Changes in the transcriptome in allograft rejection: IFN-gamma-induced transcripts in mouse kidney allografts.

We used Affymetrix Microarrays to define interferon-gamma (IFN-gamma)-dependent, rejection-induced transcripts (GRITs) in mouse kidney allografts. The algorithm included inducibility by recombinant IFN-gamma in kidneys of three normal mouse strains, increase in kidney allografts in three strain combinations and less induction in IFN-gamma-deficient allografts. We identified 40 transcripts, which were highly IFN-gamma inducible (e.g. Cxcl9, ubiquitin D, MHC), and 168 less sensitive to IFN-gamma in normal kidney. In allografts, expression of GRITs was intense and consistent at all time points (day 3 through 42). These transcripts were partially dependent on donor IFN-gamma receptors (IFN-gammars): receptor-deficient allografts manifested up to 76% less expression, but some transcripts were highly dependent (ubiquitin D) and others relatively independent (Cxcl9). Kidneys of hosts rejecting allografts showed expression similar to that observed with IFN-gamma injections. Many GRITs showed transient IFN-gamma-dependent increase in isografts, peaking at day 4-5. GRITs were increased in heart allografts, indicating them as generalized feature of alloresponse. Thus, expression of rejection-induced transcripts is robust and consistent in allografts, reflecting the IFN-gamma produced by the alloresponse locally and systemically, acting via host and donor IFN-gammar, as well as local IFN-gamma production induced by post-operative stress.

IFN-gamma alters the pathology of graft rejection: protection from early necrosis.

We studied the effect of host IFN-gamma on the pathology of acute rejection of vascularized mouse heart and kidney allografts. Organs from CBA donors (H-2k) were transplanted into BALB/c (H-2d) hosts with wild-type (WT) or disrupted (GKO, BALB/c mice with disrupted IFN-gamma genes) IFN-gamma genes. In WT hosts, rejecting hearts and kidneys showed mononuclear cell infiltration, intense induction of donor MHC products, but little parenchymal necrosis at day 7. Rejecting allografts in GKO recipients showed infiltrate but little or no induction of donor MHC and developed extensive necrosis despite patent large vessels. The necrosis was immunologically mediated, since it developed during rejection, was absent in isografts, and was prevented by immunosuppressing the recipient with cyclosporine or mycophenolate mofetil. Rejecting kidneys in GKO hosts showed increased mRNA for heme oxygenase 1, and decreased mRNA for NO synthase 2 and monokine inducible by IFN-gamma (MIG). The mRNA levels for CTL genes (perforin, granzyme B, and Fas ligand) were similar in rejecting kidneys in WT and GKO hosts, and the host Ab responses were similar. The administration of recombinant IFN-gamma to GKO hosts reduced but did not fully prevent the effects of IFN-gamma deficiency: MHC was induced, but the prevention of necrosis and induction of MIG were incomplete compared with WT hosts. Thus, IFN-gamma has unique effects in vascularized allografts, including induction of MHC and MIG, and protection against parenchymal necrosis, probably at the level of the microcirculation. This is probably a local action of IFN-gamma produced in large quantities in the allograft.

Interferon-gamma acts directly on rejecting renal allografts to prevent graft necrosis.

In transplant rejection interferon (IFN)-gamma regulates the recipient immune response but also acts directly on IFN-gamma receptors in the graft. We investigated these direct actions by comparing rejecting kidneys from donors lacking IFN-gamma receptors (GRKO mice) or control donors (129Sv/J) in CBA recipients. Beginning day 5, 129Sv/J kidneys displayed high major histocompatibility complex (MHC) expression, progressive infiltration by inflammatory cells, but no thrombosis and little necrosis, even at day 21. GRKO kidneys showed increasing fibrin thrombi in small veins, peritubular capillary congestion, hyaline casts, and patchy parenchymal necrosis, progressing to near total necrosis at day 10. Terminal dUTP nick-end labeling assays were positive only in the interstitial infiltrate, confirming that massive cell death in GRKO transplants was not apoptotic. Paradoxically, GRKO kidneys showed little donor MHC induction and less inflammatory infiltration. Both GRKO and 129Sv/J allografts evoked vigorous host immune responses including alloantibody and mRNA for cytotoxic T cell genes (perforin, granzyme B, Fas ligand), and displayed similar expression of complement inhibitors (CD46, CD55, CD59). GRKO kidneys displayed less mRNA for inducible nitric oxide synthase and monokine inducible by IFN-gamma but increased heme oxygenase-1 mRNA. Thus IFN-gamma acting on IFN-gamma receptors in allografts promotes infiltration and MHC induction but prevents early thrombosis, congestion, and necrosis.

