This study aims to investigate the immunological status of small-for-size liver allografts and possible mechanism that contributes to the accelerated immune response in these allografts. Eight experimental groups were: whole isografts; 40% isografts; whole allografts, no treatment; 40% allografts, no treatment; whole allografts with sodium salicylate intraperitoneal injection, D0-3; 40% allografts with sodium salicylate, D0-3; whole allografts with FK506 intramuscular injection D0-3, and 40% allografts with FK506, D0-3. The 40% allografts survived significantly shorter than whole allografts (p = 0.02). At 72 h after reperfusion, a higher number of macrophages infiltrated into the periportal area of small-for-size allografts than whole allografts. Remarkable up-regulation of interleukin-1β (IL-1β), interleukin-2 (IL-2), interleukin-10 (IL-10) and interferon-γ (IFN-γ) messenger RNA (mRNA) levels were detected in small-for-size allografts within 24 h after reperfusion. Sodium salicylate administration reduced IL-1β and IFN-γ mRNA in both small-for-size and whole allografts, but it could decrease IL-2 and IL-10 mRNA levels only in small-for-size allografts. In vitro study revealed that CD80, CD86 and CD11b expression on macrophages was augmented after IL-1β stimulation, whereas the up-regulation could be blocked by sodium salicylate. In conclusion, early activation of macrophages as a result of graft injury might play an important role in the accelerated acute rejection process in small-for-size allografts.
Partial liver transplantation is a valuable alternative in solving the problem of organ shortage. However, with reduction of graft size, chances of primary graft nonfunction and complications also increase (1,2). Our previous study revealed that complications might be related to microcirculatory injuries that arose from transient portal hypertension (3). However, a few studies suggested that the immune response of small-for-size allografts might be accelerated and account for suboptimal liver function and poor result (4,5).
Alloantigen presentation and subsequent T-cell activation are two crucial steps that mediate acute rejection. In small-for-size allografts, the liver regeneration process may enhance expression of alloantigens, resulting in accelerated rejection (4,5). However, regeneration in small-for-size grafts may be delayed while the immediate post-reperfusion microcirculatory injury may augment the expression of adhesion molecules, inflammatory parameters and alloantigens (6,7). We hypothesize that the accelerated immune process in small-for-size allografts may be related to the consequence of post-reperfusion injury and inflammatory reaction. To test this hypothesis, we investigated the type of cells presenting in allografts and the cytokines that were related to inflammatory and cellular responses during the early phase after reperfusion.
Sodium salicylate is a biotransformation product of aspirin that possesses similar anti-inflammatory potency as aspirin, through its inhibition of prostaglandin synthesis on inflammatory cells (8–10). In addition, a recent study demonstrated that salicylate could suppress integrin-dependent neutrophil aggregation (11). To examine the possible relationship between inflammatory cell activation and acute rejection process in small-for-size allografts, sodium salicylate was introduced into the present study to inhibit the process of inflammation and to investigate whether this approach could influence the acute rejection process.
Materials and Methods
Animals and experimental groups
Adult male Dark Agouti (DA, RT1a) and Lewis (LEW, RT1l) rats, weighing 200–300 g, purchased from the Animal Resources Centre (Murdoch), were maintained under standard conditions and cared for according to the international guidelines for animal cares. They were used as donors and recipients, respectively. The grafts were either whole size or reduced to a certain percentage of recipients' liver weight according to experimental protocols. A total of eight experimental groups were included: group A, whole isograft (LEW to LEW), n = 5; group B, 40% isograft, n = 7; group C, whole allograft (DA to LEW), no treatment, n = 7; group D, 40% allograft, no treatment, n = 7; group E, whole allograft, sodium salicylate (Santa Cruz Biotechnology, Inc.) 40 mg/kg intraperitoneal injection (i.p.), days 0–3, n = 6; group F, 40% allograft, sodium salicylate 40 mg/kg i.p., days 0–3, n = 6; group G, whole allograft, FK506 1 mg/kg intramuscular injection (i.m.), days 0–3, n = 6; and group H, 40% allograft with FK506 i.m., days 0–3, n = 6. In groups A, B, C, D, E and F, at 1, 6, 24, 48 and 72 h after reperfusion, another three animals were sacrificed, respectively, at each time point for tissue collection. Graft samples were also collected at the time when the animals died. Death of recipients was defined as complete rejection and was confirmed by histopathology.
