TRAF2 regulates peripheral CD8+ T‐cell and NKT‐cell homeostasis by modulating sensitivity to IL‐15

In this study, a critical and novel role for TNF receptor (TNFR) associated factor 2 (TRAF2) is elucidated for peripheral CD8⁺ T‐cell and NKT‐cell homeostasis. Mice deficient in TRAF2 only in their T cells (TRAF2TKO) show ∼40% reduction in effector memory and ∼50% reduction in naïve CD8⁺ T‐cell subsets. IL‐15‐dependent populations were reduced further, as TRAF2TKO mice displayed a marked ∼70% reduction in central memory CD8⁺CD44^hiCD122⁺ T cells and ∼80% decrease in NKT cells. TRAF2TKO CD8⁺CD44^hi T cells exhibited impaired dose‐dependent proliferation to exogenous IL‐15. In contrast, TRAF2TKO CD8⁺ T cells proliferated normally to anti‐CD3 and TRAF2TKO CD8⁺CD44^hi T cells exhibited normal proliferation to exogenous IL‐2. TRAF2TKO CD8⁺ T cells expressed normal levels of IL‐15‐associated receptors and possessed functional IL‐15‐mediated STAT5 phosphorylation, however TRAF2 deletion caused increased AKT activation. Loss of CD8⁺CD44^hiCD122⁺ and NKT cells was mechanistically linked to an inability to respond to IL‐15. The reduced CD8⁺CD44^hiCD122⁺ T‐cell and NKT‐cell populations in TRAF2TKO mice were rescued in the presence of high dose IL‐15 by IL‐15/IL‐15Rα complex administration. These studies demonstrate a critical role for TRAF2 in the maintenance of peripheral CD8⁺ CD44^hiCD122⁺ T‐cell and NKT‐cell homeostasis by modulating sensitivity to T‐cell intrinsic growth factors such as IL‐15.     

Interleukin‐4 downregulates CD127 expression and activity on human thymocytes and mature CD8+ T cells

Signaling via the IL‐7 receptor complex (IL‐7Rα/CD127 and IL‐2Rγ/CD132) is required for T‐cell development and survival. Decreased CD127 expression has been associated with persistent viral infections (e.g. HIV, HCV) and cancer. Many IL‐2Rγ‐sharing (γ_C) cytokines decrease CD127 expression on CD4⁺ and CD8⁺ T cells in mice (IL‐2, IL‐4, IL‐7, IL‐15) and in humans (IL‐2, IL‐7), suggesting a common function. IL‐4 is of particular interest as it is upregulated in HIV infection and in thyroid and colon cancers. The role of IL‐4 in regulating CD127 expression and IL‐7 activity in human thymocytes and mature CD8⁺ T cells is unknown and was therefore investigated. IL‐4 decreased CD127 expression on all thymocyte subsets tested and only on naïve (CD45RA⁺) CD8⁺ T cells, without altering membrane‐bound CD127 mRNA expression. Pre‐treatment of thymocytes or CD8⁺ T cells with IL‐4 inhibited IL‐7‐mediated phosphorylation of STAT5 and decreased proliferation of CD8⁺ T cells. By downregulating CD127 expression and signaling on developing thymocytes and CD8⁺ T cells, IL‐4 is a potential contributor to impaired CD8⁺ T‐cell function in some anti‐viral and anti‐tumor responses. These findings are of particular consequence to diseases such as HIV, HCV, RSV, measles and cancer, in which CD127 expression is decreased, IL‐7 activity is impaired and IL‐4 concentrations are elevated.     

Hyaluronan binding identifies the most proliferative activated and memory T cells

CD44 is expressed on T cells where its ability to bind hyaluronan is tightly regulated. Here, we investigated when T cells bind hyaluronan during an immune response. We found that naïve, murine T cells do not bind fluoresceinated hyaluronan but are induced to bind upon antigen‐induced T‐cell activation in vitro and in vivo. Hyaluronan binding occurred on proliferating T cells and the percentage of hyaluronan‐binding cells correlated with the strength of the activation stimulus. A small percentage of hyaluronan‐binding cells persisted after in vitro activation and had a memory phenotype (CD122⁺CD44^hi). This hyaluronan‐binding population increased after culture with IL‐7 or IL‐15 and proliferated more rapidly than nonbinding cells. In vivo, approximately 20–30% of antigen‐specific OT‐I CD8⁺ memory T cells in the spleen and BM bound hyaluronan. Hyaluronan binding identified memory cells that proliferated faster in IL‐7 and IL‐15, and enriched for CD62L⁺ central memory cells. In vivo homeostatic proliferation induced hyaluronan binding on a small percentage of the most rapidly dividing cells after several cell divisions. This study demonstrates that hyaluronan binding is induced upon antigen‐induced T‐cell activation and occurs on a percentage of the most proliferative activated and memory T cells.     

