NMR signals move back to the position of the free CSD1 (bottom histogram), as exemplified by the chemical shift changes observed for the His 213 and Arg 239 triple-zipper residues (zoomed spectra on the right). This shows that at least one CSD1 is pushed out of the complex with the 27-mer. d, These results were confirmed by static light scattering, where only a stable 1:1:1 complex (Sxl–CSD1–RNA) could be formed. The experimentally determined molecular weight is consistent with a 1:1:1 complex. e, Small-angle X-ray scattering (SAXS) data of the 18-mer (red) compared with the 34-mer (blue) suggest that two Sxl–CSD1 complexes are independently assembled on the 34-mer like ‘beads on a string’, that is, with no contacts between the two subcomplexes bound to site E and F, respectively. The radius of gyration is much larger and Dmax(the maximum diameter of the particle, derived from the distance distribution function P(r)) increases more than would be expected for a compact and rigid 34-mer complex. Although Kratky plots are consistent with a structured 34-mer complex, the curve does not reach down to the baseline after the peak as for the 18-mer complex, indicating a higher degree of flexibility. I, Q, r and Rg refer to the scattering intensity, the scattering vector, interatomic distance and radius of gyration, respectively. A shape analysis, using ab initio modelling (see Methods for details), demonstrates that the average of 20 models adopts an elongated shape (right) to accommodate two Sxl–CSD1 complexes bound to the 34-mer RNA. The data presented in this figure, the inability to produce crystals in extensive crystallization trials, and the difference in RNA binding affinity for sites E and F (see main text, Fig. 1), as well as the non-conserved distance between the E and F sites in different Drosophila species (Extended Data Fig. 7) suggest the lack of contacts between both single complexes (site E and site F complex), at least in the absence of other co-factors. Notably, the weaker affinity of the E site is probably due to the missing CG dinucleotide in the corresponding 3′ flanking regions (AGCAUGAA flanking E site vs AGCACGUGAA flanking F site). The C11 nucleotide in the 18-mer (flanking the F site) is specifically recognized in the 18-mer complex (Fig. 2). In the E site this contact would be formed with a U instead of a C, thereby missing the hydrogen bond between the NH2 of the base with the backbone carbonyl oxygen of Sxl Lys 161. As a result, the subsequent AA dinucleotides might not be able to form the non-canonical contacts with the α1 helix of Sxl RRM1 and thus maybe induce the flexibility observed in SAXS.

Extended Data Figure 4 Isothermal titration calorimetry analysis of the dRBD3–CSD1–msl2-mRNA interactions.

a, Isothermal titration calorimetry data of the cognate RNA (a hexamer, including the bases bound to CSD1 in the ternary complex) titrated to CSD1 can be fitted to an affinity corresponding to a dissociation constant Kd of 14 µM. b, The 18-mer can accommodate two CSD1 moieties which bind with similar affinity. The second binding site is within the exothermic part of the isothermal titration calorimetry binding curve, which arises due to the simultaneous melting of the self-complementary part of the RNA (CACGUG). This can also be observed, when titrating a hexamer oligonucleotide consisting of this sequence alone into CSD1 (f). The higher affinity compared to the hexamer is due to the increased availability of binding surface. c, Isothermal titration calorimetry data of titrating dRBD3 into an 18-mer. Fitting the data, assuming one binding site yielded dissociation constant Kd of 200 nM, similar to other observations16. d, Titration of 18-mer into a sample containing equimolar amounts of dRBD3 and CSD1 exhibits two binding sites with similar affinities (Kd = 15–20 nM). e, Titration of dRBD3 into a sample containing 18-mer–CSD1 complex resulted in one binding site with an affinity (a Kd) of 15 nM. Thus, the affinity of CSD1 in the context of the complex increased 1,000-fold compared to its affinity to RNA alone. f, Specificity of msl2 mRNA recognition by Sxl and CSD1. Isothermal titration calorimetry data shows that CSD1 has similar weak affinities to different hexamer subsets of the 18-mer in the range of Kd = 10–20 µM (the exothermic signals in the middle are due to simultaneous melting of the self-complementary oligonucleotide CACGUG), although CSD1 specifically recognizes C11 in the context of the 18-mer complex in presence of Sxl. Initial specificity seems to be governed by Sxl which binds to 18-mer RNA with higher affinity and recruits CSD1 to its position with extremely increased affinity (1,000-fold), where it, together with Sxl, specifically recognizes C11 as described in the main text. The RNA sequence recognized by UNR CSD1 in the ternary complex (G10-C11-A12-C13-G14) resembles the proposed consensus sequence (A/GAACG/A) for the human homologue21. However, the crucial C11 is not present in the consensus. A reason for this could be that a Sxl-like binding partner was not present in the SELEX experiments; alternatively, human UNR could show a different binding specificity. g, Isothermal titration calorimetry of ternary complex formation with C11U mutant. Affinity decreases by fivefold if the C11 is substituted by a uridine, which is in agreement with EMSA data (h). The two transitions observable for wild-type 18-mer are not visible here. This is probably due to similar affinities for complex formation and Sxl’s affinity to the 18-mer alone. Numbers indicated are dissociation constants (Kd). Errors indicate the fitting error. For each experimental setup at least duplicates have been acquired. i, Complex formation at site B in the 5′ UTR of msl2. Binding of dRBD4 and Unr to the E/F site and 5′B probes was assessed by EMSA (left) and crosslink-immunoprecipitation of Unr from embryo extracts (right). Probe sequences are indicated at the bottom. Probe 5′B does not support efficient complex formation in vitro or in extracts.