An orthotopic liver transplantation model was performed according to Kamada's method (12). Graft size was reduced according to the method described previously (3). In brief, the median lobe was kept as the residual graft, whereas other lobes were ligated and removed. The median ratio between the graft weight and recipient liver weight in the 40% graft group was 43% (ranging from 41 to 49%).
Liver function tests
Plasma samples from groups C and D were collected at 1, 6, 24, 48 and 72 h after reperfusion, respectively (three samples for each time point in each group). Liver biochemistry, including albumin, total bilirubin and alanine aminotransferase, was tested in the clinical biochemistry laboratory (Queen Mary Hospital, the University of Hong Kong).
When the animals were sacrificed, half of the liver grafts were fixed in 10% buffered formalin and embedded in paraffin, while the other half of the tissue was snap-frozen in liquid nitrogen. The paraffin-embedded tissue was then cut into 5-µm-thick sections for histological studies by hematoxylin-eosin staining and immunohistochemical staining of ED1 (monoclonal mouse anti-rat ED1, Serotec) and inducible nitric oxide synthase (iNOS) (polyclonal rabbit anti-rat iNOS, Stressgen Biotechnologies). The sections were incubated with 1° antibody for 1 h at room temperature, followed by a 1-h incubation of 2° antibodies (horseradish peroxidase-conjugated goat anti-mouse antibody and horseradish peroxidase-conjugated goat anti-rabbit antibody, respectively, Dako Corporation). Color development was performed in substrate solution with diaminobenzidine (Dako Corporation) for 5 min. The slides were then counter-stained with hematoxylin.
Frozen tissues from each group were homogenized in RIPA buffer [1× PBS, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), adding a cocktail tablet of proteinase inhibitor at the time of use. All reagents were purchased from Amersham Biosciences]. Protein concentration was determined by Coomassie Blue dye-binding assay (Bio-Rad Laboratories). One-sixth volume of 6× SDS/sample buffer (375 mM Tris-HCl, ph 6.8, 60% glycerol, 12% SDS, 30% mercaptoethanol, 0.06 mg/mL bromophenol blue. All reagents were purchased from Amersham Biosciences) was added to each sample and boiled for 10 min. Samples (50 µg) were subject to 12% SDS–polyacrylamide gel electrophoresis (PAGE). The gel was then transferred electrophoretically onto a polyvinylidene difluoride (PVDF) membrane (Bio-Rad Laboratories). The membrane was blocked for 1 h in 10% low-fat milk Tris-buffered saline with 0.05% Tween-20 (Amersham Biosciences) (TBST), followed by an overnight incubation at 4 °C in primary antibody (rabbit anti-rat iNOS polyclonal antibody with 1:1000 dilution, Stressgen Biotechnologies); (mouse anti-rat CD11b monoclonal antibody with 1:1000 dilution, Serotec). The membrane was then incubated in goat anti-rabbit IgG antibody for 1 h (Dako Corporation). Signals were developed using an ECL Kit (Amersham Pharmacia Biotech Incorporation). The level of protein expression was expressed by a density ratio between samples and β-actin measured by Gelworks 1D software (UVP, Inc.), and described as mean ± SD.
Total RNA was extracted from the snap-frozen tissue using an RNeasy Mini Kit (Qiagen Inc.). A total of 1 µg RNA was used as template for the first-strand DNA synthesis. Primers specific for rat interleukin-2 (IL-2), interleukin-10 (IL-10) and interferon-γ (IFN-γ) were designed according to Kita et al. (13). Primer sequences for IL-1β were: sense, 5′-GGACCTTCCAGGATGAGGACCTGAG-3′ and antisense, 5′-GGAAGGCATTAGAACCGCTCCAGCC-3′. PCR reactions were carried out through reverse transcription incubation at 94 °C for 5 min, 35 cycles of 94 °C for 1 min, 58 °C for 1 min, 72 °C for 1 min, and a single cycle at 72 °C for 7 min. PCR products were analyzed by electrophoresis in 2% agarose gel stained with ethidium bromide. Relative cytokine messenger RNA (mRNA) expression was evaluated by the OD ratio between samples and β-actin (Gelworks 1D 3.0, UVP, Inc.), and described as mean ± SD.
