 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
Accelerates Chronic Graft Injury in a Rat Renal Allograft Model
*Institute of Medical Immunology, Charité-Campus Mitte, Humboldt University, Berlin, Germany; Department of Surgery, Charité-Campus Virchow, Humboldt University, Berlin, Germany; Section of Pulmonary and Critical Care, School of Medicine, Louisiana State University Health Sciences Center, New Orleans, Louisiana, USA.
Correspondence to Dr. Thomas Ritter, Institute of Medical Immunology, Charité Campus Mitte, Humboldt University Berlin, Monbijoustrasse 2a, 10117 Berlin, Germany. Phone: 0049-30-450-524197; Fax: 0049-30-450-524907;
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Lewis rat) were ex vivo transduced in group I with control Ad-construct, group II with three different therapeutic Ad-constructs expressing the immunoregulatory molecules vIL-10, TNFRp55-Ig, and IL-12p40, and group III with AdIFN-
. Group IV served as untreated controls. Control grafts (IV) showed increasing proteinuria during the 24-wk follow-up. Chronic graft injury was accelerated by Ad-control (I) and even more by AdIFN- (III). All rats carrying the AdIFN-–transduced grafts died within 12 to 13 wk by advanced chronic renal failure associated with strong immune cell infiltration and immune gene expression. By contrast, the Ad-therapy group II showed less inflammation and improved graft histology and function if compared with the groups I and III. Moreover, significantly less infiltrating ED-1+ macrophages and an improved histologic score even if compared with untreated controls (IV) was observed. However, after disappearance of therapeutic gene expression, group II showed increasing proteinuria probably as result of late T cell activation to the Ad-encoded proteins. Ex vivo transduction of allografts with Ad-control or even more AdIFN- expression promotes intragraft inflammation and chronic graft injury. Targeting macrophage activation by a cocktail of therapeutic genes improved the results. These data support the pathogenetic role of cytokines in chronic graft injury; however, they also show the limitations of the Ad-mediated gene transfer. E-mail: thomas.ritter@charite.de |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Macrophages seem to play an important role in the pathogenesis of chronic allograft rejection (2–4). Massive macrophage infiltration is associated with a poor long-term outcome of allografts (5). Several perioperative (e.g., ischemia/reperfusion injury, donor brain death) and postoperative (e.g., T cell–mediated inflammation) events activate intragraft macrophages and induce the recruitment of further monocytic precursor cells into the graft. Pro-inflammatory cytokines are the central mediators of macrophage effector function. Tumor necrosis factor-
(TNF-) is known to be an important mediator of tissue damage during allograft rejection (6). TNF- can bind to two receptors, p55 and p75, and can induce several biologic effects, including cytotoxicity, which is predominantly linked to the p55 TNF- receptor (7). It has been recently shown that blockade of TNF- by a soluble TNF receptor reduced inflammatory responses (8). We have shown that the application of a soluble TNF receptor-Ig chimeric molecule (TNFRp55-Ig) is able to completely block TNF- and TNF-
responses and to reduce the early intragraft cellular infiltration and cytokine expression in a rat syngeneic heart transplant model (9).
In addition to the well-described graft-destructive effects of TNF-, other soluble mediators from macrophages are also thought to be involved in the pathogenesis of chronic graft rejection. Interleukin-12 (IL-12) is characterized as an inductive cytokine for T helper cell 1 (TH1) cell differentiation, production of TH1 cytokines (e.g., interferon- [IFN-]) and cell-mediated immunity (10). The biologically functional form of IL-12 is a 70-kD heterodimer (IL-12p70) that consists of disulfide-bonded 40-kD (p40) and 35-kD (p35) subunits (11). It plays a key role in the induction of cellular immunity by promoting the proliferation of natural killer cell (NK) and T cells and the differentiation to TH1 cells and by cytotoxic T lymphocyte (CTL)-mediated and NK-mediated cytolytic activity (12). It has been demonstrated that overexpression of IL-12p40 blocks the activities of IL-12p70 by competitively binding to the IL-12 receptor (13), thereby reducing IL-12p70–mediated TH1 responses (14). Moreover, it was shown that overexpression of IL-12p40 could prevent the rejection of allogeneic myoblasts (14).
