Published online before print July 6, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* University of Iowa Hospitals and Clinics & VAMC Iowa City, Department of Internal Medicine;
Gemini Sciences Inc., San Diego, California; and
Immunology Graduate Program, University of Iowa, Iowa City
1Correspondence: University of Iowa Hospitals and Clinics, C51-F, Department of Internal Medicine, 200 Hawkins Dr., Iowa City, IA 52242. E-mail: nicholas-zavazava{at}uiowa.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
production by CD4+ T cells. The binding of the dimeric MHC allowed ex vivo visualization of alloreactive T cells in peripheral blood, splenocytes, and allografts, revealing low frequency of alloreactive CD8+ T cells after establishment of permanent engraftment of cardiac allografts. Thus, these data show the potential of dimeric MHC molecules to promote graft survival and allow visualization of alloreactive T cells.
Key Words: transplantation • tolerance • soluble MHC • visualization
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There is, however, limited data in vivo on immunomodulation of T cell-mediated immune responses by sMHC. In preliminary studies, we showed that monomeric MHC class I antigens prolong graft survival in only 
50% of the transplanted allografts [9
]. Others specifically inhibited rejection of skin grafts using a transgenic mouse model, where the recipient mouse secreted a monomeric, truncated form of the donor-derived MHC class I/peptide molecule [10
]. Similarly, antigen-specific inhibition of an alloreactive T cell receptor-transgenic T cell population in vivo resulted in consequent outgrowth of an allogeneic tumor [11
], indicating that MHC molecules are powerful immunomodulatory reagents. More recently, Casares et al. [12
] have demonstrated that a class II dimeric molecule prevented the onset of disease and restored normoglycemia to animals already diabetic [12
]. The antidiabetogenic effects of the MHC class II dimer were shown to rely on the induction of anergy in splenic, autoreactive CD4+ T cells via alteration of early T cell receptor signaling and stimulation of interleukin 10 (IL-10)-secreting T regulatory type 1 cells in the pancreas. Therefore, there are promising data that dimeric MHC/peptide complexes can be effective tools for immunomodulation. Here, we developed a truncated MHC class I dimer of the Lewis rat strain and showed in vivo that it induces graft tolerance to donor-type cardiac allografts in Dark Agouti (DA) recipients and allowed quantification of alloreactive T cells.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, anti-CD4, anti-CD80, anti-CD86, OX18 (anti-RT1.A, polymorphic), anti-rat immunoglobulin G2c(IgG2c), and anti-mouse IgG alkaline phosphatase-conjugated antibody were purchased from BD PharMingen (San Diego, CA). The peroxidase-conjugated rabbit anti-mouse IgG was obtained from Jackson ImmunoResearch Laboratories (West Grove, PA). The rat IgG2c control Ig was obtained from Serotec (Oxford, UK). The anti-FLAG antibody was purchased from Sigma Chemical Co. (St. Louis, MO). Eight- to 12-week-old rats were purchased from Jackson Laboratory (Bar Harbor, ME). The DA RT1.Aa, Lewis RT1.Al, and Brown Norway (BN) RT1.Au rats were purchased and maintained at the VA Medical Center Iowa City Animal Care Center (IA).
