- Top of page
- Patients and methods
The changes in bone marrow (BM) stem cell reserve and function and stromal cell function in patients with active systemic lupus erythematosus (SLE) were investigated. The study was carried out on seven SLE patients and 28 healthy controls using flow cytometry and in vitro cell culture assays. We found that patients had low CD34+ cells, compared with the control group, reflecting the decrease of both CD34+/CD38− and CD34+/CD38+ cells. Patient CD34+/Fas+ but not CD34−/Fas+ cells were significantly increased. Apoptotic (7AADdim) cells were higher among CD34+/Fas+ than among CD34+/Fas− cells, and individual values of apoptotic CD34+ cells strongly correlated with the number of CD34+/Fas+ cells. These findings are suggestive of a Fas-mediated apoptosis accounting for the low CD34+ cells in SLE patients. Moreover, we found that patients had low numbers of granulocyte-macrophage colony-forming units (CFU-GM) and erythroid burst-forming units (BFU-E), compared with the control group, and that the generation of colony-forming cells in long-term BM cultures was significantly reduced. Patient BM stroma failed to support allogeneic progenitor cell growth. In one patient, CD34+ cells were increased, apoptotic CD34+/Fas+ cells were normalized and defective stromal cell function was restored after autologous stem cell transplantation. We concluded that defective haemopoiesis in SLE patients is probably caused, at least in part, to the presence of autoreactive lymphocytes in BM.
High-dose chemotherapy with immunoablative intent followed by autologous stem cell transplantation (ASCT) has been proposed recently as a putative treatment modality for severe autoimmune diseases including systemic lupus erythematosus (SLE) refractory to conventional medications (Tyndall & Gratwohl, 1997; van Bekkum, 1999; Burt et al, 2000; Traynor et al, 2000). The rationale for this therapeutic approach relies on the fact that a vigorous preparative regimen may eliminate autoreactive immunocompetent lymphocytic clones leading to immune deregulation, whereas the subsequent ASCT may rescue the patient from prolonged cytopenias. It has also been suggested that immune reconstitution after stem transplantation, even by autologous stem cells, may lead to ontogenic re-organization of the immune system with re-induction of tolerance to self-antigens (Emmons & Quesenberry, 1999).
Immune deregulation in autoimmune diseases may affect bone marrow (BM) progenitor cell development at several stages of differentiation and/or BM stromal cell function (Otsuka et al, 1988; Atta et al, 1994). Ikehara (1998a), based on animal models, suggested that autoimmune diseases may be seen as primary stem cell disorders. In humans, serum autoantibodies capable of inhibiting granulocyte and erythroid colony formation in vitro were detected in a number of SLE patients (Bailey et al, 1989; Liu et al, 1995). T-lymphocyte-mediated suppression of colony growth, even at the level of multipotential haemopoietic progenitors, as well as antibodies directed against fibronectin affecting cell-to-matrix interactions, have also been reported (Otsuka et al, 1988; Atta et al, 1994). However, little is known about the number and functional characteristics of BM stem cells, and even less about the capacity of BM microenvironment to support progenitor cell growth in SLE patients. In the current study, we present data suggestive of the existence of significant changes in BM stem cell reserve and function and BM stromal cell function in SLE patients. The data from a patient studied before and after ASCT are also discussed.
- Top of page
- Patients and methods
There is increasing evidence that haemopoiesis may be severely disturbed in patients with SLE. It has been suggested that many of these disturbances could be attributed to the presence of autoreactive immunocompetent lymphocytes and the action of a variety of pro-inflammatory cytokines (Otsuka et al, 1993; Tsokos, 1994). However, BM stem cell reserve and function and BM microenvironment function have not extensively been studied in SLE patients. The present study describes a stem cell defect in SLE patients indicated by the low number of CD34+ cells in flow cytometry, the low frequency of CFU-GM and BFU-E in clonogenic assays, the low progenitor cell recovery in LTBMCs and the increased proportion of apoptotic cells within the CD34+ cell compartment. from a theoretical point of view, these abnormalities could be attributed either to a primary stem cell defect (Ikehara, 1998a) or to a stem cell damage closely related to immune deregulation in SLE (Tsokos, 1994) or even to an abnormal function of BM microenvironment (Otsuka et al, 1993). The possibility for a drug-related effect cannot be ruled out for certain, but it seems unlikely here as only two of our patients had ever been exposed to cytotoxic drugs such as azathioprine or cyclophosphamide. Corticosteroids and cyclosporine can induce apoptosis in mitogen-stimulated peripheral blood mononuclear cells (Horigome et al, 1997) but, to our knowledge, they do not damage the haemopoietic stem cells. In contrast, there is strong evidence for a positive effect of corticosteroids on the proliferation and differentiation of BM progenitor cells (Rinehart et al, 1997; Von Lindern et al, 1999; Clausen et al, 2000).