Macrophage cell line CRL-2192 was purchased from the American Type Culture Collection (ATCC). Cells were maintained as half-adhering and half-suspending culture in F-12K medium with 15% fetal bovine serum (FBS) and 1% penicillin/streptomycin (Life Technologies) at 37 °C in a humidified atmosphere of 5% CO2 in air. Cells were cultured in serum-free medium for 12 h prior to rat IL-1β (BioVision, Inc.) 10 ng/mL or IL-1β 10 ng/mL plus sodium salicylate 1 mM administration and maintenance for 6, 12, and 24 h, respectively. The cells were then labeled with mouse anti-rat CD11b, CD80 and CD86 monoclonal antibodies (Serotec) and detected in a FACS Calibur (Becton Dickinson Immunocytometry Systems). Appropriate isotypes of irrelevant mAbs were used as controls.
Graft survival was analyzed by Log rank testing using GraphPad Prism software (GraphPad Software Inc.). Comparisons of the liver function parameters, cytokine mRNA levels, iNOS and CD11b protein levels between different treatment groups were performed using the Student's t-test (SPSS 10.0 for Windows, SPSS Inc.). p-values less than 0.05 were considered as statistically significant.
Small-for-size allografts survived shorter than whole allografts
When no treatment was given, 70% of 40% isografts survived indefinitely, whereas all animals in the 40% allograft group died within 8 days (median 7.25 days, p = 0.03, 40% allografts compared with 40% isografts), which was significantly shorter than that of the whole allografts (median 8.75 days, p = 0.02, 40% allografts compared with whole allografts). Sodium salicylate significantly prolonged 40% allograft survival to a median of 10.5 days (p = 0.01, compared with 40% allograft without sodium salicylate), but it did not affect the survival of whole allografts. FK506 remarkably prolonged both whole and small-for-size allograft survivals (median 75 days, p < 0.01 and 40 days, p < 0.01, compared with whole and 40% allografts without treatment), but there was no significant difference between whole and 40% allografts with FK506 (p = 0.1467) (Table 1). When the animals died, the liver grafts demonstrated massive lymphocyte infiltration, obliteration of bile ducts and destruction of hepatic architecture (Figure 1).
Table 1. Survival of animals
NT: no treatment; SS: sodium salicylate 40 mg/kg intraperitoneal injection; FK: FK506 1 mg/kg intramuscular injection; acompared with group C; bcompared with group D.
>100, >100, >100, >100,
4, 5, >100, >100, >100,
7, 7, 8, 8, 9, 9, 10
4, 5, 5, 7, 7, 8, 8
8, 8, 9, 9, 10, 11
7, 7, 10, 10, 11, 11
30, 60, 60, 90, >100, >100
5, 13, 30, 50, 50, >100
Small-for-size allografts demonstrated impaired graft function
The plasma albumin level decreased while the total plasma bilirubin and alanine amino transferase levels increased in the 40% allograft group during the early period after reperfusion, consistently from 6 to 72 h (Figure 2).
