Alloreactive memory T cells are responsible for the persistence of graft-versus-host disease

Graft-vs-host disease (GVHD) is caused by a donor T cell anti-host reaction that evolves over several weeks to months, suggesting a requirement for persistent alloreactive T cells. Using the C3H.SW anti-C57BL/6 (B6) mouse model of human GVHD directed against minor histocompatibility Ags, we found that donor CD8(+) T cells secreting high levels of IFN-gamma in GVHD B6 mice receiving C3H.SW naive CD8(+) T cells peaked by day 14, declined by day 28 after transplantation, and persisted thereafter, corresponding to the kinetics of a memory T cell response. Donor CD8(+) T cells recovered on day 42 after allogeneic bone marrow transplantation expressed the phenotype of CD44(high)CD122(high)CD25(low), were able to homeostatically survive in response to IL-2, IL-7, and IL-15 and rapidly proliferated upon restimulation with host dendritic cells. Both allogeneic effector memory (CD44(high)CD62L(low)) and central memory (CD44(high)CD62L(high)) CD8(+) T cells were identified in B6 mice with ongoing GVHD, with effector memory CD8(+) T cells as the dominant (>80%) population. Administration of these allogeneic memory CD8(+) T cells into secondary B6 recipients caused virulent GVHD. A similar allogeneic memory CD4(+) T cell population with the ability to mediate persistent GVHD was also identified in BALB/b mice receiving minor histocompatibility Ag-mismatched B6 T cell-replete bone marrow transplantation. These results indicate that allogeneic memory T cells are generated in vivo during GVH reactions and are able to cause GVHD, resulting in persistent host tissue injury. Thus, in vivo blockade of both alloreactive effector and memory T cell-mediated host tissue injury may prove to be valuable for GVHD prevention and treatment.     

Delayed expansion and contraction of CD8+ T cell response during infection with virulent Salmonella typhimurium

Ag presentation to CD8(+) T cells often commences immediately after infection, which facilitates their rapid expansion and control of infection. Subsequently, the primed cells undergo rapid contraction. We report that this paradigm is not followed during infection with virulent Salmonella enterica, serovar Typhimurium (ST), an intracellular bacterium that replicates within phagosomes of infected cells. Although susceptible mice die rapidly (approximately 7 days), resistant mice (129 x 1SvJ) harbor a chronic infection lasting approximately 60-90 days. Using rOVA-expressing ST (ST-OVA), we show that T cell priming is considerably delayed in the resistant mice. CD8(+) T cells that are induced during ST-OVA infection undergo delayed expansion, which peaks around day 21, and is followed by protracted contraction. Initially, ST-OVA induces a small population of cycling central phenotype (CD62L(high)IL-7Ralpha(high)CD44(high)) CD8(+) T cells. However, by day 14-21, majority of the primed CD8(+) T cells display an effector phenotype (CD62L(low)IL-7Ralpha(low)CD44(high)). Subsequently, a progressive increase in the numbers of effector memory phenotype cells (CD62L(low)IL-7Ralpha(high)CD44(high)) occurs. This differentiation program remained unchanged after accelerated removal of the pathogen with antibiotics, as majority of the primed cells displayed an effector memory phenotype even at 6 mo postinfection. Despite the chronic infection, CD8(+) T cells induced by ST-OVA were functional as they exhibited killing ability and cytokine production. Importantly, even memory CD8(+) T cells failed to undergo rapid expansion in response to ST-OVA infection, suggesting a delay in T cell priming during infection with virulent ST-OVA. Thus, phagosomal lifestyle may allow escape from host CD8(+) T cell recognition, conferring a survival advantage to the pathogen.     

