Additional Supporting Information may be found in the online version of this article at the publisher's website.


Supporting Figure 1: (A) Drug-induced hepatotoxicity causes cytoplasmic aberrant DNA accumulation within Lamin-A coated vesicles. 3D reconstruction and outline of the cells was made using computer-assisted image analysis and confirmed that such vesicles were contained within the cytoplasm of hepatocytes. Bar = 10µm. (B) Cytoplasmic DNA accumulation (propidium iodide staining) due to APAP challenged occurs in a dose-dependent manner. Note that part of DNA is circumscribed by lamin-positive membrane (arrow head). Bar = 10µm. * P<0.05 in comparison to controls. Statistical tests as indicated in Methods.

Supporting Figure 2: Staurosporine caused a different profile of DNA vesiculation and accumulation in hepatocytes. (A) In vitro incubation with staurosporine (STA) resulted in the condensation of chromatin, nuclear shrinkage and the formation of DNA-containing apoptotic bodies (Feulgen reaction), morphological hallmarks of apoptosis. (B) Aberrant DNA accumulation in the cytoplasm was not observed following fluorescence staining of DNA (propidium iodide) and nuclear membrane. Bars = 10µm.

Supporting Figure 3: Thioacetamide, another hepatotoxic compound, also induces aberrant DNA location in hepatocytes and intravascular DNA coating. (A) Transmission electron microscopy of primary hepatocytes incubated with thioacetamide (TAA) presenting aberrant DNA in cytoplasm vesicles (V); nucleus (N). (B-D) TAA induces liver injury and neutrophil recruitment in vivo, as shown by (B) elevated serum ALT levels and (C) increased neutrophil recruitment to TAA-induced necrotic areas in the liver (n > 5/group). (D) Representative histology sections confirming severe liver injury caused by TAA. (E) Liver confocal intravital microscopy showing intravascular DNA accumulation (Sytox green staining) and neutrophil infiltration (anti-GR1 antibody; red) following TAA challenge in vivo (n > 5/group). Bar = 100µm. (F,G) Quantification of intravascular DNA coating and liver necrosis induced by TAA, as shown by intravital microscopy (n > 5/group). * P<0.05 in comparison to controls. Statistical tests as indicated in Methods.

Supporting Figure 4: TLR9 absence does not impact on APAP metabolism or aberrant DNA accumulation. Moreover, DNA is not directly cytotoxic to hepatocytes in vitro. (A) Incubation of TLR9-/- hepatocytes with acetaminophen led to aberrant extra-nuclear DNA location (propidium iodide staining) similar to that of WT cells, as shown by the fluorescence staining of DNA and nuclear membrane (anti-lamin A antibody). Bars=10µm. (B) In parallel, no differences in cytotoxicity were observed following acetaminophen incubation in WT cells plus DNase or in TLR9-/- cells (n = 6 replicates/experiment). (C) Mice that lack TLR9 had similar hepatic GSH levels in comparison to WT, and upon APAP administration, a similar GSH depletion profile was observed, confirming that the protective effects were not derived from deficient APAP metabolism. GSH concentration was estimated in liver extracts normalized to liver wet weight. (D) Direct incubation of hepatocytes with liver-extracted DNA (50µg/ml) did not result in cytotoxic effects (n = 6 replicates/experiment). (E) Serum ALT levels of WT mice in comparison to TLR9-/- mice after 2hs and 24hs of APAP administration. Bars = 10µm. * P<0.05 in comparison to controls. Statistical tests as indicated in Methods.

Supporting Figure 5: Hepatocytes do not sense self-DNA during drug-induced hepatotoxicity. (A) No activation of NF-κB (Nfkbia ) and Interferon type-I (Isg54, Ifna and Ifnb1) pathways was observed after acetaminophen incubation, using real time-PCR. To enhance comparability, graphs were plotted using the same scale as used for neutrophil sensing studies. (B) Fluorescence staining showing absence of nuclear translocation of IRF3 during APAP-challenge, corroborating the real time-PCR data. DNA was stained by propidium iodide and anti-IRF3 antibody evidenced IRF3 location. Bar = 10µm

Supporting Figure 6: (A) Survival rate of mice challenged with APAP. Around 25% of mice succumbed to liver injury and no more deaths were observed until the 8th day after APAP treatment. * P<0.05 in comparison to controls (Log rank test)

