Background A key property of the vertebrate immune system is its ability to establish and maintain long-lasting protection and fast, adapted reaction against previously encountered pathogens, the immunological memory. However, immunological memory can also be detrimental, when directed against self antigens and we hypothezise that it is driving chronific inflammations. The cellular correlates of immunological memory are memory plasma cells, memory B cells, memory CD4+ T helper cells and memory CD8+ cytotoxic T (Tc) cells. However, it is not well understood, how immunological memory is organized and how it is maintained longterm up to the entire lifetime of an organism.
Objectives Cells of the immunological memory are not long-lived intrinsically, but dependent on extrinsic signals provided by dedicated stroma cells. We have investigated the maintenance and organization of immunological memory by the stroma of the bone marrow.
Methods We have established a murine bone marrow chimeric system which allows the visualization of the stroma cell network in the bone marrow. Together with a nearest-neighbor analysis and pulse-labeling experiments we have used this system to define the components of the survival niche and their stability for the cells of the immunological memory.
Results We could show that memory plasma cells and memory T cells are maintained in niches defined by VCAM-1+ stroma cells of the bone marrow expressing CXCL12 and IL-7, respectively. In addition to the CXCL12-expressing stroma cell, eosinophils providing the survival factor APRIL are an essential component of the survival niche for memory plasma cells. In contrast to the memory plasma cells and the stroma cells which were long-lived and did not show any indication of proliferation, eosinophils constituting the plasma cell survival niche displayed a high turnover. Apparently, the plasma cell survival niche in the bone marrow is highly dynamic and maintains its stability through the constant exchange of eosinophils with the stroma cell as central organizer.
The niche for CD4+ memory T cells seems to be organized by particular IL-7+ stroma cells, binding to which is mediated by CD49b (alpha 2 integrin). Blockade of CD49b with antibodies prevents the settling of CD4+ memory T cells in the bone marrow. In addition, blockade of CD49b with antibodies in the memory phase, i.e. after CD4+ memory T cells have settled in the bone marrow, resulted in the depletion of memory CD4+ T cells from the bone marrow. CD8+ memory T cells also accumulate in the bone marrow and are maintained there in stable numbers over long time periods as non-proliferating cells. In the bone marrow he CD8+ memory T cells also colocalized with IL-7-expressing stroma cells. However, CD8+ memory T cells showed no detectable expression of CD49b, indicating that the IL-7+ stroma cells supporting CD8+ memory T cells are distinct from those supporting CD4+ memory T cells.
Conclusions Our data indicate, that stroma cells of the bone marrow play a central role in the organization and maintenance of immunological memory. Apparently the bone marrow is a privileged site housing immunological memory. Our understanding of how cells of the immunological memory, in particular memory cells maintaining rheumatic inflammation, are maintained could help us to therapeutically interfere with their survival and to selectively eliminate rheumatic immunological memory.
Disclosure of Interest None Declared