Xenogeneic bone marrow transplantation: II. Porcine-specific growth factors enhance porcine bone marrow engraftment in an in vitro primate microenvironment

Defects in Long-Term APC Repopulation Ability of Adult Human Bone Marrow Hematopoietic Stem Cells (HSCs) Compared with Fetal Liver HSCs

Immunodeficient mice reconstituted with immune systems from patients, or personalized immune (PI) mice, are powerful tools for understanding human disease. Compared with immunodeficient mice transplanted with human fetal thymus tissue and fetal liver-derived CD34⁺ cells administered i.v. (Hu/Hu mice), PI mice, which are transplanted with human fetal thymus and adult bone marrow (aBM) CD34⁺ cells, demonstrate reduced levels of human reconstitution. We characterized APC and APC progenitor repopulation in human immune system mice and detected significant reductions in blood, bone marrow (BM), and splenic APC populations in PI compared with Hu/Hu mice. APC progenitors and hematopoietic stem cells (HSCs) were less abundant in aBM CD34⁺ cells compared with fetal liver-derived CD34⁺ cell preparations, and this reduction in APC progenitors was reflected in the BM of PI compared with Hu/Hu mice 14-20 wk posttransplant. The number of HSCs increased in PI mice compared with the originally infused BM cells and maintained functional repopulation potential, because BM from some PI mice 28 wk posttransplant generated human myeloid and lymphoid cells in secondary recipients. Moreover, long-term PI mouse BM contained functional T cell progenitors, evidenced by thymopoiesis in thymic organ cultures. Injection of aBM cells directly into the BM cavity, transgenic expression of hematopoietic cytokines, and coinfusion of human BM-derived mesenchymal stem cells synergized to enhance long-term B cell and monocyte levels in PI mice. These improvements allow a sustained time frame of 18-22 wk where APCs and T cells are present and greater flexibility for modeling immune disease pathogenesis and immunotherapies in PI mice.

Achieving tolerance in pig-to-primate xenotransplantation: reality or fantasy

Because the immunologic differences between species are far greater than those within species, it is likely that the amount of immunosuppression that would be required for successful xenografting would be so much greater than that now used for allografting, that the side-effects and complications would be unacceptable. Tolerance approaches to xenotransplantation would overcome this concern. Studies in humanized mouse models have demonstrated that human T cells can be tolerized to porcine xenografts, providing important proofs of principle of the potential feasibility of pig-to-primate xenograft tolerance. The results available from studies of pig-to-primate xenotransplantation to date have demonstrated that while chronic immunosuppressive drugs have not completely avoided either T cell responses or humoral rejection, approaches directed toward tolerance induction have been encouraging with regard to avoiding immunization at both of these levels.

Role of VLA-4 and VLA-5 in ex vivo maintenance of human and pig hematopoiesis in human stroma-supported long-term cultures

OBJECTIVE

The advantage of recipient hematopoiesis over that of xenogeneic donors poses a fundamental obstacle to the induction of xenograft tolerance through mixed hematopoietic chimerism. Here we explore the role of beta1 integrins in maintenance of human vs porcine hematopoiesis within a human hematopoietic environment.

METHODS

Porcine and human c-kit+ bone marrow cells were purified and cultured on human bone marrow stroma for 6 weeks. The role of VLA-4 and VLA-5 in the maintenance of porcine vs human hematopoiesis in this human stroma-supported long-term bone marrow culture (LTBMC) system was evaluated by using blocking mAbs that bind to both species.

RESULTS

Blocking VLA-4 with HP2/1 inhibited both human and porcine hematopoiesis, whereas anti-VLA-5 (SAM-1) suppressed the function of human, but not porcine, hematopoietic cells. In mixed LTBMC of porcine and human cells on a human stroma, porcine hematopoietic cells were at a competitive disadvantage, as seen by a rapid decline in cellularity, including clonogenic progenitors. This disadvantage was substantially overcome by the addition of SAM-1. Furthermore, human, but not porcine, cell adhesion to human fibronectin was inhibited by arginine-glycine-aspartic acid (RGD) peptides.

CONCLUSION

Taken together, these results indicate that VLA-4 plays critical role for porcine hematopoiesis in a human hematopoietic environment, and raise the possibility that porcine VLA-5 might be unable to bind the respective human ligand and/or to initiate adequate post-ligand-binding signaling. Thus, VLA-5 may provide a potential target for developing approaches to improve porcine hematopoiesis in human recipients.

T Cells from Presensitized Donors Fail to Cause Graft-versus-Host Disease in a Pig-to-Mouse Xenotransplantation Model

BACKGROUND

The ability of T cells from pigs, the most suitable donors for clinical xenotransplantation, to induce graft-versus-host disease (GVHD) and to facilitate hematopoietic cell engraftment in highly disparate xenogeneic recipients remains unclear. In this article, the authors address these questions in a presensitized pig-to-mouse transplantation model using porcine cytokine-transgenic mice.

METHODS

Swine donors were presensitized by mouse skin grafting and boosted by the injection of mouse cells after the skin graft was rejected. Porcine peripheral blood mononuclear cells (PBMC) and splenocytes were collected at various times after mouse skin grafting, and their potential to induce GVHD and to facilitate donor hematopoietic cell engraftment in conditioned murine recipients was evaluated.

RESULTS

Presensitization of donor pigs resulted in marked enhancement of anti-mouse responses, as reflected in augmented anti-mouse mixed lymphocyte responses, cell-mediated cytotoxicity, and antibody production. However, injection of high numbers of PBMC and splenocytes (>1 x 10(8)) with bone marrow cells from the presensitized pigs failed to induce detectable GVHD or long-term chimerism in mice that were treated with depleting anti-T-cell and natural killer cell antibodies, cobra venom factor, medronate-liposomes, and 4 Gy of whole-body and 7 Gy of thymic irradiation. Histologic analysis revealed no mononuclear cell infiltration or GVHD-associated lesions in the liver, intestine, lungs, or skin of the mouse recipients. Furthermore, the recipient mice had no detectable T cells or anti-pig immunoglobulin G antibodies in the blood by 6 weeks after injection of porcine cells.

