Potential conflict of interest: Nothing to report.
Supported by grants from the National Institutes of Health (NIH), National Institute of Allergy and Infectious Diseases (NIAID) R01AI057832, UCLA Academic Senate, and the Pfleger Foundation (to A.J.C.). S.D. was supported in part by a fellowship from the Fundação para a Ciência e Tecnologia (FCT), Portugal.
Hepatic ischemia/reperfusion (IRI) injury remains a major challenge in clinical orthotopic liver transplantation (OLT). Tenascin-C (Tnc) is an extracellular matrix protein (ECM) involved in various aspects of immunity and tissue injury. Using a Tnc-deficient mouse model, we present data that suggest an active role for Tnc in liver IRI. We show that Tnc-deficient mice have a reduction in liver damage and a significant improvement in liver regeneration after IRI. The inability of Tnc−/− mice to express Tnc significantly reduced the levels of active caspase-3/transferase-mediated dUTP nick end-labeling (TUNEL) apoptotic markers and enhanced the expression of the proliferation cell nuclear antigen (PCNA) after liver IRI. The lack of Tnc expression resulted in impaired leukocyte recruitment and decreased expressions of interleukin (IL)-1β, IL-6, and CXCL2 after liver reperfusion. Tnc-deficient livers were characterized by altered expression patterns of vascular adhesion molecules, such as vascular cell adhesion molecule-1 and platelet endothelial cell adhesion molecule-1 post-IRI. Moreover, matrix metalloproteinase-9 (MMP-9) synthesis, which facilitates leukocyte transmigration across vascular barriers in liver IRI, was markedly down-regulated in the absence of Tnc. We also show that Tnc is capable of inducing MMP-9 expression in isolated neutrophils through Toll-like receptor 4. Therefore, our data suggest that Tnc is a relevant mediator of the pathogenic events underlying liver IRI. The data also support the view that studies aimed at further understanding how newly synthesized ECM molecules, such as Tnc, participate in inflammatory responses are needed to improve therapeutic approaches in liver IRI. (HEPATOLOGY 2011)
Hepatic ischemia/reperfusion injury (IRI) occurs during trauma, shock, transplantation, and other surgical procedures where the blood supply to liver is temporarily interrupted. In transplantation, IRI insult can lead to a significantly higher incidence of acute and chronic rejections.1 Hepatocellular damage caused by IRI is the result of complex interactions between various inflammatory mediators, including leukocytes, proinflammatory cytokines, and free radicals.2, 3
Extracellular matrix (ECM) proteins can act as inflammatory stimuli by promoting leukocyte migration and by inducing the expression of proinflammatory cytokines and growth factors.4 These ECM molecules are considered endogenous danger signals, or damage-associated molecular patterns (DAMPs).5 Tenascin-C (Tnc) is likely a prominent DAMP molecule within the ECM.6 Tnc is a hexameric protein of 1.5 million Da consisting of a single fibrinogen-like domain (FBG) and multiple fibronectin (FN)-type III and epidermal growth factor (EGF)-like domains.7 Tnc is often called an oncofetal molecule because of its unique expression8; it is abundantly expressed during embryogenesis, and is not normally expressed in adult tissues, with the exception of specific hematopoietic and lymphoid tissues.9 Notably, Tnc emerges in adult tissues during conditions associated with high rates of cell turnover and migration, including wound repair, tumorigenesis, rheumatoid arthritis, multiple sclerosis, and hepatitis.5, 7, 10, 11
The action of Tnc is complex; it is accepted that Tnc is involved in cell-cell interactions, cell-adhesion, and deadhesion events.12 Tnc can influence cell behavior through interactions with cell surface receptors and by altering cell binding to other ECM proteins.13 Leukocytes can form low-avidity adhesions to Tnc, producing tethering and rolling under flow in a more efficient manner than on selectins.14 Although in vitro assays have supported a role for Tnc in leukocyte migration and activation, its precise function at sites of tissue injury remains mostly unclear. It has been recently demonstrated that Tnc induces cytokine release and tissue destruction in arthritic joint disease.6 The role of Tnc at sites of inflammation is defined by the type of tissue injury or by the participating cell types.13 In the present study, we used Tnc−/− mice to examine the significance of Tnc expression in hepatic IRI.
