LEUKEMIA INHIBITORY FACTOR UP-REGULATES CYTOKINE EXPRESSION BY A MURINE STROMAL CELL-LINE ENABLING THE MAINTENANCE OF HIGHLY ENRICHED COMPETITIVE REPOPULATING STEM-CELLS

Attempts to maintain or expand primitive hematopoietic stem cells in v itro without the concomitant loss of their differentiative and prolife rative potential in vivo have largely been unsuccessful. To investigat e this problem, we compared the ability of three cloned bone marrow (B M) stromal cell lines to support the growth of primitive Thy-1(lo)Sca- 1(+)H-2K(hi) cells isolated by fluorescence-activated cell sorting fro m the BM of Ly-5.2 mice treated 1 day previously with ti-fluorouracil. Sorted cells were highly enriched in cobblestone area-forming cells ( CAFC), but their frequency was dependent on the stromal cell lines use d in this assay (1 per 45 cells on SyS-1; 1 per 97 cells on PA6). In t he presence of recombinant leukemia inhibitory factor (LIF), CAFC clon ing efficiency was increased to 1 per 8 cells on SyS-1 and 1 per 11 ce lls on PA6, thus showing the high clonogenicity of this primitive stem cell population. More primitive stem cells with competitive repopulat ing potential were measured by injecting the sorted cells into lethall y irradiated Ly-5.1 mice together with 10(5) radioprotective Ly-5.1 BM cells whose long-term repopulating ability has been ''compromised'' b y two previous cycles of marrow transplantation and regeneration. Dono r-derived lymphocytes and granulocytes were detected in 66% of animals injected with 50 sorted cells. To quantitate the maintenance of compe titive repopulating units (CRU) by stromal cells, sorted cells were tr ansplanted at limiting dilution before and after being cultured for 2 weeks on adherent layers of SyS-1, PA6, or S17 cells. CRU represented 1 per 55 freshly sorted cells. CRU could be recovered from cocultures supported by all three stromal cell lines, but their numbers were simi lar to sevenfold less than on day 0. In contrast, the addition of LIF to stromal cultures improved CRU survival by 2.5-fold on S17 and PA6 c ells (similar to two-fold to threefold decline), and enabled their mai ntenance on SyS-1. LIF appeared to act indirectly, because alone it di d not support the proliferation of Thy-1(lo)Sca-1(+)H-2K(hi) cells in stroma-free cultures. Polymerase chain reaction (RT-PCR) analysis reve aled that Interleukin-1 beta (IL-1 beta) IL-2, IL-6, granulocyte-colon y stimulating factor, granulocyte macrophage-colony stimulating factor , transforming growth factor-beta, LIF, and Steel Factor (SLF) mRNAs w ere upregulated in SyS-1 within 1 to 6 hours of LIF-stimulation. To de termine if increased expression of SLF by LIF-stimulated SyS-1 cells c ould account for their capacity to support stem cells, sorted cells we re cocultured on simian CV-E cells that were transfected with an expre ssion vector encoding membrane-bound SLF, or supplemented with soluble SLF. In both cases, SLF synergized with IL-6 produced endogenously by CV-E cells enabling CAFC growth equivalent to that on LIF-stimulated SyS-1. CAFC development on LIF-stimulated SyS-1 could also be complete ly abrogated by an anti-SLF antibody. These data provide evidence for a role of LIF in the support of long-term repopulating stem cells by i ndirectly promoting cytokine expression by BM stroma. Furthermore, we have used quantitative assays to show a maintenance of CRU numbers, wi th retention of in vivo function following ex vivo culture. (C) 1996 b y The American Society of Hematology.