The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bagley, J.
Right arrow Articles by Iacomini, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bagley, J.
Right arrow Articles by Iacomini, J.
The Journal of Immunology, 2000, 165: 4842-4847.
Copyright © 2000 by The American Association of Immunologists

Defining the Requirements for Peptide Recognition in Gene Therapy-Induced T Cell Tolerance1

Jessamyn Bagley, Yin Wu, David H. Sachs and John Iacomini2

Transplantation Biology Research Center, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02129


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Expression of a retrovirally transduced MHC class I Ag, H-2Kb (Kb), in bone marrow-derived cells leads to specific prolongation of Kb disparate skin grafts. To examine the extent to which peptides derived from Kb contribute to the induction of tolerance, retroviruses carrying mutant Kb genes designed to enter separate pathways of Ag presentation were constructed. Thymectomized and CD8 T cell-depleted mice that had been irradiated and reconstituted with bone marrow cells expressing a secreted form of Kb showed prolongation of Kb disparate skin graft survival. Skin graft prolongation was not observed when similar experiments were performed using mice that were not CD8 T cell depleted. This suggests that hyporesponsiveness can be induced in CD4 T cells, but not CD8 T cells by Ags presented via the exogenous pathway of Ag processing. Modest prolongation of skin allografts was observed in mice reconstituted with bone marrow cells transduced with retroviruses carrying a gene encoding a mutant Kb molecule expressed only in the cytoplasm. Prolongation was also observed in similar experiments in mice that were thymectomized and CD4 T cell depleted following complete reconstitution, but not in mice that were reconstituted and then thymectomized and CD8 T cell depleted. Thus, hyporesponsiveness can be induced in a subset of CD8 T cells by recognition of peptides derived from Kb through both the direct and indirect pathways of Ag recognition, while CD4 T cell hyporesponsiveness to MHC class I disparate grafts occurs only through the indirect pathway of Ag recognition.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The induction of immunological tolerance remains a major goal in transplantation biology because of its potential to permit permanent graft survival without the need for lifelong immunosuppression. An effective means of inducing transplantation tolerance is the creation of mixed hemopoietic chimerism by allogeneic bone marrow transplantation (1). However, several complications associated with bone marrow transplantation across MHC barriers make its clinical utility limited. To overcome some of the difficulties associated with the induction of cellular mixed chimerism, we have developed a gene therapy approach involving the transfer of allogeneic MHC genes into autologous bone marrow hemopoietic progenitor cells to induce molecular rather than cellular chimerism (for review, see Ref. 2). Expression of a retrovirally transduced allogeneic MHC class I gene, H-2Kb (Kb),3 in bone marrow-derived cells induces hyporesponsiveness to the Kb alloantigen, resulting in specific prolongation of Kb disparate skin graft survival (3). Because this approach involves the transfer of genes rather than the transplantation of allogeneic bone marrow cells, the risk of graft-vs-host disease associated with allogeneic bone marrow transplantation is eliminated.

A major question related to the induction of tolerance by gene therapy is the mechanism by which alloreactive T cells are rendered hyporesponsive by this approach. While expression of a retrovirally transduced Kb gene in bone marrow-derived cells leads to specific prolongation of Kb disparate skin graft survival, the presence of additional allogeneic determinants expressed on skin together with Kb can restore rapid rejection (4, 5). Interestingly, either a single MHC class I or class II alloantigen expressed on skin in association with Kb is sufficient for this effect (5). Thus, the presence of additional Th determinants (6) on Kb-bearing skin grafts can overcome the effect of gene therapy in this system, suggesting that prolongation of skin graft survival induced by gene therapy may involve a peripheral mechanism controlled at the level of T cell help.

T cells recognize allogeneic MHC Ags via two distinct pathways of Ag presentation (7). The direct pathway involves the recognition of intact allogeneic MHC molecules on the surface of donor cells by host T cells. The indirect pathway involves the recognition donor MHC-derived peptides that are processed by host APCs and presented as peptides to host T cells in the context of self MHC. In the case of transplants that are matched at least at a single MHC locus, such as MHC class II, it is possible for alloreactive host T cells to recognize allogeneic MHC-derived peptides presented in the context of host or donor MHC via the indirect pathway. Previously, we have observed that CD4 T cells are able to reject skin grafts that differ by a single MHC class I allelic disparity by a mechanism that is likely to involve the recognition of donor MHC class I (Kb)-derived peptides presented to host CD4 T cells through the indirect pathway (8). Insofar as hyporesponsiveness induced by gene therapy appears to be controlled at the level of T cell help (4, 5), it seems possible that the generation alloreactive CD4 T cells that recognize Kb-derived peptides via the indirect pathway and generate T cell help are affected by this approach.

