VEGF mitigates histone‐induced pyroptosis in the remote liver injury associated with renal allograft ischemia–reperfusion injury in rats

Clinical evidence has indicated a possible link between renal injury and remote liver injury. We investigated whether extracellular histone mediates remote hepatic damage after renal graft ischemia–reperfusion injury, while vascular endothelial growth factor (VEGF) is protective against remote hepatic injury. In vitro, hepatocyte HepG2 cultures were treated with histone. In vivo, the Brown‐Norway renal graft was stored in 4°C preservation solution for 24 hours and then transplanted into a Lewis rat recipient; blood samples and livers from recipients were harvested 24 hours after surgery. Prolonged cold ischemia in renal grafts enhanced liver injury 24 hours after engraftment. Caspase‐1, ASC, NLRP3, and AIM2 expressions in hepatocyte, CD68+‐infiltrating macrophages, tissue, and serum interleukin‐1β and ‐18 were greatly elevated, indicating that pyroptosis occurred in the liver and resulted in acute liver functional impairment. Blocking the caspase‐1 pathway decreased the number of necrotic hepatocytes. VEGF treatment suppressed the hepatocyte pyroptosis and liver function was partially restored. Our data suggested that renal allograft ischemia–reperfusion injury is likely associated with acute liver damage due to hepatocyte pyroptosis induced by histone and such injury may be protected by VEGF administration. VEGF, therefore, may serve as a new strategy against other remote organ injuries related to renal transplantation.


| INTRODUC TI ON
Ischemia-reperfusion injury (IRI) is an inevitable consequence of renal transplantation and a major determinant of graft survival. Renal IRI is associated with deleterious consequences for several organs. 1,2 Accumulating clinical evidence has identified a close relationship between renal injury and injuries to other organ systems, including heart, lung, and liver. 3 Indeed, impaired liver function is often seen in patients with acute kidney injury (AKI). 4,5 In addition, in-hospital death is more likely in patients with AKI and liver failure than in those with AKI alone. 6 The pathophysiology of remote liver injury after AKI is complex and incompletely understood, although both preclinical and clinical studies have shown that inflammation is an important mediator. 2 The exact functions of the various cytokines involved in regulating the complex inflammatory events after IRI are not yet fully unraveled. 7 The liver is particularly vulnerable to inflammatory challenge from the distant renal graft because it is abundantly perfused with systematic circulation. Liver injury could develop secondary to delayed graft function in a renal graft recipient. Remote hepatic injury is believed to be initiated and sustained by proinflammatory cytokines that are released or activated during IRI after renal damage. 8,9 If liver repair and regenerative mechanism are not activated promptly, acute liver functional impairment could occur. Clinical management of such patients is difficult, and effective organ protective strategy is therefore required.
Recently, extracellular histone has been identified as a key inflammatory mediator in renal injury. 10 Histone is a highly conserved eukaryotic chromosomal protein. Under oxidative or inflammatory stress, histone undergoes translocation from the nucleus to the cytoplasm, and then it is secreted from the necrotic cells. 11 In the liver, histone has been shown to mediate hepatocyte cell death and inflammation. 12,13 It is suggested that histone mediates cell death and inflammation by binding to Toll-like receptors (eg, TLR-4 and TLR-9). 14 Activation of TLRs may trigger the downward cascade including the inflammasome to activate pyroptosis and produce proinflammatory cytokines, exacerbating the injury and causing a systemic response. 15,16 Pyroptosis is a lytic type of cell death and a form of regulated necrosis that is inherently associated with inflammation and distinguished from other forms of cell death by the associated secretion of interleukin (IL)-1β after caspase-1 activation. The cellular processes that occur during pyroptosis include nuclear condensation, DNA damage, cell swelling, and, finally, cell lysis, with the subsequent release of IL-1β and are reliant on intracellular sensors of bacterial products and formation of the inflammasome. Pyroptosis is a proinflammatory response that is triggered by a variety of pathologic stimuli, including myocardial infarction, stroke, and malignancy.
Various studies have indicated that pyroptosis is intrinsically involved in the development of infectious diseases, nervous system disorders, and atherosclerotic processes. [17][18][19] Pyroptosis is thought to be a key modulator of the immune response to microbial infection, while pathogens that have evolved and developed mechanisms to bypass pyroptosis demonstrate enhanced disease-causing and septic potential. 20 To date, there is a lack of studies about the effects of acute renal allograft injury on the liver. In the present study, we tested the hypothesis that IRI in renal allografts would initiate the distant hepatic injury. The underlying molecular mechanism, which is centered on pyroptosis, was also explored in this study.

