Targeting the Upstream Open Reading Frame of Gadd34 to Treat Sepsis‐induced Acute Kidney Injury

Abstract

Sepsis‐induced acute kidney injury (AKI) remains a major clinical problem with no effective therapies established to date. We have previously shown that bacterial sepsis causes global translation shutdown via phosphorylation of the eukaryotic translation initiation factor 2α (eIF2α). Under physiological conditions, eIF2α phosphorylation is tightly counter‐regulated by two eIF2α holophosphatases since excessive phosphorylation of eIF2α is deleterious. Of the two holophosphatases, growth arrest DNA‐inducible gene 34 (Gadd34) is the only stress‐inducible regulatory subunit. Using ribosome profiling, or Ribo‐Seq, we found that Gadd34 is translationally repressed during late phase sepsis even though eIF2α is already heavily phosphorylated. This failure of Gadd34 induction could explain the sustained phosphorylation of eIF2α and translation shutdown that contributes to delayed renal recovery in sepsis. The 5′‐untranslated region (UTR) of Gadd34 has multiple upstream open reading frames (uORFs) that are conserved across mammalian species. Our Ribo‐Seq analysis of kidneys from septic animals revealed high ribosomal occupancy of a specific Gadd34 uORF, but not the main protein coding sequence (CDS), consistent with a model in which the uORF serves as a translational inhibitor of the downstream CDS. With these findings, the objective of the investigation was to analyze the effect of disrupting translation in the uORF of Gadd34 in septic states.To investigate the inhibitory role of the Gadd34 uORF, we designed plasmid constructs, which consisted of a full length Gadd34 5′‐UTR and luciferase reporter CDS in‐frame with a single nucleotide mutation introduced to abolish the uORF start codon. The uORF point mutation led to a two‐fold increase in luciferase signal compared with the wild‐type control, confirming the inhibitory property of the Gadd34 uORF. Next, we designed antisense oligonucleotides (ASOs) complementary to the upstream portion of the uORF. Interestingly, masking the uORF with ASOs resulted in sequence‐specific increases in translation of the downstream CDS, possibly due to enhanced leaky ribosomal scanning. Finally, we tested the applicability of the ASO approach in vivo using a mouse model of endotoxin‐induced kidney injury. Despite late intervention at eight hours post‐endotoxin challenge, intravenous administration of Gadd34 uORF ASOs significantly reduced renal tissue damage as determined by the levels of hepatitis A virus cellular receptor 1/kidney injury molecule 1 (Havcr1/KIM1). Collectively, these findings indicate that translational suppression of Gadd34 in late phase sepsis is a maladaptive response that could be therapeutically modulated by targeting its uORF.Support or Funding InformationThis work was supported by NIH grants K08‐DK113223, R01‐DK080063 and a Veterans’ Affairs Merit grant 1I01BX002901.