The development of colitogenic CD4(+) T cells is regulated by IL-7 in collaboration with NK cell function in a murine model of colitis

We previously reported that IL-7(-/-)RAG(-/-) mice receiving naive T cells failed to induce colitis. Such abrogation of colitis may be associated with not only incomplete T cell maintenance due to the lack of IL-7, but also with the induction of colitogenic CD4(+) T cell apoptosis at an early stage of colitis development. Moreover, NK cells may be associated with the suppression of pathogenic T cells in vivo, and they may induce apoptosis of CD4(+) T cells. To further investigate these roles of NK cells, RAG(-/-) and IL-7(-/-)RAG(-/-) mice that had received naive T cells were depleted of NK cells using anti-asialo GM1 and anti-NK1.1 Abs. NK cell depletion at an early stage, but not at a later stage during colitogenic effector memory T cell (T(EM)) development, resulted in exacerbated colitis in recipient mice even in the absence of IL-7. Increased CD44(+)CD62L(-) T(EM) and unique CD44(-)CD62L(-) T cell subsets were observed in the T cell-reconstituted RAG(-/-) recipients when NK cells were depleted, although Fas, DR5, and IL-7R expressions in this subset differed from those in the CD44(+)CD62L(-) T(EM) subset. NK cell characteristics were the same in the presence or absence of IL-7 in vitro and in vivo. These results suggest that NK cells suppress colitis severity in T cell-reconstituted RAG(-/-) and IL-7(-/-)RAG(-/-) recipient mice through targeting of colitogenic CD4(+)CD44(+)CD62L(-) T(EM) and, possibly, of the newly observed CD4(+)CD44(-)CD62L(-) subset present at the early stage of T cell development.     

Persistence and function of central and effector memory CD4+ T cells following infection with a gastrointestinal helminth

Immunity in the gastrointestinal tract is important for resistance to many pathogens, but the memory T cells that mediate such immunity are poorly characterized. In this study, we show that following sterile cure of a primary infection with the gastrointestinal parasite Trichuris muris, memory CD4+ T cells persist in the draining mesenteric lymph node and protect mice against reinfection. The memory CD4+ T cells that developed were a heterogeneous population, consisting of both CD62L(high) central memory T cells (T(CM)) and CD62L(low) effector memory T cells (T(EM)) that were competent to produce the Th type 2 effector cytokine, IL-4. Unlike memory T cells that develop following exposure to several other pathogens, both CD4+ T(CM) and T(EM) populations persisted in the absence of chronic infection, and, critically, both populations were able to transfer protective immunity to naive recipients. CD62L(high)CD4+ T(CM) were not apparent early after infection, but emerged following clearance of primary infection, suggesting that they may be derived from CD4+ T(EM). Consistent with this theory, transfer of CD62L(low)CD4+ T(EM) into naive recipients resulted in the development of a population of protective CD62L(high)CD4+ T(CM). Taken together, these studies show that distinct subsets of memory CD4+ T cells develop after infection with Trichuris, persist in the GALT, and mediate protective immunity to rechallenge.     

Naive, effector, and memory CD8 T cells in protection against pulmonary influenza virus infection: homing properties rather than initial frequencies are crucial.

The goal of adoptive immunotherapy is to target a high number of persisting effector cells to the site of a virus infection or tumor. In this study, we compared the protective value of hemagglutinin peptide-specific CD8 T cells generated from the clone-4 TCR-transgenic mice, defined by different stages of their differentiation, against lethal pulmonary influenza infection. We show that the adoptive transfer of high numbers of Ag-specific unprimed, naive CD8 T cells failed to clear the pulmonary virus titer and to promote host survival. The same numbers of in vitro generated primary Ag-specific Tc1 effector cells, producing high amounts of IFN-gamma, or resting Tc1 memory cells, generated from these effectors, were protective. Highly activated CD62Llow Tc1 effectors accumulated in the lung with rapid kinetics and most efficiently reduced the pulmonary viral titer early during infection. The resting CD62Lhigh naive and memory populations first increased in cell numbers in the draining lymph nodes. Subsequently, memory cells accumulated more rapidly and to a greater extent in the lung lavage as compared with naive cells. Thus, effector cells are most effective against a localized virus infection, which correlates with their ability to rapidly distribute at the infected tissue site. The finding that similar numbers of naive Ag-specific CD8 T cells are not protective supports the view that qualitative differences between the two resting populations, the naive and the memory population, may play a major role in their protective value against disease.     

