Targeted T-Cell Depletion or CD154 Blockade Generates Mixed Hemopoietic Chimerism and Donor-Specific Tolerance in Mice Treated with Sirolimus and Donor Bone Marrow

Abstract

Background. 

The administration of donor specific bone marrow (DSBM) to mice conditioned with antilymphocyte serum (ALS) and sirolimus can result in stable multilineage mixed chimerism and long-term graft survival. This study seeks to determine if either the targeted depletion of CD4 and/or CD8 pos T cells or costimulation blockade can substitute for ALS and preserve the efficacy of this regimen.

Methods. 

C57BL/6 recipients of BALB/c skin allografts were treated with DSBM (150× 10⁶ cells), sirolimus (24 mg/kg intraperitonealy), and either ALS or various monoclonal antibodies (αCD4, αCD8, αCD154 alone or in combination). Recipient peripheral blood mononuclear cell (PBMC) depletion, donor chimerism, and deletion of donor reactive T cells were assessed using flow cytometry. The specificity of immunologic nonreactivity and the presence of immunoregulatory activity were assessed through a mixed lymphocyte reaction assay.

Results. 

The administration of ALS, sirolimus, and DSBM resulted in sustained recipient PBMC depletion, transient chimerism, and prolonged graft survival. The substitution of an equivalent degree and duration of targeted depletion of either CD4 or CD8 pos T cells alone for ALS failed to produce chimerism or prolonged graft survival. In contrast, depletion of both CD4 and CD8 pos T cells resulted in durable multilineage chimerism, indefinite allograft acceptance (>350 days), and donor-specific tolerance to secondary skin grafts. Substitution of αCD154 monoclonal antibody for ALS also resulted in a state of mixed chimerism and donor specific tolerance. This tolerant state appears to be maintained at least partially through clonal deletion and suppression.

Conclusion. 

Either combined CD4 and CD8 T-cell depletion or αCD154 blockade can effectively substitute for ALS in producing chimerism and tolerance in this model.

Donor specific tolerance and multilineage chimerism can be produced in a major MHC mismatched murine skin allograft model by conditioning the recipient with antilymphocyte serum (ALS), sirolimus, and donor bone marrow (DSBM) (¹). In addition, it has been determined that this state of tolerance is maintained predominantly through the same mechanism of clonal deletion that supports radiation based chimerism models (²). However, the relative contributions of each of the components of the regimen to the development of tolerance and chimerism remain undefined. In particular, the manner in which ALS contributes to the generation of chimerism in this model is unknown.

The rabbit polyclonal antilymphocyte serum originally used in this protocol is produced by immunizing rabbits with a single-cell suspension obtained from mouse lymphocytes. The use of ALS as an immunosuppressive induction agent was first reported in the 1960s when it was successfully used to prolong kidney allograft survival in a variety of preclinical models (^3–7). Since then, polyclonal antilymphocyte preparations have been used in a variety of clinical scenarios including induction immunosuppression, treatment of acute allograft rejection, graft-versus-host disease prophylaxis, and treatment of autoimmune disorders (^8–10).

Polyclonal antilymphocyte serum is a complex mixture of antilymphocyte antibodies with a multitude of specificities (¹¹). While the depletion of circulating lymphocytes is probably responsible for most of the immunosuppressive effects of ALS, receptor modulation and blockade of cell surface molecules (costimulation, adhesion, and MHC) can also contribute (¹²). The aim of the present investigation is to define the time course and magnitude of peripheral blood mononuclear cell depletion following the administration of ALS. Once defined, this study seeks to determine whether the inhibition of lymphocyte activity (either through depletion or costimulation blockade) is predominantly responsible for the therapeutic effects of polyclonal antilymphocyte serum and whether these effects are necessary and sufficient to produce donor multilineage chimerism and donor-specific tolerance in mice also treated with sirolimus and donor bone marrow.

METHODS

Animals

Male C57BL/6 (H-2^b), BALB/c (H-2^d), and C3H/HeJ (H-2^k) mice were purchased from the National Cancer Institute (Frederick, MD). Mice were used at 8–12 weeks of age. All care and handling of animals was carried out in accordance with guidelines provided in the Guide for the Care and Use of Laboratory Animals published by the U.S. Department of Health and Human Services.

