Murine hematopoietic stem cells change their surface phenotype during ex vivo expansion

Abstract

Ex vivo expansion of hematopoietic stem cells (HSCs) is important for many clinical applications, and knowledge of the surface phenotype of ex vivo-expanded HSCs will be critical to their purification and analysis. Here, we developed a simple culture system for bone marrow (BM) HSCs using low levels of stem cell factor (SCF), thrombopoietin (TPO), insulin-like growth factor 2 (IGF-2), and fibroblast growth factor-1 (FGF-1) in serum-free medium. As measured by competitive repopulation analyses, there was a more than 20-fold increase in numbers of long-term (LT)-HSCs after a 10-day culture of total BM cells. Culture of BM "side population" (SP) cells, a highly enriched stem cell population, for 10 days resulted in an approximate 8-fold expansion of repopulating HSCs. Similar to freshly isolated HSCs, repopulating HSCs after culture were positive for the stem cell markers Sca-1, Kit, and CD31 and receptors for IGF-2. Surprisingly, prion protein and Tie-2, which are present on freshly isolated HSCs, were not on cultured HSCs. Two other HSC markers, Endoglin and Mpl, were expressed only on a portion of cultured HSCs. Therefore, the surface phenotype of ex vivo-expanded HSCs is different from that of freshly isolated HSCs, but this plasticity of surface phenotype does not significantly alter their repopulation capability.

Figure 1.

A novel culture system of total BM cells dramatically expands HSCs. (A) Total BM cells (10⁶) were initiated in serum-free medium with SCF, TPO, IGF-2, and FGF-1 as described in “Materials and methods,” and total cell numbers were counted at days 7, 10, 14, 21, and 28. Results from 3 independent cultures were plotted. (B) Comparison of the long-term repopulation potential of 10-day cultured and freshly isolated BM cells. We mixed 5 × 10⁴, 1 × 10⁴, 5000, 2500, or 1250 freshly isolated CD45.2 BM cells or 3.5 × 10⁵, 7 × 10⁴, 3.5 × 10⁴, 1.75 × 10⁴, or 9000 10-day cultured BM cells (the product of 5 × 10⁴, 1 × 10⁴, 5000, 2500, or 1250 initially plated CD45.2 cells, respectively) with 10⁵ CD45.1 competitor BM cells and transplanted them into lethally irradiated recipients (n = 6 mice). In 1 case IGF-2 was not added, but the remainder of the culture conditions was unchanged. Peripheral blood cells were analyzed for the presence of CD45.2⁺ cells at 4 months after transplantation. Three independent experiments were performed that gave similar results. (C) Multilineage contribution of 3.5 × 10⁵ cultured cells (derived from 5 × 10⁴ input cells) at 4 months after transplantation (n = 6). (D) Multilineage contribution of the 3.5 × 10⁵ cultured cells (equivalent to 5 × 10⁴ input cells) at 4 months after transplantation of mice receiving a secondary transplant (n = 4). (E) Limiting dilution analysis of the repopulating ability of total BM cells before and after culture. Irradiated CD45.1 congenic mice were injected with 10⁵ CD45.1 BM competitor cells and the indicated numbers of freshly isolated CD45.2 BM cells (▪ and —) or their progeny after 10 days of culture in serum-free medium with SCF, TPO, IGF-2, and FGF-1 (▿ and —). Plotted is the percentage of recipient mice containing less than 1% CD45.2 lymphoid and myeloid subpopulations in nucleated peripheral blood cells 4 months after transplantation versus the number of injected cells. The curve was anchored by the 0 cells/100% negative mice point. Error bars indicate SEM.

Figure 2.

Culture dramatically increases in vivo repopulating stem-cell activity of BM SP cells. (A) Freshly isolated adult CD45.2 BM SP cells (2, 5, 25, or 100) or their progenies after 10 days of culture were transplanted (together with 1 × 10⁵ CD45.1 competitor BM cells per mouse, n = 5-8) into CD45.1 congenic mice. Peripheral blood engraftments at 4 weeks and 4 months after transplantation are shown. Error bars indicate SEM. (B) Limiting dilution analysis of the repopulating ability of BM SP cells before culture. Irradiated CD45.1 congenic mice were injected with 10⁵ CD45.1 BM competitor cells and 2 (n = 7 mice), 5 (n = 6), 25 (n = 8), or 100 (n = 5) freshly isolated CD45.2 BM SP cells. Similar to Figure 1E, plotted is the percentage of recipient mice containing less than 1% CD45.2 lymphoid and myeloid subpopulations in nucleated peripheral blood cells 4 months after transplantation versus the number of injected cells. Input SP cells (100) resulted in 0% of negative mice, and this data point is not plotted. (C) Limiting dilution analysis of the repopulating ability of BM SP cells after culture. The same assay as used in panel B was carried out except the progenies of the input 2 (n = 5), 5 (n = 6), 25 (n = 4), or 100 SP cells (n = 5) after 10 days of culture were injected. The cultured progeny of 25 or 100 input SP cells resulted in 0% of negative mice, and the data points are not plotted.

