Systemic silencing of Phd2 causes reversible immune regulatory dysfunction
Atsushi Yamamoto, … , Peter J. Ratcliffe, Chris W. Pugh
Atsushi Yamamoto, … , Peter J. Ratcliffe, Chris W. Pugh
Published June 4, 2019
Citation Information: J Clin Invest. 2019;129(9):3640-3656. https://doi.org/10.1172/JCI124099.
View: Text | PDF
Research Article Immunology

Systemic silencing of Phd2 causes reversible immune regulatory dysfunction

Abstract

Physiological effects of cellular hypoxia are sensed by prolyl hydroxylase (PHD) enzymes, which regulate HIFs. Genetic interventions on HIF/PHD pathways have revealed multiple phenotypes that extend the known biology of hypoxia. Recent studies have unexpectedly implicated HIF in aspects of multiple immune and inflammatory pathways. However, such studies are often limited by systemic lethal effects and/or use tissue-specific recombination systems, which are inherently irreversible, unphysiologically restricted, and difficult to time. To study these processes better, we developed recombinant mice that expressed tetracycline-regulated shRNAs broadly targeting the main components of the HIF/PHD pathway, permitting timed bidirectional intervention. We show that stabilization of HIF levels in adult mice through PHD2 enzyme silencing by RNA interference or inducible recombination of floxed alleles results in multilineage leukocytosis and features of autoimmunity. This phenotype was rapidly normalized on reestablishment of the hypoxia-sensing machinery when shRNA expression was discontinued. In both situations, these effects were mediated principally through the Hif2a isoform. Assessment of cells bearing Treg markers from these mice revealed defective function and proinflammatory effects in vivo. We believe our findings reveal a new role for the PHD2/HIF2α pathway in the reversible regulation of T cell and immune activity.

Authors

Atsushi Yamamoto, Joanna Hester, Philip S. Macklin, Kento Kawai, Masateru Uchiyama, Daniel Biggs, Tammie Bishop, Katherine Bull, Xiaotong Cheng, Eleanor Cawthorne, Mathew L. Coleman, Tanya L. Crockford, Ben Davies, Lukas E. Dow, Rob Goldin, Kamil Kranc, Hiromi Kudo, Hannah Lawson, James McAuliffe, Kate Milward, Cheryl L. Scudamore, Elizabeth Soilleux, Fadi Issa, Peter J. Ratcliffe, Chris W. Pugh

×

Figure 4

Phenotype reversal following withdrawal of doxycycline.

Options: View larger image (or click on image) Download as PowerPoint
Phenotype reversal following withdrawal of doxycycline. (A) Representati...
(A) Representative bright-field images (scale bars: 1 cm) and wet weight of pLNs from shPhd2#9 mice and their littermate controls maintained on doxycycline for 3 to 4 weeks (doxycycline-ON group) or treated with doxycycline for 3 to 4 weeks and then analyzed 7 weeks after doxycycline withdrawal (doxycycline ON/OFF group). Data are represented as the mean ± SEM (n = 4/group). ***P < 0.001. Multigroup comparisons were analyzed by 1-way ANOVA with Tukey’s multiple comparisons post hoc test. (B) Mean percentage of BW changes of shPhd2#9 and control mice, relative to day 0 of doxycycline treatment or doxycycline withdrawal. Data are represented as the mean ± SD (n = 7–8/group). ****P < 0.0001, by 2-way ANOVA. (C) Representative fluorescence images of ANAs using mouse serum from doxycycline-ON and doxycycline-ON/OFF groups of shPhd2#9 and control mice. Scale bars: 50 μm. (D–K) H&E-stained images of (D and E) skin (original magnification, ×20); (F–I) kidney (original magnification, ×2.5 for F and G and ×10 for H and I); and (J and K) lung (original magnification, ×10) from doxycycline-ON and doxycycline-ON/OFF groups of shPhd2#9 mice. Scale bars: 500 μm (at ×2.5, and proportionately shorter lengths at higher magnifications). (L) FACS analysis of cells from pLNs from doxycycline-ON and doxycycline-ON/OFF groups of shPhd2#9 and control mice, analyzed for CD45, CD3, B220, and Gr1 expression. Data are represented as the mean ± SEM (n = at least 4/group). *P < 0.05. Unpaired, independent groups of 2 were analyzed by 2-tailed Student’s t test. Each symbol in A and L represents an individual mouse. Data were consistent over 2 independent experiments.