Effector function-deficient memory CD8+ T cells clonally expand in the liver and give rise to peripheral memory CD8+ T cells

Upon adoptive transfer into histocompatible mice, naive CD8(+) T cells stimulated ex vivo by TCR+IL-4 turn into long-lived functional memory cells. The liver contains a large number of so formed memory CD8(+) T cells, referred to as liver memory T cells (T(lm)) in the form of cell clusters. The CD62L(low) expression and nonlymphoid tissue distribution of T(lm) cells are similar to effector memory (T(em)) cells, yet their deficient cytotoxicity and IFN-γ inducibility are unlike T(em) cells. Adoptive transfer of admixtures of TCR+IL-4-activated Vβ8(+) and Vβ5(+) CD8(+) T cells into congenic hosts reveals T(lm) clusters that are composed of all Vβ5(+) or Vβ8(+), not mixed Vβ5(+)/Vβ8(+) cells, indicating that T(lm) clusters are formed by clonal expansion. Clonally expanded CD8(+) T cell clusters are also seen in the liver of Listeria monocytogenes-immune mice. T(lm) clusters closely associate with hepatic stellate cells and their formation is IL-15/IL-15R-dependent. CD62L(low) T(LM) cells can home to the liver and secondary lymphoid tissues, remain CD62L(low), or acquire central memory (T(cm))-characteristic CD62L(hi) expression. Our findings show the liver as a major site of CD8(+) memory T cell growth and that T(lm) cells contribute to the pool of peripheral memory cells. These previously unappreciated T(lm) characteristics indicate the inadequacy of the current T(em)/T(cm) classification scheme and help ongoing efforts aimed at establishing a unifying memory T cell development pathway. Lastly, our finding of T(lm) clusters suggests caution against interpreting focal lymphocyte infiltration in clinical settings as pathology and not normal physiology.     

Phenotype and homing of CD4 tumor-specific T cells is modulated by tumor bulk

Technical difficulties in tracking endogenous CD4 T lymphocytes have limited the characterization of tumor-specific CD4 T cell responses. Using fluorescent MHC class II/peptide multimers, we defined the fate of endogenous Leishmania receptor for activated C kinase (LACK)-specific CD4 T cells in mice bearing LACK-expressing TS/A tumors. LACK-specific CD44(high)CD62L(low) CD4 T cells accumulated in the draining lymph nodes and had characteristics of effector cells, secreting IL-2 and IFN-gamma upon Ag restimulation. Increased frequencies of CD44(high)CD62L(low) LACK-experienced cells were also detected in the spleen, lung, liver, and tumor itself, but not in nondraining lymph nodes, where the cells maintained a naive phenotype. The absence of systemic redistribution of LACK-specific memory T cells correlated with the presence of tumor. Indeed, LACK-specific CD4 T cells with central memory features (IL-2(+)IFN-gamma(-)CD44(high)CD62L(high) cells) accumulated in all peripheral lymph nodes of mice immunized with LACK-pulsed dendritic cells and after tumor resection. Together, our data demonstrate that although tumor-specific CD4 effector T cells producing IFN-gamma are continuously generated in the presence of tumor, central memory CD4 T cells accumulate only after tumor resection. Thus, the continuous stimulation of tumor-specific CD4 T cells in tumor-bearing mice appears to hinder the systemic accumulation of central memory CD4 T lymphocytes.     

Allograft rejection by primed/memory CD8+ T cells is CD154 blockade resistant: therapeutic implications for sensitized transplant recipients

