Immunology and Cell Biology (2014) 92, 181–190 & 2014 Australasian Society for Immunology Inc. All rights reserved 0818-9641/14 www.nature.com/icb
ORIGINAL ARTICLE
Phenotypic and transcriptional profile correlates with functional plasticity of antigen-specific CD4 þ T cells Emily SJ Edwards, Corey Smith and Rajiv Khanna The role of CD4 þ T cells in the control of infectious pathogens is highly complex with a myriad of functions but how these T cells acquire differential functional potentiality remains poorly defined. Here we show that human cytomegalovirus (CMV)specific CD4 þ T cells directed towards different viral antigens expressed predominantly TNF-a alone or TNF-a and IFN-c. TNF-a þ and IFN-c þ CD4 þ T cells expressed significantly higher levels of T-box transcription factors T-bet with graded loss of Eomesodermin (Eomes) expression (T-betHiEomesHi/Lo) when compared with TNF-a þ CD4 þ T cells expressing lower levels of both T-bet and Eomes (T-bet Eomes ). Furthermore, TNF-a þ and IFN-c þ CD4 þ T cells expressed significantly higher levels of perforin and interleukin (IL)-2 and displayed a terminally differentiated phenotype (CCR7 CD27 CD45RA CD57 þ CD62L ). In contrast, TNF-a þ alone CMV-specific CD4 þ T cells were predominantly early-memory phenotype with a proportion of these cells displaying T memory stem-cell phenotype (CD95 þ CD45RA þ CCR7 þ CD27 þ ). In vitro stimulation of CMV-specific CD4 þ T cells with viral antigen in the presence of IL-12 was sufficient to dramatically change the transcriptional and functional profile of TNF-a þ CD4 þ T cells, whereas TNF-a þ and IFN-c þ CD4 þ T cells remained unaltered. These findings illustrate an intrinsic link between cytokine expression, transcriptional regulation and cellular differentiation, and their impact on functional plasticity of virus-specific CD4 þ T cells. Immunology and Cell Biology (2014) 92, 181–190; doi:10.1038/icb.2013.88; published online 3 December 2013 Keywords: antigen; cytokine; T cells; transcription factors; virus
CD4 þ T cells elicit a plethora of effector functions required for the effective elimination and control of intracellular and extracellular pathogens as well as neoplastic cells. Each naı¨ve CD4 þ T cell has the potential to differentiate into one of several distinct effector T-helper (Th) lineages. The recognition of the cognate antigen in complex with MHC class II molecules initiates the differentiation process. Strength of T-cell receptor (TCR) signal in conjunction with cytokine milieu, antigen concentration, type of antigen presenting cell and the ligation of co-stimulatory molecules determines the Th lineage that the cell is consequently committed to.1–3 As differentiation proceeds, antigenspecific Th cells acquire and maintain effector function enabling a robust response upon re-exposure to cognate antigen. These effector functions include cytokine production, ability to proliferate and expand in response to repeated exposure, provision of cognate help, and most recently described the ability to directly lyse target cells expressing cognate peptide bound to MHC II molecules.4–6 CD4 þ Th phenotypes can be characterized according to the pattern of cytokines and transcription factors expressed. Th1 immunity is characterized by the production of the interferon-gamma (IFN-g),1 in conjunction with expression of the T-box transcription factor Tbx21 (T-bet). Initial TCR stimulation upregulates T-bet expression, and
upon cessation of TCR signalling, T-bet induces expression of interleukin (IL)-12 receptor b2, which is required for responsiveness to IL-12 and reinforces Th1 commitment by amplifying T-bet expression.7 Numerous studies have identified that T-bet is essential for the generation and maintenance of Th1 cells, as well as representing a major factor for the induction of IFN-g production.8–11 T-bet has been shown to antagonize expression of other lineage specifying factors, including Th2 transcription factor GATA-3. Overexpression of T-bet in developing or fully differentiated Th2 cells has been shown to repress IL-4 production and induce IFN-g secretion.10,11 Unsurprisingly, T-bet deficiency partially inhibits IFN-g production and halts Th1 differentiation.9,11,12 A second T-box transcription factor, Eomesodermin (Eomes), is expressed in CD4 þ T cells albeit at a lower level than in their CD8 counterparts.13 Eomes expression is induced by TCR stimulation, consequent to T-bet expression. Eomes has been shown to complement T-bet for IFN-g production and Th1 commitment. However, Eomes expression alone is inadequate for inhibition of signals mediated by polarizing cytokines to other Th lineages, thereby inducing Th lineage switching.14 Both T-bet and Eomes have been shown to cooperatively induce perforin and granzyme B expression in addition to that of IFN-g.13
Centre for Immunotherapy and Vaccine Development and Tumour Immunology Laboratory, Department of Immunology, Queensland Institute of Medical Research, Brisbane, Queensland, Australia Correspondence: Professor R Khanna, Queensland Institute of Medical Research, Tumour Immunology Laboratory, Department of Immunology, 300 Herston Rd, Brisbane 4006, Queensland, Australia. E-mail: [email protected] Received 2 September 2013; revised 23 October 2013; accepted 24 October 2013; published online 3 December 2013
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 182
To date, very little has been elucidated about how T-bet and Eomes expression impacts on the acquisition of functional plasticity in antigen-specific CD4 þ T cells. Here we show that CD4 þ T cells directed towards different CMV antigens expressed during different stages of viral infection exhibit distinct cytokine expression potentiality which correlates with differential patterns of T-box transcriptional factors, and this expression is linked to their ability to produce perforin and IL-2. This differential expression of cytokines and T-box transcriptional factors was also associated with an altered differentiation status and functional plasticity in terms of their response to antigenic stimulation in presence of inflammatory cytokines. These results illustrate how CD4 þ T-cell programming during persistent viral infection modulates the transcriptional profile of antigen-specific T cells and their functional plasticity.
RESULTS Differential expression of IFN-c and TNF-a by CMV-specific CD4 þ T cells directed towards different viral antigens Previous studies have demonstrated that CMV-specific CD4 þ T cells display heterogeneous patterns of cytokine expression where populations of both single-cytokine-producing and polyfunctional T cells can be activated upon stimulation with CMV lysate or CMV-infected cells.15–21 However, to date, the functional profile of CD4 þ T-cell populations directed towards different CMV proteins remains undetermined. In the first set of experiments, we assessed IFN-g and TNF-a expression in CMV-specific CD4 þ T cells directed towards three different immunodominant CMV proteins pp65, IE-1 and gB.5,15,22,23 It has been shown previously that different expression patterns are exhibited by these three proteins, with IE-1 being expressed during early phase of replicative infection, whereas pp65 and gB expressed during late phase of the viral infection and are packaged in the viral particle.24 Peripheral blood mononuclear cells (PBMCs) from a panel of CMV-seropositive donors were stimulated with recombinant CMV proteins, and then assessed for the expression of IFN-g and TNF-a. These analyses showed that the responding frequency of CD4 þ T cells directed towards IE-1 antigen were 3.51-fold lower when compared with the CD4 þ T cells specific for pp65 and 1.75-fold lower than gB-specific CD4 þ T cells (Mean: IE-1-specific CD4 þ T cells: 0.3002% (range 0.069–0.6420%); pp65-specific CD4 þ T cells: 1.055% (range 0.152–4.302%); gB-specific CD4 þ T cells: 0.5263% (range 0.0239–1.16%)). It is important to note that in all the donors tested the IE-1 response was subdominant compared with that elicited to pp65 and gB. Interestingly, co-expression analysis revealed discordance in IFN-g and TNF-a expression in CMV-specific CD4 þ T-cell populations (Figures 1a–c). Although no statistical significance was observed between the magnitude of antigen-specific T cells expressing IFN-g alone, CD4 þ T cells directed against pp65 and gB were predominantly IFN-g þ TNF-a þ , whereas the majority of IE-1-specific CD4 þ T cells expressed only TNF-a. No inter-individual variation was observed with regards to the predominance of TNF-a only cells for IE-1, or IFN-g þ TNF-a þ for pp65 and gB. IE-1 dose escalation did not alter the predominance of the TNF-a þ only cells elicited by this protein (data not shown). Therefore, CMV-specific CD4 þ T cells displayed two distinct patterns of cytokine expression: TNF-a þ only and IFN-g þ TNF-a þ . Furthermore, CMV-specific CD4 þ T cells expressing both IFN-g and TNF-a expressed a higher level of these cytokines when compared with the CD4 þ T cells expressing IFN-g or TNF-a alone (unpublished observation). Immunology and Cell Biology