Small-for-size allografts demonstrated more severe cell infiltration during the early phase after reperfusion
At 1, 6 and 24 h after reperfusion, histology of grafts in all groups demonstrated normal liver architecture, with exception in 40% isografts and 40% allografts (groups B and D), where some grafts presented with scattered areas of necrosis at 6 and 24 h. At 48 h after reperfusion, there was a significantly increased cellular infiltrate in the periportal area in the 40% allografts compared with whole allografts, and the cell number was further augmented at 72 h (Figure 3D). Sodium salicylate decreased cell influx at both 48 and 72 h in the 40% allograft group with a predominant effect at 72 h (Figure 3F). However, there was no obvious difference in cell infiltration between the whole allografts with and without sodium salicylate (Figure 3C, E). Macrophage (labeled by ED1) was the predominant type of cells that migrated to the periportal area of grafts at 48 and 72 h after reperfusion, with an increased number in 40% isografts. Furthermore, this type of cells greatly expanded in small-for-size allografts at 72 h (Figure 4D), but the expansion was inhibited by sodium salicylate (Figure 4F). There was no difference in the number of macrophages between whole allografts with and without sodium salicylate administration (Figure 4C, E). Corresponding to the number of macrophages in allografts, CD11b (a marker of inflammatory cell activation) expression was higher in the small-for-size graft group than the whole graft group, and it could be inhibited by sodium salicylate (Figure 5).
The activity of macrophages was enhanced in small-for-size allografts
In addition to the change in the number of macrophages, the activity of these cells was also enhanced in small-for-size grafts, as represented by the exaggerated iNOS expression. In isografts, a higher iNOS protein level was detected in the 40% graft group compared with that of the whole graft group at 48 and 72 h after reperfusion. However, a more predominant up-regulation of iNOS was detected in the 40% allografts at 72 h after transplantation (Figure 6), and the majority of iNOS positive cells were macrophages (as the majority of iNOS positive cells could also be labeled by ED1, data not shown) (Figure 7D). Sodium salicylate greatly down-regulated the iNOS level in the small-for-size allografts from 24 to 72 h after reperfusion, whereas it reduced iNOS expression in the whole allografts only at 48 h (Figure 6).
Small-for-size allografts produced a higher level of cytokines
Higher levels of IL-1β (Figure 8) and IL-2 mRNA (Figure 9) were detected in the small-for-size allograft group starting from 6 h after reperfusion compared with the whole allograft group (p< 0.05), whereas IL-10 (Figure 10) and IFN-γ mRNA (Figure 11) levels were up-regulated from 24 h after reperfusion. Sodium salicylate remarkably down-regulated IL-1β and IFN-γ mRNA levels in both the 40% and whole allograft groups, whereas it remarkably decreased IL-2 and IL-10 mRNA levels in the 40% allograft group.
IL-1β stimulated CD80 and CD86 expression in macrophage cell lines
CD80 and CD86 expression on cell surfaces characterizes the activity of antigen-presenting cells. To determine whether some proportion of macrophages functioned as antigen-presenting cells under inflammatory situation, IL-1β was added to the culture medium of macrophage cell lines. The expression of CD80 on macrophages increased after 6-, 12- and 24-h culture (Figure 12A), whereas up-regulation of CD86 was detected after 12- and 24-h culture (Figure 12B). In addition, CD11b up-regulation was more predominant in the presence of IL-1β for 12 and 24 h (Figure 12C). When IL-1β was administered with sodium salicylate, the up-regulations of CD80, CD86 and CD11b were inhibited (Figure 12A–C).
The present study revealed the different immunological status in small-for-size and whole grafts during the early phase after reperfusion. The lost small-for-size allograft on post-operative day 7 displayed a similar cellular rejection picture as the whole allograft on post-operative day 10, suggesting that the immune responses were accelerated in small-for-size allografts, which might be related to adhesion molecule up-regulation (14) and inflammatory cell activation (15) subsequent to transient hemodynamic changes. A small proportion of 40% isografts were lost within 1 week after transplantation, indicating that microcirculatory injury could also account for the shorter survival time of small-for-size allografts. However, although 60% of small-for-size allografts could survive more than 1 week, their survival time was significantly shorter than that of whole allografts, suggesting that the accelerated immune process could also play an important role in determining the small-for-size allograft survival.
The transient elevation of IL-1β and IFN-γ mRNA in small-for-size isografts suggested that these two cytokines might be closely related to the injury process, whereas consistent up-regulation of pro-inflammatory, Th1 and Th2 cytokines in the small-for-size allografts during the early phase after reperfusion indicated the presence of both enhanced inflammatory and cellular rejection responses. Anti-inflammatory treatment decreased the IL-1β and IFN-γ mRNA levels in both small-for-size and whole allograft groups, while it significantly attenuated, but not diminished, the IL-2 and IL-10 mRNA expression only in small-for-size allografts, indicating that the inhibitory effects of sodium salicylate to cellular rejection might be an indirect effect of anti-inflammation.