Microchimerism in sensitized renal patients.

BACKGROUND: Patients exposed to allogeneic human tissue sometimes produce anti-HLA antibody for many years in the absence of further obvious antigen exposure. To investigate the mechanism of sustained sensitization, we identified females awaiting renal transplantation with high panel-reactive antibody but no exposure to allogeneic tissue for at least 1 month. METHODS: We analyzed peripheral blood microchimerism using nested polymerase chain reaction amplification specific for the SRY region of the Y chromosome. RESULTS: Microchimerism was detected in 3 of 10 patients but in none of 8 normal female subjects. In two cases, the amplified DNA polymerase chain reaction product was sequenced and was confirmed to be identical to the SRY gene. The estimated level of chimerism as compared with serial dilutions of DNA from male peripheral blood leukocytes was about 1/50000. CONCLUSION: These results do not establish causality but support the possibility that antigens from microchimeric donor cells may sustain the HLA antibody response in certain patients.

Mycophenolate mofetil reduces production of interferon-dependent major histocompatibility complex induction during allograft rejection, probably by limiting clonal expansion.

The immunosuppressive drug mycophenolate mofetil (MMF) acts by releasing mycophenolic acid (MPA), which inhibits the enzyme inosine monophosphate dehydrogenase (IMPDH) and thus inhibits de novo purine synthesis. Unlike cyclosporine (CsA), MMF has no direct effect on cytokine gene expression in vitro. We examined the effect of MMF, in comparison to CsA, on in vivo production of interferon-gamma (IFN-gamma) in mice. Two stimuli for IFN-gamma induction were used: (1) allogeneic P815 mastocytoma ascites tumour cells and (2) bacterial lipopolysaccharide (LPS). The allogeneic response is dependent on clonal expansion of T cells, while the LPS response is polyclonal and T cell independent. Since major histocompatibility complex (MHC) induction in mouse kidney is IFN-gamma dependent, we assessed the in vivo induction of IFN-gamma indirectly by measuring MHC induction in mouse kidneys in three systems: radiolabelled antibody binding assay, immunoperoxidase staining in tissue sections, and Northern blotting for steady-state MHC mRNA levels. IFN-gamma steady-state mRNA levels were assessed by reverse transcriptase polymerase chain reaction (RT-PCR). In the allogeneic response, MMF (40-160 mg/kg/day) reduced the production of IFN-gamma in a dose-dependent fashion. MHC class I and II induction was reduced by 35% to 74% and 30% to 74%, respectively. However, MMF had less effect on the induction of MHC by a nonimmune stimulus, bacterial LPS, whereas CsA reduced the induction of IFN-gamma in both responses. We conclude that MMF reduces the IFN-dependent induction of MHC in vivo during specific immune responses, probably by limiting clonal expansion, while preserving nonspecific cytokine production in response to LPS.

IFN regulatory factor-1 plays a central role in the regulation of the expression of class I and II MHC genes in vivo.