Expression of the transcription factors T-bet and Eomes in CMV-specific CD4 þ T cells correlates with differential cytokine expression patterns Studies carried out in murine models have demonstrated that T-box transcription factors T-bet and Eomes have a crucial role in regulating effector function(s) and the establishment of CD4 þ T-cell memory.8 However, the potential role of these transcription factors in modulating polyfunctional potentiality of antigen-specific T cells has remained unexplored. To explore the possibility that selective expression of IFN-g and/or TNF-a by antigen-specific CD4 þ T cells may be intrinsically linked to the expression of T-bet and Eomes, PBMCs from CMV-seropositive individuals were stimulated with IE-1, pp65 or gB protein and then assessed for the expression of these transcription factors in combination with TNF-a and/or IFN-g. Data presented in Figures 2a–d show that CMV-specific CD4 þ T cells exhibited three individual patterns of expression for T-bet and Eomes: T-bet Eomes , T-betLoEomesHi and T-betHiEomesHi/Lo (Figure 2a). CMV-specific CD4 þ T cells directed towards pp65, IE-1 and gB antigens expressing both IFN-g and TNF-a showed significantly higher levels of T-bet expression when compared with the T cells expressing TNF-a alone (Figures 2b–c). Interestingly, combined analyses of T-bet and Eomes showed discordance in the expression of these transcription factors between antigen-specific CD4 þ T cells expressing both IFN-g and TNF-a or TNF-a only. A significantly higher proportion of CMV-specific CD4 þ T cells expressing both IFN-g and TNF-a displayed a T-betHiEomesHi/Lo phenotype, whereas a significantly higher proportion of T cells expressing TNF-a þ only exhibited a T-bet Eomes phenotype. The magnitude of cells expressing the T-betLoEomesHi phenotype wasn’t significant between IFN-g and TNF-a or TNF-a only populations except for in response to IE-1 (Figures 2b–d). It is important to mention here that the T-betHiEomesHi/Lo phenotype of IFN-g þ TNF-a þ T cells was observed for all three different specificities (that is, IE-1, pp65 and gB). Thus two distinct profiles of T-bet/Eomes expression was observed, T-cells that displayed mainly T-bet Eomes phenotype and T-cells that expressed a predominately T-betHiEomesHi/Lo phenotype. CMV-specific CD4 þ T cells expressing IFN-g and/or TNF-a display differential cytolytic potentiality and IL-2 production To further delineate the impact of a discordant profile of T-box transcription factors on effector functions of CMV-specific CD4 þ T cells, we next assessed the expression of perforin, granzyme B and IL-2 in these effector cells. PBMCs from healthy seropositive donors were incubated with gB, pp65 and IE-1 proteins then assessed for expression of IFN-g, TNF-a, granzyme B, perforin and IL-2. Data in Figures 3a–c show the frequency of IFN-g þ TNF-a þ and TNF-a þ only CMV-specific CD4 þ T cells co-expressing perforin, granzyme B and IL-2. These analyses showed that although no significant difference was observed for granzyme B expression, a significantly higher proportion of IFN-g þ TNF-a þ CMV-specific CD4 þ T cells produced perforin and IL-2 when compared with the T cells expressing TNF-a only. Furthermore, analyses of the mean fluorescence intensity for perforin and IL-2 production showed that IFN-g þ TNF-a þ CMV-specific CD4 þ T cells produced significantly higher levels of both perforin and IL-2 when compared with the T cells expressing TNF-a only (Figures 3d–e). Taken together, these observations demonstrate that differential functional programming of CMV-specific CD4 þ T cells directed towards gB, pp65 and IE-1 correlates with transcriptional profile, cytolytic potentiality and IL-2 production.
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 183
a
Unstimulated 5
10
4
10
10
3
10
TNF-
pp65 10
10
2
5
31.90%
IE-1 5
10
65.19%
4
2
3
10
4
10
3
5
10
3
3
10
2
010
3
10
4
10
2
10 0
2.91%
79.51%
7.05%
4
10
2
10
10
4
10 0 0 10
gB 5
14.23%
10
2
10 0
84.26%
10 0
1.51%
5
2
10
0 10
3
10
4
10
5
10
13.40% 0 10
2
10
3
4
10
5
10
IFN-
Frequency cytokine producing CD4+T-cells (%)
b 100
IFN-+ alone
TNF -+ alone
ns
**
IFN-+TNF -+
*
80 60 40 20 0 pp65
IE-1
c
gB
IFN-+ alone
pp65
IE-1
TNF -+ alone
pp65
IE-1
gB
pp65
IE-1
gB
IFN-+TNF-+ gB
Figure 1 CMV-specific CD4 þ T cells directed towards pp65, IE-1 and gB antigens display differential cytokine expression potentiality. PBMCs from CMVseropositive individuals (n ¼ 10) were stimulated with pp65, IE-1 or gB protein overnight, stained with anti-CD4 and assessed for intracellular expression of IFN-g and TNF-a. Panel a shows representative analysis of IFN-g and/or TNF-a producing T cells elicited towards three different CMV antigens (n ¼ 1). Percentages represent the proportion of total cytokine-producing cells that express IFN-g, TNF-a or IFN-g and TNF-a. Panel b shows frequency of antigenspecific CD4 þ T cells directed towards three different CMV antigens expressing IFN-g, TNF-a or IFN-g and TNF-a. The mean precursor frequency is represented by a horizontal line in each graph. Panel c shows overall pattern of cytokine expression by T cells directed towards pp65, IE-1 and gB proteins. *Po0.05; **Po0.001; ns, not significant.
CMV-specific T cells expressing IFN-g and/or TNF-a display distinct maturation status On the basis of the data presented above, we hypothesized that differential functional programming of CMV-specific CD4 þ T cells may be linked to their maturation/differentiation status. To test this hypothesis, PBMCs from CMV-seropositive individuals were stimulated with viral proteins and then assessed for expression of IFN-g, TNF-a, CD27, CD62L, CD57, CD45RA and CCR7 (Figure 4a). Data presented in Figures 4b and c show that a significantly higher proportion of TNF-a þ only CMV-specific T cells were CCR7 þ , CD27 þ , CD45RA þ and CD62L þ when compared with the T cells expressing both IFN-g and TNF-a. In contrast, IFN-g þ TNF-a þ pp65-specific T cells were predominantly CD57, whereas no significant difference was observed for CD57 expression on IE-1-specific T cells expressing TNF alone or both IFN-g and TNF-a. These observations indicate that CMV-specific CD4 þ T cells expressing
TNF-a only were less differentiated when compared with the T cells expressing both IFN-g and TNF-a. Recent studies have described a unique population of human memory T cells which have been referred to as stem-cell memory like (TSCM) population.25,26 These cells have a superior self-renewal capacity as well as the potential to differentiate into central memory, effector memory and effector T cells.22,23 To determine whether the CMV-specific CD4 þ T cells displaying differential functional programming include TSCM population, these antigenspecific T cells were stained for the expression of CD95, CD45RA, CD27 and CCR7, as well as IFN-g and TNF-a (Figure 5a). These antigen-specific CD4 þ T cells were then categorized according to differential expression of these markers, into TSCM (CD95 þ CD45RA þ CCR7 þ CD27 þ ), early-differentiated (CD95 þ CD45RA þ / CCR7 þ CD27 þ ), intermediate (CD95 þ CD45RA CCR7 CD27 þ ) and late-differentiated cells (CD95 þ CD45RA þ / CCR7 CD27 ). Immunology and Cell Biology
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 184
b
T-bet- Eomes-
10
T-bet
Lo EomesHi
T-bet
Hi EomesHi/Lo
5
104
Eomes
103 102 0 0 102
103
104
Frequecy of cytokine producing CD4+ T-cells (%)
a
pp65 100 80
**
ns
**
60 40 20 0 TNF-+ alone
105
IFN-+TNF-+
T-bet
d
gB 100 80
**
ns
**
60 40 20 0
TNF-+ alone
IFN-+TNF-+
Frequecy of cytokine producing CD4+ T-cells (%)
Frequecy of cytokine producing CD4+ T-cells (%)
c
IE-1 100 80
**
**
**
60 40 20 0
TNF-+alone
IFN-+TNF-+
Figure 2 Expression of the T-box transcription factors T-bet and Eomes in CMV-specific CD4 þ T cells. PBMCs from CMV-seropositive healthy donors (n ¼ 10) were stimulated with pp65, IE-1 or gB protein overnight, stained with anti-CD4 and assessed for intracellular expression of IFN-g, TNF-a, T-bet and Eomes. Panel a shows representative analysis of T-bet and Eomes expression by CMV-specific CD4 þ T cells. IFN-g and/or TNF-a producing cells are shown as an overlay of total CD4 þ T cells. Panels b–d show frequency of TNF-a þ or IFN-g þ TNF-a þ pp65, gB and IE-1-specific T cells expressing different profiles of T-bet and Eomes (T-bet Eomes , T-betLoEomesHi or a T-betHiEomesHiLo). T-bet and Eomes profiles were compared between TNF-a þ or IFN-g þ TNF-a þ . **Po0.001; ns, not significant.