Construction and expression of dimeric RT1.Al -Fc
The full-length cDNA of RT1.Al was kindly provided by Dr. T. J. Gill III and Dr. S. K. Salgar (University of Pittsburgh, Pittsburgh, PA, GenBank-Accession No.: L26224). The first four exons of RT1.Al were amplified using polymerase chain reaction (PCR). The full-length construct served as a template. During amplification, a SmaI restriction site was attached to the 5'-end, and a XbaI restriction site was added at the 3'-end. Generated PCR fragments were ligated into the SmaI/XbaI-linearized eukaryont expression vector pRK5 (BD PharMingen). In addition, DNA encoding for the hinge-, C2-, and C3-region of a rat IgG2c molecule (GenBank-Accession No: X07189) was amplified from spleen cDNA, and the resulting PCR fragment [292–990 base pairs (bp)] was ligated into the TA-cloning site of the pCR2.1 vector (Invitrogen, Carlsbad, CA). During amplification, an XbaI restriction site was attached to the 5'-end, and a HindIII restriction site was added at the 3'-end. Subsequently, the XbaI/HindIII IgG fragment was subcloned into the RT1.Al/pRK5 vector, and DNA encoding for a FLAG tag was added by PCR at the 3'-end. The resulting SmaI/HindIII DNA fragment (see Fig. 1
, 1638 bp) was thereafter subcloned into the expression vector pcDNA3.1– (Invitrogen) and was used for generating stable transfectants expressing recombinant dimeric RT1.Al molecules (RT1.Al). Each primer set was designed with regard to the full-length sequences published in GenBank and synthesized by Life Technologies (Berlin, FRG). The generated rsRT1.Al and IgG2c sequences were confirmed by DNA sequencing using the ABI373 DNA sequencer (Perkin-Elmer, Norwalk, CT). The rat myeloma cell line Y3-Ag 1.2.3 (American Type Culture Collection, Mansassas, VA) was chosen for recombinant expression of dimeric RT1.Al-Fc, as it does not express an IgG heavy chain but ß2-microglobulin. Y3-Ag cells were transfected using effectene (Qiagen, Hilden, FRG) according to the manufacturer’s instructions.
 View larger version (29K): [in a new window] |
Figure 1. Schematic structure and characterization of the dimeric RT1.Al-Fc. (a) Schematic diagram of the dimer, which was made up of a FLAG tag and the Fc region of the rat IgG2c fragment fused through the hinge region to the MHC class I molecule. (b) The dimer was precipitated with the OX18 mAb (anti-RT1.A heavy chain, lane 1) and an anti-ß2 microglobulin polyclonal serum (lane 6). Lanes 2 and 5 represent the vector controls, and lanes 3 and 4 represent the precipitates with the respective IgG isotype controls. (c) To determine the molecular size of the dimer, gel filtration of the supernatant was performed. Two main peaks of 66 kD and 200 kD were observed. The dimeric MHC molecule used in these experiments was detected in the 200-kD peak, confirming that the molecule was indeed dimeric and had the expected molecular size. (d) A single band of 67 kD representing the single RT1.Al-Fc single chain was observed in the purified sample (lane 1). Lanes 2 and 3 represent the control supernatants of the wild-type cells and that of Lewis-derived splenocytes, respectively. MW, Molecular weight.
|
Immunoprecipitation and Western blotting of recombinant RT1.Al-Fc
Culture supernatant (1 mL) of transfected cells grown in serum-free medium was incubated with shaking at 4°C with 2 µg polyclonal anti-ß2 microglobulin serum (Dako, Danemark) or 2 µg OX18, a mAb against a conformational-dependent epitope on heavy chains of rat MHC class I molecules. After 4 h, 50 µl protein-A agarose (Pharmacia, Sweden) was added to each sample, and the mixture was shaken overnight at 4°C. Samples were washed thoroughly in phosphate-buffered saline and centrifuged. The pellets were resuspended in 100 µl sodium dodecyl sulfate-loading buffer, boiled, and centrifuged. The recovered supernantant was run on a 7.5% polyacrylamide gel and blotted on a nitrocellulose membrane. The proteins were stained using an anti-FLAG mAb. Protein bands were visualized by an anti-mouse IgG alkaline phosphatase-conjugated antibody.
Transplantation of cardiac allografts
To test the efficacy of the soluble dimer to promote graft survival, DA recipient animals received 10 µg of the dimer by intraperitoneal infusion daily for 14 days. They were transplanted with a cardiac allograft on the first day according to Ono and Lindsey [13
] and left untreated (n=6) or received a single dose of 15 mg/kg cyclosporine A (CsA; n=12). Graft function was monitored by palpation. Lewis-derived cardiac allografts were transplanted to DA recipients as we previously reported [14
]. In this strain combination, cardiac allografts are acutely rejected on Day 7.