Tnc-deficient (Tnc−/−) knockout (KO) mice (C57BL/6N-TgH) were obtained from Riken, Japan.15 Toll-like receptor 4-deficient (TLR4−/−) mice (C57Bl/10ScNJ-Tlr4) and C57Bl/10J controls (wildtype, WT) were purchased from the Jackson Laboratory. Tnc−/− mice were rederived by sterile embryo transfer to surrogate mothers and housed in the UCLA animal facility under specific pathogen-free conditions.
Model of Partial Lobar Liver IRI.
A warm hepatic IRI model was performed in 10-week-old male Tnc−/− mice and matched Tnc+/+ WT littermates, as previously described.16 Briefly, the arterial and portal venous blood supplies were interrupted to the cephalad lobes of the liver for 90 minutes using an atraumatic clip. After 90 minutes of ischemia the clip was removed, thus initiating hepatic reperfusion.
Assessment of Liver Damage.
Serum alanine transaminase (ALT) and aspartate transaminase (AST) levels were measured in blood samples with an autoanalyzer by Antech Diagnostics (Los Angeles, CA). Liver specimens were fixed in a 10% buffered formalin solution, embedded in paraffin, and processed for hematoxylin and eosin (H&E) staining. The degree of hepatic necrosis was assessed in H&E-stained paraffin sections; H&E stains were digitally photographed and the percent of necrotic was quantified using NIH ImageJ software in a blind manner to the different experimental groups as described.17 Ten random sections per slide were evaluated in duplicate to determine the percentage of necrotic area.
Myeloperoxidase Assay (MPO).
MPO activity was evaluated as described.16 Frozen tissue was homogenized in an iced solution of 0.5% hexadecyltrimethyl-ammonium (Sigma, St. Louis, MO) and 50 mmol/L of potassium phosphate buffer solution (Sigma) with pH adjusted to 5. After centrifugation the supernatants were mixed in a solution of hydrogen peroxide-sodium acetate and tetramethyl benzidine (Sigma). The quantity of enzyme degrading 1 μmol/L of peroxide per minute at 25° C per gram of tissue was defined as 1U of MPO activity.
RNA Extraction and Reverse Transcriptase Polymerase Chain Reaction (PCR).
For evaluation of gene expression, livers were harvested and RNA was extracted with Trizol (Life Technologies, Grand Island, NY) as described.16 Reverse transcription was performed using 5 μg of total RNA in a first-strand cDNA synthesis reaction with SuperScript III RNaseH Reverse Transcriptase (Invitrogen Life Technologies, CA), as recommended by the manufacturer. Densitometric quantifications were performed using the NIH ImageJ software.
Immunohistology was performed in cryostat sections as described.16 Antibodies against mouse macrophage antigen-1 (Mac-1; M1/70BD), Ly-6G (1A8), intercellular adhesion molecule (ICAM-1; 3E2), platelet endothelial cell adhesion molecule-1 (PECAM-1; MEC13.3), all from BD Biosciences (San Jose, CA); tenascin-C (Tnc; 6-6B; Calbiochem, San Diego, CA), vascular cell adhesion molecule-1 (VCAM-1; MVCAM A 429; Serotec, Raleigh NC); MMP-9 (AF909; R&D Systems, Minneapolis, MN); and proliferating cell nuclear antigen (PCNA; PC10; Lab Vision, Fremont, CA) were used at optimal dilutions. The sections were evaluated blindly by counting labeled cells in triplicate in 10 high-power fields per section.