Figure 3.

All HSCs reside in the Kit- and Sca-1–positive fraction of cultured BM cells. (A) Ten-day cultured BM cells were stained with a cocktail of biotinylated lineage-specific antibodies, followed by streptavidin-APC, anti–Sca-1–FITC, and anti-Kit–PE. Forward scatter (FSC) and side scatter (SSC) on the left plot is used to gate on hematopoietic cells. In the middle and right plots, Lin^- (negative APC-stained) and propidium iodide–negative (PI^-) cells were gated to show surface expression of Sca-1 and Kit. Numbers in the graph are the percentages of each cell fraction. (B) Expanded HSCs in cultured BM cells are Sca-1⁺Kit⁺. After 10 days of culture of total BM cells, 1.3 × 10⁴ sorted CD45.2 Sca-1⁺Kit⁺, Sca-1⁺Kit^-, Sca-1^-Kit⁺, or Sca-1^-Kit^- cells were transplanted together with 2 × 10⁵ CD45.1 competitor cells into lethally irradiated CD45.1 mice (n = 4). Peripheral blood engraftments at 3 weeks and 4 months after transplantation are shown. Error bars indicate SEM.

Figure 4.

Ex vivo–expanded BM HSCs are in both Endoglin-positive and -negative fractions. After 10 days of culture of total BM cells, 6000 sorted Endoglin⁺ and 2.4 × 10⁴ Endoglin^- cells or 3000 Lin^-Endoglin^- and 3000 Lin^-Endoglin⁺ cells were transplanted together with 2 × 10⁵ CD45.1 competitor cells into lethally irradiated CD45.1 mice (n = 4). Shown are peripheral blood engraftment at 4 weeks (left) and 4 months (right) after transplantation. Different numbers of sorted Endoglin⁺ Endoglin^- cells were used because after culture there were 3 times more Endoglin^- cells (76% of the total) than Endoglin⁺ cells (26%). Error bars indicate SEM.

Figure 5.

All ex vivo–expanded BM HSCs are PrP^-. (A) Either 10⁵ PrP^- or 2 × 10⁴ PrP⁺ freshly isolated CD45.2 donor BM cells were mixed with 10⁵ competitor CD45.1 cells and transplanted into lethally irradiated CD45.1 mice (n = 4). Donor CD45.2 contribution at 3 weeks (left) and 7 months (right) after transplantation are shown. (B) PrP⁺ and PrP^- cells (5000) sorted from 10-day cultured total BM or 3000 Lin^-Sca-1⁺PrP⁺ and Lin^-Sca-1⁺PrP^- cells sorted from 4-day cultured total BM cells were transplanted together with 2 × 10⁵ CD45.1 competitor cells into lethally irradiated CD45.1 mice (n = 4). Peripheral blood engraftments at 3 weeks (left) and 7 months (right) after transplantation are shown. Error bars indicate SEM.

Figure 6.

Ex vivo–expanded BM HSCs are CD62L^-, IGF2-hFc⁺, CD31⁺, Tie-2^-, and prominin-1^-. After 10 days of culture of total BM cells, 4 × 10⁴ sorted CD62L⁺ and CD62L^- cells, or IGF2-hFc⁺ and IGF2-hFc^- cells, or Mpl⁺ and Mpl^- cells, or CD31⁺ and CD31^- cells, or 6000 sorted Tie-2⁺ and Tie-2^- cells, or 6000 prominin-1⁺ and prominin-1^- cells were transplanted, respectively, together with 10⁵ CD45.1 competitor cells into lethally irradiated CD45.1 mice (n = 4). Peripheral blood engraftments at 4 months after transplantation are shown. Error bars indicate SEM.