CONCLUSION

These results demonstrate that porcine T-cell function is severely impaired in the xenogeneic murine microenvironment.

Induction of human T-cell tolerance to porcine xenoantigens through mixed hematopoietic chimerism

Xenotransplantation from pigs could provide a potential solution to the severe shortage of allogeneic donor organs. Because xenogeneic tissues are subject to vigorous immune rejection, tolerance induction is likely to be essential to the success of clinical xenotransplantation. Here we explore the possibility of inducing human T-cell tolerance to porcine xenografts through mixed chimerism. We previously showed that NOD/SCID-Tg mice expressing porcine cytokine transgenes permit the induction of durable porcine hematopoietic chimerism. In this study we achieved human T-cell development in these mice by engrafting human fetal thymus/liver tissues. In porcine hematopoietic chimeras, human thymus grafts were populated with porcine class II(high) cells in addition to human cells, and human T cells were tolerant of the porcine hematopoietic donor as measured by mixed lymphocyte reaction assay and skin grafting. This study proves the principle that porcine chimerism induces tolerance of xenoreactive human T cells.

Detection of porcine endogenous retrovirus in cultures of freshly isolated porcine bone marrow cells

Pigs are under consideration as possible sources of organs for xenotransplantation in humans. The induction of hematopoietic microchimerism through xenotransplantation of source animal hematopoietic cells has been suggested as a means to induce tolerance in potential recipients. Because all porcine cells contain genetic information for porcine endogenous retrovirus (PERV), coculture techniques, reverse transcriptase (RT) and reverse transcriptase-polymerase chain reaction assays were used to determine whether infectious PERV is released from fresh porcine bone marrow cells cultured in the presence or absence of porcine cytokines. Human embryonic kidney cell line, HEK-293 cells cocultured with porcine bone marrow cells were positive for PERV RNA but never became positive for viral RT activity, suggesting the PERV infection was not productive. In contrast, high levels of RT activity was detected in porcine ST-IOWA cells after coculture, demonstrating that these cells became productively infected. PERV was released from cultured porcine bone marrow cells without stimulation, and combinations of the porcine hematopoietic cytokines, interleukin-3, granulocyte macrophage-colony stimulating factor and stem cell factor had no additional effect on the infectivity or in vitro tropism of released PERV virions.

Persistent chimerism despite antidonor MHC in vitro responses in miniature swine following allogeneic hematopoietic cell transplantation

BACKGROUND

T-cell chimerism predominates in miniature swine receiving hematopoietic-cell transplantation without myelosuppressive conditioning. Several chimeric recipients have become hyporesponsive to donor-major histocompatibility complex (MHC) in vitro and accepted donor-matched renal transplants without immunosuppression. However, some retained antidonor in vitro responses and subsequently rejected donor renal allografts despite the persistence of peripheral blood chimerism. In this study, we characterize the donor cells in both "tolerant" and "nontolerant" chimeric miniature swine.

METHODS

Peripheral blood chimerism was determined by flow cytometry. In vitro antidonor responsiveness was determined by mixed lymphocyte reaction (MLR) and cell-mediated lymphocytotoxicity (CML). Donor cells were separated from chimeras by immunomagnetic bead separation and used as stimulators or targets in CML assays. Phenotypic analysis of donor cells in chimeras was performed using flow cytometry.

RESULTS

Peripheral blood chimerism stabilized beyond 100 days and was made up almost entirely of T cells. PBMC from nontolerant chimeras could be stimulated in vitro to kill donor cells isolated from the mixed chimera itself. In contrast, PBMC from tolerant chimeras hyporesponsive to donor-type cells could not be stimulated in vitro to kill their own sorted donor cells.

CONCLUSIONS

The in vivo persistence of donor T cells in mixed chimeric animals with in vitro antidonor responsiveness is not caused by an inability of these cells to be killed but rather by the poor stimulating capacity of these donor T cells. The nature of donor T cells that persist in the face of in vitro antidonor responses, has important implications for the induction of transplant tolerance by way of the generation of mixed chimerism.

Cryopreservation and mycophenolate therapy are detrimental to hematopoietic progenitor cells

BACKGROUND

The aim of the present study was to determine whether certain components of nonmyeloablative regimens for hematopoietic cell transplantation might compromise the growth of hematopoietic progenitors.

METHODS

Porcine peripheral blood progenitor cells (PBPC) were cytokine-mobilized, collected by leukapheresis, and cryopreserved using 5% dimethyl sulfoxide and 6% hydroxyethyl starch. The influence of cryopreservation on PBPC was tested in vitro by enumeration of colony-forming units (CFUs) in methylcellulose and cobblestone area-forming cell (CAFC) subsets in stromal-associated long-term cultures on fresh and frozen PBPC. The effects of mycophenolate mofetil (MMF) on porcine PBPC and baboon and human bone marrow (BM) were tested in vitro by adding varying doses of MMF to the CFU assays. One baboon was treated with increasing doses of MMF (100-500 mg/kg per day continuously intravenous), and sequential BM aspirations were tested for CFU content.

RESULTS

Fresh cytokine-mobilized PBPC had similar frequencies of progenitor cells when compared with porcine BM. Freezing-thawing of PBPC had no effect on porcine CFUs but reduced the recovery of CAFCs by more than 90%. In vitro, MMF completely inhibited the development of porcine and human CFUs at a concentration of 1 microg/mL and of baboon CFUs at levels between 10 and 100 microg/mL. Plasma-free mycophenolic acid levels of 10 to 30 microg/mL were associated with decreased CFUs in the BM.