Extended Data Figure 5 Electrophoretic mobility shift assays of dRBD4 and CSD1 variants.

a, Assessment of the RNA binding ability of dRBD4 mutants. All mutants exhibited the similar binding patterns as the wild type (WT), indicating that all mutants fully retained their RNA binding capacity. b, Assessment of the ability of dRBD4 mutants to form a ternary complex with RNA and CSD1. c, Assessment of the ability of CSD1 mutants to form a ternary complex with RNA and dRBD4. d, Translational repression assays in embryo extracts (containing endogenous Unr) and increasing amounts of dRBD4 derivatives. A luciferase reporter containing the msl2 EF region in the 3′ UTR was used (BmutLEF). Error bars represent the standard deviation of three experiments. e, Analysis of Msl2 expression in Schneider 2 cells transfected with Sxl derivatives. Msl2 and Sxl expression were monitored by western blot (upper panel). Msl2 was quantified, normalized for Sxl expression, and reported relative to the T137A negative control. Error bars represent the standard deviation of two experiments.

Extended Data Figure 6 Multiple sequence alignment of cold shock domains and human Sxl orthologues.

a, Sequence alignment of Drosophila Unr CSD1 with CSD1 of Unr of other species. Completely conserved residues are framed and have a red background. Residues which are exchanged by a similar residue are framed and have a red font. The secondary structure of CSD1 in the crystal structure presented here is indicated above the alignment. Unr CSD1 is highly conserved even between different phyla (arthropoda vs chordata) and within the different classes of chordata (actinoplerygii, amphibia, reptilia, aves and mammalia). Residues involved in complex formation (His 213, Asp 237 and Arg 239, indicated by a blue star) are fully conserved, suggesting that CSD1 of human Unr might also form a similar complex with human counterparts of Sxl in other pathways. b, Alignment of CSD1, CSD2, CSD3, CSD4 and CSD5 of Drosophila Unr. The starting residue number of each domain within full length Unr is in front of each sequence. Same colour code as in a. The highly conserved RNP1 and fully conserved RNP2 are indicated with a green square, indicating that all five cold shock domains are likely to be able to bind RNA. Asp 237 can only be found in CSD1, and Arg 239 in CSD1 and CSD5 (blue stars). c, Alignment of RRMs of cleavage stimulation factor subunit 2 (CstF), RNA binding protein X-linked 2 (RBMX2), four ELAV-like protein homologues (ELAV-like 1–4, ELAV-like 1 = HuR), Sxl and RNA binding motif 6 (RBM6). Only the sequence region containing residues involved in ternary complex formation (α1 helix, triple-zipper residues) was aligned and was used in an extensive similarity search of human proteins. RBM6 shows the best conservation of triple-zipper (blue star) and α1-helix (red triangle) residues. Arg 139, the mutation of which has the weakest effect on complex formation is not conserved in any of the listed proteins. ESPript was used in this figure with the default colour code (fully conserved, white font, red background; weakly conserved, red font)56.

Extended Data Figure 7 Conservation of the 3′ UTR of msl2 mRNA between different Drosophila species.

Alignment of a region of the 3′ UTR of msl2 mRNA of different Drosophila species indicates that the E and F sites, where the ternary complex is formed, are highly conserved, especially C11 (indicated with an arrow), which is fully conserved. The flanking regions show much less conservation and regions towards the 5′ and 3′ ends of the sequences aligned here are even less conserved (not shown). A phylogenetic tree is shown, illustrating the degree of diversion of the species used in the alignment. The variable length of the region between sites E and F is further indirect evidence for the lack of contact between complexes formed on each site (Extended Data Fig. 3). The sequences were retrieved from The multiple sequence alignment was performed using Clustal Omega58, and the phylogenetic tree was generated by Phylogeny.fr59. The alignment figure was generated using ESPript with the default colour code (fully conserved, white font, red background; weakly conserved, red font)56.

Extended Data Table 1 Crystallographic data collection and refinement statistics
Extended Data Table 2 Small-angle scattering statistics according to Jacques et al.60
Extended Data Table 3 Isothermal titration calorimetry data of the dRBD4-CSD1–msl2-mRNA complex

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Hennig, J., Militti, C., Popowicz, G. et al. Structural basis for the assembly of the Sxl–Unr translation regulatory complex. Nature 515, 287–290 (2014).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

This article is cited by


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing