Original Paper: https://www.nature.com/articles/s41586-019-1232-1
Changes in metabolic pathways may lead to abnormalities of the immune system which leads researchers to metabolism in search of targets for novel therapies. Serine hydroxymethyltransferase 2 (SHMT2), which regulates one-carbon transfer reactions essential for amino acid and nucleotide metabolism, is involved in an association which provides a potential link between metabolism and inflammation. It exists as a dimer but stabilization of the active tetrameric state occurs with the binding of pyridoxal-5’-phosphate (PLP).1 Promotion of inflammatory cytokine signaling via SHMT2 is accomplished by interacting with the deubiquitinating BRCC36 isopeptidase complex (BRISC). BRCC36 is a Zn2+-dependent deubiquitylase (DUB) and requires interaction with pseudo-DUBs Abraxas 1 or Abraxas 2 for activity.2 Interaction with Abraxas 1 leads to a complex required for DNA repair; however, BRCC36 interacts with Abraxas 2 as part of the larger BRISC-SHMT2 complex.2 BRISC-SHMT2 interaction leads to delivery to ubiquitylated type I interferon (IFN) receptors, allowing deubiquitylation by BRCC36.4 BRISC-deficient mice are protected from pathological conditions that come from increased inflammatory signaling.4
The interaction between BRISC and SHMT2 was broadly known; however, the structure of the complex and the potential connection to metabolism remained unknown. The Nature paper by Walden et al. sought to resolve these unknowns and present a mechanism by which metabolites may regulate inflammation. The authors produced a modified SHMT2, SHMT2ΔN, which contains residues 18-504 and is a mixture of dimers and tetramers in solution. Furthermore, a SHMT2(A285T) mutant that contains a substitution near the PLP-acceptor residue (K280) had already been identified and available. This allowed the authors to utilize SHMT2ΔN(A285T) which is a dimer in solution and crystal lattice. Walden et al. were thus able to directly compare the dimer and tetramer structures. The addition of SHMT2ΔN inhibited DUB activity of BRISC, with reduced inhibition occurring when preincubated with PLP. SHMT2ΔN(A285T and SHMT2ΔN(K280A/A285T), both obligate dimers, effectively inhibited BRISC. These findings revealed a previously unknown role of dimeric SHMT2 in BRISC regulation.
The authors aimed to further understand the molecular basis of BRISC-SHMT2 association and the involvement of PLP. Waden et al. accomplished this by solving the structure of BRISC bound to an SHMT2 dimer. SHMT2 bridges the two arms of BRISC, sitting directly above BRCC36 and Abraxas 2. SHMT2 primarily interacts with BRISC through the surface centered on the two α6 helices. Alteration of various SHMT2 residues in proximity to BRISC further established the most important contact surface for BRISC interaction and inhibition of DUB activity to be the SHMT2 α6 helix.
The next step of the paper was to utilize mouse embryonic fibroblasts (MEFs) in an attempt to understand BRISC-SHMT2 interaction in cells. Expressing Abraxas 2 mutants that result in loss of interaction with SHMT2 in MEFs resulted in reduced STAT1 phosphorylation. Furthermore, ectopic expression of wild-type Abraxas 2 revealed 48 genes being upregulated between 2- and 8000-fold in response to lipopolysaccharide (LPS). This is significant compared to the between 2- and 10-fold upregulation of 11 genes in Abraxas2-/- cells. These findings highlighted the importance of Abraxas 2 and revealed specific Abraxas 2 surfaces that are required for BRISC-SHMT2 complex assembly and immune signaling.
The positioning of SHMT2 within the complex sterically blocks the BRCC36 active site and kinetic data suggests that it acts as a competitive inhibitor. Wendel et al. propose that SHMT2 acts as a reversible endogenous BRISC inhibitor that prevents non-specific DUB activity which means that it would be the first example of an endogenous DUB inhibitor. In addition, it provides the first non-enzymatic role of the dimer form of SHMT2. The SHMT2 α6 helix that interacts with BRISC overlaps with the tetramerization surface which explains why the tetramer form of SHMT2 does not form the BRISC-SHMT2 complex.
To finally connect their work with metabolism, the authors examine the ability of PLP to regulate SHMT2 dimer availability and thus formation of the BRISC-SHMT2 complex. MEFs and HEK293T cells were cultured in B6-vitamer-free medium because PLP is the active form of vitamin B6. Fewer SHMT2 dimers were then seen after pyridoxal supplementation. In addition, expressing PLP-resistant SHMT2 mutants resulted in no observable PLP-dependent reduction of BRISC-SHMT2 interaction. As a way to support this finding further, levels of STAT1 phosphorylation were measured. A pyridoxal-dependent reduction of phosphorylated STAT1 was observed and PLP-resistant SHMT2 mutants once again failed to show a reduction. Reduction of nine IFN-induced genes in LPS-challenged MEFs was observed in a pyridoxal-dependent manner. Similarly as before, there was no change in MEFs that expressed a PLP-resistant mutant. The authors managed to reveal a direct link between vitamin B6 metabolism and inflammatory signaling by showing the regulatory influence of PLP on BRISC activity.
Future studies could build off this work in numerous ways. The mechanism of SHMT2 DUB inhibition can be utilized as a baseline to examine regulation of other DUBs in vivo. SHMT1 was briefly looked at by the authors because previous work suggested that it also interacts with BRISC. However, the authors could not assemble BRISC-SHMT1 complexes and thus further research into SHMT1 may uncover a currently unknown role. The finding that PLP influences BRISC activity opens important avenues for future research. Small-molecule ligands have the potential to be used as therapeutic agents by controlling ubiquitin signaling. The regulatory mechanisms of BRISC-SHMT2 interactions need to be further examined because it may reveal other methods of regulation or additional layers of regulation for immune signaling. Obtaining additional knowledge of the BRISC-SHMT2 complex may be crucial for the eventual development of drugs that treat inflammatory conditions by regulation of interferon signaling.
1. Giardina, Giorgio, Paolo Brunotti, Alessio Fiascarelli, Alessandra Cicalini, Mauricio G. S. Costa, Ashley M. Buckle, Martino L. di Salvo, et al. “How Pyridoxal 5′-Phosphate Differentially Regulates Human Cytosolic and Mitochondrial Serine Hydroxymethyltransferase Oligomeric State.” The FEBS Journal 282, no. 7 (April 1, 2015): 1225–41. https://doi.org/10.1111/febs.13211.
2. Walden, Miriam, Safi Kani Masandi, Krzysztof Pawłowski, and Elton Zeqiraj. “Pseudo-DUBs as Allosteric Activators and Molecular Scaffolds of Protein Complexes.” Biochemical Society Transactions 46, no. 2 (April 17, 2018): 453–66. https://doi.org/10.1042/BST20160268.
3. Walden, Miriam, Lei Tian, Rebecca L. Ross, Upasana M. Sykora, Dominic P. Byrne, Emma L. Hesketh, Safi K. Masandi, et al. “Metabolic Control of BRISC–SHMT2 Assembly Regulates Immune Signalling.” Nature 570, no. 7760 (June 2019): 194–99. https://doi.org/10.1038/s41586-019-1232-1.
4. Zheng, Hui, Vibhor Gupta, Jeffrey Patterson-Fortin, Sabyasachi Bhattacharya, Kanstantsin Katlinski, Junmin Wu, Bentley Varghese, et al. “A BRISC-SHMT Complex Deubiquitinates IFNAR1 and Regulates Interferon Responses.” Cell Reports 5, no. 1 (October 17, 2013): 180–93. https://doi.org/10.1016/j.celrep.2013.08.025.