The proinflammatory role of HECTD2 in innate immunity and experimental lung injury

Abstract

Invading pathogens may trigger overactivation of the innate immune system, which results in the release of large amounts of proinflammatory cytokines (cytokine storm) and leads to the development of pulmonary edema, multiorgan failure, and shock. PIAS1 is a multifunctional and potent anti-inflammatory protein that negatively regulates several key inflammatory pathways such as Janus kinase (JAK)-signal transducer and activator of transcription (STAT) and nuclear factor κB (NF-κB). We discovered a ubiquitin E3 ligase, HECTD2, which ubiquitinated and mediated the degradation of PIAS1, thus increasing inflammation in an experimental pneumonia model. We found that GSK3β phosphorylation of PIAS1 provided a phosphodegron for HECTD2 targeting. We also identified a mislocalized HECTD2 polymorphism, HECTD2(A19P), that was present in 8.5% of the population and functioned to reduce inflammation. This polymorphism prevented HECTD2/PIAS1 nuclear interaction, thus preventing PIAS1 degradation. The HECTD2(A19P) polymorphism was also protective toward acute respiratory distress syndrome (ARDS). We then developed a small-molecule inhibitor, BC-1382, that targeted HECTD2 and attenuated lipopolysaccharide (LPS)- and Pseudomonas aeruginosa-induced lung inflammation. These studies describe an unreported innate immune pathway and suggest that mutation or antagonism of the E3 ligase HECTD2 results in reduced severity of lung inflammation by selectively modulating the abundance of the anti-inflammatory protein PIAS1.

Fig. 1. HECTD2 targets PIAS1 for polyubiquitination, thereby increasing NF-κB signaling

(A) Endogenous PIAS1 protein half-life study with MG132 or leupeptin (n = 3). CHX, cycloheximide. (B) Immunoblotting showing levels of endogenous PIAS proteins and V5-HECTD2 after ectopic HECTD2 plasmid expression (n = 3). (C) Endogenous PIAS1 protein half-life determination with empty plasmid or HECTD2 expression (top panel); endogenous PIAS1 protein half-life determination with control (CON) shRNA or HECTD2 shRNA expression (bottom panel) (n = 3). (D) In vitro ubiquitination assay. Purified E1 and E2 components were incubated with V5-PIAS1, HECTD2, and the full complement of ubiquitination reaction components (second lane) showing polyubiquitinated PIAS1 proteins (n = 2). (E to G) NF-κB promoter activity assays. 293T cells were cotransfected with Cignal NF-κB dual luciferase reporter plasmids along with empty or HECTD2 plasmid or control shRNA, HECTD2 shRNA, or PIAS1 shRNA 24 hours before HECTD2 plasmid expression. Twenty-four hours later, cells were treated with LPS (10 µg/ml), TNF (10 ng/ml), or IFN-γ (10 ng/ml) for an additional 6 hours (F and G) or 18 hours. Cells were then collected and assayed for luciferase activity to evaluate NF-κB promoter activity. Means ± SEM (n = 3). (H) MLE cells were treated with LPS in a time- or dose-dependent manner; cells were then collected and assayed for HECTD2, PIAS, and actin immunoblotting. Endogenous HECTD2 was also immunoprecipitated (IP) and followed by PIAS1 and PIAS4 immunoblotting (n = 2). Source data in fig. S10 and table S2.