Western blots were performed as described.16 Briefly, proteins (50 μg/sample) in sodium dodecyl sulfate (SDS)-loading buffer were electrophoresed through 10%-15% SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to PVDF membranes (Thermo Fisher). Membranes were incubated with specific antibodies against BclxL, caspase-3, cleaved-caspase-3, all from Cell Signaling (Beverly, MA), Bcl2 (Abcam, Cambridge, MA), and Cyclin D1 (DCS-6; BD Biosciences). After development, membranes were stripped and reblotted with an antibody against actin (Santa Cruz Biotechnology). Bands were visualized using SuperSignal West Pico Chemiluminescent Substrate (Pierce). Relative quantities of protein were determined by densitometry using the NIH ImageJ software.
Protein extraction and zymography analyses were performed as described.16 Briefly, gelatinolytic activity was detected in liver extracts at a final protein content of 100 μg by 10% SDS-PAGE containing 1 mg/mL of gelatin (Invitrogen) under nonreducing conditions. After incubation in development buffer (50 mmol/L Tris-HCl, 5 mmol/L CaCl2, and 0.02% NaN3, pH 7.5), gels were stained with Coomassie brilliant blue R-250 (Bio-Rad, Hercules, CA) and destained with methanol/acetic acid/water (20:10:70). A clear zone indicated enzymatic activity. Positive controls for MMP-9 (Biomol International, Plymouth PA), and prestained molecular weight markers (Bio-Rad Laboratories) served as standards.
The TUNEL assay was performed on 5-μm cryostat sections using the In Situ Cell Death detection kit (Roche Diagnostics, Temecula, CA) according to the manufacturer's instructions and as described.18 The sections were evaluated blindly by counting labeled cells in triplicate in 10 high-power fields per section.
Isolation of adult murine neutrophils from bone marrow was performed as published.16, 18 Briefly, femurs and tibias were harvested from Tnc−/−, TLR-4−/−, or WT mice and stripped of all muscle and sinew, and bone marrow was flushed with 2.5 mL of Hanks' balanced saline solution (HBSS) containing 0.1% (wt/vol) bovine serum albumin (BSA) and 1% (wt/vol) glucose on ice. Cells were pelleted and erythrocytes were removed by hypotonic lysis. The bone marrow preparation was resuspended at 5 × 107 cells/mL in HBSS. Cells were layered on a Percoll (Sigma-Aldrich) gradient (55% Percoll, top; 65% Percoll, middle; 80% Percoll, bottom). Mature neutrophils were recovered at the interface of the 65% and 80% fractions and were more than 90% pure and more than 95% viable in the neutrophil-rich fraction. Isolated neutrophils were placed on 24-well Tnc or polylysine-coated plates at 5 × 106 cells/well and incubated at 37°C, 5% CO2 for 6 hours. In parallel, isolated neutrophils were stimulated with lipopolysaccharide (LPS) (1 ng/mL) or interleukin (IL)-6 (100 U/mL) for 6 hours as described.18 Gelatinolytic activity was detected in cell supernatants by zymography, as described above. Data are presented as fold increase over controls.
All values are expressed as the mean ± standard deviation (SD). Differences between groups were compared using Student's t test and a two-tailed P value < 0.05 was considered significant. Calculations were made using SPSS software (Chicago, IL).
Tnc Expression in Hepatic IRI.
Tnc messenger RNA (mRNA) expression was hardly detected in naive WT livers and it was significantly up-regulated in Tnc+/+ livers post-IRI (Fig. 1A). These results were correlated with our immunohistological observations, in which Tnc protein expression was virtually undetectable in naive livers, and it was readily deposited in the vascular areas of WT livers after 6 hours and 24 hours post-IRI (Fig. 1C). Tnc was undetectable in Tnc−/− livers before and after IRI (Fig. 1B,C).
Tnc Deficiency Improved Liver Function and Histology After Liver IRI.