Data presented in Figure 5b show that pp65 or IE-1-specific T cells expressing TNF-a only predominantly displayed an early-differentiated phenotype with a small proportion of these cells expressing surface markers of TSCM population. In contrast, the majority of T cells expressing both IFN-g and TNF-a exhibited a phenotypic profile of late/intermediate-differentiated cells. Low-to-negligible proportion of these effector cells expressed early-differentiated or TSCM phenotype (Figure 5b). Taken together these analyses clearly demonstrate that CMV-specific CD4 þ T cells expressing TNF-a only are less differentiated when compared with the IFN-g þ TNF-a þ T cells. CMV-specific CD4 þ T cell displaying differential functional and phenotypic profile exhibit distinct levels of functional plasticity Expression of a differential phenotypic profile by CMV-specific T cells led us to hypothesize that antigen-specific CD4 þ T cells displaying an immature profile may be more amenable to interconversion to other phenotypic and functional profiles as a consequence of cytokine milieu alteration. To explore this possibility, PBMCs from CMVseropositive individuals were stimulated with pp65 or IE-1 in the absence or presence of different combinations of cytokines (IL-2, IL-4 and/or IL-12). These T cells were then assessed for the expression of IFN-g and TNF-a, to determine whether antigenic stimulation in the presence of specific cytokines altered their functional, phenotypic and transcriptional profile. Data presented in Figure 6a show that whilst the functional profile of pp65-specific CD4 þ T cells remained unaltered following in vitro stimulation, a dramatic change in IE-1specific CD4 þ T cells was observed. Following in vitro stimulation with IE-1 protein in combination with IL-2, IL-12, IL-2 þ IL-4 or Immunology and Cell Biology
IL-2 þ IL-12, the majority of T cells specific for this protein switched from TNF-a only to both IFN-g þ and TNF-a þ . No major alteration in the functional profile was observed following stimulation with IE-1 antigens and IL-4 (Figure 6a). Consistent with this altered functional profile, we also observed a dramatic change in the expression of T-box transcription factors. The majority of the IE-1-specific T cells switched from T-bet Eomes to a dominant T-betHiEomesHi/Lo expressing population (Figure 6b). This impact was particularly evident when these cells were cultured in the presence of IL-12 or IL-2 þ IL-12. Upon IL-12 or IL-2 þ IL-12 treatment, a statistical increase in the proportion of TbetLoEomesHi and TbetHiEomesHi/Lo cells was observed, with a concomitant significant decrease in Tbet Eomes cells. Furthermore, coculture of IE-1-specific T cells with IL-12 or IL-2 þ IL-12 increased the proportion of perforin-positive cells indicating that these cells were also capable of acquiring cytolytic function under appropriate stimulation conditions (Figure 6c). Finally, we also observed a change in the differentiation status following stimulation with viral antigen in the presence of IL-12 or IL-2 þ IL-12. The majority of the T cells acquired a late/intermediate phenotypic profile. These observations further emphasize an intrinsic link between cytokine expression, transcriptional regulation and cellular differentiation and their impact on functional plasticity of antigen-specific CD4 þ T cells. DISCUSSION The CD4 þ T-cell repertoire to an individual infectious agent such as CMV, is directed towards a diverse array of antigenic peptides and it has been shown that a large frequency of antigen-specific T cells is
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 185
Frequency of cytokine prodycing CD4+ T-cells (%)
a
Granzyme B ns
100
Perforin
IL-2
**
**
80 60 pp65 40 20 0 TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+
Frequency of cytokine producing CD4+ T-cells (%)
b
GranzymeB
100
ns
Perforin
IL-2
**
**
80 60 IE-1 40 20 0 TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+
c Frequency of cytokine producing CD4+ T-cells (%)
Granzyme B 100
Perforin
ns
IL-2 ns
*
80 60 gB 40 20 0 TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+
d
e 4000
**
**
**
3000
IFN-+TNF-+
**
**
ns
pp65
IE-1 Protein
gB
3000 IL-2 MFI
Perforin MFI
TNF-+ alone 4000
2000
2000 1000
1000 0 pp65
IE-1 Protein
gB
0
Figure 3 Cytolytic potentiality and IL-2 production by CMV-specific CD4 þ T cells. PBMCs from CMV-seropositive donors (n ¼ 10) were incubated overnight with recombinant proteins to pp65, IE-1 or gB, in the presence of brefaldin A. Cells were then labelled with anti-CD4 and assessed for the intracellular expression of IFN-g, TNF-a, granzyme B, perforin and IL-2. Panels a–c show frequency of granzyme B þ , perforin þ and IL-2 þ CD4 þ T cells directed towards pp65, IE-1 and gB proteins. The mean value is represented by a horizontal line. Panels d and e show mean±SEM of the mean fluorescence intensity of perforin and IL-2 expression in CMV-specific CD4 þ T-cell populations with distinct cytokine profile (TNF-a þ or IFN-g þ TNF-a þ ). *Po0.05; **Po0.001; ns, not significant.
induced ready for secondary exposure to antigen.5,27–29 CD4 þ T-cell populations specific for certain proteins are preferentially expanded and comprises heterogeneous memory T cells.30,31 To date, the factors responsible for the generation of the functional and phenotypic
heterogeneity of T cells have yet to be elucidated. However, studies have suggested that antigen load and persistence,31–33 precursor frequency, pMHCII/TCR avidity and cytokine milieu have an impact on Th commitment and memory T cell generation. Immunology and Cell Biology
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 186
10
10
2
2
2
5
3
10
10
4
10
5
Frequency of cytokine producing CD4+ T-cells (%)
4
10
10
10
3
3
10
10
2
100
10
4
10
5
CCR7
3
4
10
5
10
5
0
3
10
4
10
0 0 10
5
10
5.8%
4
3
10
4
10
5
2
10
0 10 5
10
9.9%
4
3
4
10
10
5
10
0.72%
5
10
3
3
10
4
10
3
10
10
2
2
10
10
0
0 2
0 10
3
4
10
10
5
10
CD27
CD45RA
**
**
**
2
10 2
10
10 0
0 3
10
2
0
2
10 2
10
10 2
10
48.3%
3
10
0 10
50.0%
4
10
3
3
5
4
0 10
2
0 10
3
10
4
10
5
10
CD57
2
0 10
3
4
10
10
5
10
CD62L
**
**
80 60
pp65
40 20 0 TNF-+ alone
Frequency of cytokine producing CD4+ T-cells (%)
3
10
2
c
10
5
10 0
b
0
10
6.9%
4
10
2
10
3.3%
4
10
12.3%
4
10 0
0
CD4
10
10
10
0 10
10
10
3
3
10 0
10
10
58.3%
4
10
CD62L 5
+
10
4
CD57 5
TNF- alone
10
CD45RA 5
+
10
CD27 5
9.0%
+
CCR7 5
IFN- TNF-
a
100
IFN-+ TNF-+
CCR7
**
TNF-+ alone
IFN-+ TNF-+
TNF-α+ alone
IFN-+ TNF-+
CD27
CD45RA
**
**
TNF-α+ alone
IFN-+ TNF-+
CD57 ns
TNF-+ alone
IFN-+ TNF-+
CD62L
**
80 60 IE-1 40 20 0 TNF-+ alone
IFN-+ TNF-+
TNF-+ alone
IFN-+ TNF-+
TNF-+ alone
IFN-+ TNF-+
TNF-+ alone
IFN-+ TNF-+
TNF-+ alone
IFN-+ TNF-+
Figure 4 Differential phenotypic profile of CMV-specific CD4 þ T cells expressing TNF-a or IFN-g and TNF-a. PBMCs from CMV-seropositive donors (n ¼ 10) were incubated overnight with recombinant proteins to pp65, IE-1 or gB, stained with fluorescently labelled anti-CD4, CCR7, CD27, CD45RA, CD57 and CD62L followed by staining for the intracellular expression of IFN-g and/or TNF-a. Panel a shows representative analysis of cell surface marker expression by TNF-a þ and IFN-g þ TNF-a þ CMV-specific CD4 þ T cells. Cytokine-producing cells are represented as an overlay of total CD4 þ T cells. Panels b and c show comprehensive analysis of individual surface marker on pp65 and IE-1-specific CD4 þ T cells expressing TNF-a or IFN-g and TNF-a. Horizontal bars represent the mean frequency of CD4 þ CMV-specific cells expressing each individual marker. **Po0.001; ns, not significant.
