Co-Transplantation of Stromal Cells Enhances Engraftment Ability of MDS-Derived Hematopoietic Precursors in NOD/SCID-b2microglobulin-Deficient (b2m^null) Mice.

Propagation of hemopoietic clones derived from MDS marrow in vitro or in vivo has proven difficult. We recently showed engraftment of clonal cells identified by fluorescence in situ hybridization (FISH) markers, in b2m^null mice using MDS bone marrow mononuclear cells (BMMC) injected intramedullarly along with the human stroma-derived cell lines, HS5 and HS27a. To define the role of stromal cells for transplant success, we conducted additional in vitro and in vivo studies.

Methods:

1. Using NOD/SCID-b2m^null mice (irradiated with 350cGy), MDS BMMC (10⁷cells) or purified CD34+ MDS-derived hematopoietic precursors (10⁶cells) were transplanted into the right femur with or without the addition of HS5 and HS27a (2x10⁵) stroma cells. Engraftment was defined as presence of ≥ 0.2% human CD45+ cells in peripheral blood.

2. To document the presence of human stroma, mice were transplanted with GFP transduced HS5 and HS27a cells, and sacrificed 1h or 24h after transplant. Marrow cells from the femora were evaluated for the presence of GFP signals by flow cytometry.

3. MDS-derived hematopoietic precursors from patients with identifiable FISH markers, were sorted from the blast gate (CD45 dim, low side scatter [ SSC]), and plated on stromal feeder layers (either HS5 or HS27a, irradiated with 1800cGy), in contact or transwell cultures. After one week, cells were recovered and analyzed by FISH, and also plated in methocult. Colonies were plucked at two weeks and also analyzed by FISH.

Results: 11 of 13 mice transplanted without co-injection of stroma cells showed engraftment of human cells. At six weeks, clonal cells were detected in four of six mice with a FISH marker [−5 and del(5q)], suggesting that stromal cells were not an absolute requirement for the propagation of clonal cells. However, in mice transplanted with cells from the same marrow source with or without stromal cells, the engraftment rate was higher with stroma (4/7, 57%) than without stroma (2/6, 33%), as was the proportion of human hemopoietic cells in marrow (6.4±2.7% vs 2.2±0.4%, respectively). The proportion of human cells was higher at the injection site than in the contralateral femur. Furthermore, clonal (FISH+) cells were detected in 3 of 4 mice in the stroma group (del(5q) and +8), but not in two mice that were not injected with stroma but showed engraftment of “normal” human cells. Among mice transplanted with cells from a patient with +8, only those in the stroma group showed engraftment (clonal and non-clonal cells). Nine mice transplanted with CD34+ MDS cells (7 with and 2 without addition of stroma) did not show engraftment, suggesting that not only stroma but also other accessory cells are important for the tranplant success. GFP+ stromal cells were recognized in mouse bone marrow obtained from the injection site, but not contralaterally. In vitro, clonal cells (−7) were recovered from short-term assays on HS5 or HS27a cells, from contact and trans-well cultures. However, colonies were formed only from cells in contact cultures; HS5 stroma (a rich source of cytokines) favored clonal (FISH+) cell growth.

Conclusions: Thus, the intramedullary co-transplantation of stromal cells enhanced the engraftment ability of normal and clonal cells from human MDS marrow. In vitro data suggest that cell-to-cell contact is necessary, and that clonal cell growth may have an advantage in the presence of HS5 cells.