Recent studies suggest that IL-10 inhibits the production of IL-12 and TNF- by monocytes. TH2 cells and particularly macrophages are the major cellular source of IL-10, which inhibited the cytokine synthesis of TH1 cells, particularly IFN- (15), and the production of monokines such as IL-1, IL-6, IL-8, TNF-, and GM-CSF (16). In addition, IL-10 downregulates the expression of MHC class II and co-stimulatory molecules, such as CD40, CD80, and CD86 (16). IL-10 also impairs antigen presentation by antigen-presenting cells (APC). Interestingly, an Epstein-Barr virus homologue of cellular IL-10 (vIL-10) has been shown to be even more effective in the downregulation of immune responses than cellular IL-10, at least in rodent models (17).
For many years, it was thought that TH1 cells and their key cytokine IFN- are the key players in acute rejection, whereas chronic rejection was thought to be less T cell–dependent. Very recently it was shown, however, that acute rejection was quite normal in IFN- gene–deficient mice, whereas chronic rejection was almost absent, identifying IFN- as a key cytokine in the pathogenesis of chronic rejection (18). On the other hand, it has been shown that IFN- may also exert protective effects. It is not possible to induce tolerance by blocking co-stimulation using CTLA-Ig and/or anti-CD40L monoclonal antibody (mab) in IFN-–deficient mice (19), because IFN- seems to be important for mediating activation-induced death of T cells. In addition, IFN- seems to directly mediate graft-protective effects (20). Graft protective effects were also described for early perioperative TNF-, whereas late TNF- expression in the graft seems to mediate graft injury (Reinke P, et al., personal communication). Thus, the role of cytokines in the pathogenesis of chronic graft injury is not completely understood.
Gene therapy is an approach to introduce therapeutic genes into cells or tissues. In transplantation, gene transfer has the advantage of delivering immunoregulatory molecules ex vivo into the graft, which may result in local modulation of immunity after grafting without systemic side effects (21,22). Our laboratory (23) and others (24–26) have successfully demonstrated gene transfer with recombinant adenovirus (Ad) expressing immunomodulatory molecules into allografts and subsequent prolongation of graft survival in models of acute rejection. To our knowledge, however, it has not been investigated so far if early expression of immunomodulatory molecules may influence renal chronic allograft injury processes.
On the basis of our knowledge of the pathogenesis, macrophage-activating cytokines and cytokines released by activated macrophages seem to be an ideal target for preventing chronic allograft injury. IL-12p40, vIL-10, and TNFRp55-Ig interact with the production and/or action of the key cytokines IL-12, TNF-, and IFN-. Previous studies (21,27) and preliminary data from our group (Ritter T, et al., unpublished data) have suggested that a combination of two or more therapeutic molecules but not the sole application showed protective effects in a model of acute allograft rejection. Therefore, in the present study, we evaluated the ex vivo gene therapeutic approach using a combination of Ad-vectors encoding vIL-10, IL-12p40, and TNFRp55-Ig to prevent chronic allograft injury in a well-defined rat kidney transplant model. Moreover, we wanted to address the question of whether IFN- overexpression leads to an advanced chronic graft injury.
We could show that Ad-mediated gene transfer by itself induces an earlier onset of chronic graft injury that was dramatically accelerated by IFN- intragraft overexpression, whereas gene transfer of the therapeutic cocktail (vIL-10, IL-12p40, TNFRp55-Ig) improved the results.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Kidney Transplantation
The orthotopic kidney transplantation used in this study is a modification of the technique described by Lee (28), in which the arterial and ureteric end-to-end anastomoses are completed with an interrupted suture technique, whereas the venous anastomosis is completed with a continuous suture technique. All surgical procedures were performed under ether anesthesia. The donor kidney was perfused with cold University of Wisconsin solution (UW solution) immediately after procurement and remained preserved in cold UW solution for 45 min before transplanted into the recipient rat. One native kidney was removed during transplantation and the remaining kidney 10 d later.