A primary stem cell defect in autoimmune diseases has been suggested by Ikehara (1998a, b) in animal models, but there is no evidence for such a defect in humans. In keeping with the latter aspect is the observation that both increased proportion of Fas+ cells and increased proportion of apoptotic cells within the CD34+ cell population returned to normal range and were associated with a significant rise in the number of CD34+ cells after ASCT in one of the patients studied. On the other hand, the defective colony formation found in our patient after ASCT does not necessarily indicate a primary stem cell defect in SLE given that significant reduction in the clonogenic potential of CD34+ cells may be seen for several years after the recovery from autologous or allogeneic stem cell transplantation performed for unrelated reasons (Novitzky & Mohammed, 1997).
The role of autoreactive immunocompetent lymphocytes in the pathogenesis of SLE has been studied previously by several investigators. Because these cells are present in patient BMMCs and LTBMC adherent layers (Berneman et al, 1989; Otsuka et al, 1993), one could postulate that a mechanism of immune-mediated inhibition of colony formation, or even an immune-mediated apoptotic stem cell depletion, may be operative in SLE patients. Indeed, removal of T lymphocytes from SLE marrow samples has been reported to result in a significant rise in the progenitor cell clonogenic potential (Kiely et al, 1995). Moreover, it has been shown that Fas-mediated induction of apoptosis by immune cytokines is normally involved in the maintenance of homeostasis in BM haemopoietic progenitor cell population (Nagafuji et al, 1995). In the present study, Fas antigen was significantly upregulated in patient CD34+ cells, and a highly significant correlation was noted between the proportion of apoptotic cells and the proportion of Fas+ cells within the CD34+ cell compartment. Although a non-functional accumulation of Fas in patient CD34+ cells cannot be ruled out, this hypothesis seems unlikely because the proportion of apoptotic cells was much higher in the Fas-positive than in the Fas-negative stem cell population in the subjects studied. This suggests that the Fas pathway is actively involved in the induction of apoptosis in BM progenitor cells in SLE patients. Although Fas expression does not always lead to apoptotic cell death (Miyawaki et al, 1992), involvement of Fas antigen in the induction of apoptosis within the stem cell compartment has already been well documented in a number of other disease states (Maciejewski et al, 1995; Saheki et al, 2000).
The mechanisms leading to upregulation of Fas in CD34+ cells in SLE patients are unknown. Normal BM CD34+ cells do not express Fas antigen on their surface, or they express the molecule at a proportion < 10% (Maciejewski et al, 1995; Saheki et al, 2000). It has been reported that interferon-gamma (IFN-γ) and tumour necrosis factor-alpha (TNF-α) can induce the expression of functional Fas molecules on cultured normal haemopoietic progenitor cells (Nagafuji et al, 1995). Although IFN-γ and TNF-α have rarely been found significantly increased in the sera of SLE patients (Aderka et al, 1993; Lacki et al, 1997), they can be produced in increased amounts at sites of tissue inflammation such as in the kidney (Lacki et al, 1997; Kelley & Wuthrich, 1999). An aberrant TNF-α mRNA expression in BM has also been reported in SLE patients (Alvarado-de la Barrera et al, 1998). Fas-mediated apoptosis of haemopoietic stem cells caused by increased production of TNF-α and IFN-γ by activated BM cytotoxic T lymphocytes has been well documented in patients with aplastic anaemia (Maciejewski et al, 1995), a disease usually caused by immune-mediated inhibition of haemopoiesis (Zoumbos et al, 1985; Young, 1995).
The present study also describes a defect in BM microenvironment in SLE patients, in terms of the inability of patient LTBMC adherent layers to support haemopoietic progenitor cell growth in a two-stage cell culture system. The fact that this defective function has been restored by ASCT provides additional evidence for the secondary cause of the disturbance, which may be caused by the presence of autoreactive lymphocytes within the stromal cell population. The observation that isolated BM fibroblasts from SLE patients gave normal fibroblast colonies in vitro (Otsuka et al, 1993) is in agreement with this suggestion. On the other hand, abnormally produced cytokines in SLE (Tsokos, 1994; Horwitz et al, 1998), may also affect the ability of BM microenvironment to support haemopoiesis.
In conclusion, BM primitive CD34+/CD38− and committed CD34+/CD38+ stem cells were both reduced in our SLE patients. Patients had also increased apoptosis within the CD34+ cell compartment, reduced frequency of CFC and defective haemopoiesis supporting capacity of LTBMC adherent layers. The cause and the underlying pathogenetic mechanisms of these abnormalities are presently unknown, but it seems to be related to a defective function of BM microenvironment rather than a primary stem cell defect. The role of autoreactive lymphocytes and the changes in the cytokine network within the BM microenvironment remain to be elucidated. Our data also provide some evidence that an intensive immunosuppression to eradicate autoreactive lymphocytic clones, followed by an autologous stem cell graft to offer probably a more normal set of haemopoietic cells may restore, at least in part, some of the abnormalities of haemopoiesis found in SLE patients.