There were no apparent differences in either the very low transaminase levels or liver histology between naive Tnc−/− and Tnc+/+ mice. Tnc−/− and Tnc+/+ mice showed also comparable serum transaminase levels (U/L), vascular congestion and necrosis at 6 hours (AST: 13,977 ± 2,620 versus 14,362 ± 3,489; ALT: 44,400 ± 17,154 versus 32,200 ± 10,563; n = 4 mice/group), and 12 hours (AST: 6,170 ± 6,028 versus 11,825 ± 8,384; ALT: 37,400 ± 7,213 versus 36,400 ± 8,854; n = 4 mice/group) post-IRI (Fig. 2A,B). In contrast, the transaminase levels (U/L) were profoundly depressed in Tnc−/− mice (AST: 1,902 ± 1,435 versus 9,008 ± 1,774; P < 0.001; and ALT: 2,067 ± 1,436 versus 27,340 ± 6,834; P < 0.01; n = 4-5 mice/group) at 24 hours post-IRI (Fig. 2A). A sustained effect was observed in the Tnc−/− mice with transaminase levels (AST: 545 ± 270 versus 1,445 ± 544; P < 0.05; and ALT: 786 ± 335 versus 5,060 ± 2,022; P < 0.05; n = 3 mice/group) significantly decreased at 48 hours post-IRI (Fig. 2A). Moreover, improvement of liver function in the Tnc−/− mice was correlated with better histological preservation. Although Tnc+/+ livers were characterized by elevated sinusoidal congestion and extensive necrosis, Tnc−/− livers showed relatively modest signs of vascular changes or necrosis 24 hours (Fig. 2B). Compared with Tnc+/+ controls, Tnc−/− mice demonstrated a 5-fold lower level of hepatocellular necrosis at 24 hours (n = 4/group; P < 0.01) (Fig. 2C). These data strongly support an important role for Tnc expression in the perpetuation of liver damage after IRI.
Tnc Deficiency Impaired Caspase-3 Activation in Liver IRI.
Tnc−/− livers showed significantly lower numbers of TUNEL+ cells with hepatocyte morphology (47.2 ± 17.1 versus 128.7 ± 9.8, P < 0.01) at 6 hours post-IRI (Fig. 3A,B). Caspase-3 is an apoptotic effector caspase,19 expressed in tissues as an inactive 32-35 kDa precursor, and cleaved during apoptosis generating the 19-20 kDa fragment and the mature active 17 kDa subunit. Although the pro-caspase-3 was present in all liver samples, the active 17-kDa caspase-3 was predominantly detected in Tnc+/+ control livers at 6 hours post-IRI. The active caspase-3 was significantly depressed in Tnc−/− livers at 6 hours of IRI (P < 0.05) and virtually undetected in naive livers (Fig. 3C,D). Protection against apoptosis may also involve antiapoptotic mechanisms; however, Bcl-2, an antiapoptotic molecule, was rather enhanced in Tnc+/+ livers and modestly expressed in both naive and Tnc−/− livers post-IRI (Fig. 3C). Our results are consistent with earlier observations that expression of Bcl-2 can be induced in livers as a response to the postreperfusion apoptotic stress,20 which seems to be reduced in Tnc−/− mice. Indeed, it has been shown that ischemic preconditioning down-regulated caspase-3 activity and inhibited apoptosis in livers post-IRI, despite lower levels of Bcl-2 expression detected in the preconditioned livers.20 Morphologic alterations of apoptosis are considered to be mostly mediated by caspases and cell death can occur by way of caspase-dependent and Bcl2-independent pathways.19, 21, 22 Therefore, our results provide an indication that Tnc−/− mice are less sensitive to apoptosis induced by liver IRI, regardless of showing comparable transaminase levels at 6 hours postreperfusion. Although necrosis has been shown to correlate with serum transaminases,23 apoptosis can occur without altering transaminase levels24; this can perhaps be explained by observations that, in contrast to necrosis, apoptotic cells maintain their plasma membrane integrity until late.25
Liver Regeneration After IRI Is Improved in the Absence of Tnc.
We evaluated the impact of Tnc deficiency on the hepatic regenerative response post-IRI. Cyclin D1 is normally expressed in livers26 and reduced in impaired liver regeneration.27 Cyclin D1 expression was detected in significantly higher levels in Tnc−/− livers at 24 hours post-IRI (Fig. 4A,B). To determine whether Tnc expression interferes with proliferation after IRI, the number of S-phase cells was assessed by PCNA staining. Indeed, proliferating hepatocytes (PCNA Index %) were detected in increased numbers in the Tnc−/− livers (64.5 ± 3.9 versus 18.3 ± 6.4, P < 0.001; n = 4/group) at 24 hours after IRI, suggesting that regeneration occurs earlier in the absence of Tnc (Fig. 4C,D).
