Inhibition of Kupffer cell-mediated early proinflammatory response with carbon monoxide in transplant-induced hepatic ischemia/reperfusion injury in rats†
Article first published online: 29 OCT 2008
Copyright © 2008 American Association for the Study of Liver Diseases
Volume 48, Issue 5, pages 1608–1620, November 2008
How to Cite
Tomiyama, K., Ikeda, A., Ueki, S., Nakao, A., Stolz, D. B., Koike, Y., Afrazi, A., Gandhi, C., Tokita, D., Geller, D. A. and Murase, N. (2008), Inhibition of Kupffer cell-mediated early proinflammatory response with carbon monoxide in transplant-induced hepatic ischemia/reperfusion injury in rats. Hepatology, 48: 1608–1620. doi: 10.1002/hep.22482
Potential conflict of interest: Nothing to report.
- Issue published online: 29 OCT 2008
- Article first published online: 29 OCT 2008
- Manuscript Accepted: 10 JUN 2008
- Manuscript Received: 15 FEB 2008
- National Institutes of Health. Grant Numbers: DK071753, DK062313, CA76541, DK54411
Proinflammatory responses play critical roles in hepatic ischemia/reperfusion (I/R) injury associating with liver transplantation (LTx), and carbon monoxide (CO) can effectively down-regulate them. Using wild-type (WT) to enhanced green fluorescent protein (EGFP)-transgenic rat LTx with 18-hour cold preservation in University of Wisconsin solution, this study analyzed the relative contribution of donor and host cells during early posttransplantation period and elucidated the mechanism of hepatic protection by CO. CO inhibited hepatic I/R injury and reduced peak alanine aminotransferase levels at 24 hours and hepatic necrosis at 48 hours. Abundant EGFP+ host cells were found in untreated WT liver grafts at 1 hour and included nucleated CD45+ leukocytes (myeloid, T, B, and natural killer cells) and EGFP+ platelet-like depositions in the sinusoids. However, reverse transcription polymerase chain reaction (RT-PCR) analysis of isolated graft nonparenchymal cells (NPCs) revealed that I/R injury-induced proinflammatory mediators [for example, tumor necrosis factor alpha (TNF-α), interleukin-6 (IL-6), and inducible nitric oxide synthase (iNOS)] were not up-regulated in purified CD45+ cells of donor or host origin. Instead, TNF-α and IL-6 messenger RNA (mRNA) elevation was exclusively seen in isolated CD68+ cells, whereas iNOS mRNA up-regulation was seen in hepatocytes. Nearly all CD68+ cells at 1 hour after LTx were EGFP− donor Kupffer cells, and CO efficiently inhibited TNF-α and IL-6 up-regulation in the CD68+ Kupffer cell fraction. When graft Kupffer cells were inactivated with gadolinium chloride, activation of inflammatory mediators in liver grafts was significantly inhibited. Furthermore, in vitro rat primary Kupffer cell culture also showed significant down-regulation of lipopolysaccharide (LPS)-induced inflammatory responses by CO. Conclusion: These results indicate that CO ameliorates hepatic I/R injury by down-regulating graft Kupffer cells in early postreperfusion period. The study also suggests that different cell populations play diverse roles by up-regulating distinctive sets of mediators in the acute phase of hepatic I/R injury. (HEPATOLOGY 2008;48:1608–1620.)
Liver transplantation (LTx) becomes an effective therapy for most end-stage liver diseases. However, the transplantation procedure obligates cold preservation and warm reperfusion of liver grafts, which results in some degree of ischemia/reperfusion (I/R) injury in every transplant graft. I/R injury remains as one of the major complications affecting short-term and long-term outcomes of LTx. The process of hepatic I/R injury is a cascade of inflammatory events involving multiple interconnected factors, including hepatic sinusoidal endothelial cell injury and disturbances of microvascular circulation, production and release of reactive oxygen species and inflammatory mediators, activation of Kupffer cells, complements and coagulation pathways, and extravasation of host inflammatory cells.1–5 Thus, parenchymal hepatocellular damage caused by cold I/R injury in liver grafts is caused by the various types of cells of graft and host origins. However, the relative contribution of donor and recipient cells in each stage of hepatic I/R injury remains undetermined.
