Context-dependent contribution of peptidyl arginine deiminase 4 (PAD4) to neutrophil extracellular trap formation and liver injury in acute and chronic hepatotoxicant challenge

Abstract

Neutrophils play a complex role in the pathogenesis of chronic liver disease and have been linked to both liver damage and injury resolution. Recent reports propose that neutrophils drive liver injury and fibrosis through the formation of neutrophil extracellular traps (NETs). This study tests the hypothesis that the enzyme peptidyl arginine deiminase-4 (PAD4) drives NET formation and liver fibrosis in experimental chronic liver injury. Wild-type (PAD4+/+) and PAD4-deficient (PAD4-/-) mice were chronically challenged twice weekly with carbon tetrachloride (CCl4, 1 ml/kg, i.p) or vehicle (corn oil) for 6 weeks, and samples were collected 24 h after the final challenge. In separate studies, mice were challenged once, and samples were collected 24 to 48 h later. Circulating NET biomarkers (e.g. myeloperoxidase-DNA complexes) were elevated in chronic CCl4-challenged wild-type mice compared to vehicle, though surprisingly, intrahepatic NETs were rarely observed. In contrast to our hypothesis, PAD4 deficiency did not eliminate circulating NET markers in chronic challenge. Furthermore, PAD4 deficiency did not impact liver fibrosis assessed by picrosirius red labeling or the myofibroblast marker α-smooth muscle actin but caused a modest, sex-specific decrease in hepatic collagen type I immunolabeling. Interestingly, plasma NET biomarkers and intrahepatic NETs were both increased following acute CCl4 challenge in a PAD4-dependent manner. Furthermore, PAD4 deficiency reduced coagulation activity after 24 h and decreased hepatocellular necrosis 48 h after challenge. Our studies ultimately suggest that PAD4 affects liver injury uniquely, depending on the stage of disease and that mechanisms of NET formation may occur independent of PAD4 in chronic liver injury.

Keywords: hepatic; hepatotoxicity; inflammation; liver; pathology.

Fig. 1.

Liver injury, fibrosis, and activation of the coagulation cascade in wild-type mice following chronic carbon tetrachloride (CCl4) challenge. Male (A to D) and female (E to H) C57BL/6J mice were challenged twice weekly for 6 weeks (i.p.) with 1 ml/kg CCl4 or vehicle (corn oil). Serum, citrated plasma, and liver tissues were collected 24 h after the final injection. (A, E): Serum alanine aminotransferase (ALT) activity. (B, F): Plasma thrombin–antithrombin (TAT) complexes. (C, G): Area of picrosirius red staining quantified in formalin-fixed paraffin-embedded liver tissues represented as a percentage of area of the left lateral lobe. Representative low magnification images are shown in (D, H). Data are presented as mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

Fig. 2.

Evaluation of circulating and intrahepatic neutrophil extracellular trap (NET) biomarkers in toxicant-induced chronic liver injury. Male (A to H) and female (I to P) C57BL/6J mice were challenged twice weekly for 6 weeks (i.p.) with 1 ml/kg CCl₄ or vehicle (corn oil). Serum, citrated plasma, and liver tissues were collected 24 h after the final injection. (A, I): Plasma levels of myeloperoxidase (MPO)–DNA complexes. (B, J): Hepatic Padi4 gene expression. (C, K): Representative photomicrographs showing intrahepatic NETs in CCl₄-challenged mice, detected by co-localization of Ly6G (green), citrullinated histone H3 (CitH3, magenta), and DNA (DAPI, blue). Scale bar = 20 µm. (D, L): Intrahepatic neutrophils (Ly6G+ cells) were quantified and represented as a percentage of total cell detections. (E, M): CitH3+ neutrophils were quantified and represented as a percentage of total cells detected. Pearson correlation coefficients were calculated for MPO–DNA complexes vs. serum ALT (F, N), hepatic picrosirius red (PSR) area (G, O), and plasma TAT (H, P) for CCl₄-challenged mice. Pearson correlation coefficient (r) and P value (P) are shown in upper left of each plot. Data were fit with simple linear regression, and line of best fit with 95% CI is shown. Data are presented as mean ± SEM (n = 3 to 10/group). *P < 0.05, **P < 0.01.