Skin Grafting

Full thickness donor (for primary grafts) or donor and third party (for secondary grafts) skin grafts were transplanted on to the lateral thoracic wall of the recipients using standard techniques described previously (¹³). Skin grafts were scored as described by Billingham (¹⁴). Briefly, skin grafts were monitored daily by visual and tactile inspection after the removal of the bandage on day 7. The percent of viable donor epithelium remaining was recorded for each day. Rejection was defined as complete loss of viable donor epithelium.

Immunosuppression

Rabbit antimouse ALS (generously provided by Dr. A. P. Monaco, Boston, MA) was produced by injecting mouse lymphocytes in combination with complete Freund’s adjuvant subcutaneously into rabbits as previously described (¹⁵).

Sirolimus (generously provided by Wyeth Ayerst, Princeton, NJ) was suspended in a carboxymethyl cellulose and polysorbate vehicle as described previously and administered as a single dose (24 mg/kg intraperitonealy [IP] on day 6) (¹⁶).

Bone marrow cells were harvested from the long bones of donor mice by injecting the marrow cavity with 4°C Hanks’ balanced salt solution (HBSS). The cell suspension was filtered through a fine mesh and washed three times.

Our standard ALS/sirolimus/DSBM conditioning regimen for tolerance induction was utilized. This consists of the administration of polyclonal rabbit antimouse ALS (0.5 mL of 4 mg IgG/mL) on days −1, +2, and +5 relative to skin grafting on day 0. In experiments in which monoclonal antibodies (mAbs) were used in lieu of ALS, 0.2 mg of mAb(s) were administered in 6 doses (days −1, +2, +5, +7, +9, and +14 relative to skin grafting on day 0, respectively). When depleting monoclonal antibodies were used, this dosing schedule was shown to result in a similar level and duration of depletion of CD4 and/or CD8 positive T cells as that seen following the administration of ALS. Freshly harvested DSBM cells (150× 10⁶ viable mononuclear cells) were administered intravenously on day 7.

Monoclonal Antibodies

The following mAbs were produced by culture in Integra flasks (Argos Technologies, East Dundee, IL) with IgG depleted fetal calf serum and purified through protein G column in the core facility of our laboratory: YTS 191.1 [rat IgG2b directed towards murine CD4] (hybridoma provided by H. Waldmann, Oxford, UK), YTS 169.4 [rat IgG2b directed toward murine CD8α] (hybridoma provided by H. Waldmann, Oxford, UK), and MR1 [Syrian Hamster IgG directed toward murine CD154] (American Type Tissue Culture, Chantilly, VA). Rat IgG (Rockland Immunochemicals, Gilbertsville, PA) and Syrian Hamster IgG (Rockland Immunochemicals, Gilbertsville, PA) were used as irrelevant controls.

Flow Cytometric Analysis

Peripheral blood was collected in a heparinized tube via tail vein bleed. The blood was diluted 3:1 with HBSS, then placed in a centrifuge tube with 5 mL of Lymphocyte-m (Cedarlane Laboratories Ltd., Hornby, Ontario, Canada), and centrifuged for 20 min at 1500×g at room temperature. The lymphocytes were then harvested and washed three times.

Chimeric cells were evaluated by flow cytometric analysis of peripheral white blood cells (WBC). Flow cytometric analysis was performed on a FACScan (Becton Dickinson, Mountain View, CA). The cells were first incubated with an anti-CD16/32 monoclonal antibody for 10 min to block nonspecific FcγR binding of labeled Abs. Cells were then stained with the following: PE conjugated anti-CD3, FITC conjugated anti-CD4, CD8, CD19, PAN-NK, CD11b, Vβ11, Vβ5.1/2, Vβ8.1/2, and cychrome conjugated biotinylated anti-H-2D^d monoclonal antibodies (PharMingen, San Diego, CA). The cells were incubated with the conjugated antibodies for 30 min, then washed twice, and fixed in paraformaldehyde. Two- and three-color data were displayed as dot plot diagrams. PE and FITC conjugated and biotinylated isotype antibodies (Hamster IgG, Rat IgG2a, Rat IgM, Rat IgG2b, Mouse IgG1, and Mouse IgG2a) were used as irrelevant controls. The determination of the percent of chimerism present was made by comparison with the percentage obtained from normal donor and host-type animals, which were used as positive and negative controls. The relative percentage of staining of a chimera with anti-H2D^d mAb was calculated by using the formula: 100% × (net chimera percent positive) − (net negative control percent positive)/(net positive control percent positive) − (net negative control percent positive) where “net” percent positive refers to the percentage obtained after subtraction of staining with the appropriate isotype control.