We have previously shown that peritransplant treatment of liver transplant recipients with inhaled carbon monoxide (CO) reduces hepatic I/R injury.6, 7 Although normally considered as a toxic gas molecule involved in air pollution and poisonings, CO is endogenously produced in the body as a byproduct of heme degradation by the heme oxygenases (HO) and functions as a physiological regulator.8–11 Beneficial effects of CO have been confirmed in many in vitro and in vivo models of oxidant-induced cellular and tissue injury.12–14 We have previously demonstrated that CO renders cytoprotection via a down-regulation of inflammatory responses in hepatic I/R injury after LTx. In the current study, we investigated CO's anti-inflammatory functions by identifying the target cell population of inhaled CO. To differentiate donor and recipient populations in the syngenic transplantation model of I/R injury, we conducted transplantation of wild-type (WT) liver grafts after 18 hours' cold preservation in enhanced green fluorescence protein (EGFP) transgenic rat recipients. In this system, EGFP-tagged host infiltrating cells were easily identified from WT donor hepatic cells. Analyses of various cell types isolated from liver grafts during I/R injury revealed that different cell populations distinctively responded to ischemic injury, and CO mediated hepatic protection at least in part by inhibiting proinflammatory up-regulation in Kupffer cells.
Materials and Methods
Phycoerythrin (PE)-conjugated monoclonal antibodies used in this study included anti-CD68 (ED1), CD163 (ED2), CD11b/c (OX42), alpha E2 integrin (OX62) (Serotec, Oxford, UK), CD45 (OX1), NKR-P1 (10/78), αβ τ cell receptor (TCR) (R7.3), and rat immunoglobulin M (BD Pharmingen, San Diego, CA). Collagenase (type IV), bovine serum albumin, ethylenediamine tetraacetic acid, ethylene glycol tetraacetic acid, gadolinium chloride (GdCl3), Histodenz, and lipopolysaccharide (LPS) were obtained from Sigma, St. Louis, MO. Fetal bovine serum, phosphate-buffered saline (PBS), Ca+/Mg+-free Hank's balanced salt solution, Gay's balanced salt solution, and William's medium E were from Invitrogen, Carlsbad, CA. Anti-PE microbeads were from Miltenyi Biotec, Bergisch Gladbach, Germany. L-glutamine and gentamicin were from Life Technologies, Grand Island, NY.
EGFP-transgenic and WT Sprague Dawley rats were obtained from Japan SLC, Inc. (Hamamatsu, Japan). The expression of EGFP was under the control of the cytomegalovirus enhancer and the chicken β-actin promoter derived from an expression vector, pCAGGS.15, 16 A breeding colony of EGFP transgenic rats was maintained by brother–sister mating for more than 15 generations, and the level of homozygosity reached by individual rats in this colony was confirmed by the acceptance of skin grafts. Inbred male Lewis (RT1l) rats weighing 200 to 300 g were purchased from Harlan Sprague-Dawley Inc. (Indianapolis, IN) for the isolation of Kupffer cells and in vitro LPS stimulation. Animals were maintained in laminar flow cages in a specific pathogen-free animal facility at the University of Pittsburgh. All procedures in this experiment were performed according to the guidelines of the Council on Animal Care at the University of Pittsburgh and the National Research Council's Guide for the Humane Care and Use of Laboratory Animals.
In a WT to EGFP model, WT liver grafts were transplanted into EGFP transgenic rats. Male Lewis rats were used for both donors and recipients. Orthotopic LTx without hepatic arterial reconstruction was performed according to the method previously described by Kamada and Calne.17 Liver grafts were stored in University of Wisconsin solution at 4°C for 18 hours' cold ischemic time (CIT) and orthotopically transplanted into designated recipients as previously described.18, 19 Recipient animals were sacrificed at 1-48 hours, and liver graft samples were obtained for analyses described later.
In Vivo Treatment
Recipient animals were exposed to CO at 100 ppm for 1 hour before and for 24 hours after the transplant surgery in a CO chamber with rat chow and water ad libitum. A CO chamber (3.70 ft3) was filled with the CO gas (1%) mixed with air (21% oxygen), and CO concentration was maintained at 100 ppm by continuous monitoring using a CO analyzer (Interscan, Chatsworth, CA). During the CO exposure, carboxyhemoglobin levels in the recipient arterial blood were maintained at approximately 12%.6, 7 In separate groups, GdCl3 (7.5 mg/kg, intraperitoneally) was administered to donor animals at 24 hours before surgery to inhibit graft Kupffer cells.