CONCLUSIONS

Cryopreservation and MMF potentially prevent engraftment of porcine PBPC by reducing the content or development of progenitor cells. These results indicate that the use of fresh PBPC might improve the induction of mixed hematopoietic chimerism and raise the possibility that use of high doses of MMF in the poststem cell transplant may compromise engraftment.

Porcine mononuclear cells adhere to human fibronectin independently of very late antigen-5: implications for donor-specific tolerance induction in xenotransplantation

To combat the shortage of donor organs, transplantation across species barriers has been proposed. Induction of tolerance would overcome the substantial immunologic barriers to xenotransplantation and would avoid the chronic use of immunosuppressive agents. Successful transplantation of hematopoietic cells induces robust specific tolerance to donor antigens in allogeneic and xenogeneic models. The beta1 integrin class of adhesion molecules and their interactions with extracellular matrix components are thought to be integral to the engraftment and maturation of hematopoietic stem cells. We therefore examined the efficacy of porcine very late antigen-5 (VLA-5) and VLA-4 interactions with the human extracellular matrix (ECM) protein, fibronectin. Peripheral blood mononuclear cells (PBMCs) from humans and miniature swine were flourochrome labeled and adhesion to plates coated with whole human fibronectin (whFN) or its 120 KDa fragment containing the VLA-5 binding region was determined. Flow cytometry and immuno- precipitation were used to identify a monoclonal antibody that cross-reacted on porcine VLA-5. Human and pig PBMC adhesion to human fibronectin (hFN) or 120 kDa fragment-coated plates was assessed following incubation with control ab, anti-VLA-4, anti-VLA-5, or soluble fibronectin. Using rabbit complement, cells expressing VLA-5 were purged from PBMC preparations before performing the adhesion assay. Porcine and human PBMC both adhered to hFN in a divalent cation-dependent and activation-dependent manner. Adhesion to hFN of human but not pig PBMC was blocked by anti-VLA-5 monoclonal antibody SAM-1, although this mAb immunoprecipitated a heterodimeric cell surface molecule (155/135 kDa) resembling VLA-5 from pig PBMC. Complement-mediated depletion of VLA-5-expressing cells ablated specific binding of human but not porcine cells to hFN and its 120 kDa fragment. Addition of soluble fibronectin was capable of blocking adhesion of PBMC of both species to hFN. Anti-VLA-4 reduced the binding of PBMC from both species to hFN to a similar extent. Human and pig cells can specifically adhere to hFN and its 120 kDa fragment, suggesting that this critical cell-ECM interaction is preserved across species. While human cells exclusively use VLA-5 for binding to the 120 kDa fragment, porcine cells could not be shown to adhere to whFN or its 120 kDa fragment via VLA-5. However, porcine VLA-4 is capable of mediating adhesion to human FN. We conclude that disparities in the adhesive interactions of beta1 integrins may be a barrier to the use of porcine hematopoietic stem cell transplantation as a means of inducing donor-specific tolerance in the pig to human species combination.

Porcine hematopoiesis on primate stroma in long-term cultures: enhanced growth with neutralizing tumor necrosis factor-alpha and tumor growth factor-beta antibodies

BACKGROUND

Donor hematopoiesis is at a competitive disadvantage when bone marrow transplantation is across species barriers. This could present major limitations to xenogeneic stem cell transplantation as an approach to tolerance induction. An in vitro model of xenogeneic engraftment was established to identify inhibitors of porcine hematopoiesis in a primate environment.

METHODS

Porcine bone marrow cells (BMC), in the presence or absence of primate CD34+ positive cells, were cultured for 4-6 weeks on primate stroma with porcine cytokines. Cellularity and growth of colony-forming cells were indicators of hematopoietic growth. Effects of soluble factors were determined by using Transwell inserts to separate porcine cells from stroma. Neutralizing antibodies for human transforming growth factor-beta (TGF-beta) and tumor necrosis factor-alpha (TNF-alpha) were added to cultures.

RESULTS

Porcine hematopoiesis can be maintained in long-term cultures on primate stroma with pig cytokines. Adding BMC to the stroma below Transwell-containing porcine cells dramatically inhibited porcine hematopoiesis, showing an inhibitory role for soluble factors. Neutralizing antibodies against TNF-alpha or TGF-beta caused a modest enhancement of porcine hematopoiesis; however, the combination of both led to a substantial increase. Inhibitory effects of these cytokines were confirmed by adding TNF-alpha and TGF-beta to porcine cultures.

CONCLUSIONS

Porcine cells may be more sensitive to inhibitory effects of TNF-alpha and TGF-beta than primate cells and are at a disadvantage when in a primate environment. Potential implications of this observation are discussed in the context of establishing specific immune tolerance via mixed chimerism to facilitate xenotransplantation.

Pig hematopoietic cell chimerism in baboons conditioned with a nonmyeloablative regimen and CD154 blockade

BACKGROUND

In an attempt to induce mixed hematopoietic chimerism and transplantation tolerance in the pig-to-primate model, we have infused high-dose porcine peripheral blood progenitor cells (PBPC) into baboons pretreated with a nonmyeloablative regimen and anti-CD154 monoclonal antibody (mAb).

METHODS

Group 1 baboons (n=2) received a nonmyeloablative regimen including whole body irradiation, pharmacological immunosuppression, porcine hematopoietic growth factors, and immunoadsorption of anti-Galalpha1,3Gal (Gal) antibody before infusion of high doses of PBPC (2.7-4.6x10(10) cells/kg). In group 2 (n=5), cyclosporine was replaced by anti-CD154 mAb. Group 3 (n=3) received the group 1 regimen plus anti-CD154 mAb.