We have shown that CD8(+) CTLs are the key mediators of accelerated rejection, and that CD8(+) T cells represent the prime targets of CD154 blockade in sensitized mouse recipients of cardiac allografts. However, the current protocols require CD154 blockade at the time of sensitization, whereas delayed treatment fails to affect graft rejection in sensitized recipients. To elucidate the mechanisms of costimulation blockade-resistant rejection and to improve the efficacy of CD154-targeted therapy, we found that alloreactive CD8(+) T cells were activated despite the CD154 blockade in sensitized hosts. Comparative CD8 T cell activation study in naive vs primed hosts has shown that although both naive and primed/memory CD8(+) T cells relied on the CD28 costimulation for their activation, only naive, not primed/memory, CD8(+) T cells depend on CD154 signaling to differentiate into CTL effector cells. Adjunctive therapy was designed accordingly to deplete primed/memory CD8(+) T cells before the CD154 blockade. Indeed, unlike anti-CD154 monotherapy, transient depletion of CD8(+) T cells around the time of cardiac engraftment significantly improved the efficacy of delayed CD154 blockade in sensitized hosts. Hence, this report provides evidence for 1) differential requirement of CD154 costimulation signals for naive vs primed/memory CD8(+) T cells, and 2) successful treatment of clinically relevant sensitized recipients to achieve stable long term graft acceptance.     

OX40/OX40L costimulation affects induction of Foxp3+ regulatory T cells in part by expanding memory T cells in vivo

OX40 is a member of the TNFR superfamily and has potent T cell costimulatory activities. OX40 also inhibits the induction of Foxp3(+) regulatory T cells (Tregs) from T effector cells, but the precise mechanism of such inhibition remains unknown. In the present study, we found that CD4(+) T effector cells from OX40 ligand-transgenic (OX40Ltg) mice are highly resistant to TGF-beta mediated induction of Foxp3(+) Tregs, whereas wild-type B6 and OX40 knockout CD4(+) T effector cells can be readily converted to Foxp3(+) T cells. We also found that CD4(+) T effector cells from OX40Ltg mice are heterogeneous and contain a large population of CD44(high)CD62L(-) memory T cells. Analysis of purified OX40Ltg naive and memory CD4(+) T effector cells showed that memory CD4(+) T cells not only resist the induction of Foxp3(+) T cells but also actively suppress the conversion of naive CD4(+) T effector cells to Foxp3(+) Tregs. This suppression is mediated by the production of IFN-gamma by memory T cells but not by cell-cell contact and also involves the induction of T-bet. Importantly, memory CD4(+) T cells have a broad impact on the induction of Foxp3(+) Tregs regardless of their origins and Ag specificities. Our data suggest that one of the mechanisms by which OX40 inhibits the induction of Foxp3(+) Tregs is by inducing memory T cells in vivo. This finding may have important clinical implications in tolerance induction to transplanted tissues.     

Concurrent naive and memory CD8(+) T cell responses to an influenza A virus

Memory Thy-1(+)CD8(+) T cells specific for the influenza A virus nucleoprotein (NP(366-374)) peptide were sorted after staining with the D(b)NP(366) tetramer, labeled with CFSE, and transferred into normal Thy-1.2(+) recipients. The donor D(b)NP(366)(+) T cells recovered 2 days later from the spleens of the Thy-1.2(+) hosts showed the CD62L(low)CD44(high)CD69(low) phenotype, characteristic of the population analyzed before transfer, and were present at frequencies equivalent to those detected previously in mice primed once by a single exposure to an influenza A virus. Analysis of CFSE-staining profiles established that resting tetramer(+) T cells divided slowly over the next 30 days, while the numbers in the spleen decreased about 3-fold. Intranasal infection shortly after cell transfer with a noncross-reactive influenza B virus induced some of the donor D(b)NP(366)(+) T cells to cycle, but there was no increase in the total number of transferred cells. By contrast, comparable challenge with an influenza A virus caused substantial clonal expansion, and loss of the CFSE label. Unexpectedly, the recruitment of naive Thy-1.2(+)CD8(+)D(b)NP(366)(+) host D(b)NP(366)(+) T cells following influenza A challenge was not obviously diminished by the presence of the memory Thy-1.1(+)CD8(+)D(b)NP(366)(+) donor D(b)NP(366)(+) set. Furthermore, the splenic response to an epitope (D(b)PA(224)) derived from the influenza acid polymerase (PA(224-233)) was significantly enhanced in the mice given the donor D(b)NP(366)(+) memory population. These experiments indicate that an apparent recall response may be comprised of both naive and memory CD8(+) T cells.     

