ES-cell derived hematopoietic cells induce transplantation tolerance

Abstract

Background: Bone marrow cells induce stable mixed chimerism under appropriate conditioning of the host, mediating the induction of transplantation tolerance. However, their strong immunogenicity precludes routine use in clinical transplantation due to the need for harsh preconditioning and the requirement for toxic immunosuppression to prevent rejection and graft-versus-host disease. Alternatively, embryonic stem (ES) cells have emerged as a potential source of less immunogenic hematopoietic progenitor cells (HPCs). Up till now, however, it has been difficult to generate stable hematopoietic cells from ES cells.

Methodology/principal findings: Here, we derived CD45(+) HPCs from HOXB4-transduced ES cells and showed that they poorly express MHC antigens. This property allowed their long-term engraftment in sublethally irradiated recipients across MHC barriers without the need for immunosuppressive agents. Although donor cells declined in peripheral blood over 2 months, low level chimerism was maintained in the bone marrow of these mice over 100 days. More importantly, chimeric animals were protected from rejection of donor-type cardiac allografts.

Conclusions: Our data show, for the first time, the efficacy of ES-derived CD45(+) HPCs to engraft in allogenic recipients without the use of immunosuppressive agents, there by protecting cardiac allografts from rejection.

Figure 1. HOXB4-transduced ES cells robustly differentiate into CD45⁺.

HOXB4-transduced ES cells were allowed to form embryoid bodies. The cell clusters were dismantled and subsequently treated with a hematopoietic differentiation medium. The GFP expressing cells were >70% CD45 by day 26 (A). These cells were sorted by immunomagnetic bead separation to >98% purity (B). To determine whether HPCs stimulate allogenic T cells, MRL splenocytes were co-cultured with irradiated 129SvJ splenocytes, HPCs, or bone marrow cells. Cell proliferation was determined by ³H-thymidine uptake. Splenocytes and bone marrow cells, but not HPCs, robustly stimulated allogenic T cells to proliferate (C). Lastly, immunophenotyping of the HPCs revealed poor expression of co-stimulatory molecules CD80, CD86, and ICOSL (D).

Figure 2. HPCs induce hematopoietic mixed chimerism.

(A) 2×10⁶ HPCs were transplanted in sublethally irradiated MRL, Rag2^−/−γ_c ^−/− or 129SvJ 8–12 week old mice. Mixed chimerism was monitored by measuring by flow cytometry GFP-expressing CD45⁺ cells over 100 days. (B) The expression of MHC antigens by the chimeric hematopoietic cells was determined by flow cytometry in the Rag2^−/−γ_c ^−/− mice. Although the donor cells maintained a high CD45 expression from the start, there was a gradual increase of both class I and class II antigens.

Figure 3. Poor T and B cell development in HPC-derived hematopoietic cells.

Using flow cytometry, chimeric 129SvJ, MRL, and Rag2^−/−γ_c ^−/− mice were analyzed for donor cells at 28 days post-transplantation. In all three strains, Gr-1⁺ cells were the predominant sub-population with lower cell numbers of B220- and CD3-expressing B and T cells, respectively. The percentages are calculated within the GFP⁺ population.

Figure 4. ES-derived HOXB4-expressing HPCs induce immunological tolerance to donor type cardiac allografts.

A) Treatment protocol: MRL recipient mice were transplanted HPCs 14 days prior to transplantation of cardiac allografts which was performed between days 0 and 7. Cardiac function and survival were monitored by daily abdominal palpations. B) Graft survival: Chimeric MRL recipient mice tolerated donor type cardiac allografts, but not Balb/c third party allografts. Non-chimeric MRL mice acutely rejected allografts within 13 days. Syngeneic grafts were additionally used as controls.

Figure 5. Tolerant mice maintain bone marrow mixed chimerism post-transplantation.

The percentage of donor-derived hematopoietic cells in bone marrow cells of recipient tolerant mice was monitored using flow cytometry. A robust CD45⁺ cell population was detected (A) of which a large portion was Gr-1⁺ (B). Interestingly a small proportion of these cells was CD117-expressing, suggesting the presence of a truly bone marrow-resident stem cell population (C). At day 100 post-transplantation, the percentage of donor cells had decreased (D). Altogether, at day 100 post-transplantation, donor cells were hardly detectable in peripheral blood, 0.5–1% in the spleen, and 1–2% in the bone marrow (E, n = 6).

Figure 6. HPC-induced mixed chimerism preserves organ architecture post-transplantation.

Transplanted syngeneic and allogenic animals were sacrificed at either 40 or 100 days post-transplantation and the histology of the grafts studied after H & E staining. These grafts showed preservation of the muscle architecture and lack of mononuclear cell infiltration in the syngeneic and tolerant grafts, but massive tissue destruction in the acutely rejected control grafts.

Figure 7. Tolerance induction is accompanied by the presence of CD4FoxP3⁺ T cells in the tolerated allografts but not in the syngeneic control grafts.

A) & B) Frozen sections of tolerated cardiac allografts were stained for CD4 on day 40 post-transplantation using a peroxidase-conjugated antibody (A). A section from an acutely rejected graft was used as control (B). Positively stained cells are stained brown. C)–H) To determine whether any of the CD4⁺ T cells were Tregs, frozen sections were co-stained for CD4 (FITC) and FoxP3 (PE). In the tolerant allograft, FoxP3⁺ cells (C) and CD4⁺ (D) cells were detected, which overlapped when merged (E) suggesting that the CD4⁺ cells expressed FoxP3. In contrast, none of the syngeneic or rejected allografts showed any FoxP3-expressing cells, as reflected by the merged stains of both FoxP3 and CD4 stains in F and G, respectively. (Magnification for C–E: ×600; magnification for F and G: ×200.)H) Lastly, splenic CD4⁺ T cells from these tolerant animals were used as responder cells in an ELISPOT assay that measured IL-2 production. Controls were animals pre-sensitized with two separate IP injections of donor splenocytes and non-transplanted control animals. The sensitized animals showed a higher number of IL-2 spots against the donor splenocytes, as expected. However, the control non-sensitized and tolerant animals showed equal numbers of spots, suggesting that indeed the tolerant animals had no activated alloreactive T cells. All three groups of animals responded equally to third-party Balb/c splenocytes.