In this study, we examine the extent to which recognition of peptides is involved in the induction of allograft hyporesponsiveness induced by gene therapy. We have tested the hypothesis that presentation of Kb-derived peptides through the indirect pathway is involved in the inactivation of alloreactive T cells. Mutant Kb genes were constructed encoding MHC class I molecules that can enter the indirect pathway of Ag presentation through distinct cellular pathways. These mutant Kb genes were then cloned into retroviral vectors, and viruses carrying these genes were used to modify bone marrow cells to determine whether presentation of peptides alone can affect alloreactive CD4 and CD8 T cells.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Mice

Female B10.AKM/SnJ (H-2Kk,Ik,Dq), B10.MBR/Sx (H-2Kb,Ik,Dq), and B10.BR (H-2k) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Mice were housed using microisolator conditions in autoclaved cages and maintained on irradiated feed and autoclaved acidified drinking water. All sentinel mice housed in the same colony were viral Ab free. Six- to 12-wk-old female mice were used in all experiments.

Antibodies

The following Abs were used in this study: Y3 (IgG2b; recognizes the {alpha}1 and {alpha}2 domains of properly conformed Kb molecules (9)); Ku2 (recognizes aa 23–30 of Kb when the molecule is unfolded (10); AF6-88.5 (IgG2a, anti-Kb (11); PharMingen, San Diego, CA); 2.43 (rat IgG2b, anti-CD8 (12)); and GK1.5 (rat IgG2b, anti-CD4 (13)).

T cell subset depletion and skin allografts

Thymectomy was performed as previously described (8). To prepare T cell subset-depleted mice, 1 mg of purified 2.43 or GK1.5 was administered to mice on days 1, 7, and 14 following thymectomy. All depletions were confirmed by FACS analysis of PBMC, as described (8). Tail skin grafting was performed and evaluated as previously described (14).

Vector construction

To produce a secreted Kb molecule (KbTMS), the pkQpBG367 plasmid (a gift of Dr. Gerry Waneck, Massachusetts General Hospital, Boston, MA), which contains a cDNA encoding a GPI-anchored Kb molecule, was digested with HindIII and AccI. This fragment lacks the entire transmembrane and cytoplasmic domains of Kb. The resulting restriction fragment was then ligated to a synthetic adaptor containing an AccI-compatible overhang and a stop codon after aa 301, 3 aa before the start of the transmembrane domain. Ligation of the adaptor to the 3' end of the mutant Kb gene results in a blunt-ended fragment that was then cloned as a HindIII-HincII fragment into a modified pBluescript II KS- vector (Stratagene, La Jolla, CA) in which the KpnI site in the multiple cloning cassette was destroyed (pKS-K-) to generate pKbTMS. The plasmid pKbTMS was then sequenced to verify that the stop codon was introduced correctly. The plasmid SP95.3.5 containing the mouse phosphoglycerate kinase (PGK) promoter (a gift of Dr. Cristian LeGuern, Massachusetts General Hospital, Boston, MA) was then digested with BamHI and EcoRI, and the resulting fragment containing the PGK promoter was cloned into pKbTMS plasmid 5' of the mutant Kb gene. The resulting construct pKbTMSPGK was then digested with BamHI and BssHII, and the fragment containing the PGK promoter and mutant Kb gene was cloned into the vector N2A (15) using the compatible sites BglII and MluI to generate KbTMS.N2A.

To produce a cytoplasmic Kb molecule (KbCYT), the plasmid pKb6-2B1 (provided by Dr. Gerald Waneck) containing a cDNA-encoding wild-type Kb was digested with NotI and PvuII and cloned into pKS-K- to yield a fragment in which the first 10 aa of the signal peptide were deleted to yield pKbL-. PCR-mediated site-directed mutagenesis was then performed to insert a Kozak consensus sequence at the beginning of the mature protein (aa 21 of wild-type Kb). Mutagenesis was verified by DNA sequencing. A SspI-KpnI restriction fragment containing the 5' half of the mutant Kb gene was them isolated from pKbL- and swapped into pKbTMS to generate a mutant Kb molecule lacking a signal sequence and transmembrane domain. The PGK promoter was then cloned into the plasmid 5' of the coding region and as the entire construct introduced into the N2A retroviral vector, as described above for pKbTMSPGK to yield KbL-.N2A.

Producer cell lines

Retroviral producer cell lines were generated as described previously (16). Briefly, AM12 retroviral packaging cells (17) (American Type Culture Collection, Manassas, VA) were seeded at 70% confluency and allowed to grow overnight. The cells were then transfected with either KbTMS.N2A or KbL-.N2A by CaPO4 precipitation using the Stratagene mammalian transfection kit according to the manufacturer’s specifications. Cells were then incubated 18 h at 37°C, 5% CO2, then washed three times in HBSS without Ca2+ or Mg2+ (Life Technologies, Gaithersburg, MD) and cultured for 48 h in 15P media (DMEM (Mediatech, Herndon, VA) with 15% FCS (Sigma, St. Louis, MO)) before adding 15P media containing 0.8 mg/ml G418 (Mediatech). After 10 days in selection media, colonies were picked from the plate by scraping with a sterile pipette and grown in Falcon six-well tissue culture plates (Becton Dickinson, La Jolla, CA).