Bcl6 acts as an amplifier for the generation and proliferative capacity of central memory CD8+ T cells

Central memory CD8(+) T cells (T(CM)) are considered to be more efficient than effector ones (T(EM)) for mediating protective immunity. The molecular mechanism involved in the generation of these cells remains elusive. Because Bcl6 plays a role in the generation and maintenance of memory CD8(+) T cells, we further examined this role in the process in relation to T(CM) and T(EM) subsets. In this study, we show that T(CM) and T(EM) were functionally identified in CD62L(+) and CD62L(-) memory (CD44(+)Ly6C(+)) CD8(+) T cell subsets, respectively. Although T(CM) produced similar amounts of IFN-gamma and IL-2 to T(EM) after anti-CD3 stimulation, the cell proliferation capacity after stimulation and tissue distribution profiles of T(CM) differed from those of T(EM). Numbers of T(CM) were greatly reduced and elevated in spleens of Bcl6-deficient and lck-Bcl6 transgenic mice, respectively, and those of T(EM) were constant in nonlymphoid organs of these same mice. The majority of Ag-specific memory CD8(+) T cells in spleens of these mice 10 wk after immunization were T(CM), and the number correlated with Bcl6 expression in T cells. The proliferation of Ag-specific memory CD8(+) T cells upon secondary stimulation was dramatically up-regulated in lck-Bcl6 transgenic mice, and the adoptive transfer experiments with Ag-specific naive CD8(+) T cells demonstrated that some of the up-regulation was due to the intrinsic effect of Bcl6 in the T cells. Thus, Bcl6 is apparently a crucial factor for the generation and secondary expansion of T(CM).     

A pool of central memory-like CD4 T cells contains effector memory precursors

The L51S mutation in the D10.G4.1 TCR alpha-chain reduces the affinity of the TCR to its ligand by affecting the interactions among the TCR, the beta-chain of I-A(k), and the bound peptide. We show that this mutation drives the generation of a pool of memory CD44(high)CD62L(neg)CD45RB(neg) CD4 TCR transgenic T cells. Their activation threshold is low, such that they proliferate in response to lower concentrations of agonist peptides than naive L51S CD4 T cells. Unlike effector memory CD4 T cells, however, they lack immediate effector function in response to TCR stimulation. These cells express IL-2R alpha only after culture with specific peptide. Although they can be recovered from lymph nodes, the majority lack the expression of the lymph node homing receptor CCR7. When these cells receive a second TCR stimulation in vitro, they differentiate into potent Th2-like effector cells, producing high levels of IL-4 at doses of agonist peptide too low to stimulate cytokine release from similarly differentiated naive L51S CD4 T cells. Having these properties, the L51S TCR transgenic memory CD4 T cells cannot be classified as either strict central memory or effector memory, but, rather, as a pool of memory T cells containing effector memory precursors.     

Rapid clonal expansion and prolonged maintenance of memory CD8+ T cells of the effector (CD44highCD62Llow) and central (CD44highCD62Lhigh) phenotype by an archaeosome adjuvant independent of TLR2

