Hogge D E, Cashman J D, Humphries R K, Eaves C J
Terry Fox Laboratory, British Columbia Cancer Agency, Department of Medicine, University of British Columbia, Vancouver, Canada.
Blood. 1991 Feb 1;77(3):493-9.
The ability of granulocyte-macrophage colony-stimulating factor (GM-CSF) and G-CSF to influence hematopoiesis in long-term cultures (LTC) of human marrow was studied by cocultivating light density normal human marrow cells with human marrow fibroblast feeders engineered by retroviral infection to constitutively produce one of these growth factors. Feeders producing stable levels of 4 ng/mL GM-CSF or 20 ng/mL G-CSF doubled the output of mature nonadherent cells. The numbers of both colony forming unit-GM (CFU-GM) and erythroid burst forming unit (BFU-E) in the G-CSF LTC were also increased (twofold and fourfold, respectively, after 5 weeks in culture), but this effect was not seen with the GM-CSF feeders. At the time of the weekly half medium change 3H-thymidine suicide assays showed primitive adherent layer progenitors in LTC to be quiescent in both the control and GM-CSF cultures. In contrast, in the G-CSF cultures, a high proportion of primitive progenitors were in S-phase. A single addition of either recombinant GM-CSF or G-CSF to LTC in doses as high as 80 ng/mL and 150 ng/mL, respectively, failed to induce primitive progenitor cycling. However, three sequential daily additions of 150 ng/mL G-CSF did stimulate primitive progenitors to enter S-phase and a single addition of 5 or 12.5 ng/mL of G-CSF together with 10 ng/mL GM-CSF was able to elicit the same effect. Thus, selective elevation of G-CSF in human LTC stimulates proliferation of primitive clonogenic progenitors, which may then proceed through to the terminal stages of granulopoiesis. In contrast, the effects of GM-CSF in this system appear limited to terminally differentiating granulopoietic cells. However, when both GM-CSF and G-CSF are provided together, otherwise biologically inactive doses show strong stimulatory activity. These findings suggest that the production of both of these growth factors by normal stromal cells may contribute to the support and proliferation of hematopoietic cells, not only in LTC, but also in the microenvironment of the marrow in vivo.
通过将低密度正常人骨髓细胞与经逆转录病毒感染构建的、可组成性产生粒细胞-巨噬细胞集落刺激因子（GM-CSF）或粒细胞集落刺激因子（G-CSF）之一的人骨髓成纤维细胞饲养层共同培养，研究了GM-CSF和G-CSF在人骨髓长期培养（LTC）中影响造血的能力。产生稳定水平4 ng/mL GM-CSF或20 ng/mL G-CSF的饲养层使成熟非贴壁细胞的产量增加了一倍。G-CSF长期培养体系中集落形成单位-粒细胞-巨噬细胞（CFU-GM）和红系爆式集落形成单位（BFU-E）的数量也增加了（培养5周后分别增加了两倍和四倍），但GM-CSF饲养层未见此效应。在每周更换一半培养基时，3H-胸腺嘧啶核苷自杀试验显示，长期培养体系中原始贴壁层祖细胞在对照和GM-CSF培养体系中均处于静止状态。相反，在G-CSF培养体系中，高比例的原始祖细胞处于S期。分别向长期培养体系中单次添加高达80 ng/mL的重组GM-CSF或150 ng/mL的G-CSF，均未能诱导原始祖细胞进入细胞周期。然而，连续三天每天添加150 ng/mL G-CSF确实能刺激原始祖细胞进入S期，单次添加5 ng/mL或12.5 ng/mL G-CSF与10 ng/mL GM-CSF一起也能产生相同效果。因此，人长期培养体系中G-CSF的选择性升高可刺激原始克隆形成祖细胞的增殖，这些祖细胞随后可能会进入粒细胞生成的终末阶段。相比之下，GM-CSF在该体系中的作用似乎仅限于终末分化的粒细胞。然而，当GM-CSF和G-CSF同时存在时，原本无生物学活性的剂量也显示出强大的刺激活性。这些发现表明，正常基质细胞产生的这两种生长因子不仅可能在长期培养体系中，而且可能在体内骨髓微环境中，对造血细胞的支持和增殖起作用。