Transcription factor interferon regulatory factor-1 (IRF-1) is implicated in regulating class I MHC expression in vitro. We investigated the in vivo relationship between IRF-1 and MHC expression in kidney and other nonlymphoid organs, assessing MHC expression in mice with disrupted IRF-1 genes (IRF-1 KO) compared with mice with intact IRF-1 genes (WT). In kidneys of IRF-1 KO mice, basal class I expression was decreased, particularly on arterial endothelium, but basal class II expression was unchanged. The induction of both class I and class II expression by injected rIFN-gamma was reduced in IRF-1 KOs, compared with WT mice. Similarly, stimuli that induce endogenous IFN-gamma production (LPS or oxazolone) massively increased MHC expression in kidneys of WT mice, with little increase in IRF-1 KO mice. Impaired class II induction by rIFN-gamma in IRF-1 KO mice probably reflects the role of IRF-1 in regulating class II transactivator (CIITA) expression: rIFN-gamma induced CIITA mRNA less in kidneys of IRF-1 KO mice than in WT mice. In organs of WT mice, IRF-1 mRNA was expressed in the basal state, and rIFN-gamma treatment increased IRF-1 mRNA before the induction of class I or CIITA mRNA. Treatment of WT mice with cycloheximide plus rIFN-gamma superinduced IRF-1 mRNA expression, but partially inhibited CIITA mRNA expression, indicating that IRF-1 mRNA induction is not dependent on new protein synthesis, unlike CIITA. Thus, in vivo, IRF-1 plays a major role in basal and induced class I expression and in induction of class II by IFN-gamma, probably via CIITA induction.

In vivo class II transactivator expression in mice is induced by a non-interferon-gamma mechanism in response to local injury.

BACKGROUND: Tissue injury induces MHC class II expression, which could be important in the recognition of that tissue as an allograft. The class II transcriptional activator (CIITA) is the major regulator of basal and induced MHC class II expression and is essential for antigen presentation. The role of CIITA in the induction of class II by tissue injury is unknown. In this study, we examined CIITA induction in the course of acute ischemic or toxic renal injury in mice, including the role of interferon (IFN)-gamma and of the transcription factor, interferon regulatory factor (IRF)-1. METHODS: Kidneys were injured by ischemia or by gentamicin toxicity and were then studied for changes in gene expression using Northern blot, reverse transcriptase-polymerase chain reaction, radioimmunoassay, and tissue staining. We compared wild-type (WT) mice to IFN-gamma knockout (GKO) or IRF-1 knockout mice. RESULTS: Ischemic injury induced CIITA and class II expression in the kidney, in WT and GKO mice. Gentamicin injury also induced both CIITA and class II expression, independent of IFN-gamma, in WT and GKO mice. After ischemic injury, the induction of class II protein levels and CIITA and class II mRNA levels were induced, to a lesser degree, in IRF-1 knockout mice. CONCLUSIONS: These data indicate that CIITA is induced by tissue injury, and probably accounts for class II induction during tissue injury. CIITA induction by injury is largely IFN-gamma independent but requires IRF-1. The similarities of the pattern of CIITA and class II induction in ischemic and toxic injury suggest that this is a stereotyped response of injured tissue and not a consequence of a particular mechanism of injury.

The induction of class I and II major histocompatibility complex by allogeneic stimulation is dependent on the transcription factor interferon regulatory factor 1 (IRF-1): observations in IRF-1 knockout mice.

Hosts undergoing allograft rejection show increased MHC expression locally in the graft and systemically in the normal host organs, mediated principally by IFN-gamma. The transcription factor IRF-1 has been implicated in the regulation of MHC expression by IFNs in vitro as well as in the regulation of production of some cytokines. We investigated the role of IRF-1 in vivo in the systemic regulation of MHC expression in hosts undergoing rejection of allogeneic tumors by comparing MHC induction in mice with normal IRF-1 genes (wild type or WT mice) with mice with disrupted IRF-1 genes (IRF-1 knockout or IRF-1 KO mice). We assessed MHC product expression by immunohistology and by radiolabeled antibody binding to tissue homogenates, and MHC mRNA levels by Northern blotting. By immunohistology in mice undergoing allogeneic stimulation by the ascites tumor cells, kidneys of WT mice showed massive class I and II induction, but kidneys from IRF-1 KO mice showed almost no class I and II induction. Allograft rejection also increased class I and II product levels by radiolabeled antibody binding and steady state mRNA levels, but again IRF-1 KO mice showed severe impairment of MHC induction. Similar impaired MHC class I and II induction was seen in heart and spleen, but in liver the IRF-1 mice showed impaired class I induction but unimpaired class II induction. The results indicate that IRF-1 has an essential role in both class I and class II MHC induction in allogeneic responses, but that a component of IRF-1 independent MHC induction is also demonstrable in some tissues. The reduction in MHC induction by allogeneic stimulation probably reflects decreased response to IFN-gamma and other cytokines as well as some reduction in the amount of cytokines produced.