Recently, the transcriptional regulation of Th cells has received unprecedented attention. In particular, the impact of T-box transcription factors (for example, T-bet and Eomes) on the commitment of cells to the Th1 lineage and cytokine expression patterns has been identified. T-bet expression in CD4 þ T cells has been shown to be required for commitment of these cells to the Th1 lineage, optimal IFN-g production1,8,34,35 and proinflammatory trafficking.36 Eomes expression has also been shown to invoke IFNg production even in the absence of T-bet. Thus, as for CD8 þ T cells, Immunology and Cell Biology
cooperative T-bet and Eomes expression may be required for full effector function.13 However, the full impact of Eomes expression on the functional and phenotypic profile of T cells, as well as the impact of cooperative transcription factor expression has yet to be elucidated. The finding of a CMV-specific CD4 þ T-cell population which exhibited predominant TNF-a only expression was unexpected, as previous studies by Widmann and colleagues had indicated that simultaneous induction of Th1 cytokines occurs in CMV-specific CD4 þ T cells.21 It is important to note that Widmann and colleagues
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 187 250K
150K
CD4
SSC-A
200K
100K
105
105
104
104 CD4
a
103
103
102 0
50K
0K
0 102
25
0K
0K
20
15
0K 10
50
0
K
0
102 0 103
104
105
0 102
CD8
103 104 CD95
105
105
105
104
104
104
103 102 0
IFN-γ
105
CCR7
CD27
FSC-A
103
102 0
102 0 0 102
103
104
105
0 102
CD45RA
Frequency of cytokine producing CD4+ T-cells (%)
b
103
104
CD45RA
**
103
105
0 102
103
104
105
TNF-
**
100
Late (CD95+CD45RA+/-CCR7-CD27-) Intermediate (CD95+CD45RA+CCR7-CD27+) Early (CD95+CD45RA-CCR7+CD27+) TSCM (CD95+CD45RA+CCR7+CD27+)
50
0
TNF-+ only
TNF-+ IFN-+
pp65
TNF-+ only
TNF-+ IFN-+ IE-1
CD4 þ
Figure 5 Frequencies of CMV-specific T cells displaying differential maturation status expressing TNF-a alone or IFN-g and TNF-a. PBMCs from CMV-seropositive donors were incubated overnight with recombinant pp65 or IE-1, then stained with fluorescently labelled anti-CD4, CD95, CD27, CCR7 and CD4RA and assessed for the intracellular expression of IFN-g and TNF-a. Panel a shows gating strategy for the identification of CMV-specific CD4 þ T cells expressing different maturation status that is, late (CD95 þ CD45RA þ / CCR7 CD27 ), intermediate (CD95 þ CD45RA þ CCR7 CD27 þ ), early (CD95 þ CD45RA CCR7 þ CD27 þ ) or TSCM (CD95 þ CD45RA þ CCR7 þ CD27 þ ). Panel b shows relative proportions of TNF-a þ and IFN-g þ TNF-a þ CMV-specific CD4 þ T cells directed towards pp65 or IE-1 exhibiting a late, intermediate, early or TSCM phenotype. TSCM frequency was compared between TNF-a þ only and TNF-a þ IFN-g þ . **Po0.001.
stimulated T cells with whole CMV lysate which would primarily activate T cells specific for immunodominant antigens including pp65 and gB with minimal activation of IE-1-specific T cells. In our study we have used pp65, gB and IE-1 proteins individually which allows dissection of cytokine expression pattern in T cells specifically directed to each of these antigens. Data presented in our study also show that the predominant expression of either TNF-a only or IFN-g and TNFa by CMV-specific CD4 þ T cells was coincident with dominant expansion of either T-bet Eomes or T-betHiEomesHi/Lo antigenspecific T cells, respectively. More importantly, the T-bet Eomes phenotype of antigen-specific CD4 þ T cells correlated with poor functional capacity (specifically, the production of perforin), as well as immature status when compared with antigen-specific T cells with T-betHiEomesHiLo profile. Previous studies have shown that T-bet and Eomes expression has a concomitant impact upon the generation of memory and effector CD8 þ T cells in both humans and mice.13,37–39 A recent study by Gattinoni et al., has also shown the ability of earlydifferentiated T cells known as TSCM to express cytokines including
TNF-a. Apart from the upregulation of CD95, there is no evidence of phenotypic differences in the expression of differentiation markers compared with naı¨ve T cells.25 Our results have shown that the presence of less-differentiated populations endows T cells with functional flexibility. This indicates remarkable heterogeneity of CMV-specific CD4 þ T cells which are amenable to modulation in antiviral immune responses. In CMV infection, periodic viral reactivation occurs, which makes T cells susceptible to functional exhaustion. These T cells lose the ability to control active virus replication and thus viral control wanes.40,41 However, data presented here show that heterogeneity exists with respect to cellular differentiation of CMV-specific T cells. Thus when moredifferentiated T cells become exhausted, the less-differentiated T cells have the propensity to further differentiate and acquire the ability to upregulate effector functions such as perforin production. This heterogeneity allows less-differentiated T cells to reinforce the immune responses control of viral infection, upon exhaustion of their more-differentiated counterparts. Immunology and Cell Biology
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 188
a
IFN-+ alone IL-2
IL-4
IFN-+TNF-+ alone
IL-12
IL-2+IL-4
IL-2+IL-12
IE-1
pp65
No cytokine
TNF-+ alone
b
T-bet-Eomes-
% Cytokine Positive
100
T-betLoEomesHi
ns
ns
ns
**
**
T-betHiEomesHi/Lo
*
**
**
**
80 60 40 20 0 No cytokine
IL-2
IL-2+IL-12
50
ns
*
**
40 30 20 10 no IL-2 cytokine
IL-12 IL-2+IL-12
Frequeccy of cytokine producing cells CD4+ T-cells (%)
d
c % Perforin+
IL-12
*
100
*
80 60 40 20 0 No IL-2 cytokine
IL-12 IL-2+IL-12
Late (CD95+CD45RA+/-CCR7-CD27-) Intermediate (CD95+CD45RA+CCR7-CD27+) Early (CD95+CD45RA-CCR7+CD27+) Figure 6 Impact of antigenic stimulation and inflammatory cytokines on the functional and transcriptional profile of CMV-specific T cells. PBMCs from CMVseropositive individuals were cultured with or without pp65 and IE-1 protein in the presence of IL-2, IL-4 and/or IL-12 or with no cytokine for 7 days. Following incubations, these cells were assessed for the expression of IFN-g, TNF-a, perforin, IL-2, T-bet and Eomes as outlined in the ‘Material and Methods’ section. Panel a shows relative proportion of pp65 and IE-1-specific CD4 þ T cells exhibiting an IFN-g þ alone, TNF-a þ alone or IFN-g þ TNF-a þ expression profile following antigenic stimulation in the absence or presence of IL-2, IL-4 and/or IL-12. Panels b and c show impact of antigenic stimulation in combination with IL-2 and/or IL-12 on the transcriptional profile (T-bet and Eomes) and perforin expression in IE-1-specific CD4 þ T cells. T-bet and Eomes profiles were compared between T cells treated with no cytokine and those treated with IL-2 and/or IL-12. *Po0.05; **Po0.001; ns, not significant. Panel d shows the alteration in the maturation status (early, intermediate or late) of IE-1-specific CD4 þ T cells following incubation with IL-2 and/or IL-12 and recombinant viral protein. For maturation status analysis, T cells following incubation with viral antigens and IL-2 and/or IL-12 were stained with fluorescently labelled anti-CD4, CD95, CD27, CCR7 and CD4RA antibodies and then analysed using FACS Canto II.
Previous studies have shown that cytokines (for example, IL-12) can modulate the expression of T-bet and Eomes, which could have a concomitant impact upon cellular differentiation status. It is important to note that newly activated mature CD8a þ dendritic cells secrete significant amounts of IL-12 upon murine CMV infection.42,43 Thus IL-12 may have a role in modulating expression of T-box transcription factors post CMV infection. In this study, in vitro exposure to IE-1 antigen in combination with IL-12 and/or IL-2 induced T-bet upregulation and graded expression of Eomes, which Immunology and Cell Biology
indicates that these cytokines have a role in Th1 programming of cells. IL-12 in particular directs upregulation of both IFN-g and perforin expression and transition to a more-differentiated phenotype. Thus production of IL-12 by infected dendritic cells, particularly during the priming phase, has a role in reinforcement of T-bet expression, differentiation and perforin acquisition. It is unclear whether this change in cytokine expression pattern is as a result of phenotypic changes in a select population of IE-1 cells or expansion of the IE-1 population as a whole. Further investigations will be required to
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 189
dissect the cellular basis of this observation. Nevertheless, the discordance of phenotypic and functional programming in different CMV-specific T-cell populations suggests that antigen load of infected cells has a pivotal role in generation of T cells with different levels of functional flexibility. Recent studies have shown that to maintain the effector T-cell pool for MCMV, continuous stimulation is required after latency has been established, thereby providing protection against viral reactivation.44,45 This would account for the expansion of intermediate to late memory T-cell populations with higher perforin expression concurrent with higher T-bet expression. Antigen density upon priming has been shown to influence generation of memory CD8 þ T cells.33 Therefore, we postulate that the different expression kinetics of the CMV proteins yield an explanation for why T cells specific for the different proteins are heterogeneous in their phenotypic and transcriptional programming. For example, IE-1 is expressed intermittently compared with pp65 and gB, which are expressed throughout the viral cycle.24 This intermittent antigen expression may suggest that antigen load is lower for this protein and as such is less available for encounter with T cells. Antigen density has already been shown to influence memory generation of CD8 þ T cells at priming, thereby explaining the relative immaturity of these cells compared with those specific for other antigens. Thus, persistent virus and cytokine signals cooperate to determine the phenotype and functionality of CMV-specific T cells. These observations have important implications for the modulation of immune responses in therapeutic interventions where a protective response to virus is required. Further investigation of the precise differences in antigen processing and presentation of different CMV proteins will provide an insight into why different antigens induce phenotypic and transcriptional variation in T-cell populations. In addition, the elucidation of the precise mechanism by which cytokines modulate T-cell phenotype will improve our understanding of how cytokine milieu variation can be used to modify antiviral immune responses.