Generation of Recombinant Adenovirus (Ad)
The generation of the recombinant Ad constructs has been performed as described earlier (29,30). The following Ad constructs have been used in this study:
: Ad encoding for mouse IFN- (functional both in mice and in rats)
Experimental Protocols
Four groups of animals were studied: group I (Ad-control group, Ad expressing hAAT, human alpha-1 antitrypsin), renal allografts transduced with Ad-control; group II (Ad-therapy group containing three different Ad constructs expressing the immunoregulatory molecules vIL-10, TNFRp55-Ig, and IL-12p40), renal allografts transduced with Ad-therapy; group III (AdIFN- group), renal allografts transduced with AdIFN-; group IV, untreated control group. All animals were treated with low-dose Cyclosporine A (1.5 mg/kg per d intramuscularly; Sandimmun, Novartis, Nürnberg, Germany) for 10 d after transplantation to prevent acute rejection episodes. Urinary protein excretion and creatinine clearance were analyzed 2, 4, 8, 12, 16, and 24 wk after transplantation (post-Tx). The animals of group I, II, and IV (n = 5 to 6/group each time point) were killed at three different time points (5 d, 12 wk, and 24 wk post-Tx) to analyze intragraft histology and gene expression pattern. All six animals of group III were killed approximately 12 wk post-Tx just before dying from chronic renal failure.
Ex Vivo Gene Transfer
The donor renal artery was cannulated ex vivo immediately after procurement of the graft. All grafts were perfused with 2 ml of UW solution. Then AdhAAT (5 x 109 plaque forming units [pfu]), AdvIL-10, AdTNFRp55-Ig and AdIL-12p40 (2 x 109 pfu each) and AdIFN- (5 x 109 pfu) were administered by slow infusion via cannula in groups I, II, and III, respectively. The renal grafts remained preserved in cold UW solution for 45 min before transplantation.
TNFRp55-Ig ELISA
Serum samples were collected from the recipients’ peripheral blood. The TNFRp55-Ig concentration was quantified by using ELISA specific for human TNFRp55 (R&D Systems GmbH, Wiesbaden, Germany). Plates were read at 405 nm in an ELISA reader. The linear region of TNFRp55-Ig standard curves were obtained in a series of eight twofold dilutions of TNFRp55-Ig standard, from 5000 pg/ml to 78 pg/ml.
Functional Studies
Twenty-four–hour urine samples were taken at weeks 2, 4, 8, 12, 16, and 24 after transplantation from the rats that were housed for 24 h in individual metabolic cages. Animals were weighed, and urine volume was determined. Blood samples were taken at the end of each 24-h period. Protein excretion (mg/24 h) was measured by precipitation with 20% CCI3COOH. Turbidity was assessed at a wavelength of 415 nm using a Hitachi 911 analyzer. Creatinine concentration was measured by a standardized colorimetric method. Endogenous creatinine clearance (ml/min) was calculated using the standard formula (Urinevolume · Creaurine/Creaserum · 1440) and expressed as related to body surface area.
Histology
Renal grafts were removed at 5 d, 12 wk, or 24 wk after transplantation and divided into three parts. Two parts of them were snap-frozen in liquid nitrogen for performance of immunohistochemistry and RT-PCR. The remainder of the graft was immersion-fixed in 5% neutrally buffered formalin, which were then paraffin embedded, cut, and stained with hematoxylin/eosin and periodic acid-Schiff. Histologic evaluation and grading included glomerulosclerosis, fibrosis, tubular atrophy, and mononuclear cellular infiltration. The sclerosed glomeruli and total glomeruli were counted and expressed as a percentage to determine the extent of glomerulosclerosis. The histologic grading scale of fibrosis, tubular atrophy, and mononuclear cellular infiltration was from 0 to 4+ (0 = not present, 4+ = strongest structural deterioration). The analyses were performed in a blinded form.