Immunohistochemical and flow cytometric staining
Cryostat sections recovered on Day 5 (rejecting animals) or on Day 150 (tolerant animals) were fixed in methanol, washed in Tris-HCl (pH 7.6), and incubated with the dimer and a mouse anti-FLAG alkalkaline phosphatase-conjugated mAb. Sections stained for histology were stained according to standard procedure using hemotoxylin eosin staining. For the detection of alloreactive T cells by flow cytometry, 25 µl peripheral blood was incubated with a lysis solution containing ammonium chloride to remove the erythrocytes. The peripheral blood lymphocytes or splenocytes were incubated with the dimer for 30 min on ice. The cells were washed and then incubated with a fluorescein isothiocyanate (FITC)-conjugated anti-FLAG mAb. For CD80 and CD86 staining, peritoneal macrophages, splenocytes, or lymph node-derived cells were incubated for 30 min at 4°C with FITC-conjugated anti-CD80 or anti-CD86 mAb. The cells were then washed, and cell fluorescence was measured by a FACScan (BD PharMingen).
Proliferation and cytotoxicity assays
Here, we tested the efficacy of the soluble dimer to modulate alloresponses in vivo. Thus, DA recipient animals were immunized daily from Day 0 to Day 14 with the dimeric MHC antigen (10 µg) and Lewis-derived splenocytes (107 Lewis-derived splenocytes were infused on Day 0 and Day 7), dimer only, and splenocytes only or were left untreated (control). Animals were killed on Day 14, spleens harvested, and splenocytes used in a standard [3H]thymidine proliferation assay (Lewis and BN splenocytes were used as stimulator cells).
DA responder splenocytes from the various animal groups were also used to generate CD8+ T cells for cytotoxicity assays. DA responder splenocytes were restimulated in vitro with irradiated Lewis-derived splenocytes at a 1:1 ratio for 5 days. CD8+ T cells were isolated from bulk cultures with magnetic beads (Miltenyi Biotec, Auburn, CA) and used in a 4-h 51chromium release assay with Lewis-concanavalin A (Lew-Con A) blasts as target cells.
The inhibition assay of alloreactive T cells was performed as described previously in this lab [15 16 17 ]. Briefly, DA anti-Lewis, DA anti-BN, or Lewis anti-DA CD8+ T cells were incubated with the dimer, and their cytotoxicity against Con A target cells (Lewis, BN, and DA Con A blasts, respectively) was tested in a 4-h 51chromium release assay.
Measurement of interferon- (IFN-) production by enzyme-linked immunospot (ELISPOT)
To measure the secretion of IFN- in the treated groups of animals, splenocytes were derived from animals sensitized with splenocytes only and with splenocytes and treated with the dimer or from nontreated control animals. CD4+ T cells were separated using magnetic beads (Miltenyi Biotec) according to the manufacturer’s instructions. CD4+ T cells were used as responder cells to irradiated Lewis-derived splenocytes. IFN- secretion was measured by the use of an IFN- ELISPOT kit (BD PharMingen). Spots were counted using an automated ELISPOT reader system (Immunospot, Cellular Technology Ltd., Cleveland, OH).
Peritoneal macrophage assays
APCs were obtained by adherence separation from splenocytes, lymphn odes, or peritoneal macrophages of DA recipient animals. Separated APCs were pulsed with the indicated amount of dimer and used as stimulator cells for DA-derived CD4+ T cells in a standard [3H]thymidine proliferation assay.