Tnc Deficiency Disrupted Leukocyte Accumulation and Expressions of IL-1β and IL-6 in Hepatic IRI.
MPO activity was reduced in Tnc−/− livers at 6 hours (3.23 ± 0.74 versus 7.03 ± 1.71 U/g; P < 0.05) and 24 hours (2.25 ± 1.03 versus 11.43 ± 2.32 U/g; P < 0.01) post-IRI (Fig. 5A). Ly-6G neutrophil infiltration was clearly depressed in the portal areas of Tnc−/− livers at 6 hours (6.8 ± 2.6 versus 29.3 ± 11.2; P < 0.05) and 24 hours (21.3 ± 8.4 versus 64.7 ± 7.3; P < 0.05) post-IRI (Fig. 5B). Mac-1 leukocytes were also significantly reduced in the Tnc−/− livers at 6 hours (15.2 ± 8.9 versus 49.1 ± 13.9; P < 0.01) and 24 hours (29.2 ± 13.7 versus 85.9 ± 8.7; P < 0.05) post-IRI (Fig. 5C). Moreover, the expression of IL-1β was significantly depressed in Tnc−/− livers after 12 hours (P < 0.04) and 24 hours (P < 0.04) of reperfusion (Fig. 5D). Furthermore, the expressions of CXCL-2 (P < 0.03), a potent neutrophil chemoattractant,28 and IL-6 (P < 0.04) were significantly down-regulated in the Tnc−/− livers at 24 hours post-IRI (Fig. 5D).
VCAM-1, ICAM-1, and PECAM-1 Showed Distinct Patterns of Expression in the Absence of Tnc.
VCAM-1 expression was virtually absent in naive livers and it was up-regulated in the large vessels of Tnc+/+ livers at 6 hours and 24 hours post-IRI. In contrast, VCAM-1 was almost absent in Tnc−/− livers at 6 hours post-IRI and it was significantly reduced in these livers at 24 hours postreperfusion, suggesting that there was a disruption on VCAM-1 deposition in the absence of Tnc (Fig. 6A). PECAM-1 is normally expressed on sinusoidal cells and down-regulated under inflammatory conditions30; in our settings, PECAM-1 expression was relatively high in naive livers and down-regulated in Tnc+/+ livers at both 6 hours and 24 hours post-IRI. On the other hand, PECAM-1 expression was fully restored in Tnc−/− livers at 24 hours after reperfusion, supporting that the absence of Tnc slows, or even halts, the progression of inflammation (Fig. 6B). ICAM-1 showed a similar pattern of expression to PECAM-1. ICAM-1 expression was depressed in both Tnc−/− and Tnc+/+ livers at 6 hours post-IRI; however, whereas ICAM-1 expression remained depressed in the necrotic Tnc+/+ livers it was restored to almost normal levels in the Tnc−/− livers at 24 hours post-IRI (Fig. 6C).
Tnc Deficiency Depressed the Expression/Activity of MMP-9 in Hepatic IRI.
MMP-9 activity, assessed by zymography, was reduced in Tnc−/−-deficient livers at 6 hours (≈1.7-fold; n = 4 mice/group; P < 0.05) and 24 hours (≈4.4-fold; P < 0.003) post-IRI (Fig. 7A). Moreover, MMP-9-positive leukocytes were markedly depressed in Tnc−/− livers at 6 hours (11.8 ± 5.8 versus 31.8 ± 9.4; P < 0.05) and at 24 hours (18.8 ± 7.4 versus 59.8 ± 7.3; P < 0.05) after IRI, in contrast with high levels of MMP-9+ leukocytes in controls (Fig. 7B,C).