CD70 and IFN‐1 selectively induce eomesodermin or T‐bet and synergize to promote CD8+ T‐cell responses

CD70‐mediated stimulation of CD27 is an important cofactor of CD4⁺ T‐cell licensed dendritic cells (DCs). However, it is unclear how CD70‐mediated stimulation of T cells is integrated with signals that emanate from signal 3 pathways, such as type‐1 interferon (IFN‐1) and IL‐12. We find that while stimulation of CD27 in isolation drives weak Eomesodermin^hiT‐bet^lo CD8⁺ T‐cell responses to OVA immunization, profound synergistic expansion is achieved by cotargeting TLR. This cooperativity can substantially boost antiviral CD8⁺ T‐cell responses during acute infection. Concomitant stimulation of TLR significantly increases per cell IFN‐γ production and the proportion of the population with characteristics of short‐lived effector cells, yet also promotes the ability to form long‐lived memory. Notably, while IFN‐1 contributes to the expression of CD70 on DCs, the synergy between CD27 and TLR stimulation is dependent upon IFN‐1's effect directly on CD8⁺ T cells, and is associated with the increased expression of T‐bet in T cells. Surprisingly, we find that IL‐12 fails to synergize with CD27 stimulation to promote CD8⁺ T‐cell expansion, despite its capacity to drive effector CD8⁺ T‐cell differentiation. Together, these data identify complex interactions between signal 3 and costimulatory pathways, and identify opportunities to influence the differentiation of CD8⁺ T‐cell responses.     

IL‐12‐mediated STAT4 signaling and TCR signal strength cooperate in the induction of CD40L in human and mouse CD8+ T cells

CD40L is one of the key molecules bridging the activation of specific T cells and the maturation of professional and nonprofessional antigen‐presenting cells including B cells. CD4⁺ T cells have been regarded as the major T‐cell subset that expresses CD40L upon cognate activation; however, we demonstrate here that a putative CD8⁺ helper T‐cell subset expressing CD40L is induced in human and murine CD8⁺ T cells in vitro and in mice immunized with antigen‐pulsed dendritic cells. IL‐12 and STAT4‐mediated signaling was the major instructive cytokine signal boosting the ability of CD8⁺ T cells to express CD40L both in vitro and in vivo. Additionally, TCR signaling strength modulated CD40L expression in CD8⁺ T cells after primary differentiation in vitro as well as in vivo. The induction of CD40L in CD8⁺ T cells regulated by IL‐12 and TCR signaling may enable CD8⁺ T cells to respond autonomously of CD4⁺ T cells. Thus, we propose that under proinflammatory conditions, a self‐sustaining positive feedback loop could facilitate the efficient priming of T cells stimulated by high affinity peptide displaying APCs.     

Memory precursor phenotype of CD8+ T cells reflects early antigenic experience rather than memory numbers in a model of localized acute influenza infection

The mechanistic basis of memory T‐cell development is poorly defined. Phenotypic markers that define precursors at effector stages have been characterized for acute systemic infections with high antigen load. We asked whether such markers can identify memory precursors from early effectors (d6) to late memory (>d500) for two immunodominant CD8⁺ responses during the course of a localized low‐load influenza infection in mice. CD8⁺ T cells stained with the D^bNP₃₆₆ and D^bPA₂₂₄ tetramers were characterized as IL‐7Rα^hi, IL‐7Rα^hiCD62L^hi or IL‐7Rα^hiKLRG1^lo. While the D^bNP₃₆₆‐ and D^bPA₂₂₄‐specific responses were comparable in size, decay kinetics and memory precursor frequency, their expansion characteristics differed. This correlated with a divergence in the IL‐7Rα^hi, IL‐7Rα^hiCD62L^hi and IL‐7Rα^hiKLRG1^lo phenotypes on effector, but not naïve, CD8⁺ populations. That effect was abrogated by priming with viruses engineered to present equivalent levels of NP₃₆₆ and PA₂₂₄ peptides, indicating that memory phenotypes reflect early antigenic experience rather than memory potential. Thus, the IL‐7Rα^hiKLRG1^lo phenotype had a poor predictive value in identifying memory precursors in the spleen and at the site of infection. Greater consistency in influenza‐specific IL‐7Rα^hiKLRG1^loCD8⁺ T‐cell numbers was found in draining lymph nodes, suggesting that this may be the preferential site for memory establishment and maintenance following localized virus infections.     