Fig. 2. PIAS1 phosphorylation is required for HECTD2 targeting

(A) Endogenous PIAS1 was immunoprecipitated and followed by GSK3β and phosphoserine antibody immunoblotting (IB) (n = 2). IgG, immunoglobulin G. (B and C) MLE cells were treated with LPS in a time- or dose-dependent manner; cells were then collected and assayed for HECTD2, PIAS, and actin immunoblotting. Endogenous PIAS1 was also immunoprecipitated and followed by GSK3β, phosphoserine, and phosphothreonine immunoblotting (n = 2). (D) Endogenous PIAS1 protein half-life determination with control shRNA or GSK3β shRNA overexpression (n = 3). (E) MLE cells were transfected with increasing amounts of WT or constitutively activated GSK3β hypermutant plasmids for 18 hours before PIAS1 immunoblotting. The arrow indicates the overexpressed GSK3β (n = 2). (F) Endogenous PIAS1 protein half-life determination with WT GSK3β or hyperactive GSK3β plasmid overexpression. Arrow indicates the overexpressed GSK3β (n = 2). (G) Immunoblotting showing levels of endogenous GSK3β and PIAS1 protein in MLE cells transfected with either control shRNA or GSK3β shRNA followed by LPS treatment. Endogenous PIAS1 was also immunoprecipitated and followed by phosphoserine and phosphothreonine immunoblotting. (H) In vitro GSK3β kinase assay using PIAS1 as a substrate. GSK3β*, heat-inactivated GSK3β (n = 2). (I) Protein half-life of WT, S13A, and S17A PIAS1 (n = 3). (J) MLE cells were transfected with WT, S13A, or S17A PIAS1 before being treated with LPS for up to 6 hours. Cells were then collected and assayed for V5-PIAS1. Overexpressed V5-PIAS1 was also immunoprecipitated using V5 antibody and followed by phosphoserine immunoblotting (n = 2). (K) Four biotin-labeled peptides were prebound to streptavidin and served as the bait for HECTD2 or GSK3β binding. After washing, proteins were eluted and processed for V5-HECTD2 or GSK3β immunoblotting (n = 2). Source data in fig. S11.

Fig. 3. HECTD2 contains a naturally occurring polymorphism at A19, which mislocalizes to the cytosol

(A) In vitro ubiquitination assay using V5-PIAS1 as the substrate and HECTD2^WT and HECTD2^A19P as the E3 ligase (n = 3). (B) Endogenous PIAS1 protein was immunoprecipitated from cell lysate using PIAS1 antibody and coupled to proteinA/G beads. PIAS1 beads were then incubated with in vitro synthesized products expressing his-V5-HECTD2 mutants (top). After washing, proteins were eluted and processed for V5-HECTD2 immunoblotting (bottom) (n = 2). (C) MLE cells were transfected with increasing amounts of WT or A19P HECTD2 plasmids for 18 hours before being assayed for PIAS1 protein immunoblotting (n = 3). (D and E) MLE cells were transfected with an inducible HECTD2^WT or HECTD2^A19P plasmid under control of exogenous doxycycline. Cells were treated with doxycycline for various times and dose. Cells were then collected, and cell lysates were analyzed for HECTD2 and PIAS1 by immunoblotting (n = 2). (F) Half-life study of endogenous PIAS1 uponHECTD2^WT orHECTD2^A19P overexpression (n=2). (G) 293Tcells were cotransfected with Cignal NF-κB reporter plasmids and empty, HECTD2^WT, or HECTD2^A19P plasmid for 18 hours. Cells were then exposed to LPS or TNF for additional 18 hours before being assayed for NF-κB promoter activity. Means ± SEM (n = 3). (H) MLE cells were cotransfected with CFP-PIAS1 and YFP-HECTD2^WT or YFP-HECTD2^A19P. YFP-HECTD2^WT and YFP-HECTD2^A19P nuclear/cytosol fluorescent signals were measured and quantified (n > 20, right graph, *P < 0.0001 compared to WT). Cells were also subjected to irreversible photobleaching of YFP acceptor signal using a 514-nm laser. The emission fluorescence levels of both the donor CFP-PIAS1 and acceptor YFP-HECTD2^WT or YFP-HECTD2^A19P before and after acceptor photobleaching are shown in the lower panel; the region of interest across the nucleus is marked with a red arrow. The green arrow on the graph indicates donor CFP-PIAS1 signal before and after the photobleaching. Scale bar, 10 µm (n > 10). Source data in fig. S12 and table S3.

Fig. 4. HECTD2^A19P is a loss-of-function E3 ligase polymorphism in vivo

C57BL/6J mice were administered intratracheally with Lenticontrol, Lenti- HECTD2^WT, or Lenti-HECTD2^A19P [10⁷ plaque-forming units (PFU) per mouse] for 144 hours, and mice were inoculated intratracheally with PA103 (10⁴ PFU per mouse) for 18 hours. Mice were then euthanized, and lungs were lavaged with saline, harvested, and then homogenized. (A to C) Lavage protein, cell count, and bacteria count were measured. BAL, bronchoalveolar lavage. (D to F) Lavage cytokine secretion was measured. IL-6, interleukin-6. (G) Hematoxylin and eosin (H&E) staining was performed on lung samples from (A). Scale bar, 100 µm; original magnification, ×10. (H) PIAS1, HECTD2, and actin immunoblots from homogenized lung samples. Data are average of two experiments (Student’s t test on means ± SEM). (A to G) n = 5 to 12 mice per group. Source data in fig. S13 and table S4.