The absolute counts of peripheral blood mononuclear cells were determined on posttreatment days 0, 7, 14, 30, 60, and 90 by flow cytometry and coulter counter.

Clonal Deletion Assay

The described experiments take advantage of the knowledge that certain murine T-cell receptor (TCR) Vβ chains are recognized by endogenous superantigens that require I-E for their expression. This information was applied to a transplant combination in which the recipient strain (C57BL/6, I-E⁻) does not express the superantigens, whereas the donor strain (BALB/c, I-E⁺) expresses I-E and the superantigens that can effect the deletion of Vβ5.1/5.2 or Vβ11 chain bearing T cells during thymic maturation or in the periphery.

To determine the percentage of recipient (C57BL/6) derived T cells that are Vβ5.1/5.2, Vβ11, or Vβ8.1/8.2 positive on posttransplantation day 250, we analyzed 10,000 gated CD3⁺ cells. The percentage of these gated cells staining with isotype controls was subtracted from the percentage of these gated cells staining with the individual anti-Vβ monoclonal antibody. Finally, the absolute count of the pertinent Vβ-bearing CD3⁺ T cells was determined.

Allogenic Mixed Leukocyte Reactions and Regulatory Cell Assay

Spleens were removed and minced into a single-cell suspension in HBSS. The cells were washed three times and resuspended in RPMI with 10% fetal calf serum supplemented with 2 mM L-Glutamine, 100 U/mL penicillin-streptomycin, 20 mM HEPES, and 50 μM 2-mercaptoethanol at a concentration of 6× 10⁶ cells/mL. Sixty microliters (3.6× 10⁵ cells) each of the responder (age-matched naïve and tolerant C57Bl/6) and irradiated (3,000 rad; Nordion Int. Inc., Ontario, Canada) stimulator (C57Bl/6, BALB/c or C3H/HeJ) cell suspension were added in triplicate to a round-bottomed 96-well tissue culture plate. Splenocytes from tolerant recipients (putative regulatory cells) were harvested at posttransplantation day 350 and added to the cultures in multiple fixed responder cell:regulatory cell ratios (3:1, 3:2, 1:1). To control for possible effects of cell crowding on proliferation, these experiments were repeated using naïve responder splenocytes in place of splenocytes from tolerant mice. The reactions were incubated at 37°C and 5% CO₂ for 72 hr, at which time proliferation assays were performed by pulsing the wells with 1 μCi of treated thymidine. The cells were harvested after 8 hr of incubation and prepared for scintillation counting.

Statistics

Vβ expression, MLR, and regulatory assay data was analyzed with univariate ANOVA using SPSS 10.0 for Windows (SPSS Inc, Chicago, IL).

RESULTS

ALS Mediated Depletion, Chimerism, and Graft Survival

C57Bl/6 recipients of BALB/c skin grafts treated with ALS, sirolimus, and donor bone marrow demonstrated a sustained depletion of CD4 T cells, CD8 T cells, B cells, NK cells, and monocytes in the peripheral blood as demonstrated in Figure 1A. This depletion peaked 7 days following the first dose of ALS. These cell populations had recovered to approximately half their baseline levels by 30 days following the first dose of ALS. Full reconstitution was achieved by 90 days following the first dose of ALS.

FIGURE 1.: Time course of peripheral blood leukocyte depletion (A) and allograft survival (B). Data points represent the mean level as determined from six separate mice at each time point. Error bars represent standard deviation. C57BL/6 recipients (n=6) of BALB/c skin grafts received three doses of ALS (IgG) (2 mg IP days -1, 2, 5), sirolimus (24 mg/kg IP day 6), and DSBM alone (150× 10⁶ cells IV day 7).

These mice demonstrated a transient period (<120 days) of donor specific multilineage chimerism peaking at less than 15% of peripheral blood mononuclear cells (Table 1). The transient observed chimerism consisted predominantly of monocytes, B cells, and NK cells. T-cell chimerism was in the 2%–3% range. This transient period of chimerism was associated with transient skin graft survival. Median graft survival in mice treated with ALS, sirolimus, and DSBM was 126 days. No chimerism was detectable following graft rejection. Median graft survival in control mice and mice treated with DSBM, sirolimus, ALS, DSBM + sirolimus, ALS + sirolimus, and ALS + DSBM was 8, 11, 23, 35, 23, 45, and 47 days, respectively (Fig. 1B).

