Microparticulate Caspase 1 Regulates Gasdermin D and Pulmonary Vascular Endothelial Cell Injury

Abstract

Lung endothelial cell apoptosis and injury occur throughout all stages of acute lung injury/acute respiratory distress syndrome and impact disease progression. Caspases 1, 4, and 5 are essential for completion of the apoptotic program known as pyroptosis that also involves proinflammatory cytokines. Because gasdermin D (GSDMD) mediates pyroptotic death and is essential for pore formation, we hypothesized that it might direct caspase 1-encapsulated microparticle (MP) release and mediate endothelial cell death. Our present work provides evidence that GSDMD is released by LPS-stimulated THP-1 monocytic cells, where it is packaged into microparticles together with active caspase 1. Furthermore, only MP released from stimulated monocytic cells that contain both cleaved GSDMD and active caspase 1 induce endothelial cell apoptosis. MPs pretreated with caspase 1 inhibitor Y-VAD or pan-caspase inhibitor Z-VAD do not contain cleaved GSDMD. MPs from caspase 1-knockout cells are also deficient in p30 active GSDMD, further confirming that caspase 1 regulates GSDMD function. Although control MPs contained cleaved GSDMD without caspase 1, these fractions were unable to induce cell death, suggesting that encapsulation of both caspase 1 and GSDMD is essential for cell death induction. Release of microparticulate active caspase 1 was abrogated in GSDMD knockout cells, although cytosolic caspase 1 activation was not impaired. Last, higher concentrations of microparticulate GSDMD were detected in the plasma of septic patients with acute respiratory distress syndrome than in that of healthy donors. Taken together, these findings suggest that GSDMD regulates the release of microparticulate active caspase 1 from monocytes essential for induction of cell death and thereby may play a critical role in sepsis-induced endothelial cell injury.

Figure 1.

Cleaved gasdermin D (GSDMD) colocalizes with active caspase 1 (Casp1) in microparticles (MPs) released from LPS-stimulated THP-1 cells. THP-1 cells were cultured at a concentration of 50 million per milliliter and stimulated with LPS (1 μg/ml) for 2 hours or left untreated. Casp1 inhibitor Y-VAD and pan-caspase inhibitor Z-VAD were used both before and after LPS treatment as indicated. DMSO and Z-FA were used as inhibitor controls. MPs were isolated from each condition and normalized by total protein for analysis. (A) Immunoblots of GSDMD, p20 caspase 1, and ASC from MPs after various treatments. Densitometric scan analysis using ImageJ software (NIH) with immunoblots is representative of four experiments. (B) Cells and MP fractions were analyzed by fluorescence microscopy for colocalization of GSDMD with MPs recovered from control and LPS-treated THP-1 cells. Cells and MPs were stained with DilC16 (lipid stain, red), GSDMD (green), and DAPI (blue). (C) Colocalization of Casp1 and GSDMD in MPs visualized by fluorescence microscopy with DiLC16 (red), GSDMD (green), and Casp1 (white). Quantification of colocalization of GSDMD with MPs was calculated as the percentage of the number of MPs colocalized with GSDMD among the total number of MPs in each field. Quantification of colocalization of Casp1 with GSDMD in MPs was calculated as the percentage of the number of MPs colocalized with Casp1 and GSDMD among the total number of MPs in each field. (D) OptiPrep fractions of control and LPS MPs were analyzed for p30 GSDMD and active p20 Casp1. Casp1 activity in each fraction was determined by fluorimetry of WEHD-afc cleavage. (E) Control and LPS MPs were further fractionated into membrane (M) and MP content (C). Both membrane and content fractions from control and LPS MPs were then analyzed by immunoblotting for p30 GSDMD, Casp1, ASC, LAMP, Hsp90, and Na⁺/K⁺-ATPase. *P < 0.05; arrow shows colocalization, inset is 100× magnification. Data are representative of four experiments. AFU = arbitrary fluorescence units.

Figure 2.

Microparticulate p30 GSDMD release and human pulmonary microvascular endothelial cell (HPMVEC) death is regulated by caspase 1. (A) MPs were isolated from THP-1 stimulated with LPS (1 μg/ml) for 2 hours or left untreated and then applied to HPMVEC, which were analyzed for cell viability using a 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay. Non-MP fractions and OptiPrep fractions 7–9 used in Figure 1 were also analyzed for HPMVEC death. (B) Cell viability using the MTS assay was compared between control MP, LPS MP, and non-MP fractions in the presence or absence of Y-VAD and Z-VAD. (C) Cell extracts from THP-1, CAS9, and CAS9/Casp1-knockout (CAS9/CASP1-KO) cells stimulated with LPS in the presence or absence of Z-VAD or left untreated were analyzed for p30 GSDMD and p45 Casp1. Hsp90 was used as a loading control. MPs were then isolated from the supernatants of these cells and analyzed for GSDMD and Casp1, using Hsp90 as a loading control. Data and densitometric scan of immunoblot are representative of three experiments. (D) Casp1 activity was measured in the above fractions using a WEHD enzymatic assay. (E) Control, LPS, and LPS + Z-VAD MPs from THP-1, CAS9, and CAS9/Casp1-KO cells were then subjected to HPMVEC and analyzed for cell death using trypan blue, and cell viability was analyzed using an MTS assay. Data are representative of three experiments. CE = cell extracts; EGM2 = endothelial cell growth medium 2; O.D. = optical density; RPMI = RPMI1640.

Figure 3.

GSDMD regulates release of microparticulate active caspase 1. CE and MP fractions were isolated from CAS9 and CAS9/GSDM-knockout (CAS9/GSDMD-KO) cells in LPS (1 μg/ml) for 2 hours or were left untreated and analyzed for the presence of cleaved GSDMD and active p20 Casp1. Hsp90 was used as a loading control. Data and densitometric scan of immunoblot are representative of three experiments.

Figure 4.

p30 GSDMD is released in plasma MPs during sepsis. MPs were isolated from plasma of healthy donors (n = 6) and septic patients with acute respiratory distress syndrome on Day 1 of ICU admission (n = 7). (A) Isolated plasma MPs were analyzed for the presence of GSDMD using immunoblotting. LPS CE and LPS MPs from THP-1 cells stimulated with LPS for 2 hours were used as markers. (B) Plasma MPs from patients 1, 6, and 7, as well as LPS CE and MPs, were analyzed by immunoblotting for GSDMD using either secondary antibody alone (left panel) or standard primary and secondary antibodies (right panel). (C) Casp1 activity was measured in healthy donor and patient samples. Casp1 activity was found to be significantly higher (*P < 0.015) in septic patients than in healthy donors. HR = heart rate; P/F = Pa_O2/Fi_O2 ratio; RR = respiratory rate; SAPS II = Simplified Acute Physiology Score II; WBC = white blood cell count.