Progressive CD127 down‐regulation correlates with increased apoptosis of CD8 T cells during chronic HIV‐1 infection

Chronic HIV‐1 infection can induce a significant decrease in CD127 expression on CD8 T cells, but the underlying mechanisms and immunological consequences are unclear. In this study, we investigated CD127 expression on CD8 T cells from a total of 51 HIV‐1‐infected subjects and 16 healthy individuals and analyzed the association between CD127 expression and CD8 T‐cell apoptosis in these HIV‐1‐infected subjects. We found that CD127 expression on total CD8 T cells was significantly down‐regulated, which was correlated with the increased CD8 T‐cell apoptosis and disease progression of chronic HIV‐1 infection. The in vitro addition of IL‐7 efficiently rescued the spontaneous apoptosis of CD8 T cells from HIV‐1‐infected individuals. IL‐7 stimulation also transiently down‐regulated CD127 expression, whereas some of the CD127^? CD8 T cells regained CD127 expression soon after IL‐7 was retracted from the incubation medium. Thus, IL‐7 stimulation reduced apoptosis of both CD127⁺ and CD127^?CD8 T cells to some degree. These data indicate that CD127 loss might impair IL‐7 signaling and increase CD8 T‐cell apoptosis during HIV‐1 infection. This study, therefore, will extend the notion that IL‐7 could be a good candidate for immunotherapy in HIV‐1‐infected patients.     

T‐cell help dependence of memory CD8+ T‐cell expansion upon vaccinia virus challenge relies on CD40 signaling

Due to their capacity to differentiate into long‐lived memory cells, CD8⁺ T cells are able to resolve subsequent infections faster than during the primary response. Among other factors, CD4⁺ T cells play a crucial role during primary and secondary CD8⁺ T‐cell responses. However, the timing and mechanisms by which they influence CD8⁺ T cells may differ in primary and secondary responses. Here, we demonstrate that during both primary and secondary vaccinia virus infection, CD4⁺ T cells are necessary to promote CD8⁺ T‐cell responses. While CD4⁺ T cells contributed to memory CD8⁺ T‐cell development, they were even more important during memory recall responses during challenge, as absence of CD4⁺ T cells during challenge resulted in markedly decreased proliferation and increased apoptosis. T‐cell help during primary and secondary responses was mediated via CD40 signaling, with DCs being an integral part of that pathway. As opposed to primary CD8⁺ T‐cell responses where only a combination of agonistic CD40 signaling and provision of IL‐2 could substitute for T‐cell help, agonistic CD40 triggering alone was sufficient to rescue memory CD8⁺ T‐cell responses in absence of T‐cell help in the context of vaccinia virus infection.     

CXCR4, but not CXCR3, drives CD8+ T‐cell entry into and migration through the murine bone marrow

The BM serves as a blood‐forming organ, but also supports the maintenance and immune surveillance function of many T cells. Yet, in contrast to other organs, little is known about the molecular mechanisms that drive T‐cell migration to and localization inside the BM. As BM accumulates many CXCR3‐expressing memory CD8⁺ T cells, we tested the involvement of this chemokine receptor, but found that CXCR3 is not required for BM entry. In contrast, we could demonstrate that CXCR4, which is highly expressed on both naive and memory CD8⁺ T cells in BM, is critically important for homing of all CD8⁺ T‐cell subsets to the BM in mice. Upon entry into the BM parenchyma, both naïve and memory CD8⁺ T cells locate close to sinusoidal vessels. Intravital imaging experiments revealed that CD8 T cells are surprisingly immobile and we found that they interact with ICAM‐1+VCAM‐1+BP‐1+ perivascular stromal cells. These cells are the major source of CXCL12, but also express key survival factors and maintenance cytokines IL‐7 and IL‐15. We therefore conclude that CXCR4 is not only crucial for entry of CD8⁺ T cells into the BM, but also controls their subsequent localization toward BM niches that support their survival.     