Fig. 5. PIAS1 knockdown induces lung injury in vivo

C57BL/6J mice were infected intratracheally with Lenti-control shRNA or Lenti-PIAS1 shRNA (10⁷ PFU per mouse) for 144 hours, and mice were inoculated with LPS (intratracheally, 3 mg/kg) for 18 hours. Mice were euthanized, and lungs were lavaged with saline, harvested, and then homogenized. (A to C) Lavage protein, cell count, and lung protein immunoblots were measured. (D to F) Lavage cytokine secretion was measured. (G) H&E staining was performed on lung samples from (A). Scale bar, 100 µm; original magnification, ×10 (Student’s t test on means ± SEM). (A to G) n = 5 to 7 mice per group. Source data in fig. S13 and table S5.

Fig. 6. HECTD2 knockdown ameliorates Pseudomonas-induced lung injury in vivo

C57BL/6J mice were infected intratracheally with Lenti-control shRNA or Lenti-HECTD2 shRNA (10⁷ PFU per mouse) for 144 hours, and mice were inoculated with PA103 [10⁴ colony-forming units (CFU) per mouse] for 18 hours. Mice were euthanized, and lungs were lavaged with saline, harvested, and then homogenized. (A to C) Lavage protein, cell count, and bacteria count were measured. (D to F) Lavage cytokine secretion was measured. (G) Survival studies of mice that were administered PA103 (intratracheally, 10⁵ PFU permouse, n = 8 mice per group) were determined (time, hours). Mice were carefully monitored over time; moribund, preterminal animals were immediately euthanized and recorded as deceased. Kaplan-Meier survival curves were generated using SPSS software (P=0.001). (H) H&E staining was performed on lung samples from (A). (I) PIAS1, HECTD2, and actin immunoblots from homogenized lung samples. Scale bar, 100 µm; original magnification, ×10. Data are average of two experiments (Student’s t test on means ± SEM). (A to G) n = 6 to 9 mice per group. NS, not significant. Source data in fig. S13 and table S6.

Fig. 7. Anti-inflammatory activity of a HECTD2 small-molecule inhibitor

(A) Structural analysis of the HECTD2-HECT domain revealed a major cavity within the C terminus of the HECT domain. (B) Docking study of candidate inhibitor BC-1382 with the HECTD2-HECT domain. (C) HECTD2 protein was immunoprecipitated using HECTD2 antibody and captured with protein A/G beads from MLE lysates. HECTD2 beads were extensively washed before exposure to BC-1382 at different concentrations (10⁻⁴ to 10 µM). Purified PIAS1 protein was then incubated with drug bound to HECTD2 beads overnight. Beads were washed, and proteins were eluted and resolved on SDS–polyacrylamide gel electrophoresis. The relative amounts of PIAS1 detected in the pull-downs were normalized to loading and quantified (n = 2). (D to H) C57BL/6J mice were infected intratracheally with PA103 (10⁴ PFU per mouse). BC-1382 was given through intraperitoneal injection (10 mg/kg) at the same time. Eighteen hours later, mice were sacrificed, and lungs were lavaged with saline, harvested, and then homogenized. Lavage bacterial count, protein, cell count, and cytokine secretion were measured in (D) to (G). (H) H&E staining was performed on lung samples (scale bar, 100 µm; original magnification, ×10). (I) PIAS1, HECTD2, and actin immunoblots from homogenized lung samples. Data are average of two experiments (Student’s t test on means ± SEM). (D to H) n = 5 to 10 mice per group. Source data in fig. S13 and table S7.

Fig. 8. One mechanism by which microbial infection or stimuli can robustly trigger inflammation is to decrease anti-inflammatory proteins such as PIAS1 in cells

Specifically, during microbial infection, HECTD2 (R397) targets PIAS1 for its ubiquitination at K30; this process is facilitated by GSK3β phosphorylation of PIAS1 at S17. In this pathway, WT HECTD2 potently activates cytokine-driven inflammation, whereas we identified a naturally occurring, mislocalized, hypofunctional variant of HECTD2 (A19P) that lowers the amount of cytokine secretion. A small-molecule HECTD2 inhibitor, BC-1382, decreased inflammation in an animal model of ARDS by antagonizing the actions of HECTD2 on PIAS1, allowing PIAS1 to continue to suppress cytokine signaling.