Hepatocytes and Hepatic Nonparenchymal Cell Isolation
Hepatocytes and hepatic nonparenchymal cells (NPCs) were isolated from the liver by the collagenase digestion method of Berry and Friend20 with some modifications. Briefly, the liver was perfused in situ via the portal vein initially with 250 mL Ca+Mg+-free Hank's balanced salt solution containing 5 mM ethylene glycol tetra-acetic acid and 10 mM 4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid, then with 500 mL Hank's balanced salt solution containing 0.05% collagenase (type IV) and 5 mM 4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid. After perfusion, the liver was removed into a Petri dish, and NPCs and parenchymal cells were liberated from the connective tissue by carefully raking the liver with scissors and shaking the liver gently. The initial cell suspension was filtrated through a 70-μm nylon mesh, and hepatocytes were obtained by low-speed centrifugation (five times at 50g for 5 minutes). The supernatant was centrifuged (at 450g for 7 minutes) to obtain hepatic NPCs.
Hepatic Leukocyte and Kupffer Cell Isolation and Culture
Kupffer cells were purified from hepatic NPCs using Auto MACS. Briefly, the NPC fraction was suspended in PBS containing 0.5% (wt/vol) bovine serum albumin and 0.8% (vol/vol) ethylenediaminetetra-acetic acid. After red cell lysis with NH4Cl buffer, Kupffer cells were positively isolated by Auto MACS (Miltenyi Biotec, Germany) using PE-conjugated anti-CD68 antibody and anti-PE microbeads. CD45+ cells were also obtained from hepatic NPCs using the same method with PE-conjugated anti-CD45 antibody. Isolated cells were used for reverse transcription polymerase chain reaction (RT-PCR). A portion of cell fraction was used to determine the purity by flow cytometry.
Kupffer cells were also purified by density gradient centrifugation on a 17.5% (wt/vol) Histodenz gradient, followed by centrifugal elutriation, as described previously.21 The cell pellet was suspended in William's medium E containing 2 mM L-glutamine, 10 μg/mL gentamicin, and 10% fetal bovine serum. Cells were then plated in six-well culture plates at a concentration of 4 × 106/well. Culture medium was bubbled with CO gas for 5 minutes before use, and Kupffer cells were exposed to CO-containing medium for 2 hours in a humidified atmosphere in an incubator maintained at 37°C before being stimulated with LPS (0.1 μg/mL). Culture mediums were collected for tumor necrosis factor alpha (TNF-α) and interleukin-6 (IL-6) measurement.
Isolated hepatic NPCs were washed and incubated with each monoclonal antibody at 4°C for 30 minutes. After cells were washed with PBS, secondary antibody was added as needed. Stained cells were analyzed on an LSRII (BD Biosciences, San Jose, CA). Data were analyzed using FACSDiva software (BD Biosciences).
TNF-α and IL-6 Protein Levels
TNF-α and IL-6 levels in serum and culture medium samples were measured using enzyme-linked immunosorbent assay kits (Biosource, Camarillo, CA).
Total RNA was extracted from the liver tissues or cells using the TRIzol reagent (Life Technologies, Inc.), and RNA content was measured using 260/280 UV spectrophotometry. Messenger RNA (mRNA) expression was quantified by SYBR Green two-step, real-time RT-PCR for TNF-α, interleukin-10, IL-6, inducible nitric oxide synthase (iNOS), heme oxygenase-1 (HO-1), and glyceraldehyde 3-phosphate dehydrogenase, as previously described.22 The expression of each gene was normalized to glyceraldehyde 3-phosphate dehydrogenase mRNA content and calculated relative to normal control.
Immunohistochemistry and Image Analysis
At sacrifice, liver graft samples were obtained by perfusing animals via the abdominal aorta with PBS followed by perfusion fixation with 2% paraformaldehyde to avoid leaching of cytosolic EGFP protein out of the cells during the thaw process. The liver samples were removed and stored in 2% paraformaldehyde at 4°C overnight, then in 2.3 M sucrose in PBS overnight, embedded in optimal cutting temperature compound, and frozen in liquid nitrogen–cooled isopentane. Samples were cut into 6-μm sections on gelatin-coated slides and washed with PBS. After blocking with superblock (ScyTek Lab, Logan, UT), sections were stained with primary antibodies. Sections were nuclear DNA stained with Hoechst dye (Bisbenzimide), coverslipped with Gelvatol, and visualized with an Olympus BX51 epifluorescence microscope or a Fluoview 1000 confocal microscope (Olympus, Center Valley, PA).