TABLE 1: Mean percent chimerism

Isolated CD4 or CD8 Depletion Fails to Produce Chimerism and Long-Term Graft Survival

Substitution of αCD4 or αCD8 monoclonal antibody for ALS in C57Bl/6 recipients of BALB/c skin grafts treated with sirolimus and donor bone marrow resulted in a sustained depletion of either CD4 or CD8 T cells as demonstrated in Figures 2A and 2B. Chimerism was not detected at any time point before rejection of the skin grafts. The median graft survival for mice depleted of either CD4 or CD8 T cells was 75 days and 49 days, respectively, as shown in Figure 2C.

FIGURE 2.: Time course of peripheral blood CD4 (A) or CD8 (B) cell depletion and αCD4 or αCD8 mediated skin graft survival (C). C57BL/6 recipients (n=6) of BALB/c skin grafts received six doses (0.2 mg IP days -1, 2, 5, 7, 9, 14) of either αCD4, αCD8 mAb, or rat IgG (0.2 mg IP days -1, 2, 5, 7, 9, 14). All recipients received sirolimus (24 mg/kg IP day 6), and DSBM (150× 10⁶ cells IV day 7). Data points represent the mean level as determined from six separate mice at each time point. Error bars represent standard deviation.

Combined CD4 and CD8 Depletion Produces Chimerism and Tolerance

Substitution of the combination of αCD4 and αCD8 monoclonal antibodies for ALS in C57Bl/6 recipients of BALB/c skin grafts treated with sirolimus and donor bone marrow resulted in a sustained depletion of CD4 and CD8 T cells as demonstrated in Figure 3A. The combined administration of the αCD4 and αCD8 monoclonal antibodies resulted in a greater than 95% depletion of the respective cell populations during treatment. Both cell populations recovered to within 50% of pretreatment levels by day 30 and were fully recovered by 90 days.

FIGURE 3.: Time course of combined peripheral blood CD4 and CD8 cell depletion (A). C57BL/6 mice (n=6) received six doses of a cocktail of αCD4 and αCD8 mAb (0.2 mg each IP days -1, 2, 5, 7, 9, 14), sirolimus (24 mg/kg IP day 6), and DSBM (150× 10⁶ cells IV day 7). Data points represent the mean level as determined from six separate mice at each time point. Error bars represent standard deviation. Summary of primary allograft (B), and secondary donor, and third party graft (C) survival in C57BL/6 recipient mice of BALB/c skin graft treated as described. Secondary skin transplantation with donor and third party graft (C3H) was completed after 250 days of primary graft survival.

These mice demonstrated durable (>250 days) donor specific multilineage chimerism as demonstrated in Table 1. Overall peripheral blood mononuclear cell chimerism stabilized at a level of approximately 20% within 60 days of bone marrow infusion. This chimerism was mainly restricted to the monocytes, B cells, and NK cells as was the case in mice treated with ALS. The levels of chimerism of these lineages were 50%–100% higher than those seen in ALS treated mice however. T-cell chimerism stabilized in the 2%–3% range, equivalent to that observed in the ALS treated group.

Primary graft survival as shown in Figure 3B was indefinite (>350 days) in comparison to median graft survival of 25 days for mice receiving isotype control antibodies in lieu of αCD4 and αCD8 monoclonal antibodies. To determine the specificity of the state of immunologic nonresponsiveness in mice bearing long-term skin grafts, secondary skin grafts were placed over 200 days following donor marrow infusion. The donor specific secondary skin graft survival was 100% (>100 days), whereas the median graft survival for third party skin grafts was only 11 days (Fig. 3C).

Substitution of CD154 Blockade for Lymphocyte Depletion

We next sought to determine if CD4 T-cell inhibition through blockade of the CD40-CD154 costimulation pathway could substitute for T-cell depletion in this model and permit the induction of multilineage chimerism. The αCD154 monoclonal antibody MR1 was given at a dose of 0.2 mg IP. Administration of αCD154 monoclonal antibody for ALS in C57Bl/6 recipients of BALB/c skin grafts treated with sirolimus and donor bone marrow was not associated with depletion of T cells, B cells, NK cells, or monocytes as demonstrated in Figure 4A. This dose was administered to mice at the same six time points used for depletional antibodies. All mice treated with the αCD154 antibody, sirolimus, and donor bone marrow demonstrated multilineage chimerism. The phenotype and kinetics of observed chimerism in this model was roughly comparable with that achieved with the combination of αCD4 and αCD8 as shown in Table 1. Overall peripheral blood mononuclear cell chimerism stabilized at a level of approximately 20% within 60 days of bone marrow infusion. As observed in the mice treated with the combination of αCD4 and αCD8 antibodies, T-cell chimerism stabilized in the 2%–3% range.