RESULTS

In group 1, pig chimerism was detected in the blood by flow cytometry (FACS) for 5 days (with a maximum of 14%), and continuously up to 13 days by polymerase chain reaction (PCR). In group 2, pig chimerism was detectable for 5 days by FACS (maximum 33%) and continuously up to 28 days by PCR. In group 3, initial pig chimerism was detectable for 5 days by FACS (maximum 73%). Two of three baboons showed reappearance of pig cells on days 11 and 16, respectively. In one, in which no anti-Gal IgG could be detected for 30 days, pig cells were documented in the blood by FACS on days 16-22 (maximum 6% on day 19) and pig colony-forming cells were present in the blood on days 19-33, which we interpreted as evidence of engraftment. Microchimerism was continuous by PCR up to 33 days.

CONCLUSIONS

These results suggest that there is no absolute barrier to pig hematopoietic cell engraftment in primates, and that this may be facilitated if the return of anti-Gal IgG can be prevented.

Depletion of anti-gal antibodies in baboons by intravenous therapy with bovine serum albumin conjugated to gal oligosaccharides

BACKGROUND

Anti-Galalpha 1-3Gal (Gal) antibodies (Ab) play a key role in the rejection of pig cells or organs transplanted into primates. A course of extracorporeal immunoadsorption (EIA) of anti-Gal Ab using an immunoaffinity column of a Gal type 6 oligosaccharide depletes Ab successfully, but Ab returns during the next few days. Although therapy with an anti-CD154 monoclonal antibody (mAb) prevents an induced Ab response to Gal or non-Gal epitopes, T cell-independent natural anti-Gal IgM and IgG return to baseline (pretransplant) levels. We have investigated the capacity of continuous i.v. infusion of bovine serum albumin conjugated to Gal type 6 oligosaccharide (BSA-Gal) to deplete or maintain depletion of circulating anti-Gal Ab.

METHODS

Porcine peripheral blood mobilized progenitor cells (PBPC) obtained by leukapheresis from MHC-inbred miniature swine (n=6) were transplanted into baboons. Group 1 baboons (n=4) underwent whole body (300 cGy) and thymic (700 cGy) irradiation, T cell depletion with antithymocyte globulin, complement depletion with cobra venom factor, short courses of anti-CD154 mAb therapy (20 mg/kg i.v. on alternate days), cyclosporine (CyA) (in two baboons only), mycophenolate mofetil, and porcine hematopoietic growth factors. Anti-Gal Ab depletion by EIA was carried out before transplantation of high doses (2-4x 1010 cells/kg) of PBPC. Group 2 baboons (n=3) received the group 1 regimen (including CyA) plus a continuous i.v. infusion of BSA-Gal. To prevent sensitization to BSA, anti-CD154 mAb therapy was continued until BSA-Gal administration was discontinued.

RESULTS

In group 1, Gal-reactive Ab returned to pre-PBPC transplant levels within 15-21 days, but no induced Ab to Gal or non-Gal determinants developed while anti-CD154 mAb therapy was being administered. In group 2, anti-Gal Ab was either not measurable or minimally measurable while BSA-Gal was being administered. After discontinuation of BSA-Gal, Ab did not return to pre-PBPC transplant level for more than 40-60 days, and no sensitization developed even when all therapy was discontinued. In one baboon, however, Ab to Gal type 2, but not type 6, returned during BSA-Gal therapy.

CONCLUSIONS

Prevention of the induced humoral response to Gal and non-Gal epitopes by anti-CD154 mAb therapy has been reported previously by our group, but our studies are the first to demonstrate a therapy that resulted in an absence of natural anti-Gal Ab for a prolonged period. The combination of BSA-Gal and T cell costimulatory blockade may facilitate survival of pig cells and organs transplanted into primates. The return in one baboon of Ab reactive with the Gal type 2 oligosaccharide, but not type 6, indicates some polymorphism of anti-Gal Ab and suggests that, to be effective in all cases, the infusion of a combination of type 6 and type 2 BSA-Gal may be required.

Adult porcine islet transplantation in baboons treated with conventional immunosuppression or a non-myeloablative regimen and CD154 blockade

The aim of the present study was to assess the survival of adult porcine islets transplanted into baboons receiving either (I) conventional triple drug immunosuppressive therapy or (2) a non-myeloablative regimen and an anti-CD154 monoclonal antibody (mAb) aimed at tolerance-induction. Group 1 baboons (n = 3) were pancreatectomized prior to intraportal injection of 10,000 porcine islet equivalents (IE)/kg and immunosuppressed with anti-thymocyte globulin (ATG), cyclosporine and azathioprine. In Group 2 (n = 2), non-pancreatectomized baboons underwent induction therapy with whole body and thymic irradiation, and ATG. Extracorporeal immunoadsorption (EIA) of anti-Galalpha1,3Gal (Gal) antibody was carried out. Maintenance therapy was with cobra venom factor, cyclosporine. mycophenolate mofetil, methylprednisolone and anti-CD154 mAb. Porcine islets were injected intraportally (14,000 and 32,000 IE/kg, respectively) and high-dose pig mobilized peripheral blood progenitor cells (3 x 10(10) cells/kg) were infused into a systemic vein. Porcine islets were also implanted in the sternomastoid muscle to facilitate subsequent biopsies. In both groups. porcine C-peptide was measured, and histological examination of liver or sternomastoid muscle biopsies was performed at regular intervals. In Group 1, total pancreatectomy reduccd human C-peptide to < 0.1 ng/ml and induced insulin-requiring diabetes. The transplantation of porcine islets was followed by normalization of glycemia for 15-24 h. Porcine C-peptide was detected only transiently immediately after porcine islet injection (maximum 0.12 ng/ml). Histological examination of liver biopsies taken between days 2 and 19 did not reveal viable islets, but necrotic cell structures with mononuclear cell infiltrates were identified in portal venules. In Group 2, injection of porcine islets into non-pancreatectomized recipients induced a transient hypoglycemia (2-4 h) requiring concentrated intravenous dextrose administration. Porcine C-peptide was detectable for 5 and 3 days (maximum 2.8 and 1.0 ng/ml), respectively. Baboon #4 died on day 12 from small bowel intussusception. Liver and sternomastoid muscle biopsies showed well-preserved porcine islets, staining positive for insulin and glucacon, without signs of rejection. In baboon #5, viable islets were detected in the sternomastoid muscle biopsy on day 14, but not on day 28 or thereafter. A progressive mononuclear cell and macrophage infiltration was seen in the biopsies. In conclusion, conventional immunosuppression allowed survival of porcine islets in baboons for < 24 h. The non-myeloablative regimen prolonged survival of porcine islets for > 14 days. However, despite depletion of T cells, anti-Gal antibody and complement, and CD154-hlockade, porcine islets were rejected by day 28. These results suggest that powerful innate immune responses are involved in rejection of discordant xenogencic islets.