Scanning electron microscopy and transmission electron microscopy were performed on 2.5% glutaraldehyde in PBS perfusion-fixed liver grafts, as previously described.23 The tissue was dehydrated through a graded series of 30% to 100% ethanol, 100% propylene oxide, then infiltrated in 1:1 mixture of propylene oxide:Polybed 812 epoxy resin (Polysciences, Warrington, PA) for 1 hour. After several changes of 100% resin throughout 24 hours, tissue was embedded in molds, cured at 37°C overnight, followed by additional hardening at 65°C for 2 more days. Ultrathin (70 nm) sections were collected on 200-mesh copper grids, stained with 2% uranyl acetate in 50% methanol for 10 minutes, followed by 1% lead citrate for 7 minutes. Tissues were visualized on a JEM-6335F scanning electron microscope and a JEOL JEM 1210 transmission electron microscope (JEOL, Peabody, MA).
The results were expressed as mean ± standard deviation. Statistical analysis was performed using unpaired Student t test or analysis of variance where appropriate. A probability level of P < 0.05 was considered statistically significant.
Inhaled CO Ameliorates Hepatic I/R Injury and Down-Regulates Proinflammatory Cytokines
Hepatic preservation injury in 18 hours' cold preserved rat liver grafts was significantly ameliorated with peritransplantation recipient inhalation of 100 ppm CO; treated recipients showed reduced serum alanine aminotransferase levels and less severe hepatic necrosis in liver grafts (Fig. 1A, B), as reported previously in Lewis rats.6, 7 Hepatic I/R injury was a vigorous proinflammatory event, mRNA levels for TNF-α and IL-6 promptly increased in the liver grafts after reperfusion with a peak at 1 hour; inhaled CO showed potent anti-inflammatory actions by inhibiting an early mRNA up-regulation of these cytokines (Fig. 1C). In accordance with mRNA levels, CO reduced serum TNF-α and IL-6 levels after reperfusion (Fig. 1D).
Inhaled CO Reduces EGFP+ Host Infiltrates and Platelet Deposits in Liver Grafts
To investigate the roles of host cells during early I/R injury period and effects of CO on host cell extravasation, tissue sections from WT liver grafts transplanted into EGFP transgenic rats with 18 hours' cold preservation were analyzed by fluorescent microscopy (Fig. 2A). At 1 to 3 hours after transplantation, abundant EGFP signals were found in untreated liver grafts. Most of them were EGFP+ infiltrating cells (colocalization of EGFP and nuclear blue stain, Fig. 2A, thick arrows); however, significant numbers of small granular type EGFP signals were also found in the sinusoids. These granular EGFP signals did not show nuclear stain, and most likely were the sludge of platelets (Fig. 2A, thin arrows). The frequency of EGFP+ nucleated cells (excluding platelets) was 51 ± 6 cells/high-power field (×20) in untreated WT liver grafts at 1 hour after transplantation (Fig. 2B). Transmission electron microscopy and scanning electron microscopy analyses of liver grafts confirmed significant platelet deposition on the sinusoids during hepatic I/R injury (Fig. 2C). With CO inhalation, the frequency of EGFP+ cells was significantly decreased to 39 ± 7 cells/high-power field, and granular EGFP signals also were reduced. When liver grafts were transplanted freshly with a minimum CIT of 1 hour, EGFP+ cells were 34 ± 8/high-power field, and there was no sludge or plateletlike EGFP signal in liver grafts.
EGFP+ Host Cells Were CD45+ Leukocytes and Do Not Include Host Macrophages
To characterize EGFP+ host cells, immunohistochemistry was conducted on 1-hour liver samples using several lineage markers. As expected, nucleated EGFP+ cells were CD45+ leukocytes (Fig. 3A). EGFP+ granular signals did not express CD45, further indicating that they could be the platelets. CO inhalation efficiently reduced the extravasation of host EGFP+ CD45+ cells and platelet deposition in the liver graft. CD45+ EGFP− graft leukocytes were also identified in the graft liver. Interestingly, however, staining for macrophage markers revealed that ED1 (CD68) and ED2 (CD163) were not colocalized with EGFP, and all ED1+ or ED2+ cells appeared to be EGFP− donor Kupffer cells (Fig. 3B). The result indicates that host macrophages were not recruited at this early time point of reperfusion.