Bone marrow transduction and transplantation

Transduction of bone marrow was performed as described previously (16). Briefly, mice were injected with 5-fluorouracil (150 mg/kg) 7 days before bone marrow harvest. Bone marrow cells were then harvested, and 107 cells were added to 100-mm tissue culture dishes that contained preestablished virus producer cell lines. The cells were cultured in transduction media consisting of DMEM, containing 15% FCS, 100 ng/ml IL-6 (R&D Systems, Minneapolis, MN), 20 ng/ml IL-3 (R&D Systems), and 2 µg/ml polybrene (Sigma). In addition, all cultures were supplemented with cell-free viral supernatants (~4 x 106 CFU per 107 bone marrow cells). Transduction was conducted for 3 days, at which point the culture media were replaced with fresh transduction media and cell-free viral supernatants. The bone marrow cells were harvested 48 h later. Lethally irradiated (10.25 Gy) B10.AKM mice were then reconstituted with 1–2 x 106 transduced bone marrow cells by tail vein injection.

Cell staining

Cell surface staining was performed using directly conjugated mAbs, as described previously (8). For intracellular staining, 106 cells were washed twice in HBSS containing 5% FCS (HBSS-FCS). The cells were then incubated for 20 min at 4°C with 10 µg of unlabeled Y3 to block cell surface staining, washed twice in HBSS-FCS, fixed in 1% paraformaldehyde for 30 min, and then washed twice in HBSS-FCS. The cells were then washed once in HBSS-FCS containing 0.1% saponin (Sigma) to permeabilize them, and then resuspended in HBSS-FCS containing 0.25% saponin and a saturating concentration of FITC-labeled Ku2. The cells were then incubated for 20 min at 4°C, washed in HBSS-FCS containing 0.1% saponin, and resuspended in 100 µl HBSS-FCS. Cells were then analyzed by flow cytometry.

Statistics

All statistical calculations were performed using GraphPad Prism 2.01 software (GraphPad Software, San Diego, CA). The Kaplan and Meier method with a 95% confidence interval was used for the calculation of survival curves. Comparison of survival curves was performed using the log rank test.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Construction of mutant Kb molecules designed to be presented through the indirect pathway

Alloantigens can enter the indirect pathway of Ag recognition through either the MHC class I or class II Ag presentation pathways. In general, secreted proteins enter the MHC class II Ag-processing pathway, while cytoplasmic proteins enter the MHC class I Ag presentation pathway (reviewed in Ref. 18). We reasoned that it would be possible to direct the Kb alloantigen to the MHC class II-processing pathway by making a mutant molecule that is secreted and must be taken into the MHC class II-processing compartment to be presented to T cells through the indirect pathway. Likewise, it would be possible to direct the Kb alloantigen to the MHC class I Ag-processing pathway by making a mutant molecule that is retained in the cytoplasm and is likely to be degraded into peptides by the proteosome. These peptides then enter the MHC class I Ag-processing compartment and are presented to T cells through the indirect pathway.

To generate a secreted Kb molecule, we used standard cloning procedures to construct a mutant gene encoding a molecule that lacks both the transmembrane and cytoplasmic domains of Kb, referred to as KbTMS (Fig. 1Go). Because the molecule encoded by this gene retains its signal sequence and lacks a transmembrane domain, it cannot be expressed on the cell surface and is instead secreted. To generate a Kb molecule that is targeted to the cytoplasm, we constructed a mutant Kb gene, referred to as KbCYT, encoding a molecule lacking a signal sequence as well as both a transmembrane and cytoplasmic domain (Fig. 1Go). Because the molecule encoded for by this gene lacks all known sequences that would allow it to be expressed as a cell surface protein, and cannot be translocated across the endoplasmic reticulum, it is predicted to remain in the cytoplasm until it is degraded.



View larger version (12K):
[in this window]
[in a new window]
  		FIGURE 1. Diagram of molecules encoded for by mutant Kb genes. Top, The wild-type Kb molecule encoded for by HTD12-10. Shown are the signal sequence (S), {alpha}1 through {alpha}3 extracellular domains, and the transmembrane (TM) and cytoplasmic domains (CYT). Middle, The KbTMS mutant molecule lacking the TM and CYT regions. The asterisk indicates the location of the introduced stop codon. Bottom, The KbCYT mutant molecule lacking the S, TM, and CYT domains. ATG indicates the site of the introduced Kozak translation initiation sequence. Drawing not to scale.

 
To characterize the proteins encoded for by KbTMS and KbCYT, these mutants were cloned into retroviral vectors and stable virus producer lines were generated. As shown in Fig. 2GoA, NIH3T3 cells infected with retroviruses carrying the wild-type Kb gene, HTD12-10 (19), expressed Kb on their surface at levels detectable by cell surface staining and flow cytometry using the anti-H-2Kb mAb AF6-88.5. In contrast, we were unable to detect Kb molecules on the surface of NIH3T3 cells infected with either KbTMS, KbCYT, or control virus (AM12-Neor) carrying only the neomycin resistance gene using AF6-88.5 (Fig. 2GoA). Similar results were obtained using the conformation-specific anti-H-2Kb Ab Y3 (data not shown).