METHODS Human volunteer recruitment A panel of healthy CMV-seropositive donors (range ¼ 28–59 years) were recruited for this study. This study was approved by the QIMR Human Ethics Committee. Informed consent was obtained from each donor according to QIMR Human Ethics Committee guidelines.
Intracellular Cytokine analysis of CMV-specific
CD4 þ
T cells
To assess cytokine expression by CMV-specific T-cells, PBMCs from CMVseropositive individuals were incubated for 2 h with 25 mg ml 1 CMV proteins pp65, gB or IE-1 and then incubated overnight in the presence of Brefaldin A. After overnight incubation, cells were incubated with live/dead fixable Aqua (Invitrogen, Mulgrave, VIC, Australia), Pacific Blue-labelled anti-CD4 (BD Biosciences, Franklin Lakes, NJ, USA) and Peridinin chlorophyll cyanin-Cy5.5 labelled anti-CD8 (BD Biosciences). Cells were then fixed and permeabilised using cytofix/cytoperm kit (BD Biosciences) and incubated with R-Phycoerythrin (PE)-Cy7 anti- IFN-g (BD Biosciences), Alexa Fluor-700 or Allophycocyanin-labelled anti-TNF-a (BD Biosciences), and combinations of the following antibodies, PE-labelled anti-IL-2 (BD Biosciences), fluorescein isothiocyanate or Alexa Fluor-700-labelled anti-granzyme B (BD Biosciences), anti-perforin PE (BD Biosciences). Acquisition of data from these cells was carried out using a FACSCanto with FACSDiva software (BD Biosciences, Palo Alto, CA, USA) and post acquisition analysis was performed using FlowJo software (Treestar, Ashland, OR, USA).
In vitro expansion of CMV-specific CD4 þ T cells To expand CMV-specific CD4 þ T cells, PBMCs from CMV-seropositive donors were initially incubated with human CD8 microbeads (Miltenyi Biotech GmbH, Bergisch Gladbach Germany), and separated on an AutoMACS column to deplete CD8 þ T cells according to the manufacturer’s guidelines (Miltenyi Biotech GmbH). These CD8 þ T cells depleted PBMCs were then incubated with 25 mg ml 1 recombinant CMV proteins (pp65, IE-1 or gB) for 2 h. Following incubation, these cells were cultured in RPMI-1640 medium supplemented with 10% fetal calf serum, recombinant IL-2 (10 IU ml 1) (R&D Systems, Minneapolis, MN, USA) and 30% MLA-144 supernatant. On day 14, antigen-specific CD4 þ T cells were assessed for their production of IFN-g and/or TNF-a using intracellular cytokine assay described above. In some experiments, CMV-specific CD4 þ T cells were also expanded as outlined above in the presence or absence of recombinant IL-2 (10 IU ml 1), recombinant IL-4 (10 ng ml 1) (R&D systems) and/or recombinant IL-12 (10 ng ml 1) (R&D systems).
Phenotypic analysis of CMV-specific CD4 þ T cells To assess ex vivo phenotypic profile of CMV-specific CD4 þ T cells directed towards IE-1, pp65 or gB protein, PBMCs were stimulated for 2 h with 25 mg ml 1 CMV protein alone, followed by addition of Brefaldin A for a further 14 h. Cells were then incubated with live/dead fixable Aqua, Pacific Blue-labelled anti-CD4, Peridinin chlorophyll cyanin-Cy5.5-labelled anti-CD8 and a combination of fluorescein isothiocyanate-labelled anti-CD45RA, Alexa Fluor-700-labelled antiCCR7, fluorescein isothiocyanate-labelled anti-CD57, PE-labelled anti-CD27 and Allophycocyanin-labelled anti-CD62L (all antibodies purchased from BD Biosciences). Cells were then fixed and permeabilised using the Cytofix/Cytoperm kit (BD Biosciences), and incubated with PE-Cy7-labelled IFN-g and Allophycocyanin or AF700-labelled TNF-a. Acquisition of data from these cells was carried out using a FACSCanto with FACSDiva software (BD Biosciences) and post acquisition analysis was performed using FlowJo software (Treestar).
Intracellular T-bet and Eomes analysis in CMV-specific CD4 þ T cells To assess the expression of the transcription factors T-bet and Eomes in CMVspecific CD4 þ T-cell populations, PBMCs from CMV-seropositive individuals were incubated for 2 h with 25 mg ml 1 CMV proteins (IE-1, pp65 or gB), followed by overnight incubation in the presence of Brefaldin A. After overnight incubation, cells were incubated with live/dead fixable Aqua (Invitrogen), Pacific Blue-labelled anti-CD4 and Peridinin chlorophyll cyanin-Cy5.5-labelled anti-CD8 (BD Biosciences), then fixed and permeabilised using the FoxP3 staining kit (eBioscience, San Diego, CA, USA). Cells were then incubated with PE-Cy7 anti- IFN-g (BD Biosciences), AF700-labelled anti-TNF-a (BD Bioscience), PE-labelled anti-T-bet and Allophycocyaninlabelled anti-Eomes (eBioscience). Acquisition of data from these cells was carried out using a FACSCanto with FACSDiva software (BD Biosciences) and post acquisition analysis was performed using FlowJo software (Treestar).
Statistical analysis All statistical analysis was undertaken using GraphPad Prism 5 Software (GraphPad Software Inc., La Jolla, CA, USA). A One-way analysis of variance followed by a post-hoc Tukey test was used to determine the significant difference observed between the mean cytokine expressions of experimental samples in Figure 1 and the T-bet and Eomes profiles in Figure 6. In addition, a non-parametric Wilcoxon test was used to determine the significance of differences between experimental samples for Figure 2 onwards. In all cases, differences where Po0.05 were considered to be significant.
CONFLICT OF INTEREST The authors declare no conflict of interest.
ACKNOWLEDGEMENTS We acknowledge the help of Linda Jones and Leone Beagley who assisted with sample processing. Thanks to Leesa Wockner for her statistical advice. This work was supported by the National Health and Medical Research Council (NH&MRC). RK is supported by NH&MRC Senior Principal Research Fellowship. Immunology and Cell Biology
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 190 1 Szabo SJ, Sullivan BM, Peng SL, Glimcher LH. Molecular mechanisms regulating Th1 immune responses. Annu Rev Immunol 2003; 21: 713–758. 2 O’Shea JJ, Paul WE. Regulation of T(H)1 differentiation–controlling the controllers. Nat Immunol 2002; 3: 506–508. 3 Murphy KM, Reiner SL. The lineage decisions of helper T cells. Nat Rev Immunol 2002; 2: 933–944. 4 Marshall NB, Swain SL. Cytotoxic CD4 T cells in antiviral immunity. J Biomed Biotechnol 2011; 2011: 954602. 5 Crompton L, Khan N, Khanna R, Nayak L, Moss PA. CD4 þ T cells specific for glycoprotein B from cytomegalovirus exhibit extreme conservation of T-cell receptor usage between different individuals. Blood 2008; 111: 2053–2061. 6 Aslan N, Yurdaydin C, Wiegand J, Greten T, Ciner A, Meyer MF et al. Cytotoxic CD4 T cells in viral hepatitis. J Viral Hepat 2006; 13: 505–514. 7 Schulz EG, Mariani L, Radbruch A, Hofer T. Sequential polarization and imprinting of type 1 T helper lymphocytes by interferon-gamma and interleukin-12. Immunity 2009; 30: 673–683. 8 Szabo SJ, Kim ST, Costa GL, Zhang X, Fathman CG, Glimcher LH. A novel transcription factor, T-bet, directs Th1 lineage commitment. Cell 2000; 100: 655–669. 9 Szabo SJ, Sullivan BM, Stemmann C, Satoskar AR, Sleckman BP, Glimcher LH. Distinct effects of T-bet in TH1 lineage commitment and IFN-gamma production in CD4 and CD8 T cells. Science 2002; 295: 338–342. 