Immunohistology
The frozen sections of grafts collected at day 5, week 12, or 24, were cut (5 µm), fixed in acetone for 10 min, air-dried, and incubated with mouse anti-rat antibodies. The sections were then interacted with rabbit anti-mouse Ig by the alkaline phosphatase, antialkaline phosphatase, or peroxidase antiperoxidase (APAAP) methods and counterstained with hematoxylin. Cell populations and cell surface markers were assessed using mab to CD4+ cells (W3/25), CD8+ cells (OX-8), and monocyte/macrophage (ED-1). Negative controls included omission of the first antibody or murine control Ig (all antibodies were purchased from DAKO A/S, Denmark). Marker-positive cells were expressed as mean ± SEM of cells/field of view. More than 20 field-of-view/section were evaluated at x400. The analyses were performed in a blinded form.
Gene Expression Analyses
Intragraft gene expression was analyzed in tissues from grafts collected at day 5, week 12, and 24 by using quantitative real-time RT-PCR (ABI Prism 7700 Sequence detector "Taqman;" PE Applied Biosystems, Weiterstadt, Germany).
Expression of the following rat genes were analyzed: CD3, CD25, TNF-, IFN-, IL-12p40, IL-4, IL-10, IL-2, bcl-2, bag-1, hemoxygenase-1 (HO-1). The analyses were performed in a blinded form.
Statistical Analyses
Statistical comparisons between groups were analyzed by a parameter-free assay (Mann-Whitney Wilcoxon). Data are shown as mean ± SEM.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Survival
Like the untreated control group IV, all animals receiving grafts transduced with either the AdhAAT control vector (group I) or the therapeutic mix of IL-12p40, vIL-10, and TNFRp55-Ig (group II) survived until the end of the observation period of 24 wk, except those killed at day 5 and week 12 for graft analysis. During follow-up, all animals gained weight without differences between the groups. In contrast, all rats receiving grafts transduced with IFN- died from chronic renal failure (or were killed because of their bad condition) within 12 to 13 wk, suggesting accelerated chronic graft injury.
Functional Studies
Next we investigated if gene transfer of immunoregulatory molecules has any influence on graft function by analyzing urinary protein secretion and creatinine clearance during the observation period of 24 wk. At 2 wk post-Tx, urinary protein excretion was significantly reduced in the Ad-therapy group II compared with Ad-control group I (P < 0.05; Figure 1). Interestingly, the AdIFN- group III was also slightly better than the Ad-control group I, although this difference was NS. However, the urinary protein excretion in the Ad-therapeutic group II was comparable to the untreated control, suggesting that the Ad-control group I showed slightly higher urinary protein excretion early post-Tx, probably as result of the Ad-mediated inflammation. Obviously, this undesired side effect of Ad-mediated gene transfer was prevented by the therapeutic gene mix and in part also by IFN-.
|
, which reached peak levels of >200 mg/24 h protein at week 8, whereas the rats from the other groups excreted <30 mg/24 h until week 12 post-Tx (P < 0.01; Figure 1). The early functional deterioration of IFN-–expressing grafts was further confirmed by a strongly diminished creatinine clearance (P < 0.01 versus all other groups) during the whole observation period (Figure 2). These data demonstrate an accelerated chronic graft injury in the AdIFN- group III, resulting in death by chronic renal failure in all animals between week 12 to 13.
|
The protective effects of the immunoregulatory molecules on the development of chronic graft injury was less prominent and visible by differences of creatinine clearance, although the levels of the Ad-therapeutic group II were slightly higher at all time points after week 12 (week 12 P < 0.05) as compared with the Ad-control group I or even with the untreated control IV (Figure 2).
Histology
Chronic renal allograft injury is characterized by typical histologic changes such as tubular atrophy, interstitial fibrosis, glomerulosclerosis, and immune cell infiltrates. Tissue sections from kidney grafts of the different treatment groups were analyzed for signs of chronic rejection by histology and immunohistology.