View larger version (43K):
[in this window]
[in a new window]
  		FIGURE 2. Analysis of mutant Kb molecules by cell staining and flow cytometry. Viral supernatants were used to transduce NIH3T3 cells, and bulk populations were selected for 10–14 days using G418. A, Cell surface staining of NIH3T3 cells infected with HTD12-10 (wt), KbTMS, KbCYT, or control AM12-Neor (Neor) virus with the anti-Kb Ab AF6-88.5. B, Intracellular staining of NIH3T3 cells infected with HTD12-10 (wt), KbTMS, or control AM12-Neor (Neor) virus with the conformation-sensitive anti-Kb Ab Y3. C, Detection of secreted Kb molecules in tissue culture supernatants harvested from cell lines expressing wild-type Kb (lane 1), KbTMS (lane 2), or the Neor gene (lane 3). Approximate m.w. are shown on the left of the gel. The arrow indicates the expected m.w. of secreted Kb. Films were exposed for 7 days. The figure shown is a digitized image of the original x-ray film. D, Intracellular staining of NIH3T3 cells infected with HTD12-10 (wt), KbCYT, or control AM12-Neor (Neor) virus with Ku2, an anti-Kb Ab that recognizes unfolded Kb. Shown are representative data.

 
We next analyzed NIH3T3 cell infected with either HTD12-10, KbTMS, KbCYT, or AM12-Neor for the presence of intracellular Kb molecules by staining after permeabilization with saponin. As shown in Fig. 2GoB, saponin-permeabilized NIH3T3 cells infected with KbTMS stained positive for Kb using the Y3 Ab, indicating that these intracellular Kb molecules were properly folded. As expected, because the molecule encoded for by KbTMS is secreted, the level of staining was less than that observed using NIH3T3 cells infected with HTD12-10 virus. We were unable to detect the presence of Kb molecules in NIH3T3 cells infected with either KbCYT or AM12-Neor following permeabilization and staining with Y3 (not shown). To confirm that KbTMS-encoded molecules were secreted, AM12 cells expressing the KbTMS construct were metabolically labeled with [35S]methionine and Kb molecules were precipitated from cell culture supernatants using the Y3 Ab, as described (10). As shown in Fig. 2GoC, we were able to detect the presence of secreted Kb molecules in supernatants from cell lines expressing KbTMS, but not cell lines expressing wild-type Kb or control lines expressing Neor.

The inability of Y3 to detect Kb molecules encoded for by the KbCYT gene may arise because these molecules are trapped in the cytoplasm in an unfolded form and rapidly degraded. Therefore, we made use of the mAb Ku2, which recognizes Kb heavy chains in an unfolded state (10), to examine KbCYT-transduced cells for the presence of mutant Kb molecules. As shown in Fig. 2GoD, permeabilized NIH3T3 cells infected with KbCYT stained positive for Kb using the Ku2 Ab, indicating that the majority of these mutant Kb heavy chains exist in an unfolded state. Permeabilized NIH3T3 cells infected with HTD12-10 virus stained negative for Kb using the Ku2 Ab, consistent with the finding that the majority of normal Kb heavy chains are rapidly folded and assembled into productive complexes that are exported to the cell surface (10). Permeabilized NIH3T3 cells infected with KbTMS virus also stained negative for Kb using the Ku2 Ab (not shown).

Tolerization of CD4 and CD8 T cells by peptides through the indirect pathway occurs via distinct pathways of Ag presentation

To examine the ability of our mutant Kb molecules to induce tolerance when expressed in bone marrow-derived cells, bone marrow from 5-fluorouracil-treated B10.AKM mice was transduced with either HTD12-10, KbTMS, or KbCYT, and used to reconstitute lethally irradiated B10.AKM recipients. To ascertain whether the level of viral transduction achieved was equivalent in viruses encoding mutant and wild-type Kb proteins, lethally irradiated B10.AKM recipients were injected with a limiting dose of bone marrow cells transduced with either HTD12-10, KbTMS, or KbCYT. Twelve days after reconstitution, the mice were sacrificed and individual splenic colonies harvested and analyzed by RT-PCR to determine the percentage expressing wild-type or mutant Kb transcripts. The percentage of splenic colonies expressing wild-type Kb, KbTMS, and KbCYT was similar in all animals analyzed (greater than 70% in all cases, data not shown).