10 Mullen AC, High FA, Hutchins AS, Lee HW, Villarino AV, Livingston DM et al. Role of T-bet in commitment of TH1 cells before IL-12-dependent selection. Science 2001; 292: 1907–1910. 11 Mullen AC, Hutchins AS, High FA, Lee HW, Sykes KJ, Chodosh LA et al. Hlx is induced by and genetically interacts with T-bet to promote heritable TH1 gene induction. Nat Immunol 2002; 3: 652–658. 12 Long M, Slaiby AM, Hagymasi AT, Mihalyo MA, Lichtler AC, Reiner SL et al. T-bet down-modulation in tolerized Th1 effector CD4 cells confers a TCR-distal signaling defect that selectively impairs IFN-gamma expression. J Immunol 2006; 176: 1036–1045. 13 Pearce EL, Mullen AC, Martins GA, Krawczyk CM, Hutchins AS, Zediak VP et al. Control of effector CD8 þ T cell function by the transcription factor Eomesodermin. Science 2003; 302: 1041–1043. 14 Yang Y, Xu J, Niu Y, Bromberg JS, Ding Y. T-bet and eomesodermin play critical roles in directing T cell differentiation to Th1 versus Th17. J Immunol 2008; 181: 8700–8710. 15 Elkington R, Shoukry NH, Walker S, Crough T, Fazou C, Kaur A et al. Cross-reactive recognition of human and primate cytomegalovirus sequences by human CD4 cytotoxic T lymphocytes specific for glycoprotein B and H. Eur J Immunol 2004; 34: 3216–3226. 16 Gamadia LE, Remmerswaal EB, Weel JF, Bemelman F, van Lier RA, Ten Berge IJ. Primary immune responses to human CMV: a critical role for IFN-gamma-producing CD4 þ T cells in protection against CMV disease. Blood 2003; 101: 2686–2692. 17 Gamadia LE, Rentenaar RJ, van Lier RA, Ten Berge IJ. Properties of CD4( þ ) T cells in human cytomegalovirus infection. Hum Immunol 2004; 65: 486–492. 18 Villacres MC, Longmate J, Auge C, Diamond DJ. Predominant type 1 CMV-specific memory T-helper response in humans: evidence for gender differences in cytokine secretion. Hum Immunol 2004; 65: 476–485. 19 Rentenaar RJ, Gamadia LE, van DerHoek N, van Diepen FN, Boom R, Weel JF et al. Development of virus-specific CD4( þ ) T cells during primary cytomegalovirus infection. J Clin Invest 2000; 105: 541–548. 20 Zhu J, Shearer GM, Marincola FM, Norman JE, Rott D, Zou JP et al. Discordant cellular and humoral immune responses to cytomegalovirus infection in healthy blood donors: existence of a Th1-type dominant response. Int Immunol 2001; 13: 785–790. 21 Widmann T, Sester U, Gartner BC, Schubert J, Pfreundschuh M, Kohler H et al. Levels of CMV specific CD4 T cells are dynamic and correlate with CMV viremia after allogeneic stem cell transplantation. PLoS One 2008; 3: e3634. 22 Sylwester AW, Mitchell BL, Edgar JB, Taormina C, Pelte C, Ruchti F et al. Broadly targeted human cytomegalovirus-specific CD4 þ and CD8 þ T cells dominate the memory compartments of exposed subjects. J Exp Med 2005; 202: 673–685. 23 Slezak SL, Bettinotti M, Selleri S, Adams S, Marincola FM, Stroncek DF. CMV pp65 and IE-1 T cell epitopes recognized by healthy subjects. J Transl Med 2007; 5: 17.
Immunology and Cell Biology
24 Stinski MF. Sequence of protein synthesis in cells infected by human cytomegalovirus: early and late virus-induced polypeptides. J Virol 1978; 26: 686–701. 25 Gattinoni L, Lugli E, Ji Y, Pos Z, Paulos CM, Quigley MF et al. A human memory T cell subset with stem cell-like properties. Nat Med 2011; 17: 1290–1297. 26 Lugli E, Gattinoni L, Roberto A, Mavilio D, Price DA, Restifo NP et al. Identification, isolation and in vitro expansion of human and nonhuman primate T stem cell memory cells. Nat Protoc 2013; 8: 33–42. 27 Crough T, Beagley L, Smith C, Jones L, Walker DG, Khanna R. Ex vivo functional analysis, expansion and adoptive transfer of cytomegalovirus-specific T-cells in patients with glioblastoma multiforme. Immunol Cell Biol 2012; 90: 872–880. 28 Burrows SR, Moss DJ, Khanna R. Understanding human T-cell-mediated immunoregulation through herpesviruses. Immunol Cell Biol 2011; 89: 352–358. 29 Khan N, Hislop A, Gudgeon N, Cobbold M, Khanna R, Nayak L et al. Herpesvirusspecific CD8 T cell immunity in old age: cytomegalovirus impairs the response to a coresident EBV infection. J Immunol 2004; 173: 7481–7489. 30 Verhoeven D, Teijaro JR, Farber DL. Heterogeneous memory T cells in antiviral immunity and immunopathology. Viral Immunol 2008; 21: 99–113. 31 Harari A, Vallelian F, Meylan PR, Pantaleo G. Functional heterogeneity of memory CD4 T cell responses in different conditions of antigen exposure and persistence. J Immunol 2005; 174: 1037–1045. 32 Vanguri V, Govern CC, Smith R, Huseby ES. Viral antigen density and confinement time regulate the reactivity pattern of CD4 T-cell responses to vaccinia virus infection. Proc Natl Acad Sci 2013; 110: 288–293. 33 Leignadier J, Labrecque N. Epitope density influences CD8 þ memory T cell differentiation. PLoS One 2010; 5: e13740. 34 Afkarian M, Sedy JR, Yang J, Jacobson NG, Cereb N, Yang SY et al. T-bet is a STAT1-induced regulator of IL-12R expression in naive CD4 þ T cells. Nat Immunol 2002; 3: 549–557. 35 Szabo SJ, Sullivan BM, Stemmann C, Satoskar AR, Sleckman BP, Glimcher LH. Distinct effects of T-bet in TH1 lineage commitment and IFN-g production in CD4 and CD8 T cells. Science 2002; 295: 338–342. 36 Lord GM, Rao RM, Choe H, Sullivan BM, Lichtman AH, Luscinskas FW et al. T-bet is required for optimal proinflammatory CD4 þ T-cell trafficking. Blood 2005; 106: 3432–3439. 37 Smith C, Elhassen D, Gras S, Wynn KK, Dasari V, Tellam J et al. Endogenous antigen presentation impacts on T-box transcription factor expression and functional maturation of CD8 þ T cells. Blood 2012; 120: 3237–3245. 38 Intlekofer AM, Takemoto N, Wherry EJ, Longworth SA, Northrup JT, Palanivel VR et al. Effector and memory CD8 þ T cell fate coupled by T-bet and eomesodermin. Nat Immunol 2005; 6: 1236–1244. 39 Intlekofer AM, Takemoto N, Kao C, Banerjee A, Schambach F, Northrop JK et al. Requirement for T-bet in the aberrant differentiation of unhelped memory CD8 þ T cells. J Exp Med 2007; 204: 2015–2021. 40 Jelley-Gibbs DM, Dibble JP, Filipson S, Haynes L, Kemp RA, Swain SL. Repeated stimulation of CD4 effector T cells can limit their protective function. J Exp Med 2005; 201: 1101–1112. 41 Antoine P, Olislagers V, Huygens A, Lecomte S, Liesnard C, Donner C et al. Functional exhaustion of CD4 þ T lymphocytes during primary cytomegalovirus infection. J Immunol 2012; 189: 2665–2672. 42 Orange JS, Biron CA. Characterization of early IL-12, IFN-alphabeta, and TNF effects on antiviral state and NK cell responses during murine cytomegalovirus infection. J Immunol 1996; 156: 4746–4756. 43 Ruzek MC, Miller AH, Opal SM, Pearce BD, Biron CA. Characterization of early cytokine responses and an interleukin (IL)-6-dependent pathway of endogenous glucocorticoid induction during murine cytomegalovirus infection. J Exp Med 1997; 185: 1185–1192. 44 Seckert CK, Schader SI, Ebert S, Thomas D, Freitag K, Renzaho A et al. Antigenpresenting cells of haematopoietic origin prime cytomegalovirus-specific CD8 T-cells but are not sufficient for driving memory inflation during viral latency. J Gen Virol 2011; 92: 1994–2005. 45 Torti N, Walton SM, Brocker T, Rulicke T, Oxenius A. Non-hematopoietic cells in lymph nodes drive memory CD8 T cell inflation during murine cytomegalovirus infection. PLoS Pathog 2011; 7: e1002313.
