Abstract
The bone marrow (BM) not only serves as a primary, but also as a secondary lymphoid organ, since it can mediate primary T cell responses against blood-borne antigens and it harbors a significant portion of the memory T cell compartment. Yet, it remains unclear to what extent BM T cells affect the local hematopoietic process. This is important from a clinical perspective, since the development of BM failure and anemia is frequently associated with (chronic) T cell-mediated inflammatory diseases, such as rheumatoid arthritis and viral infections. We postulate that particularly hematopoietic stem cells (HSCs) may be susceptible to T cell activity, since HSCs are localized in endothelial BM niches and are thus in close vicinity of where T cells enter the BM parenchyma and get activated. In support of this, we have previously shown that IFNγ, one of the key cytokines produced by activated T cells, strongly impairs HSC self-renewal and enhances their differentiation towards monoctyes, at the expense of neutrophils and erythrocytes.
To examine the impact of T cells on HSC function, we performed co-culture assays and found that T cells from murine BM actually have a positive impact on HSC function, as they enhance both their differentiation and self-renewal capacity. This feature is restricted to a subset of memory CD8 T cells in the BM, since neither naïve T cells from BM nor memory CD8 T cells from the spleen showed the same effect. Correspondingly, transgenic mice with only naïve and no memory T cells have fewer HSC numbers than control mice, which can be transiently restored when memory CD8 BM T cells are injected. To test the relevance of these findings in an inflammatory setting, we infected mice with the Armstrong-strain of lymphocytic choriomeningitis virus (LCMV), which induces an acute infection that leads to a strong influx of antigen-experienced T cells in the BM. We found that LCMV-specific memory CD8 T cells isolated from BM 12 days after infection increased both the differentiation and self-renewal capacity of HSCs. Interestingly, HSCs isolated from infected mice also displayed an enhanced propensity to differentiate towards myeloid cells compared to HSCs from non-infected control mice, whereas their self-renewal capacity was not altered. To test whether chronically stimulated T cells are also able to influence HSC function, we infected mice with LCMV clone 13, which leads to a chronic infection and induces exhaustion of the virus-specific T cells. Interestingly, virus-specific T cells isolated from BM 27 days after infection, which were exhausted based on phenotype and function, did not influence HSC differentiation but they were still able to enhance the self-renewal of HSCs from non-infected control mice, although to a lesser extent than in the acute infection. These data illustrate that antigen-experienced memory CD8 T cells in BM have a positive impact on the function of HSCs. Although cytokines produced by activated T cells, such as IFNγ and TNFα, can dramatically impair HSC maintenance, it is intriguing that memory T cells can actually fulfill a positive function on the HSC compartment. We speculate that homing of memory T cells to the BM after viral infection may play an important role in restoring the damage on the hematopoietic compartment that is inflicted by the infection itself and the ensuing cytokine storm. Enhancement of such a positive feedback mechanism may be a promising new strategy for treatment of patients with BM failure.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.