Four weeks after reconstitution, the mice received both Kb disparate B10.MBR and third-party B10.BR skin grafts. The B10.MBR strain was derived from a recombination event that occurred during the backcrossing of B10.AKM to C57BL/10, and differs from the B10.AKM strain in only the MHC class I H-2K region (20). Because of this MHC class I allelic disparity, B10.AKM mice are able to reject B10.MBR skin grafts rapidly. As expected based on previous work (3, 4), survival of B10.MBR skin grafts was significantly prolonged on B10.AKM mice-reconstituted HTD12-10 bone marrow (median survival time (MST) = 50 days, n = 7) when compared with survival of B10.BR third-party control skin grafts on the same mice (MST = 15 days, p < 0.001) (Fig. 3GoA). Survival of B10.MBR grafts was modestly prolonged on B10.AKM mice reconstituted with KbCYT-transduced bone marrow cells when compared with survival of B10.BR skin (MST = 32 vs 15 days, p = 0.003, n = 9) (Fig. 3GoC). In contrast, as shown in Fig. 3GoB, survival of Kb disparate B10.MBR skin was not prolonged on B10.AKM mice reconstituted with KbTMS-transduced bone marrow (MST = 21 vs 19 days, p = 0.3, n = 7).



View larger version (9K):
[in this window]
[in a new window]
  		FIGURE 3. Survival of B10.MBR and B10.BR skin grafts on B10.AKM mice reconstituted with transduced bone marrow. A, Mice reconstituted with HTD12-10-transduced bone marrow (p < 0.001, n = 7). B, Mice reconstituted with KbTMS-transduced bone marrow (p = 0.3, n = 7). C, Mice reconstituted with KbCYT-transduced bone marrow (p = 0.003, n = 9). Survival of B10.MBR ({blacktriangledown}) and third-party B10.BR ({blacksquare}) skin is shown. One representative experiment of three is shown. All experiments contained at least five mice per group and reached statistical significance.

 
Because secreted proteins are predicted to enter the MHC class II-processing compartment, we hypothesized that the CD4 T cells in mice receiving KbTMS-transduced bone marrow would be hyporesponsive to Kb disparate skin grafts. To test this hypothesis, lethally irradiated B10.AKM mice were reconstituted with either HTD12-10-, KbTMS-, or KbCYT-transduced syngeneic bone marrow. Four weeks after reconstitution, T cell recovery was confirmed by flow cytometry (data not shown), and the mice were then thymectomized and depleted of CD8 T cells by administration of the anti-CD8 mAb 2.43. CD8 T cell depletion in peripheral blood was confirmed by cell surface staining and flow cytometry (data not shown). As shown in Fig. 4Go, CD8 T cell-depleted B10.AKM mice reconstituted with either HTD12-10- or KbTMS-transduced marrow exhibited prolonged survival of B10.MBR grafts when compared with survival of B10.BR third-party control grafts (HTD12-10, MST = 62 vs 19 days, p < 0.001, n = 8; KbTMS, MST = 52 vs 32 days, p < 0.001, n = 5). There was no significant difference in the ability of wild-type Kb and KbTMS to prolong survival of B10.MBR grafts on CD8 T cell-depleted mice, suggesting that the amount of peptide generated that can be presented to CD4 T cells is the same in both cases. Importantly, we consistently observed that in approximately one-third of CD8 T cell-depleted mice reconstituted with HTD12-10 or KbTMS bone marrow, B10.MBR grafts survived long-term (>100 days), indicating that CD4 T cell tolerance was achieved. In contrast, CD8 T cell-depleted mice reconstituted with KbCYT-transduced marrow did not demonstrate prolonged survival of B10.MBR grafts when compared with third-party controls (MST = 21 vs 19 days, p = 0.7, n = 8). These data suggest that Kb-derived peptides can enter the MHC class II compartment either as soluble or cell surface proteins and affect alloreactive CD4 cells, but that proteins expressed only in the cytoplasm do not affect this cell population.



View larger version (10K):
[in this window]
[in a new window]
  		FIGURE 4. Survival of B10.MBR and B10.BR skin grafts on thymectomized and CD8 T cell-depleted B10. AKM mice reconstituted with transduced bone marrow. A, Mice reconstituted with HTD12-10-transduced bone marrow (p < 0.001, n = 8). B, Mice reconstituted with KbTMS-transduced bone marrow (p < 0.001, n = 5). C, Mice reconstituted with KbCYT-transduced bone marrow (p = 0.7, n = 8). Survival of B10.MBR ({blacktriangledown}) and third-party B10.BR ({blacksquare}) skin is shown. One representative experiment of three is shown. All experiments contained at least five mice per group and reached statistical significance.

 
A subset of CD8 T cells recognizes MHC class I alloantigen presented through the indirect pathway