To further support our observations, we assessed whether exogenous Tnc can regulate MMP-9 expression in isolated neutrophils. Tnc−/− and Tnc+/+ neutrophils cultured in Tnc-coated plates both showed about a 2-fold increase (P < 0.05) in MMP-9 expression/activity compared with controls (Fig. 7D). However, whereas IL-6 significantly stimulated MMP-9 activity in Tnc+/+ neutrophils, it was rather inefficient in up-regulating MMP-9 activity in Tnc−/− neutrophils (Fig. 7D); future experimentation is needed to explain this result. Thus, these data show that Tnc favors a significant increase in MMP-9+ expression/activation in liver IRI.
Tnc-Mediated MMP-9 Expression Is TLR4-Dependent.
Recent findings have identified Tnc as a ligand of TLR4.6 TLR-4 is expressed by cells of the immune system and is considered to mediate inflammation and liver damage after IRI.32 Therefore, we attempted to investigate whether Tnc stimulates up-regulation of MMP-9 activity in isolated neutrophils by way of TLR4. We found that the levels of MMP-9 activity were significantly depressed in neutrophils isolated from TLR4−/− mice and stimulated with Tnc as compared with controls (Fig. 8). Moreover, whereas LPS induction of MMP-9 activity was TLR4-dependent, IL-6 stimulated similar MMP-9 activity in neutrophils isolated from TLR-4−/− and from WT mice, suggesting that IL-6 regulates MMP-9 activity independently of TLR4 (Fig. 8). Therefore, our results provide evidence that Tnc is capable of regulating MMP-9 activity through TLR4.
In the present study we investigated the functional significance of Tnc expression in liver IRI. We show that Tnc is virtually absent in naive livers and readily detected in the vascular areas of damaged WT livers after IRI. Tnc-deficient mice showed (1) improved liver transaminases and histological outcomes; (2) decreased necrosis; (3) reduced caspase-3 activity and apoptotic cell labeling; (4) enhanced liver regeneration; (5) impaired leukocyte infiltration and IL-1β and IL-6 expressions; (6) altered patterns of expression of adhesion molecules; and (7) down-regulated MMP-9 expression/activation after liver IRI. Moreover, in vitro investigations provided evidence that exogenous Tnc (8) regulated MMP-9 activity in WT and Tnc−/− neutrophils, and (9) induced MMP-9 expression through TLR-4.
Tnc is virtually not expressed in most healthy adult tissues, but its expression is specifically and rapidly induced in response to injury.33 Tnc has been identified as an endogenous ligand of TLR-46 and its expression has been linked to inflammatory conditions, which include rheumatoid arthritis,6 cancer,13 liver fibrosis,34 and chronic hepatitis.11, 35 Indeed, given the significant role that Tnc may have in inflammation, drugs targeting Tnc are currently being developed or undergoing clinical trials.13 In the present work, Tnc deposition was readily up-regulated in the liver vascular areas of WT Tnc+/+ livers post-IRI, with a similar pattern of distribution to that of chronic hepatitis.11 We used Tnc−/−-deficient mice to test the relevance of Tnc expression to liver IRI. Tnc−/−-deficient mice have previously been reported to have no grossly phenotypic abnormalities.36 Tnc−/− mice, compared with their WT counterparts, were less susceptible to liver IRI. Although Tnc deficiency did not particularly affect transaminase levels or histological preservation at 6 hours and 12 hours post-IRI, it was highly effective in ameliorating liver function/damage at 24 hours and 48 hours postreperfusion. Tnc−/− livers exhibited limited liver necrosis at 24 hours post-IRI, contrasting with extensive necrosis in controls. Hepatocyte death by both necrosis and apoptosis is a prominent feature of liver IRI.37 Indeed, caspase-3 activation was significantly reduced in Tnc−/− livers as compared with control littermates after IRI, and was accompanied by a reduced number of TUNEL-positive cells in the Tnc-deficient livers. ECM proteins can act either as survival or proapoptotic mediators.38 In this regard, it has been shown that the epidermal growth factor-like (EGF-L) domains of Tnc are capable of exerting caspase-dependent proapoptotic activities.39