In our study, we detected a remarkably increased cell infiltration located in the periportal area of small-for-size allografts, within which the majority were macrophages, indicating the potential role of macrophages in mediating an early (within 72 h after reperfusion) immune process in these grafts. In addition, early up-regulation of CD11b expression provided additional evidence of the enhanced inflammatory status in the small-for-size allografts. Moreover, the higher level of iNOS, an important marker that characterized macrophage activity, in small-for-size isografts than that of whole isografts indicated its close relationship with injury, whereas an even higher level of iNOS expression in small-for-size allografts suggested that this enzyme might also be related to cellular rejection responses. Furthermore, one study demonstrated that inflammatory mediators could stimulate CD11b expression on neutrophils (16). Our in vitro study further revealed that IL-1β induced CD80 and CD86 expression in macrophages, implying that some of the macrophages might activate as antigen-presenting cells under inflammatory circumstances, and the activation could be blocked by sodium salicylate. However, the process of early macrophage activation in small-for-size allografts remains to be determined. We postulated that up-regulation of adhesion molecules in endothelial cells due to microcirculatory injury might trap macrophage adhesion and migration. In addition, necrosis induced by injury might also stimulate macrophage activation, as necrotic cells were reported to be more immunogenic (17). Moreover, the liver regeneration processes might present more alloantigens to antigen-presenting cells (4). Some mediators released during the regeneration process, such as vascular endothelial growth factor, might also attract macrophage migration and proliferation because monocytes were found to express vascular endothelial growth factor receptor (18).
The non-steroid anti-inflammatory drug sodium salicylate achieved limited prolongation of allograft survival only in the small-for-size graft group. Although some in vitro studies demonstrated the possible immunosuppressive effects of sodium salicylate to T-cell activation (19,20), our in vivo data did not reveal any convincing evidence of its inhibition of the cellular rejection process because of its failure to induce prolongation of whole allograft survival. One possible explanation is that in vivo T-cell activation could be induced through various pathways other than TCR-dependent p38 activation. Another reason might be that not all alloantigen-presenting cells were activated through the inflammatory process in the small-for-size allograft. Therefore, sodium salicylate could not achieve indefinite allograft survival. Interestingly, a short-term therapeutic dose of FK506 induced similar prolongation of both small-for-size and whole allograft survivals. Concerning a possible higher level of FK506 in the plasma due to graft size-related deceleration of drug metabolism, the immune status might be more severe in the small-for-size allografts.
Another interesting finding in the present study was that all the small-for-size allografts survived more than 1 week when receiving sodium salicylate, indicating that the protective effects of sodium salicylate against reoxygenation injury (21,22) might also contribute to the improved allograft survival. However, the mechanism remains to be determined.
Our study suggested the importance of anti-inflammatory treatment to small-for-size allografts during the early phase after reperfusion, based not only on the possible linkage between inflammatory and acute rejection process, and the remarkably reduced dose of immunosuppressive drug when anti-inflammation combined with immunosuppression (data not shown), but also the potential role of macrophage activation in mediating the chronic rejection process. With the increased number of macrophages and elevation of the iNOS level in small-for-size allografts, up-regulation of transforming growth factor-β1 was also detected in these allografts (data not shown). However, because the model used in this study was a nonarteralized liver transplantation model, and all animals died within a short period when no immunosuppressant was given, it was difficult to identify the long-term morphology of the small-for-size allografts.
In conclusion, early activation of macrophages in small-for-size allografts might link the inflammatory process to acute rejection through their roles as antigen-presenting cells, resulting in enhancement of alloantigen recognition and subsequent accelerated immune responses in the allografts.
The study was supported by the research grant from the Distinguished Research Achievement Award and Sun Chieh Yeh Research Foundation for Hepatobiliary and Pancreatic Surgery of the University of Hong Kong.