From our initial experiments, it appeared that B10.MBR skin graft survival was modestly prolonged on B10.AKM mice reconstituted with KbCYT-transduced bone marrow cells (see Fig. 3Go). However, because expression of the KbCYT-encoded molecule in bone marrow-derived cells did not induce CD4 T cell tolerance (Fig. 4GoC), we hypothesized the effect of expression of the molecule encoded for by this construct on CD8 T cells could be obscured by CD4 T cell-mediated rejection. Therefore, to determine the effect of mutant Kb molecules on CD8 T cells, lethally irradiated B10.AKM mice were reconstituted with either HTD12-10-, KbTMS-, or KbCYT-transduced syngeneic bone marrow. Four weeks after reconstitution, the mice were thymectomized and depleted of CD4 T cells by administration of the anti-CD4 mAb GK1.5. Depletion of CD4 T cells was confirmed as described above (data not shown). As shown in Fig. 5Go, CD4 T cell-depleted B10.AKM mice reconstituted with HTD12-10- or KbCYT-transduced bone marrow showed modestly prolonged survival of B10.MBR skin grafts compared with third-party B10.BR control grafts (MST = 20 vs 12 days, p = 0.004, n = 7, and MST = 29 vs 15 days, p = 0.003, respectively, n = 6). In contrast, CD4 T cell-depleted B10.AKM mice reconstituted with KbTMS-transduced bone marrow did not exhibit prolonged survival of B10.MBR grafts when compared with third-party controls (MST = 21 vs 19 days, p = 0.7, n = 8). These data suggest that peptides generated in the cytoplasm can enter the indirect pathway and affect the ability of a subset of CD8 T cells that recognize Kb-derived peptides to mediate graft rejection.



View larger version (10K):
[in this window]
[in a new window]
  		FIGURE 5. Survival of B10.MBR and B10.BR skin grafts on thymectomized and CD4 T cell-depleted B10.AKM mice reconstituted with transduced bone marrow. A, Mice reconstituted with HTD12-10-transduced bone marrow (p = 0.004, n = 7). B, Mice reconstituted with KbTMS-transduced bone marrow (p = 0.7, n = 8). C, Mice reconstituted with KbCYT-transduced bone marrow (p = 0.003, n = 6). Survival of B10.MBR ({blacktriangledown}) and third-party B10.BR ({blacksquare}) skin is shown. One representative experiment of three is shown. All experiments contained at least five mice per group and reached statistical significance.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Induction of allograft hyporesponsiveness by genetic engineering of bone marrow to express allogeneic MHC class I molecules is controlled at the level of T cell help, and both MHC class I- and class II-restricted Th cells are affected (4, 5). One way in which a deficit in T cell help could be induced is by presentation of peptides derived from the transduced Kb gene product through the indirect pathway to host T cells by self MHC. In this study, we used mutant Kb molecules, which must enter the indirect pathway to be presented to T cells, to induce hyporesponsiveness. This approach allowed us to define the extent to which peptide recognition contributes to the induction of both CD8 and CD4 T cell tolerance to MHC class I-mismatched allografts.

Based on our analysis of mutant Kb alloantigens using conformation-sensitive Abs, we conclude that the KbTMS and KbCYT molecules exist in distinct cellular compartments as secreted or cytoplasmic molecules, respectively. Because neither is expressed on the cell surface, peptides derived from these molecules cannot be presented via the direct pathway, but are instead presented to T cells indirectly. Because these mutants exist as secreted or cytoplasmic proteins, KbTMS- and KbCYT-encoded class I heavy chain mutants are likely to be processed by MHC class II or class I machinery, respectively, before presentation to T cell through the indirect pathway. It is unlikely that peptides derived from either the transmembrane or cytoplasmic domains of Kb contribute to induction of hyporesponsiveness because GPI-anchored Kb mutants lacking transmembrane and cytoplasmic domains are able to induce hyporesponsiveness when expressed in bone marrow to a degree similar to that observed using wild-type Kb (J.I., unpublished data).

Mice reconstituted with KbCYT-transduced bone marrow exhibited modest prolongation of class I-mismatched skin grafts. Based on the analysis of skin graft survival using CD8 or CD4 T cell-depleted mice, our data suggest that peptides derived from the KbCYT-encoded heavy chains induce hyporesponsiveness in CD8, but not CD4 T cells. This finding further demonstrates that during allograft rejection, a population of CD8 T cells that recognize processed MHC class I alloantigen exists and must be tolerized to prevent graft rejection. Interestingly, our data also indicate that peptides generated from KbCYT are unable to generate CD4 T cell hyporesponsiveness, a finding consistent with what is known about Ag processing (21). Thus, expression of cytoplasmic Ags in bone marrow-derived cells can only induce CD8 T cell hyporesponsiveness.

Previous in vitro studies using T cells from transgenic mice engineered to express a secreted allogeneic MHC class I molecule have shown that alloreactive CTL precursors are able to kill targets through the direct pathway if exogenous IL-2 is present in the cultures even though MHC class I peptide-reactive T cells appear to be absent (22). However, in this study, the effect of expressing a soluble allogeneic MHC class I transgene on allograft survival was not examined. In our model, mice reconstituted with bone marrow expressing secreted Kb exhibited prolongation of class I-mismatched skin grafts only when depleted of CD8 T cells. These data suggest that, in this system, soluble proteins predominantly enter the indirect pathway through the MHC class II Ag-processing compartment and affect alloreactive CD4 T cells. Although it has been shown that exogenous Ags can enter the MHC class I compartment (23), in our system we observed no evidence that class I-restricted CD8 T cells were affected in mice reconstituted with KbTMS-transduced bone marrow. We suggest that delivery of Kb as a secreted molecule without additional inflammatory stimuli may preclude activation of APCs and processing through the MHC class I compartment. Previous studies have indicated that cellular destruction potentiates cross-presentation of exogenous peptide through MHC class I (24, 25). A lack of cellular destruction in our model may prevent presentation of peptides through the indirect pathway. Alternatively, APCs that can process exogenous proteins for presentation of MHC class I may be unable to induce T cell hyporesponsiveness.