ORIGINAL ARTICLE
Phenotypic and transcriptional profile correlates with functional plasticity of antigen-specific CD4 þ T cells Emily SJ Edwards, Corey Smith and Rajiv Khanna The role of CD4 þ T cells in the control of infectious pathogens is highly complex with a myriad of functions but how these T cells acquire differential functional potentiality remains poorly defined. Here we show that human cytomegalovirus (CMV)specific CD4 þ T cells directed towards different viral antigens expressed predominantly TNF-a alone or TNF-a and IFN-c. TNF-a þ and IFN-c þ CD4 þ T cells expressed significantly higher levels of T-box transcription factors T-bet with graded loss of Eomesodermin (Eomes) expression (T-betHiEomesHi/Lo) when compared with TNF-a þ CD4 þ T cells expressing lower levels of both T-bet and Eomes (T-bet Eomes ). Furthermore, TNF-a þ and IFN-c þ CD4 þ T cells expressed significantly higher levels of perforin and interleukin (IL)-2 and displayed a terminally differentiated phenotype (CCR7 CD27 CD45RA CD57 þ CD62L ). In contrast, TNF-a þ alone CMV-specific CD4 þ T cells were predominantly early-memory phenotype with a proportion of these cells displaying T memory stem-cell phenotype (CD95 þ CD45RA þ CCR7 þ CD27 þ ). In vitro stimulation of CMV-specific CD4 þ T cells with viral antigen in the presence of IL-12 was sufficient to dramatically change the transcriptional and functional profile of TNF-a þ CD4 þ T cells, whereas TNF-a þ and IFN-c þ CD4 þ T cells remained unaltered. These findings illustrate an intrinsic link between cytokine expression, transcriptional regulation and cellular differentiation, and their impact on functional plasticity of virus-specific CD4 þ T cells. Immunology and Cell Biology (2014) 92, 181–190; doi:10.1038/icb.2013.88; published online 3 December 2013 Keywords: antigen; cytokine; T cells; transcription factors; virus
CD4 þ T cells elicit a plethora of effector functions required for the effective elimination and control of intracellular and extracellular pathogens as well as neoplastic cells. Each naı¨ve CD4 þ T cell has the potential to differentiate into one of several distinct effector T-helper (Th) lineages. The recognition of the cognate antigen in complex with MHC class II molecules initiates the differentiation process. Strength of T-cell receptor (TCR) signal in conjunction with cytokine milieu, antigen concentration, type of antigen presenting cell and the ligation of co-stimulatory molecules determines the Th lineage that the cell is consequently committed to.1–3 As differentiation proceeds, antigenspecific Th cells acquire and maintain effector function enabling a robust response upon re-exposure to cognate antigen. These effector functions include cytokine production, ability to proliferate and expand in response to repeated exposure, provision of cognate help, and most recently described the ability to directly lyse target cells expressing cognate peptide bound to MHC II molecules.4–6 CD4 þ Th phenotypes can be characterized according to the pattern of cytokines and transcription factors expressed. Th1 immunity is characterized by the production of the interferon-gamma (IFN-g),1 in conjunction with expression of the T-box transcription factor Tbx21 (T-bet). Initial TCR stimulation upregulates T-bet expression, and
upon cessation of TCR signalling, T-bet induces expression of interleukin (IL)-12 receptor b2, which is required for responsiveness to IL-12 and reinforces Th1 commitment by amplifying T-bet expression.7 Numerous studies have identified that T-bet is essential for the generation and maintenance of Th1 cells, as well as representing a major factor for the induction of IFN-g production.8–11 T-bet has been shown to antagonize expression of other lineage specifying factors, including Th2 transcription factor GATA-3. Overexpression of T-bet in developing or fully differentiated Th2 cells has been shown to repress IL-4 production and induce IFN-g secretion.10,11 Unsurprisingly, T-bet deficiency partially inhibits IFN-g production and halts Th1 differentiation.9,11,12 A second T-box transcription factor, Eomesodermin (Eomes), is expressed in CD4 þ T cells albeit at a lower level than in their CD8 counterparts.13 Eomes expression is induced by TCR stimulation, consequent to T-bet expression. Eomes has been shown to complement T-bet for IFN-g production and Th1 commitment. However, Eomes expression alone is inadequate for inhibition of signals mediated by polarizing cytokines to other Th lineages, thereby inducing Th lineage switching.14 Both T-bet and Eomes have been shown to cooperatively induce perforin and granzyme B expression in addition to that of IFN-g.13
Centre for Immunotherapy and Vaccine Development and Tumour Immunology Laboratory, Department of Immunology, Queensland Institute of Medical Research, Brisbane, Queensland, Australia Correspondence: Professor R Khanna, Queensland Institute of Medical Research, Tumour Immunology Laboratory, Department of Immunology, 300 Herston Rd, Brisbane 4006, Queensland, Australia. E-mail: [email protected] Received 2 September 2013; revised 23 October 2013; accepted 24 October 2013; published online 3 December 2013
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 182
To date, very little has been elucidated about how T-bet and Eomes expression impacts on the acquisition of functional plasticity in antigen-specific CD4 þ T cells. Here we show that CD4 þ T cells directed towards different CMV antigens expressed during different stages of viral infection exhibit distinct cytokine expression potentiality which correlates with differential patterns of T-box transcriptional factors, and this expression is linked to their ability to produce perforin and IL-2. This differential expression of cytokines and T-box transcriptional factors was also associated with an altered differentiation status and functional plasticity in terms of their response to antigenic stimulation in presence of inflammatory cytokines. These results illustrate how CD4 þ T-cell programming during persistent viral infection modulates the transcriptional profile of antigen-specific T cells and their functional plasticity.
RESULTS Differential expression of IFN-c and TNF-a by CMV-specific CD4 þ T cells directed towards different viral antigens Previous studies have demonstrated that CMV-specific CD4 þ T cells display heterogeneous patterns of cytokine expression where populations of both single-cytokine-producing and polyfunctional T cells can be activated upon stimulation with CMV lysate or CMV-infected cells.15–21 However, to date, the functional profile of CD4 þ T-cell populations directed towards different CMV proteins remains undetermined. In the first set of experiments, we assessed IFN-g and TNF-a expression in CMV-specific CD4 þ T cells directed towards three different immunodominant CMV proteins pp65, IE-1 and gB.5,15,22,23 It has been shown previously that different expression patterns are exhibited by these three proteins, with IE-1 being expressed during early phase of replicative infection, whereas pp65 and gB expressed during late phase of the viral infection and are packaged in the viral particle.24 Peripheral blood mononuclear cells (PBMCs) from a panel of CMV-seropositive donors were stimulated with recombinant CMV proteins, and then assessed for the expression of IFN-g and TNF-a. These analyses showed that the responding frequency of CD4 þ T cells directed towards IE-1 antigen were 3.51-fold lower when compared with the CD4 þ T cells specific for pp65 and 1.75-fold lower than gB-specific CD4 þ T cells (Mean: IE-1-specific CD4 þ T cells: 0.3002% (range 0.069–0.6420%); pp65-specific CD4 þ T cells: 1.055% (range 0.152–4.302%); gB-specific CD4 þ T cells: 0.5263% (range 0.0239–1.16%)). It is important to note that in all the donors tested the IE-1 response was subdominant compared with that elicited to pp65 and gB. Interestingly, co-expression analysis revealed discordance in IFN-g and TNF-a expression in CMV-specific CD4 þ T-cell populations (Figures 1a–c). Although no statistical significance was observed between the magnitude of antigen-specific T cells expressing IFN-g alone, CD4 þ T cells directed against pp65 and gB were predominantly IFN-g þ TNF-a þ , whereas the majority of IE-1-specific CD4 þ T cells expressed only TNF-a. No inter-individual variation was observed with regards to the predominance of TNF-a only cells for IE-1, or IFN-g þ TNF-a þ for pp65 and gB. IE-1 dose escalation did not alter the predominance of the TNF-a þ only cells elicited by this protein (data not shown). Therefore, CMV-specific CD4 þ T cells displayed two distinct patterns of cytokine expression: TNF-a þ only and IFN-g þ TNF-a þ . Furthermore, CMV-specific CD4 þ T cells expressing both IFN-g and TNF-a expressed a higher level of these cytokines when compared with the CD4 þ T cells expressing IFN-g or TNF-a alone (unpublished observation). Immunology and Cell Biology
Expression of the transcription factors T-bet and Eomes in CMV-specific CD4 þ T cells correlates with differential cytokine expression patterns Studies carried out in murine models have demonstrated that T-box transcription factors T-bet and Eomes have a crucial role in regulating effector function(s) and the establishment of CD4 þ T-cell memory.8 However, the potential role of these transcription factors in modulating polyfunctional potentiality of antigen-specific T cells has remained unexplored. To explore the possibility that selective expression of IFN-g and/or TNF-a by antigen-specific CD4 þ T cells may be intrinsically linked to the expression of T-bet and Eomes, PBMCs from CMV-seropositive individuals were stimulated with IE-1, pp65 or gB protein and then assessed for the expression of these transcription factors in combination with TNF-a and/or IFN-g. Data presented in Figures 2a–d show that CMV-specific CD4 þ T cells exhibited three individual patterns of expression for T-bet and Eomes: T-bet Eomes , T-betLoEomesHi and T-betHiEomesHi/Lo (Figure 2a). CMV-specific CD4 þ T cells directed towards pp65, IE-1 and gB antigens expressing both IFN-g and TNF-a showed significantly higher levels of T-bet expression when compared with the T cells expressing TNF-a alone (Figures 2b–c). Interestingly, combined analyses of T-bet and Eomes showed discordance in the expression of these transcription factors between antigen-specific CD4 þ T cells expressing both IFN-g and TNF-a or TNF-a only. A significantly higher proportion of CMV-specific CD4 þ T cells expressing both IFN-g and TNF-a displayed a T-betHiEomesHi/Lo phenotype, whereas a significantly higher proportion of T cells expressing TNF-a þ only exhibited a T-bet Eomes phenotype. The magnitude of cells expressing the T-betLoEomesHi phenotype wasn’t significant between IFN-g and TNF-a or TNF-a only populations except for in response to IE-1 (Figures 2b–d). It is important to mention here that the T-betHiEomesHi/Lo phenotype of IFN-g þ TNF-a þ T cells was observed for all three different specificities (that is, IE-1, pp65 and gB). Thus two distinct profiles of T-bet/Eomes expression was observed, T-cells that displayed mainly T-bet Eomes phenotype and T-cells that expressed a predominately T-betHiEomesHi/Lo phenotype. CMV-specific CD4 þ T cells expressing IFN-g and/or TNF-a display differential cytolytic potentiality and IL-2 production To further delineate the impact of a discordant profile of T-box transcription factors on effector functions of CMV-specific CD4 þ T cells, we next assessed the expression of perforin, granzyme B and IL-2 in these effector cells. PBMCs from healthy seropositive donors were incubated with gB, pp65 and IE-1 proteins then assessed for expression of IFN-g, TNF-a, granzyme B, perforin and IL-2. Data in Figures 3a–c show the frequency of IFN-g þ TNF-a þ and TNF-a þ only CMV-specific CD4 þ T cells co-expressing perforin, granzyme B and IL-2. These analyses showed that although no significant difference was observed for granzyme B expression, a significantly higher proportion of IFN-g þ TNF-a þ CMV-specific CD4 þ T cells produced perforin and IL-2 when compared with the T cells expressing TNF-a only. Furthermore, analyses of the mean fluorescence intensity for perforin and IL-2 production showed that IFN-g þ TNF-a þ CMV-specific CD4 þ T cells produced significantly higher levels of both perforin and IL-2 when compared with the T cells expressing TNF-a only (Figures 3d–e). Taken together, these observations demonstrate that differential functional programming of CMV-specific CD4 þ T cells directed towards gB, pp65 and IE-1 correlates with transcriptional profile, cytolytic potentiality and IL-2 production.