CD40-CD154 pathway blockade requires host macrophages to induce humoral unresponsiveness to pig hematopoietic cells in baboons

The effect of CD154 blockade and macrophage depletion or inhibition on baboon humoral and cellular immune responses to pig antigens was studied in a pig-to-baboon peripheral blood mobilized progenitor cell (PBPC) transplantation model aimed at inducing tolerance. We infused pig PBPCs in baboons pretreated with a nonmyeloablative regimen along with murine anti-human CD154 monoclonal antibody (mAb) and macrophage-depleting or -inhibiting agents. Group 1 baboons (n=2) underwent a nonmyeloablative regimen and immunoadsorption of anti-Gal(alpha)1,3Gal (Gal) antibody (Ab) before intravenous infusion of high doses (1.3-4.6 x 10(10)cells/kg) of PBPCs. In group 2 (n=5), cyclosporine was replaced by 8 doses of anti-CD154 mAb over 14 days. Group 3 (n=3) received the group 2 regimen plus medronate liposomes (n=2) or commercially available human intravenous immunoglobulin G depleted of anti-Gal Ab (n=1) to deplete/inhibit recipient macrophages. Group 1 developed sensitization to Gal and also developed new Ab to non-Gal porcine antigens within 10 to 20 days. In group 2, no sensitization to Gal or non-Gal determinants was seen, but Gal-reactive antibodies did return to their preleukocyte transplantation levels. CD154 blockade, therefore, induced humoral unresponsiveness to pig cells. In group 3, sensitization to Gal was seen in all three baboons at 20 days, and Abs against new porcine determinants developed in one baboon. The depletion or inhibition of host macrophages, therefore, prevented the induction of humoral unresponsiveness by CD154 blockade. These results suggest that CD154 blockade induces humoral unresponsiveness by a mechanism that involves the indirect pathway of antigen presentation. In vitro investigation of baboon anti-pig mixed lymphocyte reaction confirmed that only the indirect pathway is efficiently blocked by anti-CD154 mAb. The mechanism in which blockade of the CD40-CD154 pathway induces its effect remains to be determined, but it could involve the generation of regulatory cells capable of suppressing the direct pathway.

Porcine hematopoietic cell xenotransplantation in nonhuman primates is complicated by thrombotic microangiopathy

Thrombotic microangiopathy (TM) is a serious complication of bone marrow transplantation (BMT) that resembles thrombotic thrombocytopenic purpura (TTP). In attempting to achieve hematopoietic cell chimerism in the pig-to-baboon model, we have observed TM following infusion of high doses (>10(10) cells/kg) of porcine peripheral blood mobilized progenitor cells (PBPC) into baboons. We performed investigations to analyze the pathobiology of this TM and to test therapeutic interventions to ameliorate it. PBPC were obtained by leukapheresis of cytokine-stimulated swine. The initial observations were made in two baboons that underwent a non-myeloablative regimen (NMR) prior to PBPC transplantation (TX) (group 1). We then studied three experimental groups. Group 2 (n = 2) received NMR without PBPC TX. Group 3 (n = 2) received PBPC TX alone. Group 4 (n = 6) received NMR + PBPC TX combined with prostacyclin, low-dose heparin, methylprednisolone, and cyclosporine was replaced by anti-CD40L mAb in five cases. Baboons in groups 1 and 3 developed severe thrombocytopenia (<10,000/mm3), intravascular hemolysis with schistocytosis (>10/high powered field (hpf)), increase in plasma lactate dehydrogenase (LDH) (2500-9000 U/l), transient neurologic changes, renal insufficiency, and purpura. Autopsy on two baboons confirmed extensive platelet thrombi in the microcirculation, and, similar to clinical BMT-associated TM/TTP, no unusually large vWF multimers or changes in vWF protease activity were observed in the plasma of baboons with TM. In group 2, self-limited thrombocytopenia occurred for 10-15 days following NMR. Group 4 baboons developed thrombocytopenia (<20,000/mm3) rarely requiring platelet transfusion, minimal schistocytosis (<3/hpf), minor increase in LDH (<1000 U/l), with no clinical sequelae. We conclude that high-dose porcine PBPC infusion into baboons induces a microangiopathic state with vWF biochemical parameters resembling clinical BMT-associated TM/TTP and that administration of antithrombotic and anti-inflammatory agents can ameliorate this complication.

High-dose porcine hematopoietic cell transplantation combined with CD40 ligand blockade in baboons prevents an induced anti-pig humoral response

BACKGROUND

In pig-to-primate organ transplantation, hyperacute rejection can be prevented, but the organ is rejected within days by acute vascular rejection, in which induced high-affinity anti-Gal alpha1-3Gal (alphaGal) IgG and possibly antibodies directed against new porcine (non-alphaGal) antigenic determinants are considered to play a major role. We have explored the role of an anti-CD40L monoclonal antibody in modifying the humoral response to porcine hematopoietic cells in baboons pretreated with a nonmyeloablative regimen.