IFN-–transduced grafts were analyzed around 12 wk (grafts were collected <12 h postmortem). They showed strong evidence for glomerulosclerosis (50.6 ± 10.2%), fibrosis, tubular atrophy, and cellular infiltration (Figure 3) already at this early time point (the median score for all three histologic parameters was 4.0, ranging from 3.5. to 4.0).
|
The Ad-therapy group II also had less fibrosis, tubular atrophy, and infiltration by inflammatory cells than Ad-control group I (mean scores, 2.25 and 3.75 to 4.0, respectively; P < 0.05; Figure 3). Signs of chronic rejection were also significantly improved in the Ad-therapy group II compared with animals receiving only UW-perfused allografts (group IV) without any gene therapeutic treatment (mean scores, 3.0; P < 0.05; Figure 3). Figures 4A through 4D show typical examples of graft morphology.
|
–transduced grafts was associated with strong infiltration of CD4+ (mostly T cells), CD8+ (mostly T cells), and ED-1+ (mostly macrophages) cells by week 12. By 24 wk, the intragraft infiltration by ED-1+ macrophages and CD4+/CD8+ cells were markedly increased in the Ad-control group I as compared with the untreated control, suggesting the inflammatory potency of Ad-mediated gene transfer. By contrast, the expression of therapeutic genes prevented the negative effects of the Ad vector (Table 1). Moreover, the Ad-therapy group contained less ED-1+ macrophages (P < 0.05) and CD4+ cells (P = 0.07) than the untreated control. Figures 5A through 5D illustrate the ED-1+ cell infiltration of typical samples from the four groups.
|
|
and the protective effects of vIL-10/IL-12p40/TNFRp55-Ig, we performed quantitative real-time RT-PCR analyses of graft tissues collected at 5 d (group I, II, and IV), 12 wk (all groups), and 24 wk (group I, II, and IV) post-Tx using a panel of immunologically relevant markers (rat CD3, CD25, IL-2, IL-4, IL-10, IL-12p40, IFN-, TNF-, HO-1, bcl-2, and bag-1). The results are illustrated in Figure 6.
|
The levels of intragraft CD3 and CD25 mRNA expression were comparable between the three groups (Ad-control group I, Ad-therapy group II, and untreated control IV), although the Ad-control showed a slight increase in T cell counts, suggesting an unmodified infiltration pattern by activated T cells. In comparison with the untreated control group IV, the Ad-therapy group II showed in addition to IL-12p40 higher levels (>2-fold upregulation) of IL-2, IL-4, bcl-2, and bag-1. However, all four genes were also upregulated in the Ad-control group, suggesting the effect was due to rather the Ad-vector by itself than the therapeutic genes.
Day 84 [Week 12].
The advanced chronic injury of grafts transduced with AdIFN- (group III) was reflected by significantly higher mRNA levels of CD3 (2- to 3-fold versus Ad-therapy and untreated control), CD25 (2-fold), IFN- (2-fold), IL-2 (3-fold), IL-12p40 (3- to 6-fold), and IL-4 (5-fold). In addition, gene expression of bag-1, HO-1, TNF-, and IL-10 was also increased as compared with the untreated controls group, but it was not different from the Ad-control. By contrast, in the Ad-therapy group, the Ad-mediated, slightly elevated expression of IL-10, TNF-, IL-12p40, HO-1, and bag-1 was normalized to the level of untreated control; the expression of bcl-2 and CD3 was even lower than in the untreated control, whereas IL-4 was about two times more highly expressed. In summary, these data suggest that Ad-mediated gene transfer by itself induces some inflammation, particularly of macrophages (TNF-, IL-10, IL-12p40), even long after elimination of the gene vector. IFN- overexpression dramatically accelerates this process by triggering massive cell infiltration and immune activation at both TH1 and TH2 level. The expression of therapeutic macrophage targeting genes completely prevents the negative effects of the Ad-vector but also shifts the gene expression more to a TH2-like pattern in comparison with the untreated control (lower IFN-/IL-4 ratio).
Day 168 [Week 24].