Fig. 3.

Impact of peptidyl arginine deiminase-4 (PAD4) deficiency on chronic toxicant-induced hepatocellular necrosis, coagulation activation, and hepatic fibrosis. Male (A to F) and female (G to L) wild-type (PAD4^+/+) and PAD4-deficient (PAD4^−/−) mice were challenged twice weekly for 6 weeks (i.p.) with 1 ml/kg CCl₄. Serum, citrated plasma, and liver tissues were collected 24 h after the final injection. (A, B): Serum alanine aminotransferase (ALT) activity. (B, H): Plasma thrombin–antithrombin (TAT) complex levels. (C, I) Representative photomicrographs of picrosirius red (PSR) staining captured under polarized light (top), collagen type I (Col I) immunohistochemistry (middle, red), and α-smooth muscle actin (αSMA, bottom, brown) in formalin-fixed paraffin-embedded liver tissues. Positive labeling was quantified in (D to F, J to L) and represented as a percentage of the total area of the entire left lateral lobe. Data are presented as mean ± SEM (n = 5 to 7/group). *P < 0.05.

Fig. 4.

Impact of peptidyl arginine deiminase-4 (PAD4) deficiency on circulating NET biomarkers in toxicant-induced chronic liver injury. Male (A, B) and female (C, D) wild-type (PAD4^+/+) and PAD4-deficient (PAD4^−/−) mice were challenged twice weekly for 6 weeks (i.p.) with 1 ml/kg CCl₄. Citrated plasma was collected 24 h after the final injection. (A, C): Plasma levels of citrullinated histone H3 (CitH3). (B, D): Plasma levels of myeloperoxidase (MPO)–DNA complexes. The mean ± SD of vehicle-treated mice from Fig. 2A and I is shown in the gray shaded area for comparison. Data are presented as mean ± SEM (n = 5 to 6/group). **P < 0.01, ****P < 0.0001.

Fig. 5.

Hepatocellular necrosis and activation of the coagulation cascade in wild-type mice following acute carbon tetrachloride (CCl₄) challenge. Male (A, B) and female (C, D) C57BL/6J mice were challenged once (i.p.) with 1 ml/kg CCl₄ or vehicle (corn oil). Serum, citrated plasma, and liver tissues were collected 24 h later. (A, C): Serum alanine aminotransferase (ALT) activity. (B, D): Plasma thrombin–antithrombin (TAT) complexes. Data are presented as mean ± SEM. **P < 0.01, ****P < 0.0001.

Fig. 6.

Evaluation of circulating and intrahepatic neutrophil extracellular trap (NET) biomarkers in toxicant-induced acute liver injury. Male (A to G) and female (H to N) C57BL/6J mice were challenged once (i.p.) with 1 ml/kg CCl₄ or vehicle (corn oil). Serum, citrated plasma, and liver tissues were collected 24 h later unless noted otherwise. Plasma levels of circulating NET biomarkers were measured, including nucleosomes (A, H), citrullinated histone H3 (CitH3, B, I), and myeloperoxidase (MPO)–DNA complexes (C, J). (D, K): Hepatic Padi4 gene expression detected 2 to 96 h after CCl₄ (or vehicle) challenge. Black circle = vehicle, red square = CCl₄. (E, L): Representative photomicrographs showing intrahepatic NETs in vehicle- or CCl₄-challenged mice, detected by co-localization of Ly6G (green), citrullinated histone H3 (CitH3, magenta), and DNA (DAPI, blue). Scale bar = 5 µm. (F, M): Intrahepatic neutrophils (Ly6G+ cells) were quantified and represented as a percentage of total cell detections. (G, N) CitH3+ neutrophils (NETting neutrophils) were quantified and represented as a percentage of total cells detected. Data are presented as mean ± SEM (n = 3 to 9/group).*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Fig. 7.