METHODS

Porcine peripheral blood mobilized progenitor cells obtained by leukapheresis from both major histocompatibility complex-inbred miniature swine (n=7) and human decay-accelerating factor pigs (n=3) were transplanted into baboons. Group 1 baboons (n=3) underwent whole body (300 cGy) and thymic (700 cGy) irradiation, T cell depletion with ATG, complement depletion with cobra venom factor, short courses of cyclosporine, mycophenolate mofetil, porcine hematopoietic growth factors, and anti-alphaGal antibody depletion by immunoadsorption before transplantation of high doses (2-4 x 10(10)/cells/kg) of peripheral blood mobilized progenitor cells. In group 2 (n=5), cyclosporine was replaced by eight doses of anti-CD40L monoclonal antibodies over 14 days. The group 3 baboons (n=2) received the group 1 regimen plus 2 doses of anti-CD40L monoclonal antibodies (on days 0 and 2).

RESULTS

In group 1, sensitization to alphaGal (with increases in IgM and IgG of 3- to 6-fold and 100-fold, respectively) and the development of antibodies to new non-alphaGal porcine antigens occurred within 20 days. In group 2, no sensitization to alphaGal or non-alphaGal determinants was seen, but alphaGal-reactive antibodies did return to their pre- peripheral blood mobilized progenitor cells transplant levels. In group 3, attenuated sensitization to alphaGal antigens was seen after cessation of cyclosporine and mycophenolate mofetil therapy at 30 days (IgM 4-fold, IgG 8-30-fold), but no antibodies developed against new porcine determinants. In no baboon did anti-CD40L monoclonal antibodies prevent sensitization to its own murine antigens.

CONCLUSIONS

We believe these studies are the first to consistently demonstrate prevention of a secondary humoral response after cell or organ transplantation in a pig-to-primate model. The development of sensitization to the murine elements of the anti-CD40L monoclonal antibodies suggests that nonresponsiveness to cell membrane-bound antigen (e.g., alphaGal) is a specific phenomenon and not a general manifestation of immunological unresponsiveness. T cell costimulatory blockade may facilitate induction of mixed hematopoietic chimerism and, consequently, of tolerance to pig organs and tissues.

Cross-species compatibility of intercellular adhesion molecule-1 (CD54) with its ligands

BACKGROUND

The molecular interactions of intercellular adhesion molecule-1 (ICAM-1; CD54) are potentially important in several situations in the context of pig-to-human xenotransplantation. If porcine bone marrow is to be used for the induction of xenograft tolerance in humans, the role that has been suggested for ICAM-1 in the interactions of haematopoietic stem cells makes its cross-species compatibility important. Similarly, the potential role of ICAM-1 interactions in graft rejection makes it an important molecule to study.

METHODS

An in vitro static cell-to-cell adhesion study was used to look at the successful interaction of ICAM-1 with its ligands across the pig-human species barrier in both directions. A second in vitro system, the standard long-term bone marrow culture (LT-BMC), was used to study the functional role of ICAM-1 in haematopoiesis.

RESULTS

Human ICAM-1 was able to adhere to ligands on porcine cells, including one or more ligand that contains CD18. Conversely, human CD18-containing ligands mediated adherence to porcine cells. Using the long-term bone marrow culture system, there was no evidence that blocking the interactions of ICAM-1 inhibited hematopoiesis, either in the human-human or pig-human combinations of precursor cells and marrow stroma.

CONCLUSIONS

ICAM-1 is able to interact with at least some of its ligands across the species barrier, in both pig-human and human-pig combinations. However, the interactions of ICAM-1 do not appear to be central to hematopoiesis, at least in the model system used.

Development and analysis of transgenic mice expressing porcine hematopoietic cytokines: a model for achieving durable porcine hematopoietic chimerism across an extensive xenogeneic barrier

The capacity of mixed hematopoietic chimerism to induce tolerance has not been demonstrated in discordant xenogeneic species combinations because of the difficulty in achieving lasting hematopoietic engraftment. In an effort to create a model of long-lasting disparate xenogeneic hematopoietic chimerism, we have developed transgenic (Tg) mice carrying porcine cytokines. Three lines of Tg mice were generated: one carrying porcine IL-3 and GM-CSF genes only (termed IL/GM) and the remaining two lines carrying in addition, the soluble SCF gene (termed IL/GM/sS) or membrane-bound SCF gene (termed IL/GM/mS). Sera from mice with IL/GM and IL/GM/sS transgenes markedly stimulated the proliferation of swine marrow cells in vitro. However, proliferation of swine marrow cells was not induced in cultures containing IL/GM/mS sera. Consistent with these observations, ELISA assays revealed detectable levels of porcine cytokines in the sera of IL/ GM and IL/GM/sS, but not in sera of IL/GM/mS Tg mice. Marrow stromal cells prepared from all three kinds of Tg mice, but not those from non-Tg littermates, were capable of supporting the growth of porcine hematopoietic cells in vitro. Immunodeficient Tg mice were generated by crossing Tg founders with C.B-17 SCID mice for five generations. All Tg immunodeficient mice showed improved porcine hematopoietic engraftment compared with non-Tg controls. These Tg mice provide a useful model system for studying porcine hematopoietic stem cells, and for evaluating the feasibility of donor-specific tolerance induction by mixed chimerism across highly disparate xenogeneic barriers.

Cytokine-mobilized peripheral blood progenitor cells for allogeneic reconstitution of miniature swine

BACKGROUND

Because of the relative ease of acquisition, increased yield, and improved engraftment characteristics, mobilized peripheral blood progenitor (stem) cells (PBSCs) have recently become the preferred source for hematopoietic stem cell transplantation. In our laboratory, procurement of a megadose of PBSCs is necessary for on-going studies evaluating non-myelosuppressive transplant regimens for the induction of mixed chimerism and allograft tolerance. To exploit hematopoietic growth factor synergy, we have sought to combine growth factors with proven utility to improve PBSC mobilization and maximize our PBSC procurement through an automated collection procedure.