Supporting Figure 7: Liver DNA deposition due to APAP challenge occurred in a dose-dependent manner. (A-B) Intravital microscopy showing that a non-toxic APAP dose (150mg/Kg) induced no detectable DNA accumulation within the liver. An intermediate dose (300mg/Kg) caused a mild release of DNA to the extracellular compartment, while higher doses (600mg/Kg) led to a significant accumulation of DNA within the liver. DNA was stained by i.v. Sytox green injection. (C) Serum ALT quantification confirmed different degrees of liver damage due to crescent doses of APAP administration. * P<0.05 in comparison to controls. Statistical tests as indicated in Methods.

Supporting Figure 8: APAP-induced liver necrosis and neutrophil recruitment are dependent on CXCR2 and FPR1, but independent on CD18 or CD44-hyaluronan adhesion. (A) Liver confocal intravital microscopy of controls or APAP-treated mice injected with Sytox Green and PE-conjugated anti-Ly6G antibody (to evidence neutrophils; red) (n = 5/group). Bar = 100µm. (B) APAP-induced liver necrosis was significantly reduced by blockage of FPR1 and CXCR2, but not by CD18 (n = 5/group). (C) Neutrophil recruitment to liver necrosis was dependent on CXCR2 and FPR1, but independent on CD18 or CD44-hyaluronan interaction (n = 5/group). (D) APAP induced intravascular DNA coating in the liver was abrogated by FPR1 blockage but not by CXCR2 or CD18 blockage (n = 5/group). *P<0.05 in comparison to controls and # in comparison to vehicle-treated group. HNase = hyalorunidase. Statistical tests as indicated in Methods.

Supporting Figure 9: Liver intravital microscopy images for untreated and Kupffer cell-depleted mice. (A,B) Control mice present extensive Kupffer cell coverage of liver sinusoids (blue; anti-F4/80 antibody), with a few neutrophils (red; anti-GR1 antibody) and minor DNA labelling (Sytox green; n = 3/group). (C) Intravenous injection of clodronate liposomes (CLL; 48hs before APAP challenge) effectively depletes Kupffer cells, but does not induce neutrophil recruitment or liver necrosis (n = 3/group). Bar = 150µm.

Supporting Figure 10: Intravascular DNA coating induced by APAP does not contain neutrophil elastase in vivo. (A) Liver confocal intravital microscopy of control mice showing absence of intravascular DNA (Sytox green) and elastase labeling (Alexa-555 anti-elastase antibody). Bar = 150µm. (B) APAP-treated mice presented DNA fibers in the liver, but these were not labeled by the anti-elastase antibody (n = 4/group).

Supporting Figure 11: APAP-induced liver injury and intravascular DNA deposition occurs under sterile conditions. (A) Microbiota-depleted mice due to antibiotic treatment presented DNA deposition and neutrophil recruitment when challenged with APAP as similarly to untreated controls (n = 5/group). Bars = 100µm. (B) Real-Time PCR for bacterial DNA was negative in liver and serum after APAP intoxication (n = 5/group). (C) Absence of colony forming units (CFU) in serum and liver extract of mice challenged with APAP (n = 5/group).

Supporting Figure 12: APAP-induced liver injury is significantly reduced by blockage of TLR9 or elimination of DNA deposition, as shown in liver histology. APAP induced-liver necrosis is prevented in TLR9-/- mice. Elimination of extracellular DNA by DNASE1 or blockage of TLR9 by its antagonist E6446 rescues APAP-induced liver injury.


Supporting Video 1: DNASE1 eliminates intravascular DNA coating induced by APAP intoxication. Liver intravital microscopy of APAP-treated mice showing extensive DNA deposition (green), followed by quick DNA washout in response to intravenous DNASE1 injection. DNASE1 was injected immediately after the recording started. Neutrophils (red) remain adhered and patrolling DNA-rich areas (necrosis). Total time elapsed: 7 minutes.


Supporting Video 2: Neutrophils migrate to necrotic liver areas induced by APAP, where they keep a patrolling behavior. Liver intravital microscopy of APAP-treated mice showing a necrotic area (green) and neutrophils (yellow) in constant movement inside necrosis. Total time elapsed: 18 minutes.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.