Vaccines capable of eliciting long-term T cell immunity are required for combating many diseases. Live vectors can be unsafe whereas subunit vaccines often lack potency. We previously reported induction of CD8(+) T cells to Ag entrapped in archaeal glycerolipid vesicles (archaeosomes). In this study, we evaluated the priming, phenotype, and functionality of the CD8(+) T cells induced after immunization of mice with OVA-Methanobrevibacter smithii archaeosomes (MS-OVA). A single injection of MS-OVA evoked a profound primary response but the numbers of H-2K(b)OVA(257-264)-specific CD8(+) T cells declined by 14-21 days, and <1% of primarily central phenotype (CD44(high)CD62L(high)) cells persisted. A booster injection of MS-OVA at 3-11 wk promoted massive clonal expansion and a peak effector response of approximately 20% splenic/blood OVA(257-264)-specific CD8(+) T cells. Furthermore, contraction was protracted and the memory pool (IL-7Ralpha(high)) of approximately 5% included effector (CD44(high)CD62L(low)) and central (CD44(high)CD62L(high)) phenotype cells. Recall response was observed even at >300 days. CFSE-labeled naive OT-1 (OVA(257-264) TCR transgenic) cells transferred into MS-OVA-immunized recipients cycled profoundly (>90%) within the first week of immunization indicating potent Ag presentation. Moreover, approximately 25% cycling of Ag-specific cells was seen for >50 days, suggesting an Ag depot. In vivo, CD8(+) T cells evoked by MS-OVA killed >80% of specific targets, even at day 180. MS-OVA induced responses similar in magnitude to Listeria monocytogenes-OVA, a potent live vector. Furthermore, protective CD8(+) T cells were induced in TLR2-deficient mice, suggesting nonengagement of TLR2 by archaeal lipids. Thus, an archaeosome adjuvant vaccine represents an alternative to live vectors for inducing CD8(+) T cell memory.     

Host-reactive CD8+ memory stem cells in graft-versus-host disease

Graft-versus-host disease (GVHD) is caused by alloreactive donor T cells that trigger host tissue injury. GVHD develops over weeks or months, but how this immune response is maintained over time is unknown. In mouse models of human GVHD, we identify a new subset of postmitotic CD44(lo)CD62L(hi)CD8(+) T cells that generate and sustain all allogeneic T-cell subsets in GVHD reactions, including central memory, effector memory and effector CD8(+) T cells, while self-renewing. These cells express Sca-1, CD122 and Bcl-2, and induce GVHD upon transfer into secondary recipients. The postmitotic CD44(lo)CD62L(hi)CD8(+) T cells persist throughout the course of GVHD, are generated in the initial phase in response to alloantigens and dendritic cells and require interleukin-15. Thus, their long life, ability to self-renew and multipotentiality define these cells as candidate memory stem cells. Memory stem cells will be important targets for understanding and influencing diverse chronic immune reactions, including GVHD.     

Effector and memory CD8+ T cells can be generated in response to alloantigen independently of CD4+ T cell help

There is now considerable evidence suggesting that CD8(+) T cells are able to generate effector but not functional memory T cells following pathogenic infections in the absence of CD4(+) T cells. We show that following transplantation of allogeneic skin, in the absence of CD4(+) T cells, CD8(+) T cells become activated, proliferate, and expand exclusively in the draining lymph nodes and are able to infiltrate and reject skin allografts. CD44(+)CD8(+) T cells isolated 100 days after transplantation rapidly produce IFN-gamma following restimulation with alloantigen in vitro. In vivo CD44(+)CD8(+) T cells rejected donor-type skin allografts more rapidly than naive CD8(+) T cells demonstrating the ability of these putative memory T cells to mount an effective recall response in vivo. These data form the first direct demonstration that CD8(+) T cells are able to generate memory as well as effector cells in response to alloantigen during rejection in the complete absence of CD4(+) T cells. These data have important implications for the design of therapies to combat rejection and serve to reinforce the view that CD8(+) T cell responses to allografts require manipulation in addition to CD4(+) T cell responses to completely prevent the rejection of foreign organ transplants.     

Cutting edge: L-selectin (CD62L) expression distinguishes small resting memory CD4+ T cells that preferentially respond to recall antigen

Naive CD4+ T cells use L-selectin (CD62L) expression to facilitate immune surveillance. However, the reasons for its expression on a subset of memory CD4+ T cells are unknown. We show that memory CD4+ T cells expressing CD62L were smaller, proliferated well in response to tetanus toxoid, had longer telomeres, and expressed genes and proteins consistent with immune surveillance function. Conversely, memory CD4+ T cells lacking CD62L expression were larger, proliferated poorly in response to tetanus toxoid, had shorter telomeres, and expressed genes and proteins consistent with effector function. These findings suggest that CD62L expression facilitates immune surveillance by programming CD4+ T cell blood and lymph node recirculation, irrespective of naive or memory CD4+ T cell phenotype.