To track peritoneal macrophages in vivo, DA-derived peritoneal macrophages were pulsed with the dimer for 24 h and labeled afterward with carboxyfluorescein diacetate succinimidyl ester (CFSE; Vybrant carboxyfluorescein diacetate-SE cell tracer kit, Molecular Probes, Eugene, OR), according to the manufacturer’s instructions. Cells were washed extensively and reinfused into the peritoneum of the same DA animals. After 24 h, the animals were killed, and green fluorescent cells were detected in the peritoneum, peripheral blood, spleens, and lymph nodes by flow cytometry.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Dimeric RT1.Al-Fc induces indefinite graft survival of allogeneic cardiac allografts
DA recipient animals were infused daily with 10 µg RT1.Al-Fc for 14 days. This protocol was a modification of our previous protocol using monomeric sMHC [9
]. Together with the first dimer infusion, recipient animals were transplanted Lewis-derived heterotopic cardiac transplants and left untreated or received a single dose of 15 mg/kg CsA. Treatment of recipient animals with this dose of CsA alone prolonged graft survival from 7 days to 14 days (Fig. 2a
). It is surprising that the dimeric MHC antigen equally prolonged graft survival, 8.5 ± 0.1–13.2 ± 0.4 days (P value <0.0001; n=10), demonstrating the efficacy of these molecules to abrogate graft rejection. In contrast, nine out of a total of 12 animals (75%) treated with the dimer and subtherapeutic CsA showed indefinite graft survival, >150 days post-transplantation (Fig. 2a)
. Third-party BN-derived cardiac allografts were rejected by Day 9. These results are highly superior to our own published findings with synthetic MHC peptides or with monomeric sMHC antigens [9
, 18
], where only 50% prolonged graft survival was achievable.
 View larger version (71K): [in a new window] |
Figure 2. Dimeric sMHC class I antigens induce permanent engraftment of donor-derived cardiac allografts. (a) DA recipient rats were transplanted Lewis-derived cardiac allografts and simultaneously treated with a single dose of 10 µg dimeric RT1.Al-Fc daily for 14 days. Animals that received an additional dose of CsA permanently accepted cardiac allografts (n=12). Third-party allografts were acutely rejected. The dimer and CsA used alone moderately prolonged graft survival. (b–d) Histological sections revealed normal architecture of tolerated allografts 7 and 150 days post-transplantation, whereas rejected allografts were heavily infiltrated by mononuclear cells showing significant architectural damage to the allografts.
|
Treatment with dimeric RT1.Al-Fc modulates T cell responses to alloantigen
To understand the mechanism by which donor-derived dimeric MHC protected allografts from rejection, recipient animals were treated with the dimeric MHC antigen for 14 days as described above. Together with the first dimer infusion, animals were immunized with 10 x 106 donor-derived splenocytes. This infusion was repeated on Day 7, and the animals were killed on Day 14. Spleens were harvested and their response to alloantigen analyzed. First, T cells were used as responder cells in a 4-day mixed lymphocyte assay. Control groups were animals treated with splenocytes only, the dimer only, or untreated animals. As expected, CD8+ T cells recovered from animals sensitized with splenocytes only showed strong proliferation to donor-type alloantigen (Fig. 3a
). It is surprising that proliferation of T cells recovered from animals that were treated with the dimer in addition to the splenocytes was significantly abrogated (P<0.05, n=5). The other two control groups showed limited T cell proliferation. However, alloresponse to third-party responder animals of the BN strain remained unchanged in all groups.
 View larger version (22K): [in a new window] |
Figure 3. Treatment of recipient rats with dimeric MHC antigens abrogates T cell proliferation and cytotoxic T cell cytotoxicity. (a) To test the effect of dimer treatment on T cell response to alloantigen presented by direct presentation, CD8+ T cells from animals treated with donor splenocytes only, donor splenocytes and the dimer, the dimer only, or control, untreated animals were used in a 4-day mixed lymphocyte reaction assay as responder cells. Irradiated Lewis-derived splenocytes were used as stimulator cells. Treatment with dimeric MHC strongly abrogated the alloproliferative response of T cells in animals that were sensitized with donor splenocytes and the dimer, compared with T cells in animals sensitized with splenocytes only (P<0.05, n=5). Third-party BN-derived stimulator cells remained unchanged in all groups. Error bars represent standard deviations from triplicate wells. (b) CD8+ T cells were used as effector cells in a 4-h 51chromium release assay after restimulation of the cells from the various animal groups (as indicated) for 5 days in vitro using Lewis-derived splenocytes as stimulator cells. T cells derived from animals sensitized with splenocytes only showed strong target cell (Lew-Con A blasts) killing, whereas the cytotoxicity of cells recovered from animals treated with splenocytes and the dimer was drastically abrogated, indicating a direct influence of the dimer. As expected, CD8+ T cells derived from control animals or from animals treated with the dimer only showed low cytotoxicity. (c) To determine whether the soluble dimer abrogates cytotoxic T cells directly, DA anti-Lewis, DA anti-BN, or Lewis anti-DA cytolytic T lymphocyte (CTL) were incubated with the dimer and their cytotoxicity against Con A target cells (Lewis, BN, and DA Con A blasts, respectively), tested in a 4-h 51chromium release assay. Anti-Lewis CTL were inhibited in a concentration-dependent manner but not the DA anti-BN or the Lewis anti-DA CTL used as controls. (d) To determine the effect of the dimer treatment on CD4+ T cells, an ELISPOT was used for measuring the number of IFN--secreting CD4+ T cells. Clearly, dimer treatment led to a significant reduction in IFN- secretion by the CD4+ T cells. Error bars represent standard deviations from triplicate wells.
