Impaired protein hydroxylase activity causes replication stress and developmental abnormalities in humans
Sally C. Fletcher, Charlotte Hall, Tristan J. Kennedy, Sander Pajusalu, Monica H. Wojcik, Uncaar Boora, Chan Li, Kaisa Teele Oja, Eline Hendrix, Christian A.E. Westrip, Regina Andrijes, Sonia K. Piasecka, Mansi Singh, Mohammed E. El-Asrag, Anetta Ptasinska, Vallo Tillmann, Martin R. Higgs, Deanna A. Carere, Andrew D. Beggs, John Pappas, Rachel Rabin, Stephen J. Smerdon, Grant S. Stewart, Katrin Õunap, Mathew L. Coleman
Sally C. Fletcher, Charlotte Hall, Tristan J. Kennedy, Sander Pajusalu, Monica H. Wojcik, Uncaar Boora, Chan Li, Kaisa Teele Oja, Eline Hendrix, Christian A.E. Westrip, Regina Andrijes, Sonia K. Piasecka, Mansi Singh, Mohammed E. El-Asrag, Anetta Ptasinska, Vallo Tillmann, Martin R. Higgs, Deanna A. Carere, Andrew D. Beggs, John Pappas, Rachel Rabin, Stephen J. Smerdon, Grant S. Stewart, Katrin Õunap, Mathew L. Coleman
View: Text | PDF
Research Article Cell biology Genetics

Impaired protein hydroxylase activity causes replication stress and developmental abnormalities in humans

Abstract

Although protein hydroxylation is a relatively poorly characterized posttranslational modification, it has received significant recent attention following seminal work uncovering its role in oxygen sensing and hypoxia biology. Although the fundamental importance of protein hydroxylases in biology is becoming clear, the biochemical targets and cellular functions often remain enigmatic. JMJD5 is a “JmjC-only” protein hydroxylase that is essential for murine embryonic development and viability. However, no germline variants in JmjC-only hydroxylases, including JMJD5, have yet been described that are associated with any human pathology. Here we demonstrate that biallelic germline JMJD5 pathogenic variants are deleterious to JMJD5 mRNA splicing, protein stability, and hydroxylase activity, resulting in a human developmental disorder characterized by severe failure to thrive, intellectual disability, and facial dysmorphism. We show that the underlying cellular phenotype is associated with increased DNA replication stress and that this is critically dependent on the protein hydroxylase activity of JMJD5. This work contributes to our growing understanding of the role and importance of protein hydroxylases in human development and disease.

Authors

Sally C. Fletcher, Charlotte Hall, Tristan J. Kennedy, Sander Pajusalu, Monica H. Wojcik, Uncaar Boora, Chan Li, Kaisa Teele Oja, Eline Hendrix, Christian A.E. Westrip, Regina Andrijes, Sonia K. Piasecka, Mansi Singh, Mohammed E. El-Asrag, Anetta Ptasinska, Vallo Tillmann, Martin R. Higgs, Deanna A. Carere, Andrew D. Beggs, John Pappas, Rachel Rabin, Stephen J. Smerdon, Grant S. Stewart, Katrin Õunap, Mathew L. Coleman

×

Figure 3

Patient pathogenic variants reduce JMJD5 protein stability and activity.

Options: View larger image (or click on image) Download as PowerPoint
Patient pathogenic variants reduce JMJD5 protein stability and activity....
(A) C123Y and JMJD5Δ332–362 variants reduce JMJD5 expression. Western blotting following transfection of HA-tagged JMJD5 cDNAs and a transfection control (FLAG-JMJD7) into HeLa cells. (B) C123Y and JMJD5Δ332–362 are more rapidly degraded. Doxycycline-inducible FLAG-tagged WT, C123Y, or JMJD5Δ332–362 cDNAs were introduced into Sib1WT/WT fibroblasts. Cells were treated with 1 μg/mL doxycycline for 16 hours and then 50 μg/mL cycloheximide, to inhibit protein synthesis. Because the C123Y and JMJD5Δ332–362 proteins were expressed markedly less than WT (consistent with A), longer exposures are included for the C123Y and JMJD5Δ332–362 insets (top middle and top right) to support a more direct comparison of the three JMJD5 species. (C) Reduced JMJD5 expression in fibroblasts derived from an affected patient. Western blotting for endogenous JMJD5 in immortalized fibroblast cell lines. “Low” and “high” refer to the exposure length. Also see Supplemental Figure 20A. (D) Partially purified recombinant JMJD5 C123Y has reduced hydroxylase activity in vitro. GST-tagged WT and C123Y JMJD5 were expressed in E. coli and purified before in vitro activity assays. Each reaction was also analyzed by Coomassie gel (left). Note the presence of a smaller product (~65 kDa) in the C123Y sample. This may be a cleavage product or chaperone contamination, perhaps indicative of improper folding. The amount of recombinant JMJD5 sample added to the activity assay and Coomassie gel was equalized. Activity was monitored using the Succinate-Glo assay, which measures succinate production (right). Data represent mean ± SEM from 3 independent experiments. Statistical analysis used 1-way ANOVA with Tukey’s post hoc test, with P ≤ 0.05 considered significant; **P ≤ 0.01.