The Ad-control group I showed higher levels of TNF- (4-fold), IFN- (2-fold), IL-2 (2-fold), IL-4 (4-fold), IL-10 (3-fold), IL-12p40 (2-fold), HO-1 (9-fold), bcl-2 (2.5-fold), and bag-1 (2-fold) as compared with the untreated controls, although only the differences for TNF-, IL-4, and HO-1 reached statistical significance. Whereas at the Ad-mediated effects week 12 were completely reversed by overexpression of the therapeutic genes, at this was only partly the case week 24, reaching statistical significance for TNF- and HO-1 only (2- and 2.5-fold reduction versus Ad-control). In contrast, CD25 mRNA was even significantly higher as compared with the two other groups.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Gene therapy has a powerful potential for delivery of therapeutic genes into organs and tissues before transplantation (32). Approaches have largely been directed toward the protection of grafted organs/tissues to ischemia/reperfusion injury or the induction of immune tolerance (33,34,22). Here we studied (to our knowledge, for the first time) the influence of Ad-mediated gene transfer of cytokines/cytokine antagonists on the onset of chronic allograft rejection in a well-characterized rat renal allograft model.
We observed that Ad-mediated gene transfer by itself triggers inflammatory processes early after transplantation that result in poorer graft histology and function at the end of the observation period of 24 wk. These detrimental effects were dramatically accelerated if IFN- was locally delivered by Ad-mediated gene transfer into the graft. The combination of vIL-10 (inhibiting the secretion of IL-12, TNF-, etc., downregulating antigen-presentation), IL-12p40 (antagonizing IL-12p70), and TNFRp55-Ig (antagonizing TNF) improved the functional (proteinuria, creatinine clearance) and morphologic (fibrosis, cellular graft infiltration [particularly of macrophages], tubular atrophy, and glomerulosclerosis) parameters of chronic graft injury, suggesting a complete reversal of Ad-mediated chronic graft injury. Moreover, it even improved the histologic score (less tubular atrophy, fibrosis, and cell infiltration) compared with the untreated control. On the basis of the preliminary data in acute rejection models (Kato H, et al., Ritter T, et al., unpublished data), we have selected this powerful combination because single molecules seem to have less potency. These data underline the important role of inflammatory processes in the pathogenesis of chronic graft injury.
What might be the mechanisms of the protective/detrimental effects observed? As suggested, the long-term outcome of renal allografts is influenced by numerous immunologic and non-immunologic factors underlining the complexity of the pathogenesis of chronic graft rejection. Most emphasis has been directed toward elimination of the risk factor "acute rejection." Several studies have demonstrated a strong relationship between acute rejection episodes and subsequent development of chronic allograft injury. The degree of reversal of an acute rejection has a strong influence on the late outcome. It has recently been demonstrated, that, if an acute rejection in a renal transplant is completely reversed and renal function recovered to the same level as before acute rejection, there is apparently no increased propensity for late transplant failure (35). This was also demonstrated in an experimental re-transplantation model of acute rejection (36).
The F344LEW kidney transplant model used in this study is characterized by a relative mild acute rejection within the first week, as there is only one MHC class I mismatch. Application of low-dose cyclosporine A for 10 d prevents clinically the acute rejection. The recipients have a quite normal transplant function for >12 wk before manifestation of chronic graft injury occurs resulting in irreversible graft loss after >6 mo. However, careful analysis of the early intragraft events reveals the manifestation of subclinical rejection within the first week post-Tx, which is characterized by T cell and macrophage infiltrates and expression of pro-inflammatory cytokines (e.g., TNF-, IFN-, IL-1, IL-12) and their "downstream" products (e.g., iNOS). It was further demonstrated that combating subclinical rejection by re-transplantation of the F344 graft back into the F344 donor background before 12 wk post-Tx completely reversed the ongoing chronic graft injury process (37). We wondered whether our gene therapeutic approach may accelerate (Ad-control, AdIFN-) or prevent/reverse (Ad-therapy) the subclinical acute rejection process. We did not find significant differences in serum creatinine levels early after Tx between the four groups. Even in the IFN- group that developed accelerated chronic graft injury the creatinine serum levels remained in the normal range within the first weeks. However, this group III showed a significant decrease of creatinine clearance as early as 2 wk post-Tx and urinary protein excretion dramatically increased by week 4, suggesting ongoing graft injury. Immunohistology and gene expression analysis by week 12 reveals a picture of massive cell infiltration and immune activation including upregulation of macrophage- (IL-12p40, TNF-, IL-10), TH1- (IFN-), and TH2 (IL-4)–derived gene products. These data suggest that IFN- did not boost the subclinical rejection to an acute clinical rejection (no creatinine rise within the first 10 wk) but accelerates the chronic subclinical immune process, resulting in early chronic graft failure.
