The stromal component of the marrow microenvironment is not derived from the malignant clone in MDS

Aravind Ramakrishnan, Norihiro Awaya, Eileen Bryant and Beverly Torok-Storb

Myelodysplastic syndromes (MDSs) represent a spectrum of disorders that are generally thought to arise from a defective hematopoietic stem cell leading to clonal, dysregulated hematopoiesis. Although it is generally agreed that the marrow microenvironment plays a role in the biology of MDSs, it is unclear whether this represents an intrinsically abnormal stromal compartment derived from the MDS clone. One report does suggest that stromal cells contain MDS-associated cytogenetic abnormalities,1 while others have contradicted these findings and suggest these cells function normally.2,3 These conflicting observations, along with the differing reports on hematopoietic stem cell plasticity, prompted us to evaluate this issue directly using long-term marrow cultures (LTCs) established from MDS patients.

We first addressed the issue of the origin of stroma in 1987 by using LTCs established with marrow aspirates from patients with functioning sex-mismatched allografts.4 By using fluorescence in situ hybridization (FISH), we were able to demonstrate that all of the stromal cells were host derived, whereas all of the cells containing donor sex chromatin were macrophages. We subsequently extended this observation to first document by nonspecific esterase (NSE) staining that up to 40% of cells in primary LTCs can be macrophages and, as expected, originate from the donor. In this case, we demonstrated that even after 27 years of 100% donor-derived hematopoiesis, the stromal cells still originated from the host.5

One could infer from these data that the stroma in patients with MDSs would not be derived from the malignant clone. To address this issue directly, we analyzed LTCs established from 9 patients with MDSs. LTCs were cultured for 6 to 9 weeks before cells were harvested and cytocentrifuged onto glass slides. The percentage of macrophages was determined by staining with NSE. FISH was then performed on duplicate slides to determine the percentage of clonally marked cells. In every case except 1, the percent of clonally marked cells in the LTCs did not exceed that which could be accounted for by the combination of macrophages and background levels for FISH (Table 1). In case no. 7, we noted an unusual retention of myelocytes in the LTC, which probably contributed to the increased percentage of clonally marked cells noted in this patient. These data strongly suggest that stromal cells are not part of the malignant clone in MDSs and support our previous conclusion that stromal cells and hematopoietic cells are not derived from a common marrow-derived adult stem cell.

Table 1.

Histochemical analysis (NSE) and FISH of LTCs from 9 patients with MDS


  • Correspondence: Beverly Torok-Storb, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave N, Seattle, WA 98109; e-mail: btorokst{at}
  • Supported in part by T32 CA09515 (AR), ROI HL62923, and ROI HL69144.