Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Acute cellular rejection of human renal tissue by adoptive transfer of allogeneic human peripheral blood mononuclear cells into chimeric rats: sequential gene expression of cytokines, chemokines and cytolytic effector molecules, and their regulation by CTLA-4–Ig

Acute cellular rejection of human renal tissue by adoptive transfer of allogeneic human... Abstract Th1- and Th2-related cytokines (IFN-γ, IL-2, IL-4, IL-10), β-chemokines (RANTES, macrophage inflammatory protein-1β) and their receptor [chemotatic cytokine receptor (CCR) 5], and the cytolytic effector molecule [Fas ligand (FasL)] play an essential role in regulating and co-ordinating acute renal allograft rejection. A chimeric model of acute cellular rejection which involves subcapsular grafting of human renal tissue in the kidneys of immunodeficient rats and subsequent i.p. infusion of allogeneic human peripheral blood mononuclear cells (PBMC) was used to study cellular infiltration patterns and sequential intragraft gene expression of these key inflammatory mediators. We found that while all molecules are expressed within the human renal implant at specific time points following infusion of allogeneic human PBMC, peak mRNA expression of IFN-γ, IL-2, RANTES and CCR5 is associated with a phase of human mononuclear infiltration and accumulation, prior to graft destruction (induction phase). A short burst of FasL gene expression is found at the end of induction and at the onset of graft deterioration. IL-4 mRNA, which is hardly detectable, and IL-10 mRNA, which appears early and persists throughout follow-up at high levels, both peak after the induction phase with the advent of graft destruction. Furthermore, treatment with CTLA-4–Ig, which hardly affects migration of human effector cells into graft tissue, is associated with a temporary reduction in gene transcript levels for all inflammatory mediators, especially IL-2 and IL-4, reduced apoptosis in the graft and amelioration of tissue injury. Thus, development of acute cellular rejection in our chimeric model involves a co-ordinated pattern of gene expression, in which CTLA-4–Ig promotes its effects by transient inactivation of infiltrating human cells. allograft rejection, animal model, apoptosis, RT-PCR, SCID, Th1, Th2 AEC 3-amino-9-ethylcarbazol, CCR chemotatic cytokine receptor, FasL Fas ligand, MIP macrophage inflammatory protein, PBMC peripheral blood mononuclear cells, PE phycoerythrin, SCID severe combined immunodeficiency, TBI total-body irradiation Introduction The mechanisms responsible for acute cellular rejection are primarily related to the development of alloantigen-specific cytotoxic T lymphocytes and delayed-type hypersensitivity responses, but also to non-immune mechanisms. Inflammatory mediators produced by T cells, macrophages and donor tissue have all been suggested to play an important role in the rejection process observed in murine allograft models and human transplants (1,2). There is abundant data on the expression and activity of different cytokines during acute cellular rejection in murine models and human allografts (3–9). Some of these studies advocate a paradigm of a polarized Th1/Th2 immune response in rejection and tolerance; others indicate the possible involvement of both populations in the immune response leading to graft rejection or in the induction of tolerance and prolonged graft survival (10–12). In addition, chemoattractant cytokines (chemokines), especially of the β chemokine family which are chemoattractants for monocytes and lymphocytes (13–17), and the cytolytic effector molecule, Fas ligand (FasL), have been recently identified during a rejection process (18–20). However, kinetic studies analyzing sequential intragraft expression of cytokines, chemokines and cytolytic effector molecules concurrently, their relationship to cellular infiltration, and the effect of experimental manipulation on them, are less abundant. We have recently developed a chimeric model of acute cellular rejection which involves subcapsular grafting of human renal tissue in the kidneys of SCID-like rats and subsequent i.p. infusion of allogeneic human peripheral blood mononuclear cells (PBMC) (21,22). The animals we use are normal strains of mice or rats which are converted to SCID-like with a lethal level of total-body irradiation (TBI) and radioprotection by transplantation of SCID mouse bone marrow cells (23,24). An important major difference between the SCID mouse model and the radiation chimera is associated with the engraftment rate following adoptive transfer of human PBMC, and consequently with the functionality of the engrafted human T cells (25,26). Following i.p. injection of human PBMC into SCID mice, the cells remain in the peritoneal cavity in small numbers for >30 days before expanding. In the radiation chimera recipients, the rate of appearance of human cells is greatly enhanced and engraftment, as well as formation of human follicles by T and B cells in the spleen, takes place within a few days post-transplant (23–27). The engraftment of human cells is probably improved by the lethal conditioning employed and by increasing the inoculum of human cells (300–400×106 per rat). This strong and rapid engraftment enabled us to study the rejection of the human renal grafts early after transplantation of allogeneic human PBMC. Intraperitoneal infusion of allogeneic human PBMC following kidney implantation, resulted in the presence of patchy mononuclear infiltrates in the human renal implant within 3 weeks and lead to rejection of the human renal tissue. In vivo treatment with CTLA-4–Ig fusion protein (murine CTLA-4 and human Ig) ameliorated tissue injury (21). The ability to collect from control and CTLA-4–Ig-treated chimeric rats a substantial amount of human kidney tissue, following the infusion of allogeneic human PBMC, has enabled us now to perform kinetic studies on cellular infiltration patterns and regulation of Th1 and Th2 cytokines (IFN-γ, IL-2, IL-4, IL-10), β chemokines [RANTES, macrophage inflammatory protein (MIP)-1β], the receptor for the β chemokines [chemotatic cytokine receptor (CCR) 5] (28,29), and the cytolytic effector molecule (FasL) during the different stages of the unmodified and modified cellular rejection process. Methods Experimental animals Three-month-old Lewis (from Harlan Olac, Bicester, UK) and inbred nude rats (from Harlan Sprague Dawley, Indianapolis, IN) were used. All rats were kept in small cages (one to three animals in each cage), and fed sterile food and acid water containing ciprofloxacin (20 mg/ml). Bone marrow donors were NOD.SCID mice from the Weizmann Institute Animal Breeding Center (Rehovot, Israel). Conditioning regimen Lewis rats were exposed to split dose (4 Gy followed 3 days later by 11.5 Gy) TBI, from a γ beam 150-A 60Co source (produced by the Atomic Energy of Canada, Kanata, Ontario, Canada) with a focal skin distance of 75 cm and a dose rate of 0.7 Gy/min. Nude rats were exposed to a split-dose TBI of 4 + 9.5 Gy (13). Preparation and transplantation of SCID bone marrow cells Bone marrow cells were collected into PBS by flushing the shafts of the femur and tibia obtained from SCID mice (4–8 weeks old) according to Levite et al. (30). Recipient rats were injected with 50×106 SCID bone marrow cells (i.v. in 1.0 ml PBS) 1 day after the second fraction of TBI. Human renal tissue Human kidney segments were obtained from nephrectomy specimens performed for primary localized kidney tumors. Tangential slices were recovered from the cortex of the removed kidney. In all cases, the non-tumor renal tissue used for transplantation had no histopathologic changes, as confirmed by hematoxylin & eosin and reticulin staining. The human kidney fragments were kept in sterile conditions at 4°C (for ~2 h) in either RPMI 1640 or Dulbecco's modified Eagle's medium containing 10% FCS (Biological Industries, Bett HaEmek, Israel). Transplantation of the human kidney tissue was done under general anesthesia (2.5% Avertin in PBS, 10 ml/kg i.p.). Both kidneys were exposed through a bilateral incision. A 1.5 mm incision was made at the caudal end of the kidney capsule and a 1 mm3 fragment of human kidney was implanted under each capsule. Transplanted rats were treated post-operatively with ciprofloxacin in their drinking water for 7 days. The surgical procedure was performed 7–10 days after the animals had received the SCID bone marrow. Preparation and transplantation of human PBMC Buffy coats from normal volunteers were layered onto Lymphoprep solution (Nycomed, Oslo, Norway) and spun at 2000 r.p.m. for 20 min. The interface layer was collected, washed twice, counted and resuspended in PBS (pH 7.4) to the desired cell concentration. Human PBMC (400×106 cells) were injected i.p., 1–3 days after transplantation of the human renal tissue into recipient rats, conditioned as described above. Control rats did not receive human PBMC. Treatment of animals by CTLA-4–Ig in vivo Hybridoma producing a fusion protein between mouse CTLA-4 and human IgG1 (31) was kindly provided by Dr Stephan Yung (Departemnt of Immunology, Hadassa Medical School, Jerusalem, Israel). Hybridomas were injected i.p. into pristane-treated SCID mice and ascitic fluid was recovered. For some indicated studies, rats were injected i.p. with 0.5 mg CTLA-4–Ig fusion protein on 3–5 consecutive days, beginning at 3 days after the injection of human PBMC. Control rats either did not receive human PBMC or were injected with PBS following human PBMC administration. Cell collection from human → rat chimera Peritoneal cells were obtained by lavage with 5–10 ml of PBS. Rat kidneys were removed after the animals were killed by cervical dislocation and renal grafts were carefully separated from the host's kidney. Pieces of organs and the human grafts were pressed through stainless steel sieves to make a cell suspension in PBS, and cells were then isolated using Lymphoprep, washed with PBS containing 1% BSA and stained for analysis on a FACS. FACS analysis Single-cell suspensions were incubated for 30 min on ice with labeled anti-human CD3–phycoerythrin (PE) and CD45–PerCP (pan-human leukocyte antigen) antibodies (Becton Dickinson, Mountain View, CA). After washing, two- or three-color fluorescent analysis of human antigens was performed using a FACScan analyzer (Becton Dickinson). The lymphocytes were gated according to their size on forward and side scatters. Immunohistochemistry Paraffin tissue blocks were cut 4–6 μm thick, deparaffinized in xylene, rehydrated and placed for 15 min in alcohol with 3% H2O2 to block endogenous peroxidase. Slides were thoroughly washed with tap water and transferred to PBS. Sections were then treated with 1% BSA to prevent background staining and incubated for 1 h with primary mouse anti-albumin antibody (Zymed, San Francisco, CA) at room temperature in a humidified chamber. Slides were rinsed with PBS for 3 min and incubated with the biotinylated linked antibody for 30 min and with the labeling reagent, peroxidase-conjugated streptavidin, for 30 min (StrAvigen; Biogenex, San Ramon, CA). After rinsing, the peroxidase label was visualized using 3-amino-9-ethylcarbazol (AEC), following processing instructions according to the immunohistochemical kit (Biomeda, Foster City, CA), for 15 min and counterstained with Mayer's hematoxylin. AEC produced a red product that is soluble in alcohol and can be used with an aqueous mounting medium (Kaiser's glycerol gelatin). A negative control was run using the same technique but omitting the primary antibody and adding the streptavidin–biotin complex. Following infusion of allogeneic PBMC into transplanted rats, staining of the human graft was done using rabbit anti-human CD3 (pan-T cell), mouse anti-human HLA-DR (MHC class II molecule expressed on activated T cells), mouse anti-human CD14 (monocytes), mouse anti-human CD68 (macrophages) and mouse anti-human CD56 (NK) (all from Zymed); mouse anti-human CD20 (pan-B cell) (Dakopatts, Glostrup, Denmark); and mouse anti-human CD54 (ICAM-I) (Serotec, Oxford, UK). In situ detection of apoptosis in human grafts Paraffin tissue blocks were cut 4–6 μm thick, deparaffinized in xylene, rehydrated and placed for 15 min in alcohol with 3% H2O2 to block endogenous peroxidase. Slides were thoroughly washed with tap water and transferred to PBS. Staining for apoptosis was then preformed according to the Apoptag in situ apoptosis detection kit (Oncor, Gaithersburg, MD). This kit uses the C-terminal deoxynucleotidyl TUNEL technique of using TdT to tail all fragmented DNA with digoxigenin-dUTP. The tails are labeled with an anti-digoxigenin antibody complexed to an alkaline phosphatase reporter enzyme, and the sites are visualized with nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate dyes. Molecular studies of the human renal implants Human renal tissue was dissected carefully from the subcapsular site and homogenized with a glass-Teflon tissue tearer in Tri-reagent (Molecular Research Center, Cincinnati, OH). Total RNA extraction was done using the guanidinium isothiocyanate–phenol isolation method. The isolated total RNA was air-dried, resuspended in nuclease-free water and quantified by spectrophotometry. Then 1 μg of total RNA was reverse-transcribed into cDNA using Avian myeloblastosis virus reverse transcriptase according to standard procedures. The reverse transcriptase reaction was diluted 1:50, 1:100 and 1:500 in sterile H2O, and DNA amplification was done using thermostable Tfl DNA polymerase in a 50 μl reaction mixture containing 40 μM of each dNTP, 0.4 μM of each primer, 10 mM Tris–HCl (pH 8.3) and 1.5 mM MgCl2. In all experiments the presence of possible contamination was checked by control reactions in which (i) reverse transcriptase was omitted from the reaction mixture and (ii) buffer alone was added to the reverse transcription reaction mixture. All primer sequences were run through the National Center of Biological Information (NCBI) Gene Bank library to ensure that each primer was specific to its respective human gene but not to the rat gene. In addition, all of the primers were designed to flank an intron site in the target sequence. This primer design permits detection of genomic contamination of the RNA isolated from the human renal (32). Also, to avoid cross-hybridization with unrelated or less specific genes, the annealing temperature during thermal cycling was set high (64°C). The following sense and antisense primers were used: IL-2, 5′-ACAGCTACAACTGGAGCATT-3′ and 5′-TGCTGTCTCATCAGCATATT-3′; IFN-γ, 5′-TGCAGGTCATTCAGATGTAG-3′ and 5′-AGCCATCACTTGGATGAGTT-3′; IL-4, 5′-GCGATATCACCTTACAGGAG-3′ and 5′-TTGGCTTCCTTCACAGGACA-3′; IL-10, 5′-AGATCTCCGAGATGCCTTCA-3′ and 5′-ATTCTTCACCTGCTCCACGG-3′, all according to Kirk et al. (32); RANTES, 5′-ATGAAGGTCTCCGCGGCAGCCC-3′ and 5′-CTAGCTCATCTCCAAAGAGTTG-3′; MIP-1β, 5′-ACCATCAAGCTCTGCGTGACTG-3′ and 5′-GCAGGTCAGTTCAGTTCCAGGTC-3′ (33); FasL, 5′-CAAGTCCAACTCAAGGTCCATGCC-3′ and 5′-CAGAGAGAGCTCAGATACGTTTGA-3′ (34); CCR5, 5′-CTCGGATCCGGTGGAACAAGATGGATTAT-3′ and 5′-CTCGTCGACATGTGCACAACTCTGACTG-3′ (35). With the aim of detecting PCR signals in the linear phase of product amplification, 20–35 PCR cycles were carried out in a thermocycler. The PCR product was applied to a 1.5% agarose gel, resolved by electrophoresis, visualized by ethidium bromide staining and photographed. Scanning by an Imaging Computing Laser Densitometer GS-670 (BioRad, Hercules, CA) of the negatives of the PCR products and analysis by Adobe Photoshop software (Adobe, Mountain View, CA) were used to determine the absolute absorbance of the PCR products. The densitometric values obtained for each cytokine PCR product were normalized as a percent of signal intensity of the β-actin PCR product, i.e. all PCR signals were standarized to the constitutive gene expression of β-actin included in each PCR determination (internal control) (36). Under the conditions used, there was a linear relationship between the amount of RNA in the tissue sample that was reverse-transcribed to cDNA and amplified by PCR, and the PCR product signal (36). Tissue samples to be compared were amplified in the same PCR run using reagents from a single master mix. Each sample was tested on at least three occasions. This assay is semiquantitative, i.e. it allows for the determination of the relative amount of mRNA for a given cytokine in different samples but does not allow for a precise determination of the amount of mRNA present in each sample. Similar semiquantitative PCR assays have been used to analyze cytokine mRNA expression in other conditions (19,33,37,38). Results Experimental design Groups of seven to 10 lethally irradiated Lewis rats radioprotected with SCID mouse bone marrow were transplanted with human renal tissue and were then infused with allogeneic human PBMC. Animals were left untreated or injected with the fusion protein CTLA-4–Ig, which blocks the CD28–B7 T cell co-stimulatory pathway. Some rats were implanted with human renal tissue but were not injected with human PBMC nor treated with CTLA-4–Ig. Human renal transplants were harvested at days 4, 7, 10, 13, 16, 19 and 22 following human PBMC infusion (at least three in each time point for each group). Grafts were processed for histologic and immunohistologic evaluation, and were either pooled and processed to recover human infiltrating cells or homogenized on an individual basis for RNA extraction. Inflammatory cell infiltration patterns Table 1 and Figs 1 and 2 summarize data regarding sequential analysis of the human cellular infiltrate. All of the human grafts developed a low-grade mononuclear infiltrate during the first 7 days following human PBMC infusion. There was no significant difference in the intensity of these early infiltrates between untreated grafts and those treated with CTLA-4–Ig. These early infiltrates contained a relative high number of monocytes (Table 1; Fig. 1A). After the development of the early infiltrate, the intensity of infiltration increased progressively, with the appearance of patchy mononuclear infiltrates surrounding human tubules during the second week, followed by a dense and diffuse interstitial mononuclear infiltrate and graft destruction by day 21 (Table 1). In addition, while human CD3+ and HLA-DR+ cells were found in the implant within the first week after the injection of human PBMC, intensification of these cells occurred during the second and third weeks (Table 1; Figs 1C, 1D and 2), culminating into the massive activated human T cell infiltrates previously described (21). Interestingly, after the first 13 days, hardly any monocytes were detected in the human grafts and the infiltrates were predominantly lymphocytic (Table 1; Fig. 1B). Our sequential histologic evaluation of the human renal grafts from CTLA-4–Ig-treated animals showed that the kinetics of appearance of human mononuclear infiltrates in these grafts was similar to that observed in the untreated ones (Table 1) and thus CTLA-4–Ig hardly altered the presence of dense interstitial infiltrates found during the third week (21). Furthermore, the nature of the human infiltrating cells did not change (Table 1). However, compared to the untreated transplants, in the CTLA-4–Ig treated grafts we found reduced staining for the activation markers HLA-DR and ICAM-I on leukocytes, as well as for apoptotic cells (Table 1; Figs 1E and F). Interestingly, reduced numbers of apoptotic cells were found not only in renal parenchymal structures but also within infiltrates. Overall, human renal tissue was less damaged at 3 weeks following infusion of human PBMC (histology was similar to that observed in grafts which were not challenged with PBMC and were not infiltrated at all) and graft rejection was delayed by several weeks (21). Kinetics of cytokine mRNA expression in human renal grafts subjected to allogeneic human PBMC and its regulation by CTLA-4–Ig By using repeated measurements for each time point and by normalizing each mRNA determination according to its co-expressed β-actin levels, we were able to define for each cytokine or chemokine the time point in which a relative peak of its expression is attained, as well as the inhibitory effect of CTLA-4–Ig on its expression pattern. Sequential molecular analysis of both Th1 (IFN-γ and IL-2) and Th2 (IL-4 and IL-10) type cytokines and their regulation by CTLA-4–Ig is shown in Figs 3–6. In each time point analysis of transcripts in grafts of animals that were not injected with PBMC nor treated with CTLA-4–Ig was used to determine basal mRNA level. As can be seen in Fig. 3, the expression of IL-2 started to increase during the first week following injection of human PBMC, peaked at 13 days, and then dropped sharply and remained at low plateau levels (days 16–22). A similar pattern was observed for IFN-γ (Fig. 4), with somewhat enhanced increase in transcripts level, peaking on day 10. Treatment with CTLA-4–Ig resulted in a differential effect, predominantly inhibiting expression of IL-2 and to lesser extent IFN-γ expression. Thus, IL-2 mRNA expression was completely abrogated during the second week following human PBMC infusion delaying peak expression to day 19, while IFN-γ expression was only partially inhibited (complete inhibition was not seen at any of the time points tested). IL-4 expression was hardly detectable throughout the experimental period. Nonetheless, a significant peak was reproducibly noted during the third week, decreasing to undetectable levels on day 22. Examination of the CTLA-4–Ig effect on IL-4 expression showed that the low but significant signal of IL-4 mRNA was completely inhibited by CTLA-4–Ig treatment (Fig. 5). IL-10 expression in the unmodified rejection process, as opposed to IL-4 but similar to IL-2 and IFN-γ, typically started early and persisted during the first 2 weeks. However, during the third week, as in the case of IL-4, peak expression of IL-10 was noted. Evaluation of IL-10 expression in the model shows a pattern compatible with partial inhibition by CTLA-4–Ig (Fig. 6). Thus, while the amplitude of IL-10 is reduced by CTLA-4–Ig on most of the time points during the experimental period, there is a substantial level of IL-10 mRNA, which does not seem to be affected. Kinetics of chemokine, chemokine receptor and cytolytic effector molecule mRNA expression in human renal grafts subjected to allogeneic human PBMC, and their regulation by CTLA-4–Ig In addition to testing cytokines, we examined expression of the chemokines, RANTES and MIP-1β, their β-chemokine receptor CCR5, and the cytolytic molecule FasL (Figs 7–10). As shown in Figs 7 and 8, the expression of RANTES and of MIP-1β displayed a similar pattern to that of IL-2 and IFN-γ. In contrast, expression of CCR5 and of FasL was displayed in short bursts. Thus CCR5 mRNA levels were noted in parallel to peak expression of RANTES on day 10 (Fig. 9), while FasL mRNA was found only on days 13–16 and rapidly declined to undetectable levels (Fig. 10). The effect of CTLA-4–Ig on RANTES, CCR5 and FasL was similar in that the peaks of expression were all delayed. There was a significant reduction in RANTES transcript levels, following an early peak on day 4, so that highest expression was noted on day 19, >1 week after peak levels were found in the untreated group. Nonetheless, constitutive expression of RANTES was noted in the treated group (Fig. 7). Moreover, early transcription bursts of CCR5 and FasL during the unmodified rejection process were abrogated in the treated group, with the effect on FasL being more prominent, as substantial levels of CCR5 were noted on both days 13 and 16 (Figs 9 and 10). The effect of CTLA-4–Ig on MIP-1β expression was like that found on IL-10, leading to partial inhibition of transcript levels throughout most of the follow-up, with peak levels demonstrated on the same days as in the untreated group (Fig. 8). Discussion In the present study, we attempted to investigate the kinetics of the inflammatory response during the ongoing process of cellular rejection mediated by human PBMC in a human–rat chimeric model. In particular, we tested the effect of CTLA-4–Ig on patterns of gene products. Extensive studies of the immune system generated in the mouse or rat radiation chimeras following adoptive transfer of human PBMC have shown that human T and B cells form follicles in the spleen of the animals similar to germinal centers (27). In addition, we have demonstrated that human humoral and cytotoxic responses against viral or alloantigens can be induced by systemic immunization (25,26). However, data in the present study do not address the possibility of presentation of an allogeneic peptide, derived from the kidney tissue, by antigen-presenting cells administered in the PBMC inoculum (indirect pathway of recognition) nor the migration of dendritic cells from the human implant to the human follicles in the spleen. Since mechanisms of a vascularized rejection include complex trafficking of dendritic cells from the blood to the spleen, as well as a direct and indirect pathway in allorecognition (1), interpretation of our results must take into account that our model may not represent rejection of a vascularized kidney allograft. Our evaluation of CTLA-4–Ig's action demonstrated prolonged graft survival in conjunction with preservation of intragraft mononuclear cell infiltrates, a reduction of activation markers in the graft infiltrating cells and rapid down-regulation of cytokine transcription. We found a striking differential effect on two groups of molecules, which seem to differ in their cellular origin. Thus, while the expression of IL-2, IL-4 and FasL is markedly abrogated by CTLA-4–Ig treatment, the expression of IFN-γ, IL-10, RANTES and MIP-1β which could have also originated from non T-cells, especially monocytes/macrophages (13,39), was less affected by CTLA-4–Ig. Therefore, our results suggest that CTLA-4–Ig down-regulates the immune response by influencing both Th1 and Th2 T cells, rather than by a change of balance between these subpopulations (Th1 to Th2 shift) (40). In other model systems, long-term survival of cardiac allografts and xenografts following combined administration of CTLA-4–Ig and anti-CD40 ligand antibody or treatment with cyclosporine respectively has been shown to occur in the absence of a Th2 shift (41,42). Furthermore, Josien et al. (9) showed marked reduction in the accumulation of IL-2, IFN-γ, as well as IL-4 and IL-10 in heart allografts tolerized by donor-specific transfusion. In addition to cytokine modulation, it is likely that simultaneous reduction in chemokine and FasL transcript levels, as demonstrated in our model following the administration of CTLA-4–Ig, might also prove valuable for allograft survival. However, in our model, Th1 cytokine expression was regained and human renal grafts were ultimately rejected. Such predominant production of Th1 cytokines in acute rejection episodes of human kidney allografts has been previously demonstrated (8). Since a true state of tolerance was not attained, it seems that the human T cell infiltrates were initially at a `resting' or unstimulated state, induced by CTLA-4–Ig, and were turned on with time. More detailed studies of the T cells recovered from the subcapsular site are needed to define their functional status. Our findings of advantageous but transient effects of CTLA-4–Ig are in accordance with those of Kirk et al. who studied renal allograft rejection in primates (43). Other studies showed that administration of only a single dose of CTLA-4–Ig induced donor-specific tolerance of MHC-incompatible rat renal allografts and vascularized cardiac allografts (in the latter CTLA-4–Ig was given 2 days after donor splenocytes) (40,44). Differences in the experimental approaches and also issues of optimal dosing and treatment time course of CTLA-4–Ig could be related to the varying results. In this regard the temporal expression of B7 and its ligands CD28 and CTLA-4 must be considered. It has been reported that within hours after activation, B7-2 is expressed by antigen-presenting cells and is available to bind to CD28, transmitting a co-stimulatory signal to the T cell. Only by 2–3 days after activation, antigen-presenting cells also express B7-1, whereas T cells express the CTLA-4 inhibitory receptor. Both B7-1 and B7-2 can bind to either CD28 or CTLA-4, providing continued co-stimulation or a new inhibitory signal respectively. CTLA-4–Ig can compete with CD28 and CTLA-4 for B7 binding, thus preventing co-stimulatory interactions (both the activating signal and the inhibitory one) (45). Delaying the administration of CTLA-4–Ig would allow B7 molecules time to bind to CTLA-4 (negative signal), which may be required for the induction of tolerance (46). However, delaying the administration of CTLA-4–Ig until the onset of acute rejection would not induce prolonged graft survival (47). This suggests that failure to induce tolerance in our model, despite the use of a more aggressive treatment protocol of CTLA-4–Ig, might also be related to the timing of CTLA-4–Ig administration. A second issue raised by our study is the relation between the magnitude of cellular infiltration and mRNA accumulation in the rejection process. In our analysis we found that maximal IL-2 and IFN-γ transcript levels are associated with the period preceding the arrival of massive human cellular infiltrates and histologic destruction of the human implant (induction phase). Thus determination of IL-2 and IFN-γ gene expression in implants obtained 3 weeks after infusion of human PBMC would have revealed reduced transcript levels of the Th1-type cytokines, and underscore their significance in the initiation and induction of acute cellular rejection, as noted in some studies using other experimental approaches (4,48). Furthermore, our findings might explain the lack of correlation between significantly heightened intragraft IL-2 and IFN-γ gene expression and clinicopathological acute rejection seen in human kidney biopsies (49). Previously, it was proposed that during allograft rejection, chemokines play a pivotal role in promoting the recruitment of the effector T cells into the allograft and mediate rejection (14,16). In our model, two stages of RANTES and MIP-1β expression were noted. (i) Early expression, induced before the arrival of human infiltrating cells (also partly observed in the human renal grafts obtained from rats that did not receive human PBMC) and likely associated with the inflammation caused by the experimental procedure. Interestingly, such early expression of RANTES was not found in a mouse skin allograft model (16) and it is possible that the production of RANTES in the kidney parenchyme itself, i.e. in the renal tubular epithelial cells (13), might have accounted for this discrepancy. Clearly, this early expression is most likely responsible for the early trafficking of human cells into the human renal graft. (ii) Heightened late expression during the induction phase of the rejection process (second week). This peak was found to be closely associated with the pattern of the Th1 cytokine expression, and was also accompanied by a short and significant transcription burst of the β chemokine receptor, CCR5. At this stage, the relation of RANTES gene expression, which is remarkably up-regulated, to the infiltration and accumulation of human cells in the graft, is much more obvious than that of MIP-1β. Thus, RANTES which was shown to be expressed in infiltrating allospecific activated T cells (13,50) could likely lead to further recruitment of effector cells and to their consequent amplification upon activation, culminating in the massive infiltrates and histologic destruction found during the third week following the infusion of human PBMC. In conclusion, our data confirm the beneficial role of CTLA-4–Ig in ameliorating a cellular rejection process but also clearly indicate that the ability of CTLA-4–Ig to prevent the vigorous immune response to human renal grafts by allogeneic human PBMC is only transitory and occurs in the absence of a Th2 shift. Table 1. Cellular infiltration patterns into the human renal transplants following i.p. infusion of allogeneic human PBMC Histology  Untreated animals  Animals treated with CTLA-4–Ig    4d  7d  10d  13d  16d  19d  22d  4d  7d  10d  13d  16d  19d  22d    Minor interstitial infiltrate  Focal and patchy interstitial infiltrate  Dense and diffuse infiltrate tissue destruction  Minor interstitial infiltrates  Focal and patchy interstitial infiltrates  Marked infiltrate graft tissue spared  aReactivity of human graft infiltrating cells to various anti-human mAb was semiquantatively graded: 0–1+, sparse infiltrate, focal distribution; 1–2+, dense infiltrate, focal distribution or sparse infiltrate, diffuse distribution; 2–3+, dense infiltrate, diffuse distribution. At least three grafts were analyzed in each time point. `neg', negative staining.  bCD20 was not detected at all time points.  cStaining for apoptosis in graft infiltrating cells was semiquantatively graded as stated above.  CD3b  0–1+a  0–1+  1–2+  1–2+  2–3+  2–3+  2–3+  0–1+  0–1+  1–2+  1–2+  1–2+  2–3+  2–3+  CD14  0–1+  1–2+  1–2+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  0–1+  0–1+  0–1+  CD56  0–1+  0–1+  neg  neg  neg  neg  neg  0–1+  0–1+  neg  neg  neg  neg  neg  HLA-DR  0–1+  0–1+  0–1+  1–2+  2–3+  2–3+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  ICAM-I  0–1+  0–1+  0–1+  1–2+  2–3+  2–3+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  Apoptotic cellsc  0–1+  0–1+  1–2+  1–2+  1–2+  2–3+  2–3+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  1–2+  Histology  Untreated animals  Animals treated with CTLA-4–Ig    4d  7d  10d  13d  16d  19d  22d  4d  7d  10d  13d  16d  19d  22d    Minor interstitial infiltrate  Focal and patchy interstitial infiltrate  Dense and diffuse infiltrate tissue destruction  Minor interstitial infiltrates  Focal and patchy interstitial infiltrates  Marked infiltrate graft tissue spared  aReactivity of human graft infiltrating cells to various anti-human mAb was semiquantatively graded: 0–1+, sparse infiltrate, focal distribution; 1–2+, dense infiltrate, focal distribution or sparse infiltrate, diffuse distribution; 2–3+, dense infiltrate, diffuse distribution. At least three grafts were analyzed in each time point. `neg', negative staining.  bCD20 was not detected at all time points.  cStaining for apoptosis in graft infiltrating cells was semiquantatively graded as stated above.  CD3b  0–1+a  0–1+  1–2+  1–2+  2–3+  2–3+  2–3+  0–1+  0–1+  1–2+  1–2+  1–2+  2–3+  2–3+  CD14  0–1+  1–2+  1–2+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  0–1+  0–1+  0–1+  CD56  0–1+  0–1+  neg  neg  neg  neg  neg  0–1+  0–1+  neg  neg  neg  neg  neg  HLA-DR  0–1+  0–1+  0–1+  1–2+  2–3+  2–3+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  ICAM-I  0–1+  0–1+  0–1+  1–2+  2–3+  2–3+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  Apoptotic cellsc  0–1+  0–1+  1–2+  1–2+  1–2+  2–3+  2–3+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  1–2+  View Large Fig. 1. View largeDownload slide Inflammatory cell infiltration into the human renal grafts after infusion of human PBMC into recipient rats. (A) Immunostaining for human CD14 at 7 days (×100) and (B) at 16 days, note minimal staining (×100), inset, cytoplasmic staining of few monocytes, small arrows (×400). (C) Immunostaining for human HLA-DR at 13 days (×100) and (D) at 19 days (×100), note negative staining in rat kidney. (E) In situ detection of apoptotic cells in untreated grafts at three weeks, note diffuse staining (×40), and (F) in CTLA-4–Ig-treated grafts, note reduced staining and preserved graft architecture (×40). All arrows indicate border between human grafts and rat kidney parenchyme. Fig. 1. View largeDownload slide Inflammatory cell infiltration into the human renal grafts after infusion of human PBMC into recipient rats. (A) Immunostaining for human CD14 at 7 days (×100) and (B) at 16 days, note minimal staining (×100), inset, cytoplasmic staining of few monocytes, small arrows (×400). (C) Immunostaining for human HLA-DR at 13 days (×100) and (D) at 19 days (×100), note negative staining in rat kidney. (E) In situ detection of apoptotic cells in untreated grafts at three weeks, note diffuse staining (×40), and (F) in CTLA-4–Ig-treated grafts, note reduced staining and preserved graft architecture (×40). All arrows indicate border between human grafts and rat kidney parenchyme. Fig. 2. View largeDownload slide Infiltration of human T cells into the human renal graft 1 (top), 2 (middle) and 3 weeks (bottom) after infusion of human PBMC into recipient rats. Cells were double-stained by CD3–PE (T cells) and CD45–PerCP (pan-human leukocyte antigen, denoting human origin of cells) and analyzed in a lymphogate. The percentage of double-positive cells out of the total cells is presented in each panel. Fig. 2. View largeDownload slide Infiltration of human T cells into the human renal graft 1 (top), 2 (middle) and 3 weeks (bottom) after infusion of human PBMC into recipient rats. Cells were double-stained by CD3–PE (T cells) and CD45–PerCP (pan-human leukocyte antigen, denoting human origin of cells) and analyzed in a lymphogate. The percentage of double-positive cells out of the total cells is presented in each panel. Fig. 3. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-2 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-2 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 3. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-2 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-2 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 4. View largeDownload slide Quantification of intragraft expression of mRNA encoding IFN-γ following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IFN-γ mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 4. View largeDownload slide Quantification of intragraft expression of mRNA encoding IFN-γ following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IFN-γ mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 5. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-4 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-4 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 5. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-4 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-4 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 6. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-10 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-10 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 6. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-10 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-10 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 7. View largeDownload slide Quantification of intragraft expression of mRNA encoding RANTES following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of RANTES mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 7. View largeDownload slide Quantification of intragraft expression of mRNA encoding RANTES following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of RANTES mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 8. View largeDownload slide Quantification of intragraft expression of mRNA encoding MIP1β following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of MIP1β mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 8. View largeDownload slide Quantification of intragraft expression of mRNA encoding MIP1β following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of MIP1β mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 9. View largeDownload slide Quantification of intragraft expression of mRNA encoding CCR5 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of CCR5 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 9. View largeDownload slide Quantification of intragraft expression of mRNA encoding CCR5 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of CCR5 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 10. View largeDownload slide Quantification of intragraft expression of mRNA encoding FasL following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of FasL mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 10. View largeDownload slide Quantification of intragraft expression of mRNA encoding FasL following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of FasL mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Transmitting editor: A. Fischer This study was partially supported by a grant from XTL Technologies Ltd. References 1 Cuturi, M. C., Blancho, G., Josien, R. and Soulillou, J. P. 1994. The biology of allograft rejection. Curr. Opin. Nephrol. Hypertens.  3: 578. Google Scholar 2 Pavlakis, M., Lipman, M. and Strom, T. 1996. Intragraft expression of T-cell activation genes in human renal allograft rejection. Kidney Int.  49:S. Google Scholar 3 Bugeon, L., Cuturi, M. C., Hallet, M. M., Paineau, J., Chabannes, D. and Soulillou, J. P. 1992. Peripheral tolerance of an allograft in adult rats—chracterization by low interleukin-2 and interferon-gamma mRNA levels and by strong accumulation of major histocompatibility complex transcripts in the graft. Transplantation  54: 219. Google Scholar 4 Krams, S. M., Falco, D. A., Villanueva, J. C., et al. 1992. Cytokine and T cell receptor gene expression at the site of allograft rejection. Transplantation  53: 151. Google Scholar 5 Nast, C. C., Zuo, X. J., Prehin, J., Danovitch, G. B., Wilkinson, A. and Jordan, S. C. 1994. Gamma-interferon gene expression in human renal allograft fine-needle aspirates. Transplantation  57: 498. Google Scholar 6 Farges, O., Morris, P. J. and Dallman, M. J. 1995. Spontaneous acceptance of rat liver allografts is associated with downregulation of intragraft IL-4 mRNA expression. Hepatology  21: 767. Google Scholar 7 Chan, S. Y., DeBruyne, L. S., Goodman, R. E., Eichwald, E. J. and Bishop, K. D. 1995. In vivo depletion of CD8+ T cells results in Th2 cytokine production and alternate mechanisms of allograft rejection. Transplantation  59: 1155. Google Scholar 8 D'Elios, M. M., Josien, R., Manghetti, M., Amedei, A., de-Carli, M., Cuturi, M. C., Blanch, G., Buzelin, F., del-Prete, G. and Soulillou, J. P. 1997. Predominant Th1 cell infiltration in acute rejection episodes of human kidney grafts. Kidney Int.  51: 1876. Google Scholar 9 Josien, R., Pannetier, C., Douillard, P., Cantarovich, D., Menoret, S., Bugeon, L., Kourilsky, P., Soulillou, J. P. and Cuturi, M. C. 1995. Graft-infiltrating T helper cells, CD45RC phenotype, and Th1/Th2 related cytokines in donor-specific transfusion-induced tolerance in adult rats. Transplantation  60: 1131. Google Scholar 10 Dallman, M. J. 1995. Cytokines and transplantation: Th1/Th2 regulation of the immune response to solid organ transplants in the adult. Curr. Opin. Immunol.  7: 632. Google Scholar 11 Strom, T. B., Roy-Chaudhury, P., Manfro, R., Zheng, X., Nickerson, P. W., Wood, K. and Bushell, A. 1996. The Th1/Th2 paradigm and the allograft response. Curr. Opin. Immunol.  8: 688. Google Scholar 12 Piccotti, J. R., Chan, S. Y., VanBuskirk, A. M., Eichwald, E. J. and Bishop, K. D. 1997. Are Th2 helper lymphocytes beneficial, deleterious, or irrelevant in promoting allograft survival? Transplantation  63: 619. Google Scholar 13 Pattison, J., Nelson, P. J., Huie, P., von ??, L. I., Farshid, G., Sibley, R. K. and Krensky, A. M. 1994. RANTES chemokine expression in cell-mediated transplant rejection of the kidney. Lancet  343: 209. Google Scholar 14 Nadeau, K. C., Azuma, H. and Tinley, N. L. 1995. Sequential cytokine dynamics in chronic rejection of rat renal allografts: roles for cytokines RANTES and MCP-1. Proc. Natl Acad. Sci. USA  92: 8729. Google Scholar 15 Adams, D. H., Hubscher, S., Fear, J., Johnston, J., Shaw, S. and Afford, S. 1996. Hepatic expression of macrophage inflammatory protein-1α and macrophage inflammatory protein-1β after liver transplantation. Transplantation  61: 817. Google Scholar 16 Kondo, T., Novick, A. C., Toma, H. and Fairchild, R. L. 1996. Induction of chemokine gene expression during allogeneic skin graft rejection. Transplantation  61: 1750. Google Scholar 17 Nagano, H., Nadeau, K. C., Takada, M., Kusaka, M. and Tinley, N. L. 1997. Sequential cellular and molecular kinetics in acutely rejecting renal allografts in rats. Transplantation  63: 1101. Google Scholar 18 Krams, S. M., Egawa, H., Quinn, M. F., Villaneuva, J. C., Garcia-Kennedy, R. and Martinez, O. M. 1995. Apoptosis as a mechanism of cell death in liver allograft rejection. Transplantation  59: 621. Google Scholar 19 Sharma, V. K., Bologa, R. M., Li, B., et al. 1996. Molecular executors of cell death-Differential intrarenal expression of Fas ligand, Fas, granzyme B, and perforin during acute and/or chronic rejection of human renal allografts. Transplantation  62: 1860. Google Scholar 20 Josien, R. Muschen, M., Gilbert, E., Douillard, P., Heslan. J. M., Soulillou, J. P. and Cuturi, M. C. 1998. Fas ligand, tumor necrosis factor-alpha expression, and apoptosis during allograft rejection and tolerance. Transplantation  66: 887. Google Scholar 21 Dekel, B., Burakova, T., Marcus, H., Shezen, E., Pollack, S., Cannan, A., Passwell, J. H. and Reisner, Y. 1997. Engraftment of human kidney tissue in rat radiation chimera: I. a new model of human kidney allograft rejection. Transplantation  64: 1541. Google Scholar 22 Dekel, B., Burakova, T., Ben-Hur, H., Marcus, H., Oren, R., Laufer, J., Reisner, Y. 1997. Engraftment of human kidney tissue in rat radiation chimera: II. Human fetal kidneys display reduced immunogenecity to adoptively transferred human peripheral blood mononuclear cells and exhibit rapid growth and development. Transplantation  64: 1550. Google Scholar 23 Lubin, I., Segall, H., Marcus, H., David, M., Kulova, L., Steinitz, M., Erlich, P., Gan, J. and Reisner, Y. 1994. Engraftment of human peripheral blood lymphocytes in normal strains of mice. Blood  83: 2368. Google Scholar 24 Lubin, I., Segall, H., Erlich, P., David, M., Marcus, H., Fire, G., Burakova, T., Kulova, L. and Reisner, Y. 1995. Conversion of normal rats into SCID-like animals by means of bone marrow transplantation from SCID murine donor allows engraftment of human peripheral blood lymphocytes. Transplantation  60: 740. Google Scholar 25 Marcus, H., David, M., Canaan, A., Lubin, I., Segall, H. and Reisner, Y. 1995. Human/mouse radiation chimera are capable of mounting a human primary humoral response. Blood  86: 398. Google Scholar 26 Segall, H., Lubin, I., Marcus, H., Canaan, A. and Reisner, Y. 1996. Generation of primary antigen specific human CTLs in human/mouse radiation chimera. Blood  88: 721. Google Scholar 27 Burakova, T., Marcus, H., David, M., Dekel, B., Shezen, E., Canaan, A., Lubin, I., Segall, H. and Reisner, Y. 1997. Engrafted human T and B lymphocytes form mixed follicles in lymphoid organs of human/mouse and human/rat radiation chimera. Transplantation  63: 1166. Google Scholar 28 Premack, B. A. and Schall, T. J. 1996. Chemokine receptors: gateways to inflammation and infection. Nat. Med.  2: 1174. Google Scholar 29 Raport, C. J., Gosling, J., Schweickart, V. L., Gray, P. W. and Charo, I. F. 1996. Molecular cloning and functional characterization of a novel human CC chemokine receptor (CCR5) for RANTES, MIP-1beta, and MIP-1alpha. J. Biol. Chem.  271: 17161. Google Scholar 30 Levite, M., Zinger, H., Zisman, E., Reisner, Y. and Mozes, E. 1995. Beneficial effects of bone marrow transplantation on the serological manifestations and kidney pathology of experimental systemic lupus erythematosus. Cell. Immunol.  162: 138. Google Scholar 31 Lane, P., Gerhard, W., Hubele, S. A. and McConnell, F. 1993. Expression and functional properties of mouse B7/BB1 using a fusion protein between mouse CTLA4 and human γ 1. Immunology  80: 56. Google Scholar 32 Kirk, A. D., Bollinger, R. R. and Finn, O. J. 1995. Rapid, comprehensive analysis of human cytokine mRNA and its application to the study of acute renal allograft rejection. Hum. Immunol.  43: 113. Google Scholar 33 Robinson, E., Keystone, E. C., Schall, T. J., Gillett, N. and Fish, E. N. 1995. Chemokine expression in rheumatoid arthritis (RA): evidence of RANTES and macrophage inflammatory protein (MIP)-1β production by synovial T cells. Clin. Exp. Immunol.  101: 398. Google Scholar 34 Giordano, C., Stassi, G., De-Maria, R., et al. 1997. Potential involvement of Fas and its ligand in the pathogenesis of Hashimoto's thyroiditis. Science  275: 960. Google Scholar 35 Deng, H., Liu, R., Ellmeier, W., et al. 1996. Identification of a major co-receptor for primary isolates of HIV-1. Nature.  381: 661. Google Scholar 36 Rabinovitch, A., Suarez-Pinzon, W. L., Sorensen, O., Bleackley, R. C., Power, R. F. and Rajotte, R. V. 1995. Combined therapy with interleukin-4 and interleukin-10 inhibits autoimmune diabetes recurrence in syngeneic islet-transplanted nonobese diabetic mice. Transplantation  60: 368. Google Scholar 37 Baur, F. M., Brenner, B., Goetze-Speer, B., Neu, S. and Speer, C. P. 1998. Natural porcine surfactant down-regulates mRNA of TNF-alpha and TNF-alpha type II receptor in lipopolysaccharide-stimulated monocytes. Pediatr. Res.  44: 32. Google Scholar 38 Kennedy, M. K., Torrance, D. S., Picha, K. S. and Mohler, K. M. 1992. Analysis of cytokine mRNA expression in central nervous system of mice with experimental autoimmune encephalomyelitis reveals that IL-10 mRNA expression correlates with recovery. J. Immunol.  149: 2496. Google Scholar 39 Blancho, G., Gianello, P., Germana, S., Baetscher, M., Sachs, D. H. and LeGuern, C. 1995. Molecular identification of porcine interleukin 10: regulation of expression in a kidney allograft model. Proc. Natl Acad. Sci. USA  92: 2800. Google Scholar 40 Sayegh, M. H., Akalin, E., Hancock, W. W., Russell, M. E., Carpenter, C. B., Linsley, P. S. and Turka, L. A. 1995. CD28–B7 blockade after alloantigenic challenge in vivo inhibits Th1 cytokines but spares Th2. J. Exp. Med.  181: 1869. Google Scholar 41 Larsen, C. P., Elwood, E. T., Alexander, D. Z., et al. 1996. Long-term acceptance of skin and cardiac allografts after blocking CD40 and CD28 pathways. Nature.  381: 434. Google Scholar 42 Brouard, S., Blancho, G., Morreau, A., Heslan, J. M., Cuturi, M. C. and Soulillou, J. P. 1998. Long-term survival of hamster-to-rat cardiac xenografts in the absence of a Th2 shift. Transplantation  65: 1555. Google Scholar 43 Kirk, A. D., Harlan, D. M., Armstrong, A. A., et al. 1997. CTLA4–Ig and anti-CD40 ligand prevent renal allograft rejection in primates. Proc. Natl Acad. Sci. USA  94: 8789. Google Scholar 44 Lin, H., Bolling, S. F., Linsley, P. S., et al. 1993. Long-term acceptance of major histocompatibility complex mismatched cardiac allografts induced by CTLA4–Ig plus donor-specific transfusion. J. Exp. Med.  178: 1801. Google Scholar 45 Sayegh, M. H. and Turka, L. A. 1998. The role of T-cell costimulatory activation pathways in transplant rejection. N. Engl. J. Med.  338: 1813. Google Scholar 46 Perez, V. L., Van Parijs, L., Biuckians, A., Zheng, X., Strom, T. B. and Abass, A. K. 1997. Induction of peripheral T cell tolerance in vivo requires CTLA-4 engagement. Immunity  6: 411. Google Scholar 47 Perico N., Amuchastegui, S., Bontempelli, M. and Remuzzi, G. 1996. CTLA-4–Ig alone or in combination with low-dose cyclosporine fails to reverse acute rejection of renal allograft in the rat. Transplantation  61: 1320. Google Scholar 48 Lipman, M. L., Stevens, A. C., Bleakley, R. C., et al. 1992. The strong correlation of cytotoxic T-lymphocyte-specific serine protease gene transcripts with renal allograft rejection. Transplantation  53: 73. Google Scholar 49 Strehlau, J., Palvakis, M., Lipman, M., et al. 1997. Quantitative detection of immune activation transcripts as a diagnostic tool in kidney transplantation. Proc. Natl Acad. Sci. USA  94: 695. Google Scholar 50 Schall, T. J., Bacon, K., Toy, K. J. and Goeddel, D. V. 1990. Selective attraction of monocytes and T lymphocytes of the memory phenotype by cytokine RANTES. Nature  347: 669. Google Scholar © 1999 Japanese Society for Immunology http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Immunology Oxford University Press

Acute cellular rejection of human renal tissue by adoptive transfer of allogeneic human peripheral blood mononuclear cells into chimeric rats: sequential gene expression of cytokines, chemokines and cytolytic effector molecules, and their regulation by CTLA-4–Ig

Download PDF
11 pages

Loading next page...
 
/lp/oxford-university-press/acute-cellular-rejection-of-human-renal-tissue-by-adoptive-transfer-of-r9CDSG5rcb

References (66)

Publisher
Oxford University Press
Copyright
© 1999 Japanese Society for Immunology
ISSN
0953-8178
eISSN
1460-2377
DOI
10.1093/intimm/11.10.1673
Publisher site
See Article on Publisher Site

Abstract

Abstract Th1- and Th2-related cytokines (IFN-γ, IL-2, IL-4, IL-10), β-chemokines (RANTES, macrophage inflammatory protein-1β) and their receptor [chemotatic cytokine receptor (CCR) 5], and the cytolytic effector molecule [Fas ligand (FasL)] play an essential role in regulating and co-ordinating acute renal allograft rejection. A chimeric model of acute cellular rejection which involves subcapsular grafting of human renal tissue in the kidneys of immunodeficient rats and subsequent i.p. infusion of allogeneic human peripheral blood mononuclear cells (PBMC) was used to study cellular infiltration patterns and sequential intragraft gene expression of these key inflammatory mediators. We found that while all molecules are expressed within the human renal implant at specific time points following infusion of allogeneic human PBMC, peak mRNA expression of IFN-γ, IL-2, RANTES and CCR5 is associated with a phase of human mononuclear infiltration and accumulation, prior to graft destruction (induction phase). A short burst of FasL gene expression is found at the end of induction and at the onset of graft deterioration. IL-4 mRNA, which is hardly detectable, and IL-10 mRNA, which appears early and persists throughout follow-up at high levels, both peak after the induction phase with the advent of graft destruction. Furthermore, treatment with CTLA-4–Ig, which hardly affects migration of human effector cells into graft tissue, is associated with a temporary reduction in gene transcript levels for all inflammatory mediators, especially IL-2 and IL-4, reduced apoptosis in the graft and amelioration of tissue injury. Thus, development of acute cellular rejection in our chimeric model involves a co-ordinated pattern of gene expression, in which CTLA-4–Ig promotes its effects by transient inactivation of infiltrating human cells. allograft rejection, animal model, apoptosis, RT-PCR, SCID, Th1, Th2 AEC 3-amino-9-ethylcarbazol, CCR chemotatic cytokine receptor, FasL Fas ligand, MIP macrophage inflammatory protein, PBMC peripheral blood mononuclear cells, PE phycoerythrin, SCID severe combined immunodeficiency, TBI total-body irradiation Introduction The mechanisms responsible for acute cellular rejection are primarily related to the development of alloantigen-specific cytotoxic T lymphocytes and delayed-type hypersensitivity responses, but also to non-immune mechanisms. Inflammatory mediators produced by T cells, macrophages and donor tissue have all been suggested to play an important role in the rejection process observed in murine allograft models and human transplants (1,2). There is abundant data on the expression and activity of different cytokines during acute cellular rejection in murine models and human allografts (3–9). Some of these studies advocate a paradigm of a polarized Th1/Th2 immune response in rejection and tolerance; others indicate the possible involvement of both populations in the immune response leading to graft rejection or in the induction of tolerance and prolonged graft survival (10–12). In addition, chemoattractant cytokines (chemokines), especially of the β chemokine family which are chemoattractants for monocytes and lymphocytes (13–17), and the cytolytic effector molecule, Fas ligand (FasL), have been recently identified during a rejection process (18–20). However, kinetic studies analyzing sequential intragraft expression of cytokines, chemokines and cytolytic effector molecules concurrently, their relationship to cellular infiltration, and the effect of experimental manipulation on them, are less abundant. We have recently developed a chimeric model of acute cellular rejection which involves subcapsular grafting of human renal tissue in the kidneys of SCID-like rats and subsequent i.p. infusion of allogeneic human peripheral blood mononuclear cells (PBMC) (21,22). The animals we use are normal strains of mice or rats which are converted to SCID-like with a lethal level of total-body irradiation (TBI) and radioprotection by transplantation of SCID mouse bone marrow cells (23,24). An important major difference between the SCID mouse model and the radiation chimera is associated with the engraftment rate following adoptive transfer of human PBMC, and consequently with the functionality of the engrafted human T cells (25,26). Following i.p. injection of human PBMC into SCID mice, the cells remain in the peritoneal cavity in small numbers for >30 days before expanding. In the radiation chimera recipients, the rate of appearance of human cells is greatly enhanced and engraftment, as well as formation of human follicles by T and B cells in the spleen, takes place within a few days post-transplant (23–27). The engraftment of human cells is probably improved by the lethal conditioning employed and by increasing the inoculum of human cells (300–400×106 per rat). This strong and rapid engraftment enabled us to study the rejection of the human renal grafts early after transplantation of allogeneic human PBMC. Intraperitoneal infusion of allogeneic human PBMC following kidney implantation, resulted in the presence of patchy mononuclear infiltrates in the human renal implant within 3 weeks and lead to rejection of the human renal tissue. In vivo treatment with CTLA-4–Ig fusion protein (murine CTLA-4 and human Ig) ameliorated tissue injury (21). The ability to collect from control and CTLA-4–Ig-treated chimeric rats a substantial amount of human kidney tissue, following the infusion of allogeneic human PBMC, has enabled us now to perform kinetic studies on cellular infiltration patterns and regulation of Th1 and Th2 cytokines (IFN-γ, IL-2, IL-4, IL-10), β chemokines [RANTES, macrophage inflammatory protein (MIP)-1β], the receptor for the β chemokines [chemotatic cytokine receptor (CCR) 5] (28,29), and the cytolytic effector molecule (FasL) during the different stages of the unmodified and modified cellular rejection process. Methods Experimental animals Three-month-old Lewis (from Harlan Olac, Bicester, UK) and inbred nude rats (from Harlan Sprague Dawley, Indianapolis, IN) were used. All rats were kept in small cages (one to three animals in each cage), and fed sterile food and acid water containing ciprofloxacin (20 mg/ml). Bone marrow donors were NOD.SCID mice from the Weizmann Institute Animal Breeding Center (Rehovot, Israel). Conditioning regimen Lewis rats were exposed to split dose (4 Gy followed 3 days later by 11.5 Gy) TBI, from a γ beam 150-A 60Co source (produced by the Atomic Energy of Canada, Kanata, Ontario, Canada) with a focal skin distance of 75 cm and a dose rate of 0.7 Gy/min. Nude rats were exposed to a split-dose TBI of 4 + 9.5 Gy (13). Preparation and transplantation of SCID bone marrow cells Bone marrow cells were collected into PBS by flushing the shafts of the femur and tibia obtained from SCID mice (4–8 weeks old) according to Levite et al. (30). Recipient rats were injected with 50×106 SCID bone marrow cells (i.v. in 1.0 ml PBS) 1 day after the second fraction of TBI. Human renal tissue Human kidney segments were obtained from nephrectomy specimens performed for primary localized kidney tumors. Tangential slices were recovered from the cortex of the removed kidney. In all cases, the non-tumor renal tissue used for transplantation had no histopathologic changes, as confirmed by hematoxylin & eosin and reticulin staining. The human kidney fragments were kept in sterile conditions at 4°C (for ~2 h) in either RPMI 1640 or Dulbecco's modified Eagle's medium containing 10% FCS (Biological Industries, Bett HaEmek, Israel). Transplantation of the human kidney tissue was done under general anesthesia (2.5% Avertin in PBS, 10 ml/kg i.p.). Both kidneys were exposed through a bilateral incision. A 1.5 mm incision was made at the caudal end of the kidney capsule and a 1 mm3 fragment of human kidney was implanted under each capsule. Transplanted rats were treated post-operatively with ciprofloxacin in their drinking water for 7 days. The surgical procedure was performed 7–10 days after the animals had received the SCID bone marrow. Preparation and transplantation of human PBMC Buffy coats from normal volunteers were layered onto Lymphoprep solution (Nycomed, Oslo, Norway) and spun at 2000 r.p.m. for 20 min. The interface layer was collected, washed twice, counted and resuspended in PBS (pH 7.4) to the desired cell concentration. Human PBMC (400×106 cells) were injected i.p., 1–3 days after transplantation of the human renal tissue into recipient rats, conditioned as described above. Control rats did not receive human PBMC. Treatment of animals by CTLA-4–Ig in vivo Hybridoma producing a fusion protein between mouse CTLA-4 and human IgG1 (31) was kindly provided by Dr Stephan Yung (Departemnt of Immunology, Hadassa Medical School, Jerusalem, Israel). Hybridomas were injected i.p. into pristane-treated SCID mice and ascitic fluid was recovered. For some indicated studies, rats were injected i.p. with 0.5 mg CTLA-4–Ig fusion protein on 3–5 consecutive days, beginning at 3 days after the injection of human PBMC. Control rats either did not receive human PBMC or were injected with PBS following human PBMC administration. Cell collection from human → rat chimera Peritoneal cells were obtained by lavage with 5–10 ml of PBS. Rat kidneys were removed after the animals were killed by cervical dislocation and renal grafts were carefully separated from the host's kidney. Pieces of organs and the human grafts were pressed through stainless steel sieves to make a cell suspension in PBS, and cells were then isolated using Lymphoprep, washed with PBS containing 1% BSA and stained for analysis on a FACS. FACS analysis Single-cell suspensions were incubated for 30 min on ice with labeled anti-human CD3–phycoerythrin (PE) and CD45–PerCP (pan-human leukocyte antigen) antibodies (Becton Dickinson, Mountain View, CA). After washing, two- or three-color fluorescent analysis of human antigens was performed using a FACScan analyzer (Becton Dickinson). The lymphocytes were gated according to their size on forward and side scatters. Immunohistochemistry Paraffin tissue blocks were cut 4–6 μm thick, deparaffinized in xylene, rehydrated and placed for 15 min in alcohol with 3% H2O2 to block endogenous peroxidase. Slides were thoroughly washed with tap water and transferred to PBS. Sections were then treated with 1% BSA to prevent background staining and incubated for 1 h with primary mouse anti-albumin antibody (Zymed, San Francisco, CA) at room temperature in a humidified chamber. Slides were rinsed with PBS for 3 min and incubated with the biotinylated linked antibody for 30 min and with the labeling reagent, peroxidase-conjugated streptavidin, for 30 min (StrAvigen; Biogenex, San Ramon, CA). After rinsing, the peroxidase label was visualized using 3-amino-9-ethylcarbazol (AEC), following processing instructions according to the immunohistochemical kit (Biomeda, Foster City, CA), for 15 min and counterstained with Mayer's hematoxylin. AEC produced a red product that is soluble in alcohol and can be used with an aqueous mounting medium (Kaiser's glycerol gelatin). A negative control was run using the same technique but omitting the primary antibody and adding the streptavidin–biotin complex. Following infusion of allogeneic PBMC into transplanted rats, staining of the human graft was done using rabbit anti-human CD3 (pan-T cell), mouse anti-human HLA-DR (MHC class II molecule expressed on activated T cells), mouse anti-human CD14 (monocytes), mouse anti-human CD68 (macrophages) and mouse anti-human CD56 (NK) (all from Zymed); mouse anti-human CD20 (pan-B cell) (Dakopatts, Glostrup, Denmark); and mouse anti-human CD54 (ICAM-I) (Serotec, Oxford, UK). In situ detection of apoptosis in human grafts Paraffin tissue blocks were cut 4–6 μm thick, deparaffinized in xylene, rehydrated and placed for 15 min in alcohol with 3% H2O2 to block endogenous peroxidase. Slides were thoroughly washed with tap water and transferred to PBS. Staining for apoptosis was then preformed according to the Apoptag in situ apoptosis detection kit (Oncor, Gaithersburg, MD). This kit uses the C-terminal deoxynucleotidyl TUNEL technique of using TdT to tail all fragmented DNA with digoxigenin-dUTP. The tails are labeled with an anti-digoxigenin antibody complexed to an alkaline phosphatase reporter enzyme, and the sites are visualized with nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate dyes. Molecular studies of the human renal implants Human renal tissue was dissected carefully from the subcapsular site and homogenized with a glass-Teflon tissue tearer in Tri-reagent (Molecular Research Center, Cincinnati, OH). Total RNA extraction was done using the guanidinium isothiocyanate–phenol isolation method. The isolated total RNA was air-dried, resuspended in nuclease-free water and quantified by spectrophotometry. Then 1 μg of total RNA was reverse-transcribed into cDNA using Avian myeloblastosis virus reverse transcriptase according to standard procedures. The reverse transcriptase reaction was diluted 1:50, 1:100 and 1:500 in sterile H2O, and DNA amplification was done using thermostable Tfl DNA polymerase in a 50 μl reaction mixture containing 40 μM of each dNTP, 0.4 μM of each primer, 10 mM Tris–HCl (pH 8.3) and 1.5 mM MgCl2. In all experiments the presence of possible contamination was checked by control reactions in which (i) reverse transcriptase was omitted from the reaction mixture and (ii) buffer alone was added to the reverse transcription reaction mixture. All primer sequences were run through the National Center of Biological Information (NCBI) Gene Bank library to ensure that each primer was specific to its respective human gene but not to the rat gene. In addition, all of the primers were designed to flank an intron site in the target sequence. This primer design permits detection of genomic contamination of the RNA isolated from the human renal (32). Also, to avoid cross-hybridization with unrelated or less specific genes, the annealing temperature during thermal cycling was set high (64°C). The following sense and antisense primers were used: IL-2, 5′-ACAGCTACAACTGGAGCATT-3′ and 5′-TGCTGTCTCATCAGCATATT-3′; IFN-γ, 5′-TGCAGGTCATTCAGATGTAG-3′ and 5′-AGCCATCACTTGGATGAGTT-3′; IL-4, 5′-GCGATATCACCTTACAGGAG-3′ and 5′-TTGGCTTCCTTCACAGGACA-3′; IL-10, 5′-AGATCTCCGAGATGCCTTCA-3′ and 5′-ATTCTTCACCTGCTCCACGG-3′, all according to Kirk et al. (32); RANTES, 5′-ATGAAGGTCTCCGCGGCAGCCC-3′ and 5′-CTAGCTCATCTCCAAAGAGTTG-3′; MIP-1β, 5′-ACCATCAAGCTCTGCGTGACTG-3′ and 5′-GCAGGTCAGTTCAGTTCCAGGTC-3′ (33); FasL, 5′-CAAGTCCAACTCAAGGTCCATGCC-3′ and 5′-CAGAGAGAGCTCAGATACGTTTGA-3′ (34); CCR5, 5′-CTCGGATCCGGTGGAACAAGATGGATTAT-3′ and 5′-CTCGTCGACATGTGCACAACTCTGACTG-3′ (35). With the aim of detecting PCR signals in the linear phase of product amplification, 20–35 PCR cycles were carried out in a thermocycler. The PCR product was applied to a 1.5% agarose gel, resolved by electrophoresis, visualized by ethidium bromide staining and photographed. Scanning by an Imaging Computing Laser Densitometer GS-670 (BioRad, Hercules, CA) of the negatives of the PCR products and analysis by Adobe Photoshop software (Adobe, Mountain View, CA) were used to determine the absolute absorbance of the PCR products. The densitometric values obtained for each cytokine PCR product were normalized as a percent of signal intensity of the β-actin PCR product, i.e. all PCR signals were standarized to the constitutive gene expression of β-actin included in each PCR determination (internal control) (36). Under the conditions used, there was a linear relationship between the amount of RNA in the tissue sample that was reverse-transcribed to cDNA and amplified by PCR, and the PCR product signal (36). Tissue samples to be compared were amplified in the same PCR run using reagents from a single master mix. Each sample was tested on at least three occasions. This assay is semiquantitative, i.e. it allows for the determination of the relative amount of mRNA for a given cytokine in different samples but does not allow for a precise determination of the amount of mRNA present in each sample. Similar semiquantitative PCR assays have been used to analyze cytokine mRNA expression in other conditions (19,33,37,38). Results Experimental design Groups of seven to 10 lethally irradiated Lewis rats radioprotected with SCID mouse bone marrow were transplanted with human renal tissue and were then infused with allogeneic human PBMC. Animals were left untreated or injected with the fusion protein CTLA-4–Ig, which blocks the CD28–B7 T cell co-stimulatory pathway. Some rats were implanted with human renal tissue but were not injected with human PBMC nor treated with CTLA-4–Ig. Human renal transplants were harvested at days 4, 7, 10, 13, 16, 19 and 22 following human PBMC infusion (at least three in each time point for each group). Grafts were processed for histologic and immunohistologic evaluation, and were either pooled and processed to recover human infiltrating cells or homogenized on an individual basis for RNA extraction. Inflammatory cell infiltration patterns Table 1 and Figs 1 and 2 summarize data regarding sequential analysis of the human cellular infiltrate. All of the human grafts developed a low-grade mononuclear infiltrate during the first 7 days following human PBMC infusion. There was no significant difference in the intensity of these early infiltrates between untreated grafts and those treated with CTLA-4–Ig. These early infiltrates contained a relative high number of monocytes (Table 1; Fig. 1A). After the development of the early infiltrate, the intensity of infiltration increased progressively, with the appearance of patchy mononuclear infiltrates surrounding human tubules during the second week, followed by a dense and diffuse interstitial mononuclear infiltrate and graft destruction by day 21 (Table 1). In addition, while human CD3+ and HLA-DR+ cells were found in the implant within the first week after the injection of human PBMC, intensification of these cells occurred during the second and third weeks (Table 1; Figs 1C, 1D and 2), culminating into the massive activated human T cell infiltrates previously described (21). Interestingly, after the first 13 days, hardly any monocytes were detected in the human grafts and the infiltrates were predominantly lymphocytic (Table 1; Fig. 1B). Our sequential histologic evaluation of the human renal grafts from CTLA-4–Ig-treated animals showed that the kinetics of appearance of human mononuclear infiltrates in these grafts was similar to that observed in the untreated ones (Table 1) and thus CTLA-4–Ig hardly altered the presence of dense interstitial infiltrates found during the third week (21). Furthermore, the nature of the human infiltrating cells did not change (Table 1). However, compared to the untreated transplants, in the CTLA-4–Ig treated grafts we found reduced staining for the activation markers HLA-DR and ICAM-I on leukocytes, as well as for apoptotic cells (Table 1; Figs 1E and F). Interestingly, reduced numbers of apoptotic cells were found not only in renal parenchymal structures but also within infiltrates. Overall, human renal tissue was less damaged at 3 weeks following infusion of human PBMC (histology was similar to that observed in grafts which were not challenged with PBMC and were not infiltrated at all) and graft rejection was delayed by several weeks (21). Kinetics of cytokine mRNA expression in human renal grafts subjected to allogeneic human PBMC and its regulation by CTLA-4–Ig By using repeated measurements for each time point and by normalizing each mRNA determination according to its co-expressed β-actin levels, we were able to define for each cytokine or chemokine the time point in which a relative peak of its expression is attained, as well as the inhibitory effect of CTLA-4–Ig on its expression pattern. Sequential molecular analysis of both Th1 (IFN-γ and IL-2) and Th2 (IL-4 and IL-10) type cytokines and their regulation by CTLA-4–Ig is shown in Figs 3–6. In each time point analysis of transcripts in grafts of animals that were not injected with PBMC nor treated with CTLA-4–Ig was used to determine basal mRNA level. As can be seen in Fig. 3, the expression of IL-2 started to increase during the first week following injection of human PBMC, peaked at 13 days, and then dropped sharply and remained at low plateau levels (days 16–22). A similar pattern was observed for IFN-γ (Fig. 4), with somewhat enhanced increase in transcripts level, peaking on day 10. Treatment with CTLA-4–Ig resulted in a differential effect, predominantly inhibiting expression of IL-2 and to lesser extent IFN-γ expression. Thus, IL-2 mRNA expression was completely abrogated during the second week following human PBMC infusion delaying peak expression to day 19, while IFN-γ expression was only partially inhibited (complete inhibition was not seen at any of the time points tested). IL-4 expression was hardly detectable throughout the experimental period. Nonetheless, a significant peak was reproducibly noted during the third week, decreasing to undetectable levels on day 22. Examination of the CTLA-4–Ig effect on IL-4 expression showed that the low but significant signal of IL-4 mRNA was completely inhibited by CTLA-4–Ig treatment (Fig. 5). IL-10 expression in the unmodified rejection process, as opposed to IL-4 but similar to IL-2 and IFN-γ, typically started early and persisted during the first 2 weeks. However, during the third week, as in the case of IL-4, peak expression of IL-10 was noted. Evaluation of IL-10 expression in the model shows a pattern compatible with partial inhibition by CTLA-4–Ig (Fig. 6). Thus, while the amplitude of IL-10 is reduced by CTLA-4–Ig on most of the time points during the experimental period, there is a substantial level of IL-10 mRNA, which does not seem to be affected. Kinetics of chemokine, chemokine receptor and cytolytic effector molecule mRNA expression in human renal grafts subjected to allogeneic human PBMC, and their regulation by CTLA-4–Ig In addition to testing cytokines, we examined expression of the chemokines, RANTES and MIP-1β, their β-chemokine receptor CCR5, and the cytolytic molecule FasL (Figs 7–10). As shown in Figs 7 and 8, the expression of RANTES and of MIP-1β displayed a similar pattern to that of IL-2 and IFN-γ. In contrast, expression of CCR5 and of FasL was displayed in short bursts. Thus CCR5 mRNA levels were noted in parallel to peak expression of RANTES on day 10 (Fig. 9), while FasL mRNA was found only on days 13–16 and rapidly declined to undetectable levels (Fig. 10). The effect of CTLA-4–Ig on RANTES, CCR5 and FasL was similar in that the peaks of expression were all delayed. There was a significant reduction in RANTES transcript levels, following an early peak on day 4, so that highest expression was noted on day 19, >1 week after peak levels were found in the untreated group. Nonetheless, constitutive expression of RANTES was noted in the treated group (Fig. 7). Moreover, early transcription bursts of CCR5 and FasL during the unmodified rejection process were abrogated in the treated group, with the effect on FasL being more prominent, as substantial levels of CCR5 were noted on both days 13 and 16 (Figs 9 and 10). The effect of CTLA-4–Ig on MIP-1β expression was like that found on IL-10, leading to partial inhibition of transcript levels throughout most of the follow-up, with peak levels demonstrated on the same days as in the untreated group (Fig. 8). Discussion In the present study, we attempted to investigate the kinetics of the inflammatory response during the ongoing process of cellular rejection mediated by human PBMC in a human–rat chimeric model. In particular, we tested the effect of CTLA-4–Ig on patterns of gene products. Extensive studies of the immune system generated in the mouse or rat radiation chimeras following adoptive transfer of human PBMC have shown that human T and B cells form follicles in the spleen of the animals similar to germinal centers (27). In addition, we have demonstrated that human humoral and cytotoxic responses against viral or alloantigens can be induced by systemic immunization (25,26). However, data in the present study do not address the possibility of presentation of an allogeneic peptide, derived from the kidney tissue, by antigen-presenting cells administered in the PBMC inoculum (indirect pathway of recognition) nor the migration of dendritic cells from the human implant to the human follicles in the spleen. Since mechanisms of a vascularized rejection include complex trafficking of dendritic cells from the blood to the spleen, as well as a direct and indirect pathway in allorecognition (1), interpretation of our results must take into account that our model may not represent rejection of a vascularized kidney allograft. Our evaluation of CTLA-4–Ig's action demonstrated prolonged graft survival in conjunction with preservation of intragraft mononuclear cell infiltrates, a reduction of activation markers in the graft infiltrating cells and rapid down-regulation of cytokine transcription. We found a striking differential effect on two groups of molecules, which seem to differ in their cellular origin. Thus, while the expression of IL-2, IL-4 and FasL is markedly abrogated by CTLA-4–Ig treatment, the expression of IFN-γ, IL-10, RANTES and MIP-1β which could have also originated from non T-cells, especially monocytes/macrophages (13,39), was less affected by CTLA-4–Ig. Therefore, our results suggest that CTLA-4–Ig down-regulates the immune response by influencing both Th1 and Th2 T cells, rather than by a change of balance between these subpopulations (Th1 to Th2 shift) (40). In other model systems, long-term survival of cardiac allografts and xenografts following combined administration of CTLA-4–Ig and anti-CD40 ligand antibody or treatment with cyclosporine respectively has been shown to occur in the absence of a Th2 shift (41,42). Furthermore, Josien et al. (9) showed marked reduction in the accumulation of IL-2, IFN-γ, as well as IL-4 and IL-10 in heart allografts tolerized by donor-specific transfusion. In addition to cytokine modulation, it is likely that simultaneous reduction in chemokine and FasL transcript levels, as demonstrated in our model following the administration of CTLA-4–Ig, might also prove valuable for allograft survival. However, in our model, Th1 cytokine expression was regained and human renal grafts were ultimately rejected. Such predominant production of Th1 cytokines in acute rejection episodes of human kidney allografts has been previously demonstrated (8). Since a true state of tolerance was not attained, it seems that the human T cell infiltrates were initially at a `resting' or unstimulated state, induced by CTLA-4–Ig, and were turned on with time. More detailed studies of the T cells recovered from the subcapsular site are needed to define their functional status. Our findings of advantageous but transient effects of CTLA-4–Ig are in accordance with those of Kirk et al. who studied renal allograft rejection in primates (43). Other studies showed that administration of only a single dose of CTLA-4–Ig induced donor-specific tolerance of MHC-incompatible rat renal allografts and vascularized cardiac allografts (in the latter CTLA-4–Ig was given 2 days after donor splenocytes) (40,44). Differences in the experimental approaches and also issues of optimal dosing and treatment time course of CTLA-4–Ig could be related to the varying results. In this regard the temporal expression of B7 and its ligands CD28 and CTLA-4 must be considered. It has been reported that within hours after activation, B7-2 is expressed by antigen-presenting cells and is available to bind to CD28, transmitting a co-stimulatory signal to the T cell. Only by 2–3 days after activation, antigen-presenting cells also express B7-1, whereas T cells express the CTLA-4 inhibitory receptor. Both B7-1 and B7-2 can bind to either CD28 or CTLA-4, providing continued co-stimulation or a new inhibitory signal respectively. CTLA-4–Ig can compete with CD28 and CTLA-4 for B7 binding, thus preventing co-stimulatory interactions (both the activating signal and the inhibitory one) (45). Delaying the administration of CTLA-4–Ig would allow B7 molecules time to bind to CTLA-4 (negative signal), which may be required for the induction of tolerance (46). However, delaying the administration of CTLA-4–Ig until the onset of acute rejection would not induce prolonged graft survival (47). This suggests that failure to induce tolerance in our model, despite the use of a more aggressive treatment protocol of CTLA-4–Ig, might also be related to the timing of CTLA-4–Ig administration. A second issue raised by our study is the relation between the magnitude of cellular infiltration and mRNA accumulation in the rejection process. In our analysis we found that maximal IL-2 and IFN-γ transcript levels are associated with the period preceding the arrival of massive human cellular infiltrates and histologic destruction of the human implant (induction phase). Thus determination of IL-2 and IFN-γ gene expression in implants obtained 3 weeks after infusion of human PBMC would have revealed reduced transcript levels of the Th1-type cytokines, and underscore their significance in the initiation and induction of acute cellular rejection, as noted in some studies using other experimental approaches (4,48). Furthermore, our findings might explain the lack of correlation between significantly heightened intragraft IL-2 and IFN-γ gene expression and clinicopathological acute rejection seen in human kidney biopsies (49). Previously, it was proposed that during allograft rejection, chemokines play a pivotal role in promoting the recruitment of the effector T cells into the allograft and mediate rejection (14,16). In our model, two stages of RANTES and MIP-1β expression were noted. (i) Early expression, induced before the arrival of human infiltrating cells (also partly observed in the human renal grafts obtained from rats that did not receive human PBMC) and likely associated with the inflammation caused by the experimental procedure. Interestingly, such early expression of RANTES was not found in a mouse skin allograft model (16) and it is possible that the production of RANTES in the kidney parenchyme itself, i.e. in the renal tubular epithelial cells (13), might have accounted for this discrepancy. Clearly, this early expression is most likely responsible for the early trafficking of human cells into the human renal graft. (ii) Heightened late expression during the induction phase of the rejection process (second week). This peak was found to be closely associated with the pattern of the Th1 cytokine expression, and was also accompanied by a short and significant transcription burst of the β chemokine receptor, CCR5. At this stage, the relation of RANTES gene expression, which is remarkably up-regulated, to the infiltration and accumulation of human cells in the graft, is much more obvious than that of MIP-1β. Thus, RANTES which was shown to be expressed in infiltrating allospecific activated T cells (13,50) could likely lead to further recruitment of effector cells and to their consequent amplification upon activation, culminating in the massive infiltrates and histologic destruction found during the third week following the infusion of human PBMC. In conclusion, our data confirm the beneficial role of CTLA-4–Ig in ameliorating a cellular rejection process but also clearly indicate that the ability of CTLA-4–Ig to prevent the vigorous immune response to human renal grafts by allogeneic human PBMC is only transitory and occurs in the absence of a Th2 shift. Table 1. Cellular infiltration patterns into the human renal transplants following i.p. infusion of allogeneic human PBMC Histology  Untreated animals  Animals treated with CTLA-4–Ig    4d  7d  10d  13d  16d  19d  22d  4d  7d  10d  13d  16d  19d  22d    Minor interstitial infiltrate  Focal and patchy interstitial infiltrate  Dense and diffuse infiltrate tissue destruction  Minor interstitial infiltrates  Focal and patchy interstitial infiltrates  Marked infiltrate graft tissue spared  aReactivity of human graft infiltrating cells to various anti-human mAb was semiquantatively graded: 0–1+, sparse infiltrate, focal distribution; 1–2+, dense infiltrate, focal distribution or sparse infiltrate, diffuse distribution; 2–3+, dense infiltrate, diffuse distribution. At least three grafts were analyzed in each time point. `neg', negative staining.  bCD20 was not detected at all time points.  cStaining for apoptosis in graft infiltrating cells was semiquantatively graded as stated above.  CD3b  0–1+a  0–1+  1–2+  1–2+  2–3+  2–3+  2–3+  0–1+  0–1+  1–2+  1–2+  1–2+  2–3+  2–3+  CD14  0–1+  1–2+  1–2+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  0–1+  0–1+  0–1+  CD56  0–1+  0–1+  neg  neg  neg  neg  neg  0–1+  0–1+  neg  neg  neg  neg  neg  HLA-DR  0–1+  0–1+  0–1+  1–2+  2–3+  2–3+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  ICAM-I  0–1+  0–1+  0–1+  1–2+  2–3+  2–3+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  Apoptotic cellsc  0–1+  0–1+  1–2+  1–2+  1–2+  2–3+  2–3+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  1–2+  Histology  Untreated animals  Animals treated with CTLA-4–Ig    4d  7d  10d  13d  16d  19d  22d  4d  7d  10d  13d  16d  19d  22d    Minor interstitial infiltrate  Focal and patchy interstitial infiltrate  Dense and diffuse infiltrate tissue destruction  Minor interstitial infiltrates  Focal and patchy interstitial infiltrates  Marked infiltrate graft tissue spared  aReactivity of human graft infiltrating cells to various anti-human mAb was semiquantatively graded: 0–1+, sparse infiltrate, focal distribution; 1–2+, dense infiltrate, focal distribution or sparse infiltrate, diffuse distribution; 2–3+, dense infiltrate, diffuse distribution. At least three grafts were analyzed in each time point. `neg', negative staining.  bCD20 was not detected at all time points.  cStaining for apoptosis in graft infiltrating cells was semiquantatively graded as stated above.  CD3b  0–1+a  0–1+  1–2+  1–2+  2–3+  2–3+  2–3+  0–1+  0–1+  1–2+  1–2+  1–2+  2–3+  2–3+  CD14  0–1+  1–2+  1–2+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  0–1+  0–1+  0–1+  CD56  0–1+  0–1+  neg  neg  neg  neg  neg  0–1+  0–1+  neg  neg  neg  neg  neg  HLA-DR  0–1+  0–1+  0–1+  1–2+  2–3+  2–3+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  ICAM-I  0–1+  0–1+  0–1+  1–2+  2–3+  2–3+  1–2+  0–1+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  Apoptotic cellsc  0–1+  0–1+  1–2+  1–2+  1–2+  2–3+  2–3+  0–1+  0–1+  0–1+  1–2+  1–2+  1–2+  1–2+  View Large Fig. 1. View largeDownload slide Inflammatory cell infiltration into the human renal grafts after infusion of human PBMC into recipient rats. (A) Immunostaining for human CD14 at 7 days (×100) and (B) at 16 days, note minimal staining (×100), inset, cytoplasmic staining of few monocytes, small arrows (×400). (C) Immunostaining for human HLA-DR at 13 days (×100) and (D) at 19 days (×100), note negative staining in rat kidney. (E) In situ detection of apoptotic cells in untreated grafts at three weeks, note diffuse staining (×40), and (F) in CTLA-4–Ig-treated grafts, note reduced staining and preserved graft architecture (×40). All arrows indicate border between human grafts and rat kidney parenchyme. Fig. 1. View largeDownload slide Inflammatory cell infiltration into the human renal grafts after infusion of human PBMC into recipient rats. (A) Immunostaining for human CD14 at 7 days (×100) and (B) at 16 days, note minimal staining (×100), inset, cytoplasmic staining of few monocytes, small arrows (×400). (C) Immunostaining for human HLA-DR at 13 days (×100) and (D) at 19 days (×100), note negative staining in rat kidney. (E) In situ detection of apoptotic cells in untreated grafts at three weeks, note diffuse staining (×40), and (F) in CTLA-4–Ig-treated grafts, note reduced staining and preserved graft architecture (×40). All arrows indicate border between human grafts and rat kidney parenchyme. Fig. 2. View largeDownload slide Infiltration of human T cells into the human renal graft 1 (top), 2 (middle) and 3 weeks (bottom) after infusion of human PBMC into recipient rats. Cells were double-stained by CD3–PE (T cells) and CD45–PerCP (pan-human leukocyte antigen, denoting human origin of cells) and analyzed in a lymphogate. The percentage of double-positive cells out of the total cells is presented in each panel. Fig. 2. View largeDownload slide Infiltration of human T cells into the human renal graft 1 (top), 2 (middle) and 3 weeks (bottom) after infusion of human PBMC into recipient rats. Cells were double-stained by CD3–PE (T cells) and CD45–PerCP (pan-human leukocyte antigen, denoting human origin of cells) and analyzed in a lymphogate. The percentage of double-positive cells out of the total cells is presented in each panel. Fig. 3. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-2 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-2 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 3. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-2 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-2 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 4. View largeDownload slide Quantification of intragraft expression of mRNA encoding IFN-γ following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IFN-γ mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 4. View largeDownload slide Quantification of intragraft expression of mRNA encoding IFN-γ following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IFN-γ mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 5. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-4 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-4 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 5. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-4 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-4 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 6. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-10 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-10 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 6. View largeDownload slide Quantification of intragraft expression of mRNA encoding IL-10 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of IL-10 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 7. View largeDownload slide Quantification of intragraft expression of mRNA encoding RANTES following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of RANTES mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 7. View largeDownload slide Quantification of intragraft expression of mRNA encoding RANTES following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of RANTES mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 8. View largeDownload slide Quantification of intragraft expression of mRNA encoding MIP1β following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of MIP1β mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 8. View largeDownload slide Quantification of intragraft expression of mRNA encoding MIP1β following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of MIP1β mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 9. View largeDownload slide Quantification of intragraft expression of mRNA encoding CCR5 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of CCR5 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 9. View largeDownload slide Quantification of intragraft expression of mRNA encoding CCR5 following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of CCR5 mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 10. View largeDownload slide Quantification of intragraft expression of mRNA encoding FasL following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of FasL mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Fig. 10. View largeDownload slide Quantification of intragraft expression of mRNA encoding FasL following infusion of allogeneic human PBMC (shaded bars), following infusion of allogeneic human PBMC and treatment with CTLA-4–Ig fusion protein (crossed bars), and in the absence of allogeneic human PBMC or treatment with CTLA-4–Ig (bright bars). The steady-state levels of FasL mRNA was measured by semiquantitative RT-PCR and normalized against the β-actin mRNA control. Transmitting editor: A. Fischer This study was partially supported by a grant from XTL Technologies Ltd. References 1 Cuturi, M. C., Blancho, G., Josien, R. and Soulillou, J. P. 1994. The biology of allograft rejection. Curr. Opin. Nephrol. Hypertens.  3: 578. Google Scholar 2 Pavlakis, M., Lipman, M. and Strom, T. 1996. Intragraft expression of T-cell activation genes in human renal allograft rejection. Kidney Int.  49:S. Google Scholar 3 Bugeon, L., Cuturi, M. C., Hallet, M. M., Paineau, J., Chabannes, D. and Soulillou, J. P. 1992. Peripheral tolerance of an allograft in adult rats—chracterization by low interleukin-2 and interferon-gamma mRNA levels and by strong accumulation of major histocompatibility complex transcripts in the graft. Transplantation  54: 219. Google Scholar 4 Krams, S. M., Falco, D. A., Villanueva, J. C., et al. 1992. Cytokine and T cell receptor gene expression at the site of allograft rejection. Transplantation  53: 151. Google Scholar 5 Nast, C. C., Zuo, X. J., Prehin, J., Danovitch, G. B., Wilkinson, A. and Jordan, S. C. 1994. Gamma-interferon gene expression in human renal allograft fine-needle aspirates. Transplantation  57: 498. Google Scholar 6 Farges, O., Morris, P. J. and Dallman, M. J. 1995. Spontaneous acceptance of rat liver allografts is associated with downregulation of intragraft IL-4 mRNA expression. Hepatology  21: 767. Google Scholar 7 Chan, S. Y., DeBruyne, L. S., Goodman, R. E., Eichwald, E. J. and Bishop, K. D. 1995. In vivo depletion of CD8+ T cells results in Th2 cytokine production and alternate mechanisms of allograft rejection. Transplantation  59: 1155. Google Scholar 8 D'Elios, M. M., Josien, R., Manghetti, M., Amedei, A., de-Carli, M., Cuturi, M. C., Blanch, G., Buzelin, F., del-Prete, G. and Soulillou, J. P. 1997. Predominant Th1 cell infiltration in acute rejection episodes of human kidney grafts. Kidney Int.  51: 1876. Google Scholar 9 Josien, R., Pannetier, C., Douillard, P., Cantarovich, D., Menoret, S., Bugeon, L., Kourilsky, P., Soulillou, J. P. and Cuturi, M. C. 1995. Graft-infiltrating T helper cells, CD45RC phenotype, and Th1/Th2 related cytokines in donor-specific transfusion-induced tolerance in adult rats. Transplantation  60: 1131. Google Scholar 10 Dallman, M. J. 1995. Cytokines and transplantation: Th1/Th2 regulation of the immune response to solid organ transplants in the adult. Curr. Opin. Immunol.  7: 632. Google Scholar 11 Strom, T. B., Roy-Chaudhury, P., Manfro, R., Zheng, X., Nickerson, P. W., Wood, K. and Bushell, A. 1996. The Th1/Th2 paradigm and the allograft response. Curr. Opin. Immunol.  8: 688. Google Scholar 12 Piccotti, J. R., Chan, S. Y., VanBuskirk, A. M., Eichwald, E. J. and Bishop, K. D. 1997. Are Th2 helper lymphocytes beneficial, deleterious, or irrelevant in promoting allograft survival? Transplantation  63: 619. Google Scholar 13 Pattison, J., Nelson, P. J., Huie, P., von ??, L. I., Farshid, G., Sibley, R. K. and Krensky, A. M. 1994. RANTES chemokine expression in cell-mediated transplant rejection of the kidney. Lancet  343: 209. Google Scholar 14 Nadeau, K. C., Azuma, H. and Tinley, N. L. 1995. Sequential cytokine dynamics in chronic rejection of rat renal allografts: roles for cytokines RANTES and MCP-1. Proc. Natl Acad. Sci. USA  92: 8729. Google Scholar 15 Adams, D. H., Hubscher, S., Fear, J., Johnston, J., Shaw, S. and Afford, S. 1996. Hepatic expression of macrophage inflammatory protein-1α and macrophage inflammatory protein-1β after liver transplantation. Transplantation  61: 817. Google Scholar 16 Kondo, T., Novick, A. C., Toma, H. and Fairchild, R. L. 1996. Induction of chemokine gene expression during allogeneic skin graft rejection. Transplantation  61: 1750. Google Scholar 17 Nagano, H., Nadeau, K. C., Takada, M., Kusaka, M. and Tinley, N. L. 1997. Sequential cellular and molecular kinetics in acutely rejecting renal allografts in rats. Transplantation  63: 1101. Google Scholar 18 Krams, S. M., Egawa, H., Quinn, M. F., Villaneuva, J. C., Garcia-Kennedy, R. and Martinez, O. M. 1995. Apoptosis as a mechanism of cell death in liver allograft rejection. Transplantation  59: 621. Google Scholar 19 Sharma, V. K., Bologa, R. M., Li, B., et al. 1996. Molecular executors of cell death-Differential intrarenal expression of Fas ligand, Fas, granzyme B, and perforin during acute and/or chronic rejection of human renal allografts. Transplantation  62: 1860. Google Scholar 20 Josien, R. Muschen, M., Gilbert, E., Douillard, P., Heslan. J. M., Soulillou, J. P. and Cuturi, M. C. 1998. Fas ligand, tumor necrosis factor-alpha expression, and apoptosis during allograft rejection and tolerance. Transplantation  66: 887. Google Scholar 21 Dekel, B., Burakova, T., Marcus, H., Shezen, E., Pollack, S., Cannan, A., Passwell, J. H. and Reisner, Y. 1997. Engraftment of human kidney tissue in rat radiation chimera: I. a new model of human kidney allograft rejection. Transplantation  64: 1541. Google Scholar 22 Dekel, B., Burakova, T., Ben-Hur, H., Marcus, H., Oren, R., Laufer, J., Reisner, Y. 1997. Engraftment of human kidney tissue in rat radiation chimera: II. Human fetal kidneys display reduced immunogenecity to adoptively transferred human peripheral blood mononuclear cells and exhibit rapid growth and development. Transplantation  64: 1550. Google Scholar 23 Lubin, I., Segall, H., Marcus, H., David, M., Kulova, L., Steinitz, M., Erlich, P., Gan, J. and Reisner, Y. 1994. Engraftment of human peripheral blood lymphocytes in normal strains of mice. Blood  83: 2368. Google Scholar 24 Lubin, I., Segall, H., Erlich, P., David, M., Marcus, H., Fire, G., Burakova, T., Kulova, L. and Reisner, Y. 1995. Conversion of normal rats into SCID-like animals by means of bone marrow transplantation from SCID murine donor allows engraftment of human peripheral blood lymphocytes. Transplantation  60: 740. Google Scholar 25 Marcus, H., David, M., Canaan, A., Lubin, I., Segall, H. and Reisner, Y. 1995. Human/mouse radiation chimera are capable of mounting a human primary humoral response. Blood  86: 398. Google Scholar 26 Segall, H., Lubin, I., Marcus, H., Canaan, A. and Reisner, Y. 1996. Generation of primary antigen specific human CTLs in human/mouse radiation chimera. Blood  88: 721. Google Scholar 27 Burakova, T., Marcus, H., David, M., Dekel, B., Shezen, E., Canaan, A., Lubin, I., Segall, H. and Reisner, Y. 1997. Engrafted human T and B lymphocytes form mixed follicles in lymphoid organs of human/mouse and human/rat radiation chimera. Transplantation  63: 1166. Google Scholar 28 Premack, B. A. and Schall, T. J. 1996. Chemokine receptors: gateways to inflammation and infection. Nat. Med.  2: 1174. Google Scholar 29 Raport, C. J., Gosling, J., Schweickart, V. L., Gray, P. W. and Charo, I. F. 1996. Molecular cloning and functional characterization of a novel human CC chemokine receptor (CCR5) for RANTES, MIP-1beta, and MIP-1alpha. J. Biol. Chem.  271: 17161. Google Scholar 30 Levite, M., Zinger, H., Zisman, E., Reisner, Y. and Mozes, E. 1995. Beneficial effects of bone marrow transplantation on the serological manifestations and kidney pathology of experimental systemic lupus erythematosus. Cell. Immunol.  162: 138. Google Scholar 31 Lane, P., Gerhard, W., Hubele, S. A. and McConnell, F. 1993. Expression and functional properties of mouse B7/BB1 using a fusion protein between mouse CTLA4 and human γ 1. Immunology  80: 56. Google Scholar 32 Kirk, A. D., Bollinger, R. R. and Finn, O. J. 1995. Rapid, comprehensive analysis of human cytokine mRNA and its application to the study of acute renal allograft rejection. Hum. Immunol.  43: 113. Google Scholar 33 Robinson, E., Keystone, E. C., Schall, T. J., Gillett, N. and Fish, E. N. 1995. Chemokine expression in rheumatoid arthritis (RA): evidence of RANTES and macrophage inflammatory protein (MIP)-1β production by synovial T cells. Clin. Exp. Immunol.  101: 398. Google Scholar 34 Giordano, C., Stassi, G., De-Maria, R., et al. 1997. Potential involvement of Fas and its ligand in the pathogenesis of Hashimoto's thyroiditis. Science  275: 960. Google Scholar 35 Deng, H., Liu, R., Ellmeier, W., et al. 1996. Identification of a major co-receptor for primary isolates of HIV-1. Nature.  381: 661. Google Scholar 36 Rabinovitch, A., Suarez-Pinzon, W. L., Sorensen, O., Bleackley, R. C., Power, R. F. and Rajotte, R. V. 1995. Combined therapy with interleukin-4 and interleukin-10 inhibits autoimmune diabetes recurrence in syngeneic islet-transplanted nonobese diabetic mice. Transplantation  60: 368. Google Scholar 37 Baur, F. M., Brenner, B., Goetze-Speer, B., Neu, S. and Speer, C. P. 1998. Natural porcine surfactant down-regulates mRNA of TNF-alpha and TNF-alpha type II receptor in lipopolysaccharide-stimulated monocytes. Pediatr. Res.  44: 32. Google Scholar 38 Kennedy, M. K., Torrance, D. S., Picha, K. S. and Mohler, K. M. 1992. Analysis of cytokine mRNA expression in central nervous system of mice with experimental autoimmune encephalomyelitis reveals that IL-10 mRNA expression correlates with recovery. J. Immunol.  149: 2496. Google Scholar 39 Blancho, G., Gianello, P., Germana, S., Baetscher, M., Sachs, D. H. and LeGuern, C. 1995. Molecular identification of porcine interleukin 10: regulation of expression in a kidney allograft model. Proc. Natl Acad. Sci. USA  92: 2800. Google Scholar 40 Sayegh, M. H., Akalin, E., Hancock, W. W., Russell, M. E., Carpenter, C. B., Linsley, P. S. and Turka, L. A. 1995. CD28–B7 blockade after alloantigenic challenge in vivo inhibits Th1 cytokines but spares Th2. J. Exp. Med.  181: 1869. Google Scholar 41 Larsen, C. P., Elwood, E. T., Alexander, D. Z., et al. 1996. Long-term acceptance of skin and cardiac allografts after blocking CD40 and CD28 pathways. Nature.  381: 434. Google Scholar 42 Brouard, S., Blancho, G., Morreau, A., Heslan, J. M., Cuturi, M. C. and Soulillou, J. P. 1998. Long-term survival of hamster-to-rat cardiac xenografts in the absence of a Th2 shift. Transplantation  65: 1555. Google Scholar 43 Kirk, A. D., Harlan, D. M., Armstrong, A. A., et al. 1997. CTLA4–Ig and anti-CD40 ligand prevent renal allograft rejection in primates. Proc. Natl Acad. Sci. USA  94: 8789. Google Scholar 44 Lin, H., Bolling, S. F., Linsley, P. S., et al. 1993. Long-term acceptance of major histocompatibility complex mismatched cardiac allografts induced by CTLA4–Ig plus donor-specific transfusion. J. Exp. Med.  178: 1801. Google Scholar 45 Sayegh, M. H. and Turka, L. A. 1998. The role of T-cell costimulatory activation pathways in transplant rejection. N. Engl. J. Med.  338: 1813. Google Scholar 46 Perez, V. L., Van Parijs, L., Biuckians, A., Zheng, X., Strom, T. B. and Abass, A. K. 1997. Induction of peripheral T cell tolerance in vivo requires CTLA-4 engagement. Immunity  6: 411. Google Scholar 47 Perico N., Amuchastegui, S., Bontempelli, M. and Remuzzi, G. 1996. CTLA-4–Ig alone or in combination with low-dose cyclosporine fails to reverse acute rejection of renal allograft in the rat. Transplantation  61: 1320. Google Scholar 48 Lipman, M. L., Stevens, A. C., Bleakley, R. C., et al. 1992. The strong correlation of cytotoxic T-lymphocyte-specific serine protease gene transcripts with renal allograft rejection. Transplantation  53: 73. Google Scholar 49 Strehlau, J., Palvakis, M., Lipman, M., et al. 1997. Quantitative detection of immune activation transcripts as a diagnostic tool in kidney transplantation. Proc. Natl Acad. Sci. USA  94: 695. Google Scholar 50 Schall, T. J., Bacon, K., Toy, K. J. and Goeddel, D. V. 1990. Selective attraction of monocytes and T lymphocytes of the memory phenotype by cytokine RANTES. Nature  347: 669. Google Scholar © 1999 Japanese Society for Immunology

Journal

International Immunology – Oxford University Press

Published: Oct 1, 1999

There are no references for this article.