Our data confirm the observations on the key role of IFN- in chronic rejection, but it is less significant in acute rejection generated in IFN- genetically deficient mice (18,38). In a rat heart transplant model (LEWF344), chronic rejection was strongly associated with upregulation of AIF-1, an inflammatory factor induced by IFN- (39). The mechanisms by which IFN- promotes chronic rejection are not clear so far. It promotes antigen presentation but also induces adhesion molecules and chemokines in the tissue that might be important for tracking immune cells to the graft. Very recently, Savinov et al. (40) observed that IFN- supports the homing of diabetogenic T cells into the pancreas. Our data (strong intragraft accumulation of T cells and macrophages along with overexpression of several inflammatory genes) support this view.
Although less severe, however, we found signs of mild acceleration of chronic rejection in the Ad-control group as well. In contrast to the IFN-–overexpressing group, graft injury (proteinuria) was not seen before week 16 in this group. Moreover, serum creatinine and creatinine clearance remained normal during the whole 24-wk observation period. However, histology showed more signs of chronic injury than the untreated controls. We observed an early intragraft upregulation of several genes, including IFN-, IL-2, IL-12, and CD3 as early as by day 5 post-Tx, suggesting an early acceleration of ischemia/reperfusion-mediated inflammatory processes. Although several protective genes such as IL-4, IL-10, bag-1, and bcl-2 are also upregulated in Ad-control grafts in comparison with untreated control grafts, the early inflammation might trigger persistent subclinical inflammation, resulting in long-term graft injury. In fact, the mRNA gene expression of TNF- and IL-12 persists at 2- to 5-fold higher levels for the whole 24-wk observation period in the Ad-control group versus untreated controls, although the graft infiltration by ED-1+ and CD4/8+ cells was upregulated less than twofold. By week 24, IFN- expression was also higher (twofold). The persistently active inflammation is associated with compensating upregulation of protective genes such as IL-10, bcl-2, bag-1, and HO-1 in those grafts.
In summary, mild intragraft stimulation of inflammation (by Ad-control) results in long-lasting overexpression of macrophage products (IL-12, TNF-) and promotes chronic graft injury (poorer histology, proteinuria). Overexpression of IFN- further accelerates the inflammatory processes (macrophages, TH1 and TH2 like cells) resulting in chronic graft failure and death within 12 to 13 wk.
Interestingly, intragraft gene transfer of therapeutic genes (vIL-10, TNFRp55-Ig, IL-12p40) directly or indirectly targeting macrophages improved the clinical outcome. Urinary protein excretion increased 8 wk later compared with the Ad-control, and the histology showed significantly less tubular atrophy, glomerulosclerosis, and graft infiltration. What are the mechanisms of protection by this therapeutic cocktail?
Although the expression of therapeutic genes is detectable within few days post-Tx, at day 5 the intragraft gene expression pattern was quite similar to the Ad-control. Like in Ad-controls, IL-2 and IFN- (and IL-12p40 as therapeutic gene), but also the protective genes bcl-2, bag-1, and IL-4, were up-regulated. Early post-Tx functional data were also comparable. However, the grafts from the Ad-therapy group expressed less CD3 and TNF- but more IL-4 mRNA transcripts at 12 wk. Whereas the Ad-control showed signs of persistent intragraft inflammation (including compensatory upregulation of protective genes such as IL-10, HO-1, bcl-2, and bag-1), the grafts from the Ad-therapy group looked similar to the untreated controls even with less CD3 and more IL-4 mRNA transcripts. The downregulation of inflammation was reflected by the highest creatinine clearance among the groups and low proteinuria by week 12. These data suggest that targeting macrophages by intragraft gene transfer of regulatory molecules (1) did not prevent normal ischemia/reperfusion injury–mediated inflammation (untreated control), (2) did not prevent early inflammatory adverse effects of adenoviral gene transfer, but (3) did reverse the early inflammatory process without graft injury.
A similar observation we recently made in a strong acute transplant rejection model (WFLEW) (Kato et al., in preparation), where immunomodulatory molecules did not prevent acute rejection (with creatinine rise) but induced spontaneous reversal of graft injury with long-term survival.
Compared with the Ad-control (I) and the AdIFN- (III) group, the protective effects of the therapeutic mix could be observed over the whole 24-wk observation period at both inflammatory and functional levels. Analyzing the data in comparison with the untreated control group (IV), however, offers a more complex picture. Although the Ad-therapeutic group (II) showed less tubular atrophy, fibrosis, and cell infiltration (particularly of ED-1+ macrophages), the glomerulosclerosis was similar to the untreated control (group IV) and the functional data were even less promising; the urinary protein excretion increased faster than in the untreated control during the last weeks. Although the T cell infiltration was comparable between the two groups (II and IV), the intragraft gene expression of TNF-, IFN-, and IL-2, as well as of HO-1, bcl-2, and bag-1, increased in the Ad-therapy group (II) between weeks 12 and 24 and reached higher levels than in the untreated controls (IV), although still lower than in the Ad-control group I. This inflammatory boost might explain the increasing proteinuria despite fewer signs of chronic rejection compared with the untreated controls. This view is further supported by the significant lower numbers of graft-infiltrating ED-1+ macrophages in group II versus IV; accumulation of ED1+ macrophages is typical for chronic graft injury.
It is difficult to predict which processes trigger this late "subacute" intragraft inflammation in the the Ad-therapy group. We speculate that the anti-Ad immune response plays an important role, which is then downregulated as long as the therapeutic genes are expressed (up to 10 to 12 wk). Whereas the intragraft CD3 mRNA expression decreased by 22% between weeks 12 and 24 in the untreated control group (and the Ad-control group), it increased more than twofold in the Ad-therapy group. Late increase of T cell infiltration is not typical for a normal chronic rejection course in this model (and in other models and patients as well), but it might reflect acute T cell events accelerating chronic graft injury. As the rats had no infections or other immunologic challenges during weeks 12 and 24, we speculate that the anti-Ad immune response exacerbates chronic graft injury due to the disappearance of therapeutic gene expression. This inflammatory adaptive immune response-dependent process triggers graft injury in the absence of massive signs of chronic rejection.
In summary, treatment of kidney grafts with Ad encoding for reporter gene or for IFN- accelerates chronic graft injury by triggering a persistent inflammatory process in the graft. Overexpression of immunomodulatory gene constructs targeting macrophage activation markedly improves functional, morphologic, and immunohistochemical changes after renal allograft transplantation in comparison to the reporter gene or IFN- groups. More complex, however, is the comparison of untreated controls. Although the 24-wk histology of the Ad-therapy group looks also significantly better than in the untreated controls, this improvement is not reflected by better long-term graft function. This discrepancy is probably due to the ongoing anti-Ad adaptive immune response after downregulation of the therapeutic gene expression that triggers late acute inflammation in the graft. Our data support the important role of cytokines and inflammation in the pathogenesis of chronic graft injury and demonstrate the power of gene therapy but also the limitations of presently available vectors.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
deficiency prevents coronary arteriosclerosis but not myocardial rejection in transplanted mouse hearts J Clin Invest 100: 550–557, 1997[Medline]
This article has been cited by other articles:
 |
|
 |
|
  N. Beckmann, C. Cannet, S. Zurbruegg, R. Haberthur, J. Li, C. Pally, and C. Bruns Macrophage Infiltration Detected at MR Imaging in Rat Kidney Allografts: Early Marker of Chronic Rejection? Radiology, September 1, 2006; 240(3): 717 - 724. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Lucas and G. McFadden Secreted Immunomodulatory Viral Proteins as Novel Biotherapeutics J. Immunol., October 15, 2004; 173(8): 4765 - 4774. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
, fibrosis; 
, tubular atrophy; 
, cell infiltration.
P < 0.05 versus Ad-control; $ P < 0.05 versus Ad-therapy.