CMV‐specific central memory T cells reside in bone marrow

Abstract

CMV-specific CD8 + T cell responses in peripheral blood (PB) are characterized by a preponderance of effector and effector memory T cells. CMV-specific central memory T cells (T CM ), which are considered crucial in maintaining long-term immunity, are rarely detectable in PB. In this study we have analyzed differentiation and function of CMV pp65-specific CD8 + T cells in paired samples of human PB and BM using intracellular cytokine and tetramer staining. Overall frequencies of CMV pp65-specific T cells were similar in PB compared to BM; however, CMV-specific CD45RA -CCR7 + T CM were almost exclusively detectable in BM, which was not related to a general accumulation of T CM in BM. In vitro, CMV-specific T cells could be more efficiently expanded from BM (median 128-fold, n=6) than from PB (median 72-fold, p=0.01). Taken together, these data show that the BM is a compartment harboring CMV-specific T CM and underline the concept of the BM as a secondary immune organ. CMV specific BM-derived T CM might be a valuable source for generating T cells for adoptive transfer. IntroductionBased on phenotypic and functional characteristics two main populations of memory T cells can be distinguished: central memory T cells (T CM ), expressing the chemokine receptor CCR7, and effector memory T cells (T EM ), lacking lymph node homing receptors and migrating to peripheral tissues [1]. Both subsets can differentiate into effector T cells (T Eff ). The generally accepted concept holds that T CM have a higher proliferative potential and a greater capacity than T EM for long-lasting persistence in vivo [2][3][4].There is growing evidence suggesting that CD8 + T CM play a key role in long-term protection against acute infections in vivo [5][6][7]. In the chronic phase of persistent infections, differences in T cell differentiation were found dependent on viral specificity [8]. In general, specific CD8 + T cell responses against CMV and other chronic viral infections found in peripheral blood (PB) were CCR7 -belonging to T EM and T EFF [8][9][10][11]. It is unclear yet, whether the failure to detect CMV-specific CCR7 + T CM is related to the specific situation of chronic infection, where continuous stimulation of T cells occurs, or is due to their retention in lymphoid tissues. Information concerning the anatomical distribution of T cell differentiation subsets of virusspecific T cells in humans is very limited. In the study by Ellefsen et al.[10], however, several fold lower * These two authors contributed equally to this work. frequencies of CMV pp65-specific tetramer + CD8 + T cells were found in lymph nodes compared to PB. Recently the BM was identified as an important site for accumulation and proliferation of memory T cells [12][13][14]. Furthermore, there is growing evidence of the BM as secondary lymphoid organ priming naive T cells (T N ) and also efficiently stimulating memory T cells [15,16].In this study we comparatively analyzed the differentiation subsets of CMV pp65-specific CD8 + T cells in human PB and BM samples o...

Human peripheral blood and bone marrow Epstein–Barr virus‐specific T‐cell repertoire in latent infection reveals distinct memory T‐cell subsets

“…In accordance with the proposed differentiation model of Seder the CD8 1 triple cytokine producers in our study had an intermediate differentiation memory phenotype based on the CCR7 expression level (Fig. 4C) [48].In previous studies we could detect enhanced frequencies of tumor-reactive and CMV-specific central memory T cells in BM [44,45,57]. Palendira et al…”

“…In previous studies we could detect enhanced frequencies of tumor-reactive and CMV-specific central memory T cells in BM [44,45,57]. Palendira et al…”

“…In previous studies we could detect enhanced frequencies of tumor-reactive and CMV-specific central memory T cells in BM [44,45,57]. Palendira et al [58] already compared frequencies of EBVspecific T cells in PB and BM showing threefold higher BZLF-1-specific T cells in BM by tetramer staining.…”

“…Using 15-mer overlapping proteinspanning peptide pools of latent EBNA-1 and lytic BZLF-1 we characterized EBV-specific memory CD4 1 and CD8 1 T cells based on cytokine-producing capability (IL-2, TNF, IFN-g), and differentiation marker expression (CCR7, CD45RA). As we and others could show that BM is an important compartment for harboring and homeostasis of memory T cells, for both human CMV [44,45] and mouse LCMV and OVA T-cell memory [46,47], we comparatively analyzed the T-cell response in PB and BM samples. Our data revealed the existence of six different EBVspecific memory T-cell subsets at distinct differentiation stages.…”

Heteroclitic XBP1 peptides evoke tumor-specific memory cytotoxic T lymphocytes against breast cancer, colon cancer, and pancreatic cancer cells

“…[51][52][53] However, there remain controversies in defining the immune response of the specific memory T subsets within antigen-specific CTL against tumor cells. 54,55 In these studies, we characterized the antitumor activities of the two different memory cell subsets, CM and EM, from XBP1 antigen-specific CTL against a variety of solid tumor cells.…”