Connecting E3 ligase and mRNA decay functions of Roquin proteins

• Applicant: Professor Dr. Vigo Heissmeyer
• Subject Area: Immunology ; Cell Biology
• Term: from 2016 to 2019
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 287078900

• Final Report Year: 2020

Final Report Abstract

The domain structure of Roquin-1 includes the RNA-binding ROQ domain and the ubiquitin-ligase RING domain. While the function of the ROQ domain was not known when Roquin-1 was identified as an autoimmune inhibitor more than a decade ago, it was soon unraveled and mRNA targets were discovered by us and others. In contrast, substrates for its RING domain remained unknown, with the notable exception of Roquin-1 itself (autoubiquitination). To learn more about the E3 ligase function of Roquin-1 we approached the problem from three angles. First, we performed BioID proximity labelling and PTMscan to identify potential Roquin-1 substrates. Second, we analyzed the function of the RING and the ROQ domain in cell growth assays. Third, we validated ubiquitin-modifying enzymes that were candidates derived from a whole genome CRISPR/Cas9 KO screen investigating Roquin-1-mediated cell death in MEF cells. To our big surprise we found in all MEF and T cell BioID experiments that the lists of proteins that came into close proximity of Roquin-1 were absolutely dominated by other RNA-binding proteins, including known interactors, such as Edc4, Ddx6, Cnot1,2,3 and Nufip2. These data sets will be part of a manuscript that is in preparation. Unfortunately, no reproducible results were obtained using PTMscan and hence our efforts to identify E3 ligase substrates for Roquin-1 failed, except for the identification of K136 as the potential site of auto-ubiquitination. Cell growth assays hinted at a function for the RING domain in Roquin-1-mediated cell death that goes into the same direction (suppression of cell growth), but succeeded the RNA-binding effect, but this finding needs further investigation. Induced Roquin-1-mediated cell death in MEF cells led to morphological changes accompanied by reduced cell proliferation and increased numbers of apoptotic cells. This phenotype was not observed overexpressing the Malt-1 cleavage product of Roquin-1 (AA 1-510) in the same setting. Based on these findings a whole genome knockout screen identified ~30 potential co-factors of Roquin-1- mediated cell death, including Roquin-1 itself, factors of protein glycosylation enzymes and ubiquitinmodifying E3 ligases and DUBs. Gene editing of the E3 ligases Roquin-1 (positive control) and Trim28 as well as of the DUBs Usp8, Usp7 clearly reduced Roquin-1-mediated MEF cell death. Usp8, a factor with previously reported functions in cell death and autoimmunity, partially reduced Roquin-1 ubiquitination, but did not alter its half-life. Interestingly, gene-editing or conditional knockout of Usp8 in T cells increased the expression levels of the Roquin-1 targets Ox40, Ctla4 and Irf4. However, at this point we cannot exclude that this occurs secondary to the observed activation of T cells in the absence of Usp8. In the future it will be most exiting to find out if and how Roquin-1 and Usp8 interact to repress specific targets, or to find mRNA targets which are regulated by Roquin and also reveal contributions of the RING finger or involve auto-ubiquitination of K136. Likewise, it will be important to investigate if and how Roquin in combination with the different candidates (Trim28, Usp8 and Usp7) can induce apoptotic T cell death and through which targets such regulation is achieved. Although this project has not yet clarified how the E3 ligase domain regulates Roquin function as an RNA-binding protein, the work on this topic has been very stimulating for a number of research projects of our group. It has contributed critical knowledge and data to several publications and the insights generated so-far will direct our efforts in the future and guide us how to address the molecular mechanism in sophisticated approaches and to finally find an answer for this important question.

Publications

• 2017. Roquin Suppresses the PI3K-mTOR Signaling Pathway to Inhibit T Helper Cell Differentiation and Conversion of Treg to Tfr Cells. Immunity 47: 1067-82 e12
  Essig K, Hu D, Guimaraes JC, Alterauge D, Edelmann S, Raj T, Kranich J, Behrens G, Heiseke A, Floess S, Klein J, Maiser A, Marschall S, Hrabe de Angelis M, Leonhardt H, Calkhoven CF, Noessner E, Brocker T, Huehn J, Krug AB, Zavolan M, Baumjohann D, Heissmeyer V
  
• 2018. A translational silencing function of MCPIP1/Regnase-1 specified by the target site context. Nucleic Acids Res 46:4256-4270
  Behrens G, Winzen R, Rehage N, Dorrie A, Barsch M, Hoffmann A, Hackermuller J, Tiedje C, Heissmeyer V, Holtmann H
  
• 2018. Binding of NUFIP2 to Roquin promotes recognition and regulation of ICOS mRNA. Nat Commun 9: 299
  Rehage N, Davydova E, Conrad C, Behrens G, Maiser A, Stehklein JE, Brenner S, Klein J, Jeridi A, Hoffmann A, Lee E, Dianzani U, Willemsen R, Feederle R, Reiche K, Hackermuller J, Leonhardt H, Sharma S, Niessing D, Heissmeyer V
  
• 2018. Posttranscriptional regulation of T helper cell fate decisions. J Cell Biol 217:2615-2631
  Hoefig KP, Heissmeyer V.
  
• 2018. Roquin targets mRNAs in a 3'-UTR-specific manner by different modes of regulation. Nat Commun 9: 3810
  Essig K, Kronbeck N, Guimaraes JC, Lohs C, Schlundt A, Hoffmann A, Behrens G, Brenner S, Kowalska J, Lopez-Rodriguez C, Jemielity J, Holtmann H, Reiche K, Hackermuller J, Sattler M, Zavolan M, Heissmeyer V