To confirm the immunohistochemical findings, we performed flow cytometry of hepatic NPCs isolated from liver grafts at 1 hour after transplantation. Double stain with CD68 (or CD163) and CD45 revealed that CD68+ (or CD163+) cells only faintly expressed CD45, indicating that CD45+ cells and CD68+ cells were two distinct populations (Fig. 3C). In the CD45+ population, 75.4% ± 3.0% were EGFP+ host cells in air control grafts. With CO treatment, the percentages of EGFP+ host cells in CD45+ population slightly reduced to 72.0% ± 5.8%. Nearly all of CD68+ cells were EGFP− donor phenotype in both air-treated and CO-treated groups.
Further flow cytometry analysis of CD45+ hepatic NPCs in liver grafts revealed that both EGFP− donor and EGFP+ host leukocytes included αβTCR+ cells, natural killer cells, B cells, and CD11 b/c+ myeloid cells. There was no apparent difference in the composition of CD45+ NPCs between air-treated and CO-treated groups, and αβTCR+ cells were the predominant cell type in the host population, followed by immunoglobulin M–positive B cells and natural killer cells (Fig. 4).
Hepatocytes and NPCs Respond Differently to I/R Injury by Up-Regulating Distinctive Sets of Mediators
In hepatic I/R injury, mRNA for TNF-α and IL-6 promptly increased at 1 hour after reperfusion in whole liver graft samples, and CO showed potent inhibitory effects. To determine the mechanisms of CO's cytoprotective action, we next identified the cell population responsible for proinflammatory cytokine up-regulation and regulated by CO treatment. Hepatocytes and hepatic NPCs were isolated, and RT-PCR was conducted using each of isolated cell populations. TNF-α and IL-6 mRNA levels increased exclusively in NPCs, but not hepatocytes, and inhaled CO efficiently inhibited these cytokine mRNA levels in NPCs (Fig. 5). These results confirm that the prompt increase of TNF-α and IL-6 mRNA observed in the whole liver samples is a reflection of NPC responses to hepatic I/R injury and that inhaled CO inhibits up-regulation of these cytokines by regulating NPCs. Although isolated hepatocytes did not show proinflammatory cytokine increases, they showed strong up-regulation of iNOS at 1 hour after reperfusion. Other stress responsible enzyme HO-1 up-regulation was seen in the hepatic NPC fraction and was efficiently inhibited with CO (Fig. 5). The result showed that various types of cells in the liver responded differently to I/R injury by up-regulating distinctive sets of mediators, and early proinflammatory down-regulation with CO appeared to be attributable to CO's action on hepatic NPCs.
Kupffer Cells Induce Early Proinflammatory Responses
To further identify the CO's target cell population for proinflammatory down-regulation, NPC fraction was separated into CD45+ and CD45− populations using Auto MACS. RT-PCR analysis showed that TNF-α and IL-6 mRNA levels increased in the CD45− fraction and that inhaled CO down-regulated proinflammatory responses of CD45− cells (Fig. 6). Host CD45+ cells did not appear to play roles in prompt up-regulation of TNF-α and IL-6 after reperfusion.
Because Kupffer cells are known to be the major cell type to produce cytokines, we next isolated CD68+ NPCs from liver grafts using Auto MACS. In untreated liver grafts at 1 hour, mRNA levels for TNF-α and IL-6 increased in the CD68+, but not the CD68−, NPC fraction. Inhaled CO significantly inhibited these cytokine increases in CD68+ cells (Fig. 7A). Because there were no host CD68+ cells at this early time point after transplantation (Fig. 3C), the result validated the effects of CO in down-regulating CD68+ Kupffer cells for the hepatic protection.
We next inhibited graft Kupffer cells with GdCl3 and examined TNF-α mRNA levels in whole liver grafts. GdCl3 treatment significantly decreased ED1 expression in the liver (Fig. 7B). At 1 hour after transplantation, GdCl3-treated liver grafts with 18 hours' cold preservation showed significant reduction of TNF-α mRNA levels, which were similar to those seen in CO 100 ppm inhalation group (Fig. 7B). Combination of GdCl3 donor pretreatment and recipient inhaled CO (100 ppm) further decreased TNF-α mRNA up-regulation, suggesting that these methods had additive effects in inhibiting TNF-α up-regulation.