FIGURE 4.: Level of peripheral blood mononuclear cell depletion. The relative level of PBMC depletion (A) present over time in mice treated with six doses of αCD154 mAb (0.2 mg IP days -1, 2, 5, 7, 9, 14), sirolimus (24 mg/kg IP day 6), and DSBM (150× 10⁶ cells IV day 7). Data points represent the mean level as determined from six separate mice at each time point. Error bars represent standard deviation. Summary of primary allograft (B), second allograft, and third party graft (C) survival in C57BL/6 recipient mice of BALB/c skin graft treated with six doses of αCD154 mAb (0.2 mg IP days -1, 2, 5, 7, 9, 14) or hamster IgG (0.2 mg IP days -1, 2, 5, 7, 9, 14), sirolimus (24 mg/kg IP day 6), and DSBM (150× 10⁶ cells IV day 7). Secondary skin transplantation with donor and third party graft (C3H) was completed after 250 days of primary graft survival.

Primary graft survival was indefinite (>350 days) compared with the median graft survival of 24 days in mice treated with an isotype control antibody in lieu of αCD154 (Fig. 4B). Secondary skin grafts were placed over 200 days following donor marrow infusion. The donor specific secondary skin graft survival was 100% (>100 days), whereas the median graft survival for third party skin grafts was only 13 days (Fig. 4C).

Mechanism of Tolerance

To determine whether the donor-specific unresponsive state was associated with selective deletion of donor-reactive T cells, we compared the absolute number of Vβ11, Vβ5.1/5.2, and Vβ8.1/8.2 bearing CD3+ T cells in PBMC from C57BL/6 recipients in the experimental group at >250 days after transplantation and from naïve control groups. BALB/c mice delete Vβ11 and Vβ5.1/5.2 bearing T cells in the thymus due to their high affinity for endogenous retroviral superantigens (mouse mammary tumor virus, MMTV) presented by I-E MHC class II molecules. C57BL/6 mice do not express I-E and express Vβ11 on ∼4%–5% of CD4+ T cells and Vβ5.1/5.2 on ∼2%–3% of CD4+ T cells (^17,18). As anticipated, control C57BL/6 mice failed to delete donor (BALB/c)-reactive Vβ11+ or Vβ5+ T cells (Table 2). In contrast, tolerant recipients of BALB/c marrow demonstrated near complete deletion of CD3+ Vβ11+ and CD3+ Vβ5+ T cells. For comparison, we also determined the absolute number of Vβ8-bearing CD3+ T cells, which are expressed on ∼15%–20% of BALB/c and C57BL/6 CD4+ T cells. C57BL/6 mice in the experimental group demonstrated a preserved population of Vβ8+CD3+ T cells, indicating that the T cell deletion was donor specific in nature.

TABLE 2: Clonal deletion

To determine whether lymphocytes harvested from tolerant mice were capable of responding against donor and/or third party stimulators, we performed one-way mixed lymphocyte reactions. At >300 days after transplantation, single-cell suspensions were prepared from the spleens of mice from each experimental group. Although splenocytes from mice treated with either αCD4 + αCD8 or αCD154 antibodies generated high responses to C3H (third party) stimulators, the response to BALB/c stimulators was diminished to approximately the level of the syngeneic response (Figs. 5A and 5B).

FIGURE 5.: Analysis of the splenocytes from the tolerant recipients in mixed lymphocyte reactions was performed approximately 350 days after initial transplantation. Splenocytes from C57BL/6 mice treated with BALB/c bone marrow, sirolimus, and either a cocktail of αCD4 and αCD8 (A) or αCD154 (B) generated high responses to third party C3H stimulators. In contrast, the response to BALB/c stimulators was diminished to the level of the syngeneic C57BL/6 response. Splenocytes from control C57BL/6 mice in each experiment generated vigorous responses to both BALB/c and C3H stimulators as compared with syngeneic C57BL/6 stimulators. * Indicates P<0.05 (ANOVA) compared with C57BL/6 (syngeneic) stimulators. n=6 for each experiment.