METHODS

Mobilization characteristics of PBSCs were determined in 2-5-month-old miniature swine. Animals received either swine recombinant stem cell factor (pSCF, 100 microg/kg) and swine recombinant interleukin 3 (pIL-3, 100 microg/kg), administered intramuscularly for 8 days, or pSCF, pIL-3, and human recombinant granulocyte-colony stimulating factor (hG-CSF), at 10 microg/kg. Leukapheresis was performed beginning on day 5 of cytokine treatment and continued daily for 3 days.

RESULTS

Collection of PBSCs from cytokine-mobilized animals via an automated leukapheresis procedure demonstrated a 10-fold increase in the number of total nucleated cells (TNC) (20-30 x 10(10) TNC) compared to bone marrow harvesting (2-3 x 10(10) total TNC). A more rapid rise in white blood cells (WBCs) was seen after administration of all three cytokines compared to pSCF and pIL-3 alone. An increase in colony-forming unit granulocyte-macrophage frequency measured daily from peripheral blood during cytokine treatment, was seen with the addition of hG-CSF to pSCF/pIL-3 correlating well with the rise in WBCs. Similarly, the addition of hG-CSF demonstrated a notable increase in the median progenitor cell yield from the 3-day leukapheresis procedure. Cytokine-mobilized PBSCs were capable of hematopoietic reconstitution. PBSCs mobilized with pSCF/pIL-3 were infused into an SLA-matched recipient conditioned with cyclophosphamide (50 mg/kg) and total body irradiation 1150 cGy. Neutrophil and platelet engraftment occurred on days 5 and 7, respectively, with minimal evidence of graft-versus-host disease. Complete donor chimerism has been demonstrated 331 days after transplant.

CONCLUSIONS

Our preliminary results show that in this well-defined miniature swine model, recombinant swine cytokine combinations (pSCF, pIL-3 with or without hG-CSF) successfully mobilize a high yield of progenitor cells for allogeneic transplantation. Furthermore, these cytokine-mobilized PBSCs demonstrate the potential to reconstitute hematopoiesis and provide long-term engraftment in miniature swine.

Enhancement of swine progenitor chimerism in mixed swine/human bone marrow cultures with swine cytokines

OBJECTIVE

The induction of transplantation tolerance across xenogeneic barriers by bone marrow transplantation holds great promise, but engraftment of xenogeneic stem cells has been difficult to achieve. Part of this difficulty is due to species-specific differences in regulatory cytokines and elements of the stromal microenvironment, which we studied here.

MATERIALS AND METHODS

We developed a system where fresh bone marrow cells from swine and human are cultured on human bone marrow stroma in order to study these limiting factors in a clinically relevant species combination.

RESULTS

We report here the ability of recombinant swine interleukin (IL)-3 and c-kit ligand (KL) to specifically enhance swine hematopoietic chimerism in this system. In the absence of exogenous swine cytokines, there were about half as many swine progenitors as human progenitors at 1, 2, and 4 weeks of culture. When used alone, swine IL-3 led to a notable but transient increase in the relative ratio of swine progenitors, while addition of swine KL increased the ratio of swine progenitors only modestly and only at later time points. In contrast, when swine IL-3 and KL were added together, there was a two- to fourfold increase in the ratio of swine to human progenitors at all times tested.

CONCLUSION

These data demonstrate that both swine IL-3 and KL are needed for prolonged enhancement of swine progenitor chimerism under these conditions, and suggest that the species specificity of either one or both of these cytokines may represent an important barrier to prolonged engraftment of swine bone marrow in humans.

Long-term discordant xenogeneic (porcine-to-primate) bone marrow engraftment in a monkey treated with porcine-specific growth factors

BACKGROUND

Mixed allogeneic hematopoietic chimerism has previously been reliably achieved and shown to induce tolerance to fully MHC-mismatched allografts in mice and monkeys. However, the establishment of hematopoietic chimerism has been difficult to achieve in the discordant pig-to-primate xenogeneic model.

METHODS

To address this issue, two cynomolgus monkeys were conditioned by whole body irradiation (total dose 300 cGy) 6 and 5 days before the infusion of pig bone marrow (BM). Monkey anti-pig natural antibodies were immunoadsorbed by extracorporeal perfusion of monkey blood through a pig liver, immediately before the intravenous infusion of porcine BM (day 0). Cyclosporine was administered for 4 weeks and 15-deoxyspergualin for 2 weeks. One monkey received recombinant pig cytokines (stem cell factor and interleukin 3) for 2 weeks, whereas the other received only saline as a control.

RESULTS

Both monkeys recovered from pancytopenia within 4 weeks of whole body irradiation. Anti-pig IgM and IgG antibodies were successfully depleted by the liver perfusion but returned to pretreatment levels within 12-14 days. Methylcellulose colony assays at days 180 and 300 revealed that about 2% of the myeloid progenitors in the BM of the cytokine-treated recipient were of pig origin, whereas no chimerism was detected in the BM of the untreated control monkey at similar times. The chimeric animal was less responsive by mixed lymphocyte reaction to pig-specific stimulators than the control monkey and significantly hyporesponsive when compared with a monkey that had rejected a porcine kidney transplant.

CONCLUSION

To our knowledge, this is the first report of long-term survival of discordant xenogeneic BM in a primate recipient.