Association of circulating neutrophil extracellular trap (NET) biomarkers with circulating biomarkers of hepatocellular necrosis and coagulation cascade activation following acute toxicant challenge. Male (A to F) and female (G to L) C57BL/6J mice were challenged once (i.p.) with 1 ml/kg CCl₄. Serum and citrated plasma were collected after 24 h. Pearson correlation coefficients were calculated for NET biomarkers (MPO–DNA complexes [A, D, G, J], CitH3 [B, E, H, K], and nucleosomes [C, F, I, L]) vs. serum ALT (A to C, G to I) or plasma TAT (D to F, J to L). Pearson correlation coefficient (r) and P value (P) are shown in upper left of each plot. Data were fit with simple linear regression, and line of best fit with 95% CI is shown. *P < 0.05.

Fig. 8.

Impact of peptidyl arginine deiminase-4 (PAD4) deficiency on circulating NET biomarkers and intrahepatic NET formation in experimental acute liver injury. Male (A to E) and female (F to J) wild-type (PAD4^+/+) and PAD4-deficient (PAD4^−/−) mice were challenged once (i.p.) with 1 ml/kg CCl₄. Citrated plasma was collected after 24 h. Circulating NET biomarkers, including citrullinated histone H3 (CitH3, A, F), myeloperoxidase (MPO)–DNA complexes (B, G), and nucleosomes (C, H) were measured. For B and G, the mean ± SD of MPO–DNA complex levels in vehicle-treated mice from Fig. 6C and J is shown in the gray shaded area for comparison. (D, I): Intrahepatic neutrophils (Ly6G+ cells) were quantified and represented as a percentage of total cell detections. (E, J): Representative photomicrographs showing intrahepatic NETs in CCl₄-challenged mice, detected by co-localization of Ly6G (green), citrullinated histone H3 (CitH3, magenta), and DNA (DAPI, blue). Scale bar = 20 µm. Data are presented as mean ± SEM (n = 3 to 10/group). *P < 0.05, **P < 0.01.

Fig. 9.

Impact of peptidyl arginine deiminase-4 (PAD4) deficiency on initial CCl₄-induced acute injury and coagulation activation. Male (A to D) and female (E to G) wild-type (PAD4^+/+) and PAD4-deficient (PAD4^−/−) mice were challenged once (i.p.) with 1 ml/kg CCl₄. Serum, citrated plasma, and liver tissues were collected 24 h after the final injection. (A, E): Serum alanine aminotransferase (ALT) activity. (B, F): Plasma thrombin–antithrombin (TAT) complex levels. (C): The area of hepatocellular necrosis was quantified in hematoxylin and eosin (H&E)-stained formalin-fixed paraffin-embedded (FFPE) liver tissues and is represented as a percentage of the total area of the entire left lateral lobe. (D, G): Representative photomicrographs of H&E-stained FFPE liver tissues illustrating centrilobular hepatocellular necrosis (arrow heads). Data are presented as mean ± SEM (n = 10 to 19/group). **P < 0.01.

Fig. 10.

Impact of peptidyl arginine deiminase-4 (PAD4) deficiency on peak CCl₄-induced acute injury and hepatic stellate cell activation. Male (A to F) and female (G to L) wild-type (PAD4^+/+) and PAD4-deficient (PAD4^−/−) mice were challenged once (i.p.) with 1 ml/kg CCl₄. Serum, citrated plasma, and liver tissues were collected 48 h after the final injection. (A, G): Serum alanine aminotransferase (ALT) activity. (B, H): Plasma thrombin–antithrombin (TAT) complex levels. (C, I): The area of hepatocellular necrosis was quantified in hematoxylin and eosin (H&E)-stained formalin-fixed paraffin-embedded (FFPE) liver tissues and is represented as a percentage of the total area of the entire left lateral lobe. Representative photomicrographs of H&E-stained liver tissues (D, J) are shown. (E, K): Representative photomicrographs of immunohistochemical labeling for α-smooth muscle actin (αSMA) in FFPE liver tissues (brown). (F, L): The area of positive αSMA labeling was quantified and is represented as a percentage of the total area of the entire left lateral lobe. Data are presented as mean ± SEM (n = 6 to 16/group). **P < 0.01, ****P < 0.0001.