In addition to hepatocyte cell death, impaired liver regeneration/repair is common in acute liver failure. Here we show that the number of PCNA-positive hepatocytes appeared increased in the Tnc−/− mice after IRI, suggesting that hepatocytes progress faster into the S phase of the cell cycle in the absence of Tnc. Moreover, cyclin D1, important in the transition from the G1 to S phase, was significantly depressed in controls and almost normally expressed in the Tnc−/−-deficient livers post-IRI. Inhibition of cyclin D1 leads to growth arrest40 and to impaired hepatic regeneration.27 However, the role of PECAM-1 in liver IRI is still not clear; PECAM-1 is constitutively expressed on sinusoidal endothelial cells and its loss is regarded as a marker of sinusoidal injury.41 In our settings, PECAM-1 expression was depressed in livers post-IRI and it was restored to normal levels sooner in the Tnc−/− livers. The intact expression of PECAM-1 along the sinusoids has been associated with less sinusoidal congestion/inflammation,30 suggesting that preventing PECAM down-regulation may be valuable in the treatment of early stages of liver damage.42 These observations support the view that hepatic regeneration/repair post-IRI is enhanced in the absence of Tnc. Moreover, they are consistent with the reduction of liver necrosis observed earlier in the Tnc−/−-deficient mice post-IRI. Therefore, our results agree with previous findings that Tnc mediates a persistent inflammation,6 which possibly interferes with liver regeneration and contributes to the perpetuation and aggravation of necrosis post-IRI.
Leukocyte infiltration is a hallmark feature in hepatic IRI. Indeed, neutrophils, critical mediators in acute inflammatory liver injury,43 were significantly decreased in Tnc−/− livers post-IRI. Macrophages were also depressed in the Tnc−/− livers post-IRI. Tnc is a ligand for several integrin receptors present on leukocytes, and it has been linked to diverse effects on cell migration that result from differences in cell type and in vitro assays.13 CXCL2, a cytokine-induced neutrophil chemoattractant,46 was rather down-regulated in the Tnc−/− livers post-IRI, suggesting its participation in neutrophil recruitment in this model. Notably, VCAM-1 expression, which we and others have detected on the portal track vessels of inflamed liver,16, 47 was significantly depressed in the Tnc−/− livers postreperfusion.
One of the most striking effects observed in the Tnc−/− livers was a marked decrease in MMP-9 expression/activation. Leukocyte transmigration, across endothelial and ECM barriers, results from a complex series of adhesive and focal matrix degradation events. Although adhesion molecules are essential to promote leukocyte attachment to the vascular endothelium, MMPs are important for facilitating leukocyte transmigration across vascular barriers. We previously demonstrated that MMP-9 is predominantly expressed by leukocytes in damaged livers16 and mediates leukocyte transmigration in liver IRI.16, 31 Our results showing that MMP-9 is up-regulated by Tnc in leukocytes are in agreement with other reports that demonstrated the induction of MMP-9 by Tnc in RAW264.7-macrophages,48 fibroblasts,49 and cancer cells.50 Furthermore, our data also suggest that TLR4 signaling mediates Tnc-induced up-regulation of MMP-9 activity in neutrophils. In this regard, studies using mice that lack TLR4 have shown that TLR4 mediates inflammation in hepatic IRI32 and that MMP-9 expression is reduced in TLR4-deficient mice after experimental stroke.51 However, Tnc mediated MMP-9 up-regulation in liver IRI is possibly due to a combination of factors, which may also include the expression of proinflammatory cytokines. In our settings, IL-6, which regulates MMP-9 activity in neutrophils,18 was significantly depressed in Tnc−/−-deficient livers postreperfusion.
In summary, our studies demonstrate an active role for Tnc in liver IRI. Tnc deficiency interfered with VCAM-1 vascular deposition and down-regulation of PECAM-1, disrupted MMP-9-positive leukocyte infiltration, hampered apoptosis and necrosis, and favored liver repair/regeneration after IRI. Thus, this work supports the view that further understanding of how newly synthesized ECM molecules, such as Tnc, participate in inflammatory responses may lead to potential valuable therapies in liver IRI.