Human antigen‐specific CD4+CD25+CD134+CD39+ T cells are enriched for regulatory T cells and comprise a substantial proportion of recall responses

Human Ag‐specific CD4⁺ T cells can be detected by their dual expression of CD134 (OX40) and CD25 after a 44 hours stimulation with cognate Ag. We show that surface expression of CD39 on Ag‐specific cells consistently identifies a substantial population of CD4⁺CD25⁺CD134⁺CD39⁺ T cells that have a Treg‐cell‐like phenotype and mostly originate from bulk memory CD4⁺CD45RO⁺CD127^lowCD25^highCD39⁺ Treg cells. Viable, Ag‐specific CD25⁺CD134⁺CD39⁺ T cells could be expanded in vitro as cell lines and clones, and retained high Forkhead Box Protein 3, CTLA‐4 and CD39 expression, suppressive activity and Ag specificity. We also utilised this combination of cell surface markers to measure HIV‐Gag responses in HIV⁺ patients before and after anti‐retroviral therapy (ART). Interestingly, we found that the percentage of CD39^? cells within baseline CD4⁺ T‐cell responses to HIV‐Gag was negatively correlated with HIV viral load pre‐ART and positively correlated with CD4⁺ T‐cell recovery over 96 weeks of ART. Collectively, our data show that Ag‐specific CD4⁺CD25⁺CD134⁺CD39⁺ T cells are highly enriched for Treg cells, form a large component of recall responses and maintain a Treg‐cell‐like phenotype upon in vitro expansion. Identification and isolation of these cells enables the role of Treg cells in memory responses to be further defined and provides a development pathway for novel therapeutics.     

Developing and maintaining protective CD8+ memory T cells

Summary: A critical aim of vaccine‐related research is to identify the mechanisms by which memory T cells are formed and maintained over long periods of time. In recent years, we have designed experiments aimed at addressing two key questions: (i) what are the factors that maintain functionally responsive CD8⁺ memory cells over long periods of time, and (ii) what are the signals during the early stages of infection that drive the differentiation of long‐lived CD8⁺ memory T cells? We have identified a role for CD4⁺ T cells in the generation of CD8⁺ T‐cell‐mediated protection from secondary challenge. While CD4⁺ T cells appear to play a role in the programme of CD8 memory, we find that they are also required for the long‐term maintenance of CD8⁺ memory T‐cell numbers and function. This property is independent of CD40–CD40L interactions, and we propose a role for CD4⁺ T cells in maintaining the ability of CD8⁺ memory T cells to respond to interleukin‐7 (IL‐7) and IL‐15. By manipulating both the time course of infection and the timing of antigen presentation to newly recruited CD8⁺ T cells, we also demonstrate that the programming of effector and memory potential are at least partially distinct processes.     

Interleukin‐4‐induced loss of CD8 expression and cytolytic function in effector CD8 T cells persists long term in vivo

Activation of naive CD8⁺ T cells in the presence of interleukin‐4 modulates their CD8 co‐receptor expression and functional differentiation, resulting in the generation of CD8^low cells that produce type 2 cytokines and display poor cytolytic and anti‐tumour activity. Although this CD8^low phenotype becomes stable after about a week and can persist with further stimulation in vitro, it is not known whether it can be maintained long term in vivo. Here we report that CD8^low cells derived from oval‐bumin_257–264‐specific T‐cell receptor‐transgenic CD8⁺ T cells activated in the presence of interleukin‐4 could be detected in the spleen for at least 4 months after adoptive transfer into normal mice. A significant proportion of the long‐term surviving cells retained their CD8^low phenotype in vivo and after clonal re‐activation in vitro. Although long‐term surviving CD8^low cells lacked detectable cytolytic activity or perforin expression, they showed some anti‐tumour function in vivo. The persistence of functional cells with a CD8^low phenotype in vivo raises the possibility that such cells can contribute to effector or regulatory responses to tumours or pathogens.     

IL‐27 enhances the survival of tumor antigen‐specific CD8+ T cells and programs them into IL‐10‐producing,memory precursor‐like effector cells

IL‐27 is a member of the IL‐12 family of cytokines that is comprised of an IL‐12 p40‐related protein subunit, EBV‐induced gene 3, and a p35‐related subunit, p28. IL‐27 functions through IL‐27R and has been shown to have potent antitumor activity via activation of a variety of cellular components, including antitumor CD8⁺ T‐cell responses. However, the exact mechanisms of how IL‐27 enhances antitumor CD8⁺ T‐cell responses remain unclear. Here we show that IL‐27 significantly enhances the survival of activated tumor antigen‐specific CD8⁺ T cells in vitro and in vivo, and programs tumor antigen‐specific CD8⁺ T cells into memory precursor‐like effector cells, characterized by upregulation of Bcl‐6, SOCS3, Sca‐1, and IL‐10. While STAT3 activation and the CTL survival‐enhancing effects can be independent of CTL IL‐10 production, we show here that IL‐27‐induced CTL IL‐10 production contributes to memory precursor cell phenotype induction, CTL memory, and tumor rejection. Thus, IL‐27 enhances antitumor CTL responses via programming tumor antigen‐specific CD8⁺ T cells into a unique memory precursor type of effector cells characterized by a greater survival advantage. Our results have important implications for designing immunotherapy against human cancer.     