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 183
a
Unstimulated 5
10
4
10
10
3
10
TNF-
pp65 10
10
2
5
31.90%
IE-1 5
10
65.19%
4
2
3
10
4
10
3
5
10
3
3
10
2
010
3
10
4
10
2
10 0
2.91%
79.51%
7.05%
4
10
2
10
10
4
10 0 0 10
gB 5
14.23%
10
2
10 0
84.26%
10 0
1.51%
5
2
10
0 10
3
10
4
10
5
10
13.40% 0 10
2
10
3
4
10
5
10
IFN-
Frequency cytokine producing CD4+T-cells (%)
b 100
IFN-+ alone
TNF -+ alone
ns
**
IFN-+TNF -+
*
80 60 40 20 0 pp65
IE-1
c
gB
IFN-+ alone
pp65
IE-1
TNF -+ alone
pp65
IE-1
gB
pp65
IE-1
gB
IFN-+TNF-+ gB
Figure 1 CMV-specific CD4 þ T cells directed towards pp65, IE-1 and gB antigens display differential cytokine expression potentiality. PBMCs from CMVseropositive individuals (n ¼ 10) were stimulated with pp65, IE-1 or gB protein overnight, stained with anti-CD4 and assessed for intracellular expression of IFN-g and TNF-a. Panel a shows representative analysis of IFN-g and/or TNF-a producing T cells elicited towards three different CMV antigens (n ¼ 1). Percentages represent the proportion of total cytokine-producing cells that express IFN-g, TNF-a or IFN-g and TNF-a. Panel b shows frequency of antigenspecific CD4 þ T cells directed towards three different CMV antigens expressing IFN-g, TNF-a or IFN-g and TNF-a. The mean precursor frequency is represented by a horizontal line in each graph. Panel c shows overall pattern of cytokine expression by T cells directed towards pp65, IE-1 and gB proteins. *Po0.05; **Po0.001; ns, not significant.
CMV-specific T cells expressing IFN-g and/or TNF-a display distinct maturation status On the basis of the data presented above, we hypothesized that differential functional programming of CMV-specific CD4 þ T cells may be linked to their maturation/differentiation status. To test this hypothesis, PBMCs from CMV-seropositive individuals were stimulated with viral proteins and then assessed for expression of IFN-g, TNF-a, CD27, CD62L, CD57, CD45RA and CCR7 (Figure 4a). Data presented in Figures 4b and c show that a significantly higher proportion of TNF-a þ only CMV-specific T cells were CCR7 þ , CD27 þ , CD45RA þ and CD62L þ when compared with the T cells expressing both IFN-g and TNF-a. In contrast, IFN-g þ TNF-a þ pp65-specific T cells were predominantly CD57, whereas no significant difference was observed for CD57 expression on IE-1-specific T cells expressing TNF alone or both IFN-g and TNF-a. These observations indicate that CMV-specific CD4 þ T cells expressing
TNF-a only were less differentiated when compared with the T cells expressing both IFN-g and TNF-a. Recent studies have described a unique population of human memory T cells which have been referred to as stem-cell memory like (TSCM) population.25,26 These cells have a superior self-renewal capacity as well as the potential to differentiate into central memory, effector memory and effector T cells.22,23 To determine whether the CMV-specific CD4 þ T cells displaying differential functional programming include TSCM population, these antigenspecific T cells were stained for the expression of CD95, CD45RA, CD27 and CCR7, as well as IFN-g and TNF-a (Figure 5a). These antigen-specific CD4 þ T cells were then categorized according to differential expression of these markers, into TSCM (CD95 þ CD45RA þ CCR7 þ CD27 þ ), early-differentiated (CD95 þ CD45RA þ / CCR7 þ CD27 þ ), intermediate (CD95 þ CD45RA CCR7 CD27 þ ) and late-differentiated cells (CD95 þ CD45RA þ / CCR7 CD27 ). Immunology and Cell Biology
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 184
b
T-bet- Eomes-
10
T-bet
Lo EomesHi
T-bet
Hi EomesHi/Lo
5
104
Eomes
103 102 0 0 102
103
104
Frequecy of cytokine producing CD4+ T-cells (%)
a
pp65 100 80
**
ns
**
60 40 20 0 TNF-+ alone
105
IFN-+TNF-+
T-bet
d
gB 100 80
**
ns
**
60 40 20 0
TNF-+ alone
IFN-+TNF-+
Frequecy of cytokine producing CD4+ T-cells (%)
Frequecy of cytokine producing CD4+ T-cells (%)
c
IE-1 100 80
**
**
**
60 40 20 0
TNF-+alone
IFN-+TNF-+
Figure 2 Expression of the T-box transcription factors T-bet and Eomes in CMV-specific CD4 þ T cells. PBMCs from CMV-seropositive healthy donors (n ¼ 10) were stimulated with pp65, IE-1 or gB protein overnight, stained with anti-CD4 and assessed for intracellular expression of IFN-g, TNF-a, T-bet and Eomes. Panel a shows representative analysis of T-bet and Eomes expression by CMV-specific CD4 þ T cells. IFN-g and/or TNF-a producing cells are shown as an overlay of total CD4 þ T cells. Panels b–d show frequency of TNF-a þ or IFN-g þ TNF-a þ pp65, gB and IE-1-specific T cells expressing different profiles of T-bet and Eomes (T-bet Eomes , T-betLoEomesHi or a T-betHiEomesHiLo). T-bet and Eomes profiles were compared between TNF-a þ or IFN-g þ TNF-a þ . **Po0.001; ns, not significant.
Data presented in Figure 5b show that pp65 or IE-1-specific T cells expressing TNF-a only predominantly displayed an early-differentiated phenotype with a small proportion of these cells expressing surface markers of TSCM population. In contrast, the majority of T cells expressing both IFN-g and TNF-a exhibited a phenotypic profile of late/intermediate-differentiated cells. Low-to-negligible proportion of these effector cells expressed early-differentiated or TSCM phenotype (Figure 5b). Taken together these analyses clearly demonstrate that CMV-specific CD4 þ T cells expressing TNF-a only are less differentiated when compared with the IFN-g þ TNF-a þ T cells. CMV-specific CD4 þ T cell displaying differential functional and phenotypic profile exhibit distinct levels of functional plasticity Expression of a differential phenotypic profile by CMV-specific T cells led us to hypothesize that antigen-specific CD4 þ T cells displaying an immature profile may be more amenable to interconversion to other phenotypic and functional profiles as a consequence of cytokine milieu alteration. To explore this possibility, PBMCs from CMVseropositive individuals were stimulated with pp65 or IE-1 in the absence or presence of different combinations of cytokines (IL-2, IL-4 and/or IL-12). These T cells were then assessed for the expression of IFN-g and TNF-a, to determine whether antigenic stimulation in the presence of specific cytokines altered their functional, phenotypic and transcriptional