Interestingly, in CD8 T cell-depleted mice, wild-type and KbTMS heavy chains were able to induce the same degree of hyporesponsiveness to MHC class I-mismatched allografts. Thus, both wild-type and secreted Kb heavy chains can induce CD4 T cell hyporesponsiveness when expressed in bone marrow-derived cells. Given what is known about processing of cell surface Ags and exogenous proteins that are destined to be presented to T cell by MHC class II molecules, it is likely that wild-type Kb and KbTMS enter the MHC-processing pathway through distinct mechanisms. Nevertheless, both pathways appear to lead to efficient presentation of Kb-derived peptides to CD4 T cells. Importantly, in both cases, we were able to reproducibly observe that one-third of the mice in each group accepted a Kb disparate B10.MBR graft long-term, even though third-party control grafts on the same mice were rapidly rejected. We suggest that this result indicates that long-term CD4 T cell tolerance can be achieved using gene therapy.

To achieve lifelong allograft survival without immunosuppression, tolerance through the direct and indirect pathways must be achieved. While adequate levels of MHC class I cell surface expression will probably be able to tolerize CD8 T cells that recognize Kb through the direct pathway, to induce tolerance to Kb-derived peptides presented through the indirect pathway Ag must enter the MHC class I and class II Ag presentation pathways to affect CD8 and CD4 T cells, respectively. One limitation to achieving lifelong tolerance to allografts by gene therapy is the requirement for long-term expression of retrovirally transduced MHC genes in bone marrow cells. However, based on our data, we also suggest that the ability to efficiently deliver peptides to each Ag presentation compartment may also be a limiting factor. For example, while wild-type Kb is clearly able to generate the appropriate peptides to induce tolerance in CD4 T cells as well as CD8 T cells that recognize Kb directly, it is not clear whether it can give rise to peptides that can enter the MHC class I compartment and induce tolerance in CD8 T cells using the indirect pathway, a possibility we are currently investigating. We suggest that vectors designed to deliver Ags to the indirect as well as direct pathways may make it possible to achieve lifelong tolerance to allografts using gene therapy.


   Acknowledgments
 
We thank Dr. Hidde Ploegh for his gift of the Ku2 mAb, Dr. Tokihido Sawada for assistance with skin grafts, and members of the Iacomini laboratory for helpful discussions. In addition, we thank Drs. Gilles Benichou and Gerry Waneck for critical review of the manuscript.


   Footnotes
 
1 This work was supported by Grant RO1 AI43619 from the National Institutes of Health (to J.I.). Back

2 Address correspondence and reprint requests to Dr. John Iacomini, Transplantation Biology Research Center, Massachusetts General Hospital, MGH-East, Building 149, 13th Street, Boston, MA 02129. Back