Evidence for the in vivo role of class II transactivator in basal and IFN-gamma induced class II expression in mouse tissue.

The class II transactivator (CIITA) is a protein that induces the transcription of MHC class II genes. We studied the expression of CIITA in vivo, comparing steady state levels of CIITA and class II mRNA in various mouse tissues. Many tissues in normal mice contained mRNA for CIITA, correlating with class II mRNA. The basal expression of CIITA and class II mRNA in mice with disrupted IFN-gamma genes (GKO mice) was similar to that in wild-type mice. Injection of rIFN-gamma strongly induced CIITA and class II mRNA: CIITA mRNA increased at 2 hr and declined to baseline by 48 hr, whereas class II mRNA increased at 24 hr and returned to baseline at 7 days. Proinflammatory stimuli that induce IFN-gamma production (allogeneic cells and LPS) induce CIITA and class II expression in wild-type mice, but not in GKO mice. CIITA induction by IFN-gamma was partially sensitive to cycloheximide, suggesting that another protein is required for CIITA induction. The data suggest that CIITA is a major regulator of basal and induced class II expression in vivo.

Effect of recombinant human insulin-like growth factor-1 on the inflammatory response to acute renal injury.

Renal ischemic injury evokes an inflammatory response with increased cytokine and major histocompatibility complex (MHC) expression and a mild interstitial infiltrate. This "injury response" could contribute to the tendency of ischemically injured renal transplants to reject. The studies presented here evaluated the ability of recombinant human insulin-like growth factor-1 (rhlGF-1) given after renal injury to prevent renal inflammation. The left renal pedicle of CBA and BALB/c mice was clamped for 60 min, and rhlGF-1 (25, 50, 100 micrograms) was administered sc at 2, 24, 48, 72, and 96 h after reflow. Cytokine and MHC expression was monitored in the injured kidney, compared with the contralateral kidney. In untreated mice, a single episode of injury induced the expression of MHC mRNA and products and tumor necrosis factor-alpha (TNF-alpha) mRNA, and depressed preproepidermal growth factor (ppEGF) mRNA, for up to 5 wk. With immunohistology, epithelial Class I and II MHC expression was shown to be increased for 2 wk, and Class II positive interstitial cells were shown to be increased for up to 5 wk. The ischemically injured kidneys from mice treated with rhlGF-1 and examined at 5 days showed a dose-dependent normalization of all of the changes of the injury response. This included prevention of the increased expression of MHC and cytokines and the Class II positive interstitial cells, and restoration of ppEGF mRNA. Thus the complex and long-lasting increase in proinflammatory cytokines and MHC expression that follow renal ischemia can be interrupted by treatment with rhlGF-1 beginning 2 h after the injury. This therapy may have applications to the injury response in renal transplants.

Identification of a calcium-inducible, cyclosporine sensitive element in the IFN-gamma promoter that is a potential NFAT binding site.