CO Inhibits In Vitro Kupffer Cell Activation With LPS
The effect of CO in regulating Kupffer cells was directly examined in a primary rat Kupffer cell culture system. Kupffer cells were obtained from naïve Lewis rat livers and stimulated with LPS. Incubation of Kupffer cells with culture medium containing CO for 2 hours before LPS exposure inhibited TNF-α and IL-6 production at 6 hours in a dose-dependent manner (Fig. 8). Viability of Kupffer cells was not influenced by brief exposure to CO.
Cold I/R injury in liver grafts is characterized as a cascade of prominent inflammatory events involving various types of cells of graft and host origins. Prompt inflammatory response in liver grafts was evident in our model of cold hepatic I/R injury by the elevation of mRNA levels for proinflammatory mediators (for example, TNF-α, IL-6, iNOS) at 1 to 3 hours of reperfusion. To determine the relative contribution of host and donor cells in the early proinflammatory responses in I/R injury, we have conducted liver transplantation with 18 hours' cold storage in the EGFP rat system. Recruitment of host cells was clearly visible as abundant EGFP-tagged CD45+ host cells in WT liver grafts at 1 hour after reperfusion. However, when cell fractions were isolated from liver grafts and analyzed for inflammatory responses, the CD45+ NPC population, including host CD45+ cells, did not contribute to early mRNA up-regulation for TNF-α or IL-6. Instead, isolated CD68+ NPC population (CD 45dim) showed a significant increase of mRNA for these cytokines, indicating that graft CD68+ Kupffer cells were responsible for the initial proinflammatory reaction of hepatic cold I/R injury in this model. Importantly, inhaled CO in this study inhibited CD68+ Kupffer cell activation and proinflammatory up-regulation. Together with in vitro Kupffer cell culture experiment in this study, the results indicate that CO protects liver grafts against I/R injury through the down-regulation of Kupffer cell responses. The finding is well supported by the previous in vitro studies of murine RAW264.7 macrophages in which CO efficiently modulates the proinflammatory reaction elicited by LPS.24–26
Prominent roles of Kupffer cells in inflammatory reactions in hepatic I/R injury have been recognized because of the reactive features of this cell population. To directly determine the role of Kupffer cells in hepatic cold I/R injury, several previous studies have attempted to deplete or inactivate Kupffer cells using liposome-encapsulated dichloromethylene diphosphonate (LPDD) or GdCl3. However, these studies demonstrated conflicting results, depending on the model and endpoints used in the study. Imamura et al.27 and Reinders et al.28 showed that Kupffer cell depletion with LPDD did not influence the function or survival of rat liver grafts stored for 24 hours. In contrast, using the liver perfusion circuit, Kupffer cell depletion with LPDD attenuated I/R injury-induced hepatocyte dysfunction (for example, biliary anion transport, detoxification capacity)29 but did not alter the reperfusion-induced periportal oxidative changes or improve the recovery of bile output.30 Using GdCl3-treated livers, Schauer et al.31 showed improved sinusoidal microcirculation after liver transplantation but failed to reverse early graft dysfunction. In the study of Bradham et al.,32 Kupffer cell blockade with GdCl3 reduced graft IL-6 levels but did not affect TNF-α up-regulation at 3 hours after transplantation of 24 hours' preserved liver grafts.32 Furthermore, in the mouse 30% liver transplant model, Tian et al.33 showed that depletion of Kupffer cells reduced liver injury, enhanced regeneration, and improved recipient survival. In the pig liver transplantation model, Kupffer cell depletion by GdCl3 pretreatment was effective in preventing liver graft dysfunction and nonfunction during the follow-up period of 24 hours.34 Interestingly, during hepatic warm I/R injury, Kupffer cells are consistently identified as a major contributor of the early postischemic oxidant stress and inflammatory reactions, and inactivation/depletion of Kupffer cells results in attenuation of injury.35–38 Thus, roles of Kupffer cells in hepatic cold I/R injury are puzzling, and this might suggest multiple functions of Kupffer cells. Although Kupffer cells have potent proinflammatory activities and could promote hepatic injury, they also have an important protective function in the liver through the production of a variety of modulating factors (for example, interleukin-10, cyclooxygenase-derived mediators) that may counteract with inflammatory responses or stimulate liver regeneration, collagen metabolism, and resolution of fibrosis.39–41 Furthermore, roles of proinflammatory mediators (for example, TNF-α, IL-6) in I/R injury are also complicated. Conzelmann et al.42 conducted liver transplants in tumor necrosis factor receptor 1–negative mice and showed that the lack of TNF signals in liver grafts in fact increased graft injury, whereas the lack of tumor necrosis factor receptor-1 in recipients ameliorated hepatic injury. Exogenously administered IL-6 also prevented hepatic I/R injury.43, 44 Further investigations of the relative roles and contributions of various cell types are required for precise understandings of the complicated proinflammatory events of I/R injury.