3 Abbreviations used in this paper: Kb, H-2Kb; MST, median survival time. Back

Received for publication March 31, 2000. Accepted for publication August 1, 2000.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Ildstad, S. T., D. H. Sachs. 1984. Reconstitution with syngeneic plus allogeneic or xenogeneic bone marrow leads to specific acceptance of allografts or xenografts. Nature 307:168.[Medline]
  2. Bagley, J., J. Iacomini. 1998. Gene therapy in organ transplantation. Cancer Res. Therapy Control 7:33.
  3. Sykes, M., D. H. Sachs, A. W. Nienhuis, D. A. Pearson, A. D. Moulton, D. M. Bodine. 1993. Specific prolongation of skin graft survival following retroviral transduction of bone marrow with an allogeneic major histocompatibility complex gene. Transplantation 55:197.[Medline]
  4. Fraser, C. C., M. Sykes, R. Stanton-Lee, D. H. Sachs, C. LeGuern. 1995. Specific unresponsiveness to a retrovirally transferred class I antigen is controlled through the helper pathway. J. Immunol. 154:1587.[Abstract]
  5. Mayfield, R. S., H. Hayashi, T. Sawada, K. Bergen, C. LeGuern, M. Sykes, D. H. Sachs, J. Iacomini. 1997. The mechanism of specific prolongation of class I mismatched skin grafts induced by retroviral gene therapy. Eur. J. Immunol. 27:1177.[Medline]
  6. Rosenberg, A. S., A. Singer. 1992. Cellular basis of skin allograft rejection: an in vivo model of immune-mediated tissue destruction. Annu. Rev. Immunol. 10:333.[Medline]
  7. Jr Auchincloss, H., H. Sultan. 1996. Antigen processing and presentation in transplantation. Curr. Opin. Immunol. 8:681.[Medline]
  8. Sawada, T., Y. Wu, D. H. Sachs, J. Iacomini. 1997. CD4+ T cells are able to reject class I disparate allografts. Transplantation 64:335.[Medline]
  9. Hammerling, G. J., E. Rusch, N. Tada, S. Kimura, U. Hammerling. 1982. Localization of allodeterminants on H-2Kb antigens determined with monoclonal antibodies and H-2 mutant mice. Proc. Natl. Acad. Sci. USA 79:4737.[Abstract/Free Full Text]
  10. MacHold, R. P., H. L. Ploegh. 1996. Intermediates in the assembly and degradation of class I major histocompatibility complex (MHC) molecules probed with free heavy chain-specific monoclonal antibodies. J. Exp. Med. 184:2251.[Abstract/Free Full Text]
  11. Loken, M., A. Stall. 1982. Flow cytometry as an analytical and preparative tool in immunology. J. Immunol. Methods 50:R85.[Medline]
  12. Sarmiento, M., A. L. Glasebrook, F. W. Fitch. 1980. IgG or IgM monoclonal antibodies reactive with different determinants on the molecular complex bearing Lyt2 antigen block T cell-mediated cytolysis in the absence of complement. J. Immunol. 125:2665.[Abstract]
  13. Dialynas, D. P., Z. S. Quan, K. A. Wall, A. Pierres, J. Quintans, M. R. Loken, M. Pierres, F. W. Fitch. 1984. Characterization of the murine T cell surface molecule designated L3T4, identified by monoclonal antibody GK1.5: similarity of L3T4 to the human Leu 3/T4 molecule. J. Immunol. 131:2445.[Abstract]
  14. Ildstad, S. T., S. M. Wren, J. A. Bluestone, S. A. Barbieri, D. H. Sachs. 1985. Characterization of mixed allogeneic chimeras: immunocompetence, in vitro reactivity, and genetic specificity of tolerance. J. Exp. Med. 162:231.[Abstract/Free Full Text]
  15. Armentano, D., S. F. Yu, P. W. Kantoff, T. VonRuden, W. F. Anderson, E. Gilboa. 1987. Effect of internal viral sequences on the utility of retroviral vectors. J. Virol. 61:1647.[Abstract/Free Full Text]
  16. Bagley, J., K. Aboody-Guterman, X. Breakefield, J. Iacomini. 1998. Long-term expression of the gene encoding green fluorescent protein in murine hematopoietic cells using retroviral gene transfer. Transplantation 65:1233.[Medline]
  17. Markovitz, D., S. Goff, A. Bank. 1988. Construction and use of a safe and efficient amphotropic packaging cell line. Virology 167:400.[Medline]
  18. Germain, R. N.. 1999. Antigen processing and presentation. W. E. Paul, ed. Fundamental Immunology 4th Ed.287. Lippincott-Raven, Philadelphia.
  19. Hayashi, H., R. S. Lee, S. Germana, C. Fraser, M. Sykes, D. H. Sachs, C. LeGuern. 1995. Retroviral vectors for long-term expression of allogeneic major histocompatibility complex transduced into syngeneic bone marrow cells. Transplant. Proc. 27:179.
  20. Sachs, D. H., J. S. Arn, T. H. Hansen. 1979. Two new recombinant H-2 haplotypes, one of which juxtaposes Kb and Ik alleles. J. Immunol. 123:1965.[Abstract/Free Full Text]
  21. Watts, C.. 1997. Capture and processing of exogenous antigens for presentation on MHC molecules. Annu. Rev. Immunol. 15:821.[Medline]
  22. Arnold, B., M. Messerle, L. Jatsch, G. Küblbeck, U. Koszinowski. 1990. Transgenic mice expressing a soluble foreign H-2 class I antigen are tolerant to allogeneic fragments presented by self class I but not to the whole membrane-bound alloantigen. Proc. Natl. Acad. Sci. USA 87:1762.[Abstract/Free Full Text]
  23. Bevan, M. J.. 1976. Cross-priming for a secondary cytotoxic response to minor H antigens with H-2 congenic cells which do not cross react in the cytotoxic assay. J. Exp. Med. 143:1283.[Abstract/Free Full Text]
  24. Albert, M. L., B. Sauter, N. Bhardwaj. 1998. Dendritic cells acquire antigen from apoptotic cells and induce class I-restricted CTLs. Nature 392:86.[Medline]
  25. Kurts, C., J. F. A. P. Miller, R. M. Subramaniam, F. R. Carbone, W. R. Heath. 1998. Major histocompatibility complex class I-restricted cross-presentation is biased towards high dose antigens and those released during cellular destruction. J. Exp. Med. 188:409.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 BloodHome page
J. Bagley, C. Tian, D. H. Sachs, and J. Iacomini
Induction of T-cell tolerance to an MHC class I alloantigen by gene therapy
Blood, May 29, 2002; 99(12): 4394 - 4399.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bagley, J.
Right arrow Articles by Iacomini, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bagley, J.
Right arrow Articles by Iacomini, J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS