Introduction
Shigella flexneri is a gram-negative bacterial pathogen that causes communicable human bacillary dysentery (Trofa et al., 1999). During infection, S. flexneri manipulate various host processes by delivering effector proteins into host cells via its type III secretion system (T3SS) to promote development of its intracellular replicative niche and intercellular transmission (Mattock et al., 2017). Bacterial effectors are always found to be enzymes that often target host substrates for post-translational modifications (PTM) (Chambers et al., 2020). Among S. flexneri T3SS effectors, OspC family proteins, including OspC1, OspC2, and OspC3, have recently been demonstrated to possess ADP-riboxanase activity and catalyze ADP-riboxanation (541.06 Da), a derivative of ADP-ribosylation (524.03 Da) (Li et al., 2021). In the same study, human caspase-4 and mouse caspase-11 were identified to be the host targets of OspC3, which blocks the host pyroptosis-mediated defence. However, the host targets of its paralogues, OspC1 and OspC2, are still unknown and remain to be further investigated.
Accumulating evidence indicates that secreted bacterial nucleic acids could also be recognized by the traditionally believed cytosolic viral RNA and DNA sensors RIG-I, MDA5, and STING thereby triggering IFN-β production and anti-bacterial defence (Abdullah et al., 2012; Dobbs et al., 2015). To establish a vacuolar niche inside host cells, bacterial pathogens have evolved effectors to hijack these innate immune pathways. For instance, the Coxiella burnetii effector EmcB acts as a deubiquitinase and thwarts host RIG-I signaling by preferentially cleaving its K63-linked ubiquitin chains (Duncan-Lowey et al., 2023), whereas the S. flexneri effector IpaJ blocks STING translocation from endoplasmic reticulum (ER) to ER-Golgi intermediate compartments (ERGIC) thereby inhibiting its activation (Dobbs et al., 2015).
In the present study, we identified HERC5 as a host substrate of ADP-riboxanation catalyzed by S. flexneri effector OspC1 using a previously described strategy involving eAf1521 enrichment and mass spectrometry (MS)-based proteomics (Jin et al., 2023). HERC5 is highly inducible upon type I interferon stimulation and bacterial infection (Han et al., 2018), and functions as an E3 ISG15 ligase catalyzing the ISGylation of proteins (i.e., RIG-I, MDA5, and IRF3) implicated in innate immunity (Kim et al., 2008; Liu et al., 2021; Shi et al., 2010). Given that ISG15 and ISGylation themselves counteract Listeriamonocytogenes and mycobacteria infection via different mechanisms (Radoshevich et al., 2015; Swaim et al., 2017), we proposed thatS. flexneri effector OspC1 may subvert host anti-bacterial defence by targeting HERC5 for ADP-riboxanation and thus inhibiting HERC5-catalyzed ISGylation. However, to our surprise, OspC1-catalyzed HERC5 ADP-riboxanation promotes its E3 ISG15 ligase activity and the overall ISGylation levels in the whole cell lysates slightly. To understand this confusing experimental result, we extensively reviewed the related literatures in detail, and found that not all ISGylation positively regulate host innate immune response. Indeed, HERC5-catalyzed ISGylation of MDA5 and IRF3 promote the host antiviral immunity (Liu et al., 2021; Shi et al., 2010), whereas HERC5-catalyzed RIG-I ISGylation induced by Type I interferon negatively regulates RIG-I-mediated antiviral signaling as a feedback mechanism (Kim et al., 2008). It is reasonable that HERC5 targets different proteins for ISGylaton to positively and negatively regulate the host immunity precisely and thus avoid excessive inflammatory damage. Inspired by this finding, we determined two documented HERC5-catalyzed ISGylation substrates (i.e., HERC5 itself and RIG-I) with ISG15-conjugating system in the presence of wild-type or catalytically mutated OspC1 (Wong et al., 2006; Kim et al., 2008; Zhao et al., 2005), and found that OspC1-mediated HERC5 ADP-riboxanation upregulates the ISGylation of HERC5 and RIG-I, resulting in the downregulation of host anti-Shigella defence.