|
after alloantigen stimulation. Indeed, the cells derived from animals treated with the dimer and splenocytes showed significant reduction in IFN- production after alloantigen stimulation (Fig. 3d) , as compared with the cells derived from animals sensitized with the splenocytes only. These results altogether indicated that dimer treatment effectively modulated CD4+ and CD8+ T cells. To explain these unexpected sMHC-induced effects, we hypothesized that peritoneal macrophages are involved in the presentation of the dimer, leading to down-regulation of alloreactive T cells. CD80 and CD86 were measured on the DA-derived peritoneal macrophages after a peritoneal lavage. Expression of CD80 and CD86 was surprisingly low on peritoneal macrophages compared with splenocytes and lymph node-derived cells (Fig. 4a ). Therefore, peritoneal macrophages were likely to provide weak costimulation, leading to T cell anergy rather than stimulation. Indeed, when the peritoneal macrophages were used as stimulator cells in a proliferation assay with the dimer used as alloantigen, they failed to activate alloreactive CD4+ T cells, whereas lymph node-derived as well as splenic APCs were capable of stimulating the T cells (Fig. 4b) . To determine whether peritoneal macrophages trafficked to peripheral lymphoid organs after picking up alloantigen, we harvested peritoneal macrophages and pulsed them with the dimer over 24 h ex vivo. The cells were washed and subsequently labeled with CFSE, reinfused into the peritoneum of DA recipient animals, and killed 24 h later to determine the presence of fluorescent green cells in the peritoneum, peripheral blood, spleens, and lymph nodes. About 1% of circulating leukocytes were CFSE-labeled compared with 12.9% in the peritoneal lavage (Fig. 4c) . Less than 0.5% of splenocytes or lymph node-derived cells was CFSE-positive. Although the percentage of CFSE-labeled macrophages in splenocytes or lymph node-derived cells appears to be small, the total cell numbers could be sufficient to interfere with alloantigen presentation to T cells. These results suggest that peritoneal macrophages trafficked into the circulation and into peripheral lymphoid organs after engagement with alloantigen, where they likely interacted with T cells.
 View larger version (23K): [in a new window] |
Figure 4. Peritoneal macrophages present dimeric sMHC molecules and fail to induce CD4+ T cell proliferation. (a) DA-derived peritoneal macrophages, splenocytes, and lymph node-derived cells were stained with anti-CD80 or anti-CD86 mAb. Peritoneal macrophages show the lowest expression of CD80 and CD86, compared with splenocytes and lymph node-derived cells, indicating that they can only provide weak costimulation, possibly leading to T cell anergy. (b) To determine the ability of peritoneal macrophages to present the dimeric sMHC molecules, APC derived by adherence separation from splenocytes, lymph nodes, or peritoneal macrophages of DA recipient animals were used to present the soluble dimer to DA-derived CD4+ T cells. APC derived from peritoneal macrophages failed to stimulate T cells. (c) To determine whether peritoneal macrophages traffic into the circulation after picking up alloantigen in the peritoneum, DA-derived peritoneal macrophages were pulsed with the dimer for 24 h ex vivo and reinfused into the same DA animals after labeling with CFSE. After another 24 h, the animals were killed, and green fluorescent cells were detected in the peritoneum and peripheral blood (12.9% and 1%, respectively). FSC, Forward scatter; SSC, side scatter.