To determine whether splenocytes from mice rendered tolerant using either αCD4 + αCD8 or αCD154 antibodies possessed immunoregulatory properties in vitro we conducted coculture experiments using splenocytes from tolerant mice (>350 days following donor marrow infusion) as putative regulatory cells. A specific suppressor-like effect was noted when splenocytes from tolerant C57BL/6 mice treated with αCD4 and αCD8 monoclonal antibodies were added to the cultures of naïve C57BL/6 responder splenocytes and either BALB or C3H irradiated stimulator splenocytes. In this experimental system, the addition of splenocytes from tolerant mice reduced lymphocyte proliferation nonspecifically in a dose-dependent fashion (Figs. 6A-D). While the same trend was noted in mice treated with αCD154 monoclonal antibody, it appeared to be less specific and it was not statistically significant.

FIGURE 6.: Regulatory cell coculture assay. Splenocytes were harvested from mice treated with a cocktail of αCD4 and αCD8 (A and B) or αCD154 alone (C and D) (days -1, 2, 5, 7, 9, 14), sirolimus at 24 mg/kg, and 150× 10⁶ donor bone marrow cells at 350 days after skin grafting on day 0. The splenocytes were incubated with naïve responder cells at fixed ratios of 1:3, 2:3, and 1:1 in standard mixed lymphocyte cultures against either donor, third-party, or recipient strain irradiated stimulator splenocytes. The control column indicates proliferation in wells containing only responder and stimulator cells. “Added tolerant” indicate proliferation in wells where splenocytes from tolerant mice were added to the mixed lymphocyte reactions at the indicated tolerant:responder ratios; “added naive” indicate proliferation in wells where additional naïve responder cells substituted for tolerant splenocytes to control for the effects of cell crowding on proliferation. n=6 for each group, error bars represent standard deviation. * Denotes P<0.05 by ANOVA.

DISCUSSION

While there are dozens of approaches available for inducing tolerance in the generally permissive murine vascularized allograft models, far fewer have been successful in promoting long-term survival in the stringent skin allograft model. In addition, of these approaches, only three have been translated to higher animal models with any degree of success. These include extensive lymphoid depletion, costimulation blockade, and mixed chimerism (^19–21). Of these three approaches, there seems to be a growing appreciation that the induction of mixed chimerism through the adjuvant administration of donor hematopoietic cells represents a practical and promising approach to human solid organ transplant tolerance. This appreciation is based upon experimental evidence indicating that a mixed chimerism approach to tolerance generates a robust form of tolerance mediated predominantly through the mechanism of centrally and peripherally mediated clonal deletion (^22,23). In addition, this approach has been successfully translated to progressively more complex and stringent experimental animal models (²⁴). Finally, there have been numerous reports of successful organ allografts without the need for chronic immunosuppression in patients previously treated with bone marrow transplants when the solid organs were obtained from the original bone marrow donor (²⁵).

While it is very effective at maintaining graft survival, there are some disadvantages associated with the mixed chimerism approach however. The morbidity associated with the conditioning regimens necessary to achieve partial engraftment of donor hematopoietic stem cells constitutes the major disadvantage of this approach. The successful induction of even low, transient levels of chimerism requires conditioning regimens that accomplish two major manipulations of the recipient. First, the host immune system must be suppressed to prevent attack against and destruction of the donor marrow once introduced into the host. This can be accomplished through the administration of conventional immunosuppressive drugs or biological agents. In the present study, this is accomplished through the administration of ALS and sirolimus. The second manipulation that must be accomplished is a facilitation of engraftment of the donor stem cells into the host marrow. Engraftment of stem cells is a competitive process and any manipulation that provides an advantage to the donor stem cells (i.e., administration of large numbers of stem cells, host myeloablation) can facilitate the ultimate production of chimerism (^26,27). In general this competitive nature of the engraftment process determines that an inverse relationship exists between the intensity of the conditioning regimen and the dose of donor marrow required to achieve partial engraftment (i.e., a lower dose of bone marrow is required as progressively higher degrees of myeloablation are employed). An unfortunate consequence of this is that approaches that have been most successful in producing a state of mixed chimerism in higher animal models are generally considered excessively morbid for routine use in solid organ transplant although some possible exceptions have recently been reported (^28,29). The present study uses no conventional myeloablation and instead relies upon a relatively high dose of bone marrow to promote engraftment.