Cyclosporine (CsA) inhibits cytokine transcription by preventing the activation of key promoter sites, in particular the binding of nuclear factor of activated T cells (NFAT) to the IL-2 NFAT site and the "P" site in IL-4. To identify potential NFAT-like sites in the IFN-gamma promoter, we sought areas of homology with the known sites in other promoters. In the promoter region of the mouse and human IFN-gamma gene, we identified two repeats of a consensus sequence ATTTCCnnT, designated P1 and P2 because of their homology to the calcium-inducible and CsA-sensitive "P" sequences in the IL-4 promoter. In electrophoretic mobility shift assay (EMSA), a probe containing the second P sequence "P2" in the human IFN-gamma gene bound nuclear proteins from stimulated, but not unstimulated, humans T cells. The cytosol of unstimulated cells contained similar binding activity that decreased after stimulation, indicating that this binding activity translocated to the nucleus after stimulation. CsA inhibited nuclear translocation. Competition studies demonstrated that oligomers containing the sequences P1 and P2 in IFN-gamma gene, the NFAT site in the IL-2 gene, and the IL-4 P site competed with the P2 probe for protein binding, whereas an oligomer containing mutations in the P2 site did not. Addition of anti-NFAT antiserum altered protein binding to P2, indicating that the proteins were either identical or related to NFAT. Stimulation of T cells transfected with constructs containing three copies of the P2 sequence enhanced CAT activity in response to ionomycin, and this effect was blocked by CsA. These results suggest that the P2 sequence, and probably the P1 sequence, in the IFN-gamma promoter are NFAT binding sites and contribute to the calcium inducibility and CsA sensitivity of IFN-gamma production.

Disturbed MHC regulation in the IFN-gamma knockout mouse. Evidence for three states of MHC expression with distinct roles for IFN-gamma.

We compared the expression of MHC class I and II products in tissues of IFN-gamma knockout (GKO) vs normal (wild-type) BALB/c mice. We studied expression in the basal state, after local tissue injury, after stimuli that induce systemic MHC expression (allogeneic cells, oxazolone skin painting, or LPS), and after rIFN-gamma. Basal class II expression in interstitial cells was not reduced in GKO mice. However, GKO mice had less basal class I expression in kidney, liver, heart, and arterial endothelium than wild-type mice. Local renal ischemic injury increased class I and II expression in kidney tubules of both GKO and wild-type mice, but induction in GKO was less than in wild-type. Potent inflammatory stimuli increased systemic MHC class I and II markedly in kidney, liver, and heart of wild-type mice, but induced no increase in GKO mice. rIFN-gamma induced class I and II equally in GKO and wild-type mice. Thus, three states of MHC expression can be defined that differ in their dependencies on IFN-gamma: basal, locally induced, and systemically induced. Basal class II expression in interstitial cells is IFN-gamma independent, but basal class I expression, particularly in arterial endothelium, is partially dependent on IFN-gamma. The local increase in MHC class I and II in parenchymal cells in response to injury reflects both IFN-gamma and a non-IFN-gamma factor. Systemic MHC class I and II induction is almost exclusively due to IFN-gamma.

Ischemic acute tubular necrosis induces an extensive local cytokine response. Evidence for induction of interferon-gamma, transforming growth factor-beta 1, granulocyte-macrophage colony-stimulating factor, interleukin-2, and interleukin-10.

We noted previously that ischemic acute tubular necrosis (ATN) induces local expression of MHC products in renal epithelium. The present investigations were conducted to establish the role of IFN-gamma in the regulation of MHC antigen expression in ATN and to explore the changes in cytokine and growth factor expression induced by ischemic renal injury. We produced unilateral ischemic ATN in mice by clamping the left renal pedicle. MHC class I and II steady state mRNA induction was assessed by northern blot analysis, and MHC product was quantified by the extent of binding of radiolabeled monoclonals to tissue homogenates. The steady state mRNA levels for IFN-gamma, IL-2, IL-10, and granulocyte-macrophage CSF were assessed by reverse transcriptase polymerase chain reaction and the levels for transforming growth factor-beta 1 and prepro-epidermal growth factor (ppEGF) were assessed by Northern blot analysis. In the injured kidneys, steady state mRNA levels for IFN-gamma, IL-2, IL-10, granulocyte-macrophage CSF, and transforming growth factor beta-1 were increased, whereas ppEGF mRNA was markedly decreased. The MHC expression was inhibited by treatment of mice with an anti-IFN-gamma mAb (R4-6A2). Murine EGF, administered in an attempt to accelerate recovery, did not reduce the cytokine and MHC changes. These data indicate that ischemic injury, and possibly other forms of injury, triggers a complex circuit of proinflammatory cytokines. This "injury response" could be relevant to clinical renal transplants, where ATN is associated with poor graft outcome.

Regulation of IFN-gamma expression in vivo. IFN-gamma up-regulates expression of its mRNA in normal and lipopolysaccharide-stimulated mice.

Immune responses to such stimuli as tissue injury, infection, and allografting result in localized IFN-gamma expression. Autoregulation of cytokine expression has been described for some cytokines in vitro, but whether this occurs in vivo is unknown. We therefore examined the role of murine IFN-gamma in the regulation of its own expression in vivo after stimulation with bacterial LPS. This agent is known to induce IFN-gamma expression in both spleen and kidney in a T cell-independent, cyclosporine-sensitive manner. We found that concomitant administration of a neutralizing mAb to IFN-gamma inhibited not only the MHC expression induced by LPS but also the increased IFN-gamma mRNA expression, suggesting an autoregulatory role for IFN-gamma. Inhibition was dose dependent and observed in both spleen and kidney. The effect was not seen with a neutralizing anti-IL-3 mAb, demonstrating specificity. The inhibition of IFN-gamma mRNA expression by the anti-IFN-gamma mAb occurred in both T cell-deficient athymic nude mice and their normal controls, suggesting that the autoamplification of IFN-gamma mRNA in vivo is T cell independent. Administration of rIFN-gamma to unstimulated mice induced IFN-gamma mRNA expression in spleen and kidney, supporting the conclusion that IFN-gamma up-regulates expression of its mRNA. Exposure of resting murine splenocytes to concentrations of rIFN-gamma as low as 10 U/ml in vitro induced expression of IFN-gamma mRNA. Thus, in vivo IFN-gamma may participate in an autoregulatory loop to amplify the amount of IFN-gamma expressed both at the site of local inflammation and at remote sites. This would have relevance in the mechanism by which the host defends itself against and prevents dissemination of an infectious agent.

Regulation of IFN-gamma and tumor necrosis factor-alpha expression in vivo. Effects of cycloheximide and cyclosporine in normal and lipopolysaccharide-treated mice.

Despite accumulating information about cytokine expression in vitro, relatively little is known about the regulation and biologic relevance of these mediators in vivo. In order to study the effects of inhibition of protein synthesis and cyclosporine in vivo, we made use of systemically administered LPS, which induces the expression of a variety of cytokines. The expression of IFN-gamma and TNF-alpha mRNA in normal and LPS-treated mice was examined by Northern blot analysis and amplification using the polymerase chain reaction. IFN-gamma activity was monitored using the biologic end point of MHC induction. TNF-alpha activity in serum was assessed using a L929 cytotoxicity assay. Messenger RNA for IFN-gamma and TNF-alpha could not be reliably detected by Northern analysis in spleens or kidneys of normal mice. After treatment with cycloheximide, a protein synthesis inhibitor, IFN-gamma and TNF-alpha mRNA could be detected in both sites in otherwise normal mice. The level of both IFN-gamma and TNF-alpha mRNA increased after LPS, although the temporal patterns of expression were different. The concurrent administration of cycloheximide led to marked superinduction of both cytokine mRNA levels. Similar effects were seen in T cell-deficient nude mice, suggesting that these responses are T cell independent. Cyclosporin A blocked induction of IFN-gamma in a dose dependent manner, but failed to significantly inhibit TNF-alpha mRNA or protein expression. Thus at least part of the immunosuppressive effect of cyclosporin A in vivo may be caused by its ability to inhibit the expression of certain cytokine genes, as has been found in vitro systems. However, the cellular target for this effect may extend to cell populations other than T cells.