Effect of pig-specific cytokines on mobilization of hematopoietic progenitor cells in pigs and on pig bone marrow engraftment in baboons

Mixed hematopoietic chimerism has been found to be a requirement for achieving specific immunologic hyporesponsiveness. Some of the requirements for in vitro and in vivo coexistence of discordant hematopoietic systems in the pig-to-baboon (or human) model have been investigated. We have tested the efficacy of pig-specific cytokines (PSC) (IL3, SCF, GM-CSF) in the mobilization of porcine bone marrow (BM) progenitors in vivo (i) in the pig and (ii) in baboons that underwent a conditioning regimen and porcine BM transplantation. In a preliminary in vitro study, porcine BM cells were incubated in various media to assess the effect of human plasma on pig progenitors in a colony-forming unit (CFU) assay. In in vivo studies, four pigs received PSC and one control pig did not. Six baboons underwent natural antibody removal, with subsequent pig BM transplantation. Four of these six underwent nonmyeloablative (n=2) or myeloablative (n=2) conditioning and all received PSC treatment. Two baboons did not receive PSC, one of which underwent a nonmyeloablative regimen. Sequential blood samples and BM biopsies in pigs and baboons were analyzed by CFU assay for the detection of porcine cells. Baboon samples were analyzed by polymerase chain reaction (PCR) to detect porcine DNA. In the case of the in vitro tests, colony forming by porcine progenitors was not inhibited by media containing human plasma and for the in vivo tests, PSC increased the number of progenitors in pig BM; mobilization of progenitors into the peripheral blood was observed. PSC-treated baboons which experienced transient depletion of leukocytes < 1,000/ml (as an effect of the conditioning regimen) had porcine BM cells detectable by PCR for as long as day 316 after BM transplantation. In conclusion we found that: (i) under the conditions of these studies, in vitro porcine progenitor cell growth was not inhibited by human plasma containing natural antibody and complement; (ii) PSC treatment led to an increased number of progenitors in pig BM and peripheral blood; (iii) the combination of an effective conditioning regimen and treatment with PSC was capable of inducing long-term survival of pig progenitors in baboons, although only a low level of engraftment was achieved.

T cell-mediated xenograft rejection: specific tolerance is probably required for long term xenograft survival

T cell-mediated mechanisms of xenograft rejection appear resistant to standard immunosuppression protocols used to prevent allograft rejection and, consequently, higher doses of immunosuppressive drugs are required to promote xenograft compared to allograft survival. Evidence from recent studies suggests that porcine xenografts may be especially immunogenic in humans because of a prominent and vigorous indirect xenoresponse and because of the ability of porcine endothelium to activate human T cells. This has led to an anxiety that systemic immunosuppressives, used as the mainstay of therapy for clinical xenotransplantation, may not allow the long-term survival of porcine organs transplanted into human recipients. This article will review the biology of T cell xenoresponses, present the case for the development of novel graft-specific immunosuppressive regimes in clinical xenotransplantation, and review recent experimental progress in this area.

Anti-Gal(alpha)1-3Gal antibody response to porcine bone marrow in unmodified baboons and baboons conditioned for tolerance induction

BACKGROUND

Mixed lymphohematopoietic chimerism can provide an effective means of inducing longterm immunological tolerance and has been documented in a monkey allograft model. A conditioning regimen including nonmyeloablative or myeloablative irradiation and splenectomy has been used to induce chimerism in a pig-to-primate transplantation model. Since the presence of anti-Gal(alpha)1-3Gal (alphaGal) natural antibodies leads to the hyperacute rejection of pig organs transplanted into primates, extracorporeal immunoaffinity adsorption (EIA) of anti-alphaGal antibodies is also included in the regimen. The effect of the tolerance induction protocol on the anti-alphaGal antibody response has been assessed.

METHODS

Anti-alphaGal antibody was measured after the EIA of plasma through an alphaGal immunoaffinity column in baseline studies involving two unmodified baboons, one splenectomized baboon, and one baboon that received a challenge with porcine bone marrow (BM), and in three groups of baboons (n=2 in each group) that received different conditioning regimens for tolerance induction. Group 1 received a nonmyeloablative conditioning regimen without porcine BM transplantation. Group 2 received nonmyeloablative conditioning with pig BM transplantation and pig cytokine therapy. Group 3 received myeloablative conditioning, an autologous BM transplant (with BM depleted of CD2+ or CD2+/CD20+ cells), and pig BM transplantation.

RESULTS

In the baseline studies, a single EIA of anti-alphaGal antibodies in an unmodified animal initially depleted anti-alphaGal antibody, followed by a mild rebound. Nonmyeloablative conditioning (group 1) in the absence of pig cell exposure reduced the rate of anti-alphaGal antibody return. Pig BM cells markedly stimulated anti-alphaGal antibody production in an unmodified baboon (alphaGal IgM and IgG levels increased 40- and 220-fold, respectively). This response was significantly reduced (to an only 2- to 5.5-fold increase of IgM and IgG) in baboons undergoing nonmyeloablative conditioning (group 2). A myeloablative conditioning regimen (group 3) prevented the antibody response to pig BM, with the reduction in response being greater in the baboon that received autologous BM depleted of both CD2+ and CD20+ cells. No new antibody directed against pig non-aGal antigens was detected in any baboon during the 1 month follow-up period.

CONCLUSIONS

(i) EIA of anti-alphaGal antibody in unmodified baboons results in a transient depletion followed by a mild rebound of antibody; (ii) exposure to pig BM cells results in a substantial increase in anti-alphaGal antibody production; (iii) a nonmyeloablative conditioning regimen reduces the rate of antibody return and (iv) markedly reduces the response to pig BM cells; (v) the anti-alphaGal response is completely suppressed by a myeloablative regimen if CD2+ and CD20+ cells are eliminated from the autologous BM inoculum. Furthermore, (vi) challenge with pig BM cells appears to stimulate only an anti-alphaGal antibody response without the development of other (non-alphaGal) anti-pig antibodies. We conclude that regimens used for T-cell tolerance induction can be beneficial in reducing the anti-alphaGal antibody response to porcine BM.

The use of tolerance for transplantation across xenogeneic barriers

The success of human organ transplantation as a clinical treatment has created a conundrum for the transplant community. It has caused a shortage of human donor organs and uncovered problems of chronic immunosuppression in those lucky enough to receive organ transplants due to their use of chronic immunosuppressive drugs. Our aim is to attempt to approach both issues by establishing specific transplantation tolerance to pig organ grafts.