Although GdCl3 has been frequently used in experimental animals to investigate the role of Kupffer cells in normal and pathological conditions, precise mechanisms of its actions and efficacy are not well defined. Numerous studies agree that GdCl3 inhibits phagocytosis of colloidal carbon, 51Cr-labeled erythrocytes, and fluorescent latex beads by Kupffer cells.45–47 GdCl3 appears to affect the surface attachment and engulfment phases of phagocytosis. In the physiological pH condition in the blood circulation, injected GdCl3 is expected to form colloidal aggregates, which are further integrated into serum proteins and taken up by Kupffer cells. In the endosomal–lysosomal compartment in Kupffer cells, gadolinium ions are released from internalized aggregates, resulting in the building of GdCl3 molecules into the compartment and subsequently into the surface membrane site. GdCl3 is suggested to impair the first phase of phagocytosis via the inhibition of membrane calcium movement.46 These actions may explain the loss of binding capability of rat Kupffer cells for monoclonal antibody ED2 (CD 163) with GdCl3.47 GdCl3 additionally has been shown to influence other hepatic functions and activity. In vivo administration of GdCl3 marginally increases alanine aminotransferase. It also releases cytokines and increases serum cytokine levels (for example, TNF-α, IL-6).48 Electron microscopy shows significant morphological alterations of Kupffer cells with GdCl3, such as the loss of typical Kupffer cell features (peroxidase activity in endoplasmic reticulum and nuclear envelope) and the shift to a rounded appearance, suggesting that GdCl3 may eliminate some Kupffer cells.47 Thus, GdCl3 certainly inhibits Kupffer cells' phagocytotic function with some deletional effect; however, it does not provide a clean investigation of Kupffer cell–eliminated liver grafts. Likewise, LPDD has also been used to eliminate Kupffer cells; however, this method is shown to affect dendritic cell populations.49, 50 Thus, there is no clean method currently available to selectively eliminate Kupffer cells. Although the GdCl3 experiment in this study suggests the involvement of Kupffer cells in early proinflammatory responses of hepatic I/R injury, further studies are needed to confirm the finding.
Brief recipient CO inhalation during and immediately after surgery could be applicable via a ventilation device in clinical transplantation; however, significant obstacles may exist in applying in vivo CO treatment, because of CO's possible adverse effects. Accordingly, we have been investigating the usage of CO for excised grafts to minimize the concerns associated with in vivo CO administration. We have recently shown the efficacy of ex vivo treatment of excised intestinal grafts with gaseous CO in preventing I/R injury.51 The study is underway to examine whether the same ex vivo CO application is effective in preventing hepatic I/R injury.
The current study using the EGFP transgenic rat system clearly demonstrated the recruitment of abundant host CD45+ EGFP+ leukocytes in the liver as early as 1 hour after reperfusion. Moreover, the striking observation in this study was the presence of numerous platelet-like EGFP-tagged sludge/debris (CD45−) in the sinusoidal spaces of untreated grafts. Sinusoidal endothelial cell injury and increased expression of adhesion molecules (such as intracellular adhesion molecule-1) and P-selectin in I/R injury likely activate and recruit host inflammatory cells/platelets.52, 53 Alternatively, host cells might be directly sequestered out of the vasculature because of the loss of the sinusoidal lining.1, 23 Although host cells were found in liver grafts, they did not appear to contribute to early proinflammatory reactions. However, they certainly deteriorate microvascular perfusion and might have critical roles during the later phase of hepatic I/R injury.54
In summary, using the WT-to-EGFP transgenic rat liver transplant model, the current study detected early infiltration of numerous host leukocytes and platelets in WT liver grafts. Isolation and analyses of different populations of cells from liver grafts showed that early proinflammatory response was initiated in graft Kupffer cells, and CO ameliorated hepatic I/R injury by down-regulating graft Kupffer cells.
The authors thank Dr. Masaru Okabe, Osaka University, Japan, for the EGFP rats, Mike Tabacek and Lisa Chedwick for their excellent technical support, and Carla Forsythe for preparation and organization of the manuscript.