| In vitro cell culture
Human hepatocyte HepG2 cells were cultured in EMEM, and human monocyte/macrophage U937 cells (European Cell Culture Collection, Porton Down, UK) were cultured in RPMI 1640 medium.

| Flow cytometry
Reactive oxygen species production was monitored by the measurement of superoxide (O 2− ) generation by using the fluorescent dyes dihydroethidium (DHE). 22 Cells were incubated in DHE (2 μmol/L) for 30 minutes at 37°C in the dark. The cells were washed with PBS.
Fluorescence intensity was acquired and analyzed by using flow cytometry (FACSCalibur; Becton Dickinson, Sunnyvale, CA). Each assay included at least 10 000 gated events. Propidium iodide (PI; Sigma Aldrich, St. Louis, MO) staining was used to examine cell death as described. Cells were harvested in a fluorescence-activated cell sorting (FACS) tube and washed twice before resuspension in FACS buffer. PI was added to make the final concentration to 1 μg/mL and incubated in dark for 5 minutes. PI fluorescence was detected by using flow cytometry.

| Renal transplantation
Inbred adult male Brown-Norway rats BN, RT 1n ) and Lewis (LEW, RT1 1 ) rats weighing 225 to 250 g were purchased from Harlan, Bicester, UK and bred in temperature-and humidity-controlled cages in a specific pathogen-free facility at Chelsea-Westminster Campus, Imperial College London. This study was approved by the Home Office, United Kingdom, and all animal procedures were carried out in accordance with the United Kingdom Animals (Scientific Procedures) Act of 1986. BN-to-LEW rat renal transplantation was used. Rat donor kidneys were transplanted orthotopically into recipients by using standard microvascular techniques. 21 Briefly, the donor's left kidney, aorta, and inferior vena cava were carefully exposed. The aorta was clamped below the renal vessel. An elliptical aortic patch was created. The kidney graft was then extracted, flushed, and stored in 4°C heparinized Soltran Preserving Solution (Baxter Healthcare, Newbury, UK). After the cold ischemia, the recipient's renal artery and vein were carefully isolated and cross-clamped, the left kidney was extracted, and the donor renal vein was connected to the recipient renal vein through end-to-end anastomosis with continuous 8-0 sutures. The arterial anastomosis between the donor aortic patch was connected to recipient aorta in an end-to-side manner. The successful perfusion of renal graft was confirmed by an instant color change of the kidney and rapid expansion of renal arteries and vein. Urinary reconstruction was performed by ureter-to-bladder anastomosis. The total surgical ischemia time was restricted to less than 45 minutes. The contralateral native kidney was excised immediately after surgery.

| Hydrodynamic tail vein injection
Rat VEGF siRNA (Qiagen, SI01994454), rat VEGF R2 (Qiagen, SI01528415), or scrambled siRNAs (negative control) (Qiagen) were dissolved in siRNA suspension buffer and further diluted in RNasefree PBS before use. siRNA targeting rat VEGF was administered via hydrodynamic tail vein injection according to our established pro-

| Hematoxylin-eosin staining
Liver samples obtained at various determination points were fixed in 4% buffered formalin and then embedded in paraffin, in accordance with standard procedures. Sections (5 μm) were stained with hematoxylin-eosin and examined microscopically. All samples were evaluated by an experienced pathologist who was blinded to the experiment. All fields in each section were examined for grading of steatosis and necroinflammation according to the Colantoni criteria. 25

| TUNEL staining
Hepatic cell death was detected by using the in situ terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay (Millipore) according to the manufacturer's instructions. TUNELpositive nuclei were visualized by green FITC fluorescence.

| Immunohistochemistry
For in vivo fluorescence staining, the liver sample was fixed in 4% paraformaldehyde in 0.1 mol/L PBS solution overnight for 16 hours at 4°C. This was followed by incubation in 30% sucrose solution for 24 hours at Santa Cruz), followed by secondary antibody for 1 hour. For doublelabeled immunofluorescence, cells and tissue samples were incubated with the first primary antibody overnight, followed by the first secondary antibody, and then the second primary antibody and the second secondary antibody. The slides were counterstained with nuclear dye 4′,6-diamidino-2-phenylindole and mounted with Vectashield mounting medium (Vector Laboratories, Burlingame, CA). Immunofluorescence was quantified by using ImageJ (National Institutes of Health, Bethesda, MA). Ten high-power fields at ×20 magnification were first photographed by using an AxioCam digital camera (Zeiss, Welwyn Garden City, UK) mounted on an Olympus BX60 microscope (Olympus, Middlesex, UK) with Zeiss KS-300 software (Zeiss, Welwyn Garden City, UK). The average density per section was calculated (ImageJ). Values were then calculated as percentages of the mean value for naïve controls and expressed as percent fluorescence intensity (FI).

| Western blotting
Liver samples were mechanically homogenized in lysis buffer. The lysates were centrifuged, the supernatant was collected, and total protein concentration in the supernatant was quantified according to the Bradford protein assay (Bio-Rad, Hemel Hempstead, UK). The protein extracts (40 μg/sample) were heated, denatured, and loaded on a NuPAGE 4%-12% Bis-Tris gel (Invitrogen) for electrophoresis and then transferred to a polyvinylidene difluoride membrane. The membrane was treated with blocking solution (5% dry milk in TBS with 0.1% Tween-20) for 2 hours and probed with the following primary antibodies: mouse anti-caspase-1 p20 (sc-398715, 1:1000; Santa Cruz), in TBS-T overnight at 4°C, followed by HRP-conjugated secondary antibody for 1 hour. The loading control was the constitutively expressed protein glyceraldehyde 3-phosphate dehydrogenase (1:1000; Abcam). The blots were visualized with use of the enhanced chemiluminescence system (Santa Cruz) and analyzed with the use of GeneSnap (Syngene, Cambridge, UK).

| Kidney and liver function
Blood samples were collected when animals were killed. After centrifugation, asparate aminotransferase (AST) and alanine aminotransferase (ALT) concentrations were measured using AST and ALT Activity Assay Kit (Sigma Aldrich). Serum urea and creatinine concentrations were measured by using an Olympus AU2700 analyzer (Diamond Diagnostics, Watford, UK).

| Statistical analysis
All numerical data were expressed as mean ± SD. Data were analyzed with ANOVA followed by Kruskal-Wallis nonparametric (scoring) or Newman-Keuls (measurement) test for comparisons. A P value of <.05 was considered to be of statistical significance.

| IRI in renal graft caused remote liver injury after grafting
To evaluate the effect of IRI of renal grafts on the liver, renal grafts were stored in cold preservation solution for up to 24 hours and then transplanted into the recipient. There were minimal histologic changes in the liver of recipient with live transplantation. Kidney sections of rats with ischemic injury showed extensive, multifocal, neutrophil-infiltrated areas of coagulative necrosis scattered throughout the renal parenchyma ( Figure 1A

| Enhanced inflammation was found in remote hepatic injury
Renal IRI was associated with a significant increase in CD68 + macrophages (Figure 2A

| Pyroptosis was induced in remote hepatic injury in recipient with ischemic allografts
Transplantation of ischemic renal grafts was associated with increased hepatic pyroptosis in remote hepatic injury, indicated by a significant increase in caspase-1 ( Figure 3A

| Histone-induced pyroptosis in cultured hepatocytes and activation of monocyte
Histone-induced pyroptosis was investigated in cultured HepG2 hepatocytes to determine the effects of extracellular histone in remote hepatic injury, as demonstrated in Figure 4. Our findings indicate that histone H3 treatment is associated with an increase in caspase-1 ( Figure 4A

| Recombinant histone protein treatment exacerbates the remote hepatic injury
As indicated in Figure 5, recombinant histone H3 protein treatment was found to exacerbate hepatic injury, with both low-dose and highdose histone administration associated with increased hepatic injury ( Figure 5A) and increased liver injury score ( Figure 5D

| VEGF-mediated cytoprotection in the remote liver injury
The capability for VEGF to attenuate pyroptosis in remote liver injury was assessed. In cultured hepatocytes treated with histone and in liver tissue with remote injury, expression of caspase-1 ( Figure 6A,B,E) and ASC ( Figure 6A,E) was increased, while treatment with VEGF siRNA enhanced the expression of these markers, thus indicating hepatic protection. VEGF receptors 1 and 2 siRNA, blocking the expression of receptors 1 and 2, respectively, were administered. Unlike receptor 1, blocking receptor 2 F I G U R E 2 Enhanced inflammation during remote hepatic injury in recipient rats. Renal graft from the Brown-Norway rat donor was stored in 4°C Soltran preservation solution for 24 h (cold ischemia rCI24 h), which was subsequently transplanted into Lewis rat recipients. The livers were harvested on day 1 after transplantation. (A) Immunofluroscence of CD68 + (red) macrophages. (B) Number of CD68 + macrophages. Dual labeling of (C) histone H3 (green) and TLR-4 (red) and (D) histone H3 (green) and NF-κB (red). Fluorescence intensity of (E) TLR-4 and (F) NF-κB. Concentration of histone H3 in (G) tissue and (H) serum accessed by ELISA. Concentration of IL-1β in (I) tissue and (J) serum accessed by ELISA. Concentration of IL-18 in (K) tissue and (L) serum accessed by ELISA. Scale bar: 50 μm. Data expressed as mean ± SD (n = 6) (*P < .05, **P < .01 and ***P < .001). NC, naive control; rCI, renal graft cold ischemia [Color figure can be viewed at wileyonlinelibrary.com] significantly enhanced the expression of caspase-1 ( Figure 6A,B) and ASC ( Figure 6A) and increased the percentage of PI-positive cells ( Figure 6C), suggesting that VEGF acts via receptor 2. After transplantation, treatment with VEGF siRNA or VEGF receptor 2 siRNA exacerbated hepatic injury ( Figure 6G), enhanced caspase-1 ( Figure 6B,H) and ASC expression ( Figure 6E) and increased the number of TUNEL-positive cells ( Figure 6I). This indicates that blocking the VEGF signaling pathway would suppress the VEGF protective effects.

| Recombinant VEGF protein confers cytoprotection against histone-induced pyroptosis
Recombinant VEGF protein was found to confer cytoprotection against pyroptosis, as demonstrated in Figure 7. After being challenged with histone, HepG2 cells, when treated with VEGF, exhibited reduced caspase-1 ( Figure 7A Furthermore, administration of recombinant VEGF was also associated with significantly reduced expression of caspase-1 ( Figure 7F,I) and ASC ( Figure 7F,J) by one-half and two-thirds, respectively, as well as an improvement in liver injury score ( Figure 7G). The expression of cleaved caspase-1 was also reduced by this treatment ( Figure 7K). Recombinant VEGF also caused a significant improvement in hepatic function, as evidenced by a reduction in both ALT and AST levels in Figure 7L and M, respectively. These findings indicate that VEGF may be a potential therapeutic agent to attenuate pyroptosis in hepatic F I G U R E 3 Pyroptosis in remote hepatic injury in recipient with ischemic allografts. Renal graft from the Brown-Norway rat donor was stored in 4°C Soltran preservation solution for 24 h (cold ischemia rCI24 h), which was subsequently transplanted into Lewis rat recipients. Dual labeling of (A) caspase-1 (red) and ASC (green), (B) caspase-1 (red) and NLRP3, and (C) caspase-1 (red) and AIM2 (green). Nuclei were counterstained by DAPI.

| Recombinant VEGF protein attenuates inflammation in remote hepatic injury
The administration of VEGF protein is found to be associated with a reduction in inflammation-induced hepatic injury. Administration of VEGF causes significant reduction in the number of CD68 + cells ( Figure 8A,B), as well as reduced production of pyroptosis-related cytokines, IL-1β ( Figure 8C,D), and IL-18 ( Figure 8E,F), in both liver tissue and serum. ELISA indicates reduced tissue and serum concentration of histone ( Figure 8G,H), suggesting an attenuation of histone release after VEGF protein administration. Overall, these findings indicate that treatment with VEGF protein results in a reduction in hepatic inflammation, macrophage infiltration, pyroptosis-mediated hepatocyte injury, and histone release ( Figure 8I).  The onset of liver injury after IRI in renal allografts appeared in early stages of kidney injury and was associated with increased levels of proinflammatory cytokine and activated oxidative stress.

| D ISCUSS I ON
Histones are the proteins in chromatin that play an important role in controlling gene expression. 27 There are 2 kinds of histones: core histones (including H2A, H2B, H3, and H4) and linker histones (namely, H1 and H5). 27 Despite their vital role in gene regulation, histones in the extracellular space can cause inflammation and organ damage as reported in liver, lung, and kidney injury. 10,12,28 When necrosis takes place, the cell membrane ruptures and the cellular content, also known as damage-associated molecular patterns (DAMPs) including histones, would be released, causing inflammatory response. 29,30 The release of histones may also be associated with the formation of neutrophil extracelluar traps, which has shown to cause cell death in the lungs and endothelium. 31 Allam et al. showed that histone is released and causes inflammatory response in renal tubular epithelial injury, by binding with the TLR-4 and TLR-2. 10 Histone has been shown to be closely associated with the NLRP3 inflammasome. The activation of NLRP3 leads to the assembly of the NLRP3 inflammasome, which includes pro-caspase-1, resulting in the production of proinflammatory cytokines IL-1β. NLRP3 inflammasome is formed after the oligomerization of NLRP3 and subsequent recruitment of ASC and pro-caspase-1. 32 On activation of NLRP3, ASC proteins assemble into fiberlike structures; this culminates in the production of a large protein aggregate that amplifies the activation of caspase-1. 33 The NLRP3 inflammasome is activated in response to a variety of infectious stimuli or to cellular stress caused by various sterile danger signals, including histone. 32 The inflammasome induces the release of IL-1β, IL-18, which exacerbate the inflammation. 33,34 Pyroptosis is a caspase-1-dependent programmed cell In the previous studies, an increased level of histone was toxic to liver and cause liver injury. 14 We have also demonstrated that naive animal and recipient animal, when receiving both low and high doses of histone, had significant liver damage, and blocking the its receptor TLR-4 through inhibitor reduced the injury; all these data indicated that the remote injury was effectively caused by extracellular histones. In addition, it is difficult to distinguish histone release from either kidney or liver, or both. Therefore, it remains an open question whether histones released from kidney are a primary trigger for liver injury or other renal failure-related factors (eg, toxins trigger histone F I G U R E 6 Cytoprotection mediated by vascular endothelial growth factor in the remote liver injury. Cultures of HepG2 cells were treated with VEGF siRNA, VEGFR1 siRNA, VEGFR2 siRNA, or scrambled siRNA 6 hours before histone H3 treatment. In rat transplant recipient, VEGF siRNA or VEGF R2 siRNA or scrambled SiRNA was also administered to recipient rats. This warrants further investigation in future studies.
While it was initially believed that VEGF receptors were solely confined to the vascular endothelium, hence the name of the protein family, the presence of VEGF receptors has been acknowledged in a variety of cell types, including epithelial cells. 37 As a result, VEGF is capable of eliciting its effects on both epithelial and endothelial surfaces. VEGF-A binds with a high affinity to both VEGF-R1 and VEGF-R2, with VEGF-R2 undergoing more potent tyrosine phosphorylation after ligand activation, while VEGF-B solely interacts with VEGF-R1. 38 Although VEGF-R1 has shown a higher affinity to VEGF, approximately 10-fold higher than VEGF-R2, VEGF-R2 is considered a significant positive mitogenic signal transducer due to its strong kinase activity in comparison to VEGF-R1. 39 Various studies have demonstrated that VEGF-R2 upregulation promotes organoprotection. Activation of the VEGF-R2 pathway has been shown to mediate lung protection against oxidant-induced acute lung injury. 40 Furthermore, the lungs of patients with sepsis, which is considered an important trigger of acute lung injury, have been shown to possess a significantly lower level of VEGF-R2, 41 while observational studies of lung injury in humans demonstrate a significant reduction in intrapulmonary VEGF during the early stages of acute respiratory distress syndrome. 42 Our findings demonstrate that VEGF has the potential to induce organoprotective effects, possibly by activating various signaling cascades. Inhibition of VEGF during reperfusion of the highly ischemic allografts, through either VEGF siRNA or VEGF-R2 siRNA, exacerbates the hepatic injury observed. These processes may, at least in part, explain the mechanism by which VEGF attenuates extracellular histone-induced pyroptosis in remote hepatic injury.
There are limitations on our studies. First, the exact resource of extracellular histones after renal cold ischemia-reperfusion is unclear; shown to exacerbate hepatic injury, 12,13 while histone neutralization has been shown to be a potential therapeutic avenue to attenuate its hepatotoxic effects. Recent studies have demonstrated that nuclear histone proteins are closely associated with the upregulation of DAMPs, including DNA and HMGB-1, which are responsible for contributing to hepatic IRI, 12 as well as promoting cytotoxicity via TLR-9 and MyD88 pathways. 12,15 The administration of neutralizing antibodies to extracellular histones (anti-H3 and anti-H4 antibodies) confers significant protection after hepatic IRI. 12 This is thought to occur due to the attenuation of tissue tumor necrosis factorα and IL-6. Further investigation into the hepatoprotective effects of histone neutralization after hepatic IRI is warranted. Finally, human hepatocellular carcinoma cell line was used in in vitro study. It is very different in cell phenotype compared with primary hepatocytes, which are considered to be used for future study.
Patients with AKI often have liver dysfunction and may be associated with higher mortality. 4,6 Although the effect of renohepatic crosstalk in renal graft recipients remains fully elucidated, our study indicates the role of circulating histones and their inflammatory effect on the liver. Extracellular histones, therefore, may be the target for therapy or prophylaxis against liver damage in kidney transplant patients.
In conclusion, our data suggest that a substantial release of histone in recipient after receiving ischemic renal allografts leads to F I G U R E 8 Reduced inflammation by recombinant VEGF protein in remote hepatic injury. Renal graft from the Brown-Norway rat donor was stored in 4°C Soltran preservation solution for 24 hours (cold ischemia rCI24 h), which was subsequently transplanted into Lewis rat recipients. The recipient rats were administered with VEGF recombinant protein intravenously. (A) Immunofluroscence labeling of CD68 + macrophages. (B) Number of CD68 + cells. Concentration of IL-1β in (C) tissue and (D) serum, accessed by ELISA. Concentration of IL-18 in (E) tissue and (F) serum, accessed by ELISA. Concentration of histone H3 in (G) tissue and (H) serum, accessed by ELISA. (I) Proposed mechanisms of pyroptosis induced by histone in remote liver injury after transplantation; extracellular histone binds to TLR-4 receptor, activating both NLRP3 and AIM 2 inflammasome. The production of IL-1β and IL-18 is increased and cells die by pyroptosis. In addition, pyroptosis enhanced activation of monocytes through histone release. VEGF protects against remote liver injury through attenuating the pyroptosis. Scale bar: 50 μm. Data expressed as mean ± SD (n = 6) (*P < .05, **P < .01 and ***P < .001). Ve, vehicle [Color figure can be viewed at wileyonlinelibrary.com] remote hepatic injury during early the postoperative period and the protective effects of VEGF through blocking histone-induced pyroptosis in the hepatic remote injury.

D I SCLOS U R E
The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.