Quantitative assessment of the functional plasticity of memory CD8+ T cells

While the functional plasticity of memory CD4⁺ T cells has been studied extensively, less is known about this property in memory CD8⁺ T cells. Here, we report the direct measurement of plasticity by paired daughter analysis of effector and memory OT‐I CD8⁺ T cells primed in vivo with ovalbumin. Naïve, effector, and memory OT‐I cells were isolated and activated in single‐cell culture; then, after the first division, their daughter cells were transferred to new cultures with and without IL‐4; expression of IFN‐γ and IL‐4 mRNAs was measured 5 days later in the resultant subclones. Approximately 40% of clonogenic memory CD8⁺ T cells were bipotential in this assay, giving rise to an IL‐4^? subclone in the absence of IL‐4 and an IL‐4⁺ subclone in the presence of IL‐4. The frequency of bipotential cells was lower among memory cells than naïve cells but markedly higher than among 8‐day effectors. Separation based on high or low expression of CD62L, CD122, CD127, or Ly6C did not identify a phenotypic marker of the bipotential cells. Functional plasticity in memory CD8⁺ T‐cell populations can therefore reflect modulation at the level of a single memory cell and its progeny.     

Staphylococcus aureus convert neonatal conventional CD4+ T cells into FOXP3+ CD25+ CD127low T cells via the PD‐1/PD‐L1 axis

The gut microbiota provides an important stimulus for the induction of regulatory T (Treg) cells in mice, whether this applies to newborn children is unknown. In Swedish children, Staphylococcus aureus has become a common early colonizer of the gut. Here, we sought to study the effects of bacterial stimulation on neonatal CD4⁺ T cells for the induction of CD25⁺ CD127^low Treg cells in vitro. The proportion of circulating CD25⁺ CD127^low Treg cells and their expression of FOXP3, Helios and CTLA‐4 was examined in newborns and adults. To evaluate if commensal gut bacteria could induce Treg cells, CellTrace violet‐stained non‐Treg cells from cord or peripheral blood from adults were co‐cultured with autologous CD25⁺ CD127^low Treg cells and remaining mononuclear cells and stimulated with S. aureus. Newborns had a significantly lower proportion of CD25⁺ CD127^low Treg cells than adults, but these cells were Helios⁺ and CTLA‐4⁺ to a higher extent than in adults. FOXP3⁺ CD25⁺ CD127^low T cells were induced mainly in neonatal CellTrace‐stained non‐Treg cells after stimulation with S. aureus. In cell cultures from adults, S. aureus induced CD25⁺ CD127^low T cells only if sorted naive CD45RA⁺ non‐Treg cells were used, but these cells expressed less FOXP3 than those induced from newborns. Sorted neonatal CD25⁺ CD127^low T cells from S. aureus‐stimulated cultures were still suppressive. Finally, blocking PD‐L1 during stimulation reduced the induction of FOXP3⁺ CD25⁺ CD127^low T cells. These results suggest that newborns have a higher proportion of circulating thymically derived Helios⁺ Treg cells than adults and that S. aureus possess an ability to convert neonatal conventional CD4⁺ T cells into FOXP3⁺ CD25⁺ CD127^low Treg cells via the PD‐1/PD‐L1 axis.     

Inflammation enhances the vaccination potential of CD40‐activated B cells in mice

Vaccination with antigen‐pulsed CD40‐activated B (CD40‐B) cells can efficiently lead to the in vivo differentiation of naive CD8⁺ T cells into fully functional effectors. In contrast to bone marrow‐derived dendritic cell (BMDC) vaccination, CD40‐B cell priming does not allow for memory CD8⁺ T‐cell generation but the reason for this deficiency is unknown. Here, we show that compared to BMDCs, murine CD40‐B cells induce lower expression of several genes regulated by T‐cell receptor signaling, costimulation, and inflammation (signals 1–3) in mouse T cells. The reduced provision of signals 1 and 2 by CD40‐B cells can be explained by a reduction in the quality and duration of the interactions with naive CD8⁺ T cells as compared to BMDCs. Furthermore, CD40‐B cells produce less inflammatory mediators, such as IL‐12 and type I interferon, and increasing inflammation by coadministration of polyriboinosinic‐polyribocytidylic acid with CD40‐B‐cell immunization allowed for the generation of long‐lived and functional CD8⁺ memory T cells. In conclusion, it is possible to manipulate CD40‐B‐cell vaccination to promote the formation of long‐lived functional CD8⁺ memory T cells, a key step before translating the use of CD40‐B cells for therapeutic vaccination.