This study focuses on the immunosuppressive requirement for the induction of mixed chimerism. It demonstrates that the administration of 150× 10⁶ donor bone marrow cells to mice conditioned with rabbit antimouse antilymphocyte serum and sirolimus results in transient multilineage chimerism and a median skin graft survival of 126 days. Substitution of depletional monoclonal antibodies against either CD4 or CD8 T cells in lieu of the administration of ALS did not support the development of chimerism. In contrast, substitution of either monoclonal antibody mediated depletion of both CD4 and CD8 T cells or CD40-CD154 costimulation blockade for ALS results in durable multilineage chimerism, indefinite skin graft survival, and donor specific unresponsiveness both in vivo and in vitro. In both groups where skin graft survival was indefinite, the phenotypic distribution of chimeric cells was similar (i.e., predominantly monocyte, B cell and NK cell with low levels of T cells) to that observed in mice treated with ALS. In contrast to that similarity, indefinite skin graft survival was associated with increased levels of mixed chimerism over that observed in mice treated with ALS. In addition, this higher level of chimerism was stable insofar as it has persisted in all tolerant mice for over 250 days. Whether the presence of tolerance is permissive for higher levels of stable chimerism or whether the increased levels of chimerism facilitate tolerance has yet to be determined.

The findings of this study are consistent with observations made in an earlier version of this treatment regimen that excluded the adjuvant use of sirolimus. Using a murine skin allograft model disparate for only one MHC class I antigen, De Fazio et al. demonstrated that targeted lymphocyte depletion using monoclonal antibodies against CD4 and CD8 produced at least an equal duration of skin graft survival as ALS in mice that also received 20× 10⁶ donor bone marrow cells 7 days following skin grafting (³⁰). Our findings are also partially consistent with some recent observations in a murine mixed chimerism model in which partially myeloablative radiation is used to facilitate bone marrow engraftment. This radiation facilitated chimerism study demonstrated that, as in the present study, depletion of both CD4 and CD8 T cells provided sufficient immunosuppression and resulted in consistent production of mixed chimerism (³¹). However, while CD40-CD154 costimulation blockade was effective at inhibiting CD4 T-cell function, the consistent induction of chimerism required the concomitant depletion of recipient CD8 T cells. The observation that blockade of the CD40-CD154 pathway permitted the induction of mixed chimerism without the need for simultaneous CD8 T-cell depletion in the present model is most likely related to the additional T-cell inhibition provided by sirolimus.

Mechanistically, there is abundant evidence that clonal deletion (both centrally and peripherally mediated) contributes to tolerance maintenance in mixed chimerism models (^32,33). Using a superantigen system as a surrogate for alloreactivity, our studies suggest that this mechanism is also present in our model. While superantigens do not behave in precisely the same manner as alloantigens in terms of mediating clonal deletion, its relevance has been confirmed using mice with transgenic T-cell receptors in other models (²²). In contrast, there is little consensus regarding the role of regulatory cells in the maintenance of tolerance in mixed chimerism models. The observation of nonspecific inhibition of mixed lymphocyte reactions mediated by splenocytes harvested from mice rendered tolerant more than 300 days previously suggests the possibility that T regulatory cells may also contribute to the maintenance of tolerance in this model. Further studies are currently underway to identify the cells responsible for this activity and the mechanism through which it is mediated.

In summary, these experiments suggest that the predominant mechanism whereby polyclonal antilymphocyte serum contributes to the establishment of mixed chimerism in this model is by inhibiting the host T-cell response and preventing rejection of the hematopoietic progenitor cells in the donor marrow allograft as they compete with the recipient progenitor cells for engraftment. While the depletional effects of ALS on CD4 and CD8 T cells are sufficiently immunosuppressive for the induction of chimerism, they are not absolutely necessary since inhibition of CD4 T-cell function with CD40-CD154 blockade alone also serves to induce an equivalent pattern and degree of multilineage chimerism in this model. These findings are important insofar as they will permit the use of better defined monoclonal antibody preparations in the further development and translation of this approach to tolerance induction and overcome the major practical limitation associated with the use of polyclonal preparations, that being the significant batch to batch variability in potency and pattern of depletion that is characteristic of polyclonal antibodies generated in vivo.

Acknowledgments

This work was funded by the Division of Intramural Research of The National Institute of Diabetes, Digestive and Kidney Diseases of the National Institute of Health.