|
 View larger version (73K): [in a new window] |
Figure 5. Dimeric RT1.Al-Fc visualizes intragraft and circulating alloreactive T cells. Immunohistochemical sections were incubated with the dimeric MHC molecule, and binding was visualized by red alkaline phosphatase staining (anti-FLAG-AP). Low staining of peripheral blood lymphocytes of tolerated allografts (a, original x200) and strong staining in rejected allografts (b, original x200) were observed. It is interesting that in some sections from tolerated allografts, such as in Figure 2
, we observed perivascular infiltrates, which, however, were not accompanied by any pathology, as the tissue was well maintained. Control staining from third-party BN allografts was negative for dimer staining (c, original x200). In peripheral blood of animals that rejected allografts, there were more dimer-staining CD8+ T cells (d) than in tolerated allografts (e). T lymphocytes derived from nontransplanted control animals showed no significant dimer binding (f).
|
The MHC dimer detects a high frequency of alloreactive CD8+ T cells in splenocytes of rejecting animals
To further characterize alloreactive T cells using the dimeric MHC molecules, Lewis-derived cardiac allografts were transplanted into DA recipient animals and left untreated. The organs were rejected on Day 7, and the animals were killed for the recovery of spleens. CD8+ T cells were separated from splenocytes by immunomagnetic beads and stained with the dimer. Clearly, 14.6% CD8high of the splenocytes in rejecting animals were positive for the dimer compared with only 3.3% in tolerant animals (n=6), killed on Day 40 post-transplantation (Fig. 6a
). To further confirm the specificity of this stain, splenocytes recovered from DA animals rejecting DA allografts (syngeneic control) and from third-party controls (BN transplanted in Lewis) were used as controls (Fig. 6a)
. In both control groups, only 2% of the cells were stained. As the dimer has an IgG backbone and can potentially bind to the FcR on T cells, we preincubated splenocytes from animals rejecting a Lewis allograft with antibodies against CD4, CD8, FcR, or CD28. After 1 h incubation, the cells were washed to remove excess antibody and then stained with the dimer to detect any changes in the amount of dimer binding. None of these antibodies interfered with the dimer staining, clearly showing that the stains obtained were not a result of the dimer binding to the cells via the FcR or any of the tested molecules (Fig. 6b)
. Further, dimer-binding cells were isolated by immunomagnetic separation and used as responder cells in a proliferation assay. Control cells were CD8+ cells obtained from a control animal. Indeed, dimer-binding cells from a rejecting animal responded to alloantigen three to four times more strongly than nondimer-binding cells (data not shown), clearly confirming that the cells binding the dimer were alloreactive. Collectively, the data presented here further confirmed that the dimeric MHC molecule indeed visualizes alloreactive T cells in an allospecific manner.
 View larger version (28K): [in a new window] |
Figure 6. Acutely rejecting animals show a high frequency of dimer-binding cells in the spleen. Splenocytes of acutely rejecting and those of control animals were separated for CD8+ T cells by immunomagnetic beads and stained with the dimer (a). Cell fluorescence of dimer-binding cells was highest in acutely rejecting animals (15%) and significantly lower in spleens of tolerated allografts and in the syngeneic or third-party (BN to DA) graft controls (2–3%). Preincubation of isolated CD8+ T cells from rejecting animals with anti-Fc receptor (FcR), -CD4, or -CD8 or with anti-CD28 antibodies had no significant effect on dimer binding by the CD8+ T lymphocytes of rejecting animals as shown in the overlays (b).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animals treated with sMHC dimers showed prolonged graft survival but failed to prevent graft loss by acute rejection. However, when the dimer was supplemented with a subtherapeutic single dose of CsA, which does not prevent acute graft rejection on its own, graft-specific tolerance characterized by lack of mononuclear cell infiltration and lack of signs of chronic rejection was achieved. This effect was observed in the treated animals as early as Day 7 post-transplantation and was the same 150 days post-transplantation. This unexpected result was shown to be mediated by abrogated response of CD8+ to direct presentation of alloantigen and significant reduction of IFN- production by CD4+ T cells. Our experiments indicated that peritoneal macrophages express low levels of costimulatory molecules and also the ability to activate T cells by presentation of soluble alloantigen. Therefore, the soluble dimer likely induced T cell nonresponsiveness, which favored engraftment of the cardiac allografts, revealing an interesting route for antigen delivery in the induction of transplantation tolerance. In this model, CsA appears to prevent the initial rejection process, allowing the dimer to induce T cell nonresponsiveness.
To study whether the dimeric MHC molecules could be used to track alloreactive T cells, the dimer was used to stain histological sections. In peripheral blood, 6.2 ± 1.2% (n=6) of blood lymphocytes were donor MHC-binding and CD8+. Assuming that only 25% of peripheral blood lymphocytes are in fact CD8+, the number of alloreactive T cells detected by the dimeric MHC is relatively high. In contrast, only 1.4 ± 1.1% (n=5) CD8+ cells were detected in tolerant animals. The presence of the alloreactive T cells after induction of permanent engraftment of cardiac allografts indicated persistence of alloreactive T cells with no apparent reactivity toward the allograft. These data were strongly supported by the immunohistochemical stains, which indicated specific binding to T cells and significant differences between the tolerant and rejected allografts. Although the majority of recombinant MHC used in the literature is generated in Escherichia coli and loaded with desired specific peptides thereafter, here, we chose to use the expression of recombinant MHC in mammalian B cells. That way, the MHC molecules are peptide-loaded by endogenous peptides. As the majority of alloreactive CD8+ T cells is against peptide-loaded allo-MHC molecules [3
, 23
], rather than against specific peptides, we anticipated that this strategy could be successful. Indeed, these dimeric MHC class I/peptide complexes modulated and visualized alloreactive T cells. Thus, expression of allo-MHC molecules in mammalian B cells gives the secreted MHC the complete repertoire of endogenous peptides, mimicking those presented by allogeneic organs. Analysis of peptides presented by secreted, recombinant sMHC in our lab revealed little difference between their peptides and those presented by membrane-bound MHC molecules [24
]. However, we cannot rule out the possibility that the total number of alloreactive T cells may in fact be higher than those estimated here, as a small percentage of alloreactive T cells may be specific for Lewis-derived and tissue-specific allo-peptides not expressed in the Y3-Ag 1.2.3 cell line, which was used for transfecting the sMHC alloantigen. Nonetheless, the results still allow differentiation of tolerant from acute-rejecting animals. To definitively elucidate the role of the peptide, we are working on a mouse model of these experiments that will allow us to test MHC dimers loaded with different peptides, as in the rat model, hardly any peptides have been defined.
Most recently, Casares et al. [12 ] demonstrated in a mouse model of diabetes that administration of dimeric sMHC class II molecules loaded with the hemagglutinin peptide (110–120) to prediabetic, double-transgenic mice prevented the onset of the disease or in animals that were already diabetic, restored normoglycemia. This effect was found to be mediated by induction of T cell anergy via alteration of early T cell signaling and stimulation of IL-10-secreting T regulatory cells. Clearly, our data in a transplantation model using a chimeric MHC class I molecule agree with these published data. Here, the dimeric MHC class I antigens used abrogated the development of alloreactive T cells, as revealed by their inability to kill target cells (Fig. 3b) . Altogether, these data point to a therapeutic potential of dimeric MHC/peptide complexes. We previously published data on prolonged graft survival using synthetic MHC peptides or monomeric MHC molecules [9 , 15 , 25 ]. However, none of these strategies increased graft survival in more than 50% of the cases. Additionally, graft quality was poor, as there was evidence of chronic rejection, which was not seen here, showing that the dimer is a powerful and superior regulatory molecule in this transplantation model. Indeed, compared with published data with MHC class I- or class II-derived, synthetic peptides [25 26 27 28 ], the dimer used here is more effective in modulating alloresponses. Further, the ability to stain alloreactive CD8+ by MHC-oligomers using flow cytometry and immunhistology [29 ] might allow the establishment of new, diagnostic tools, in particular, for the detection of established tolerance.
Received February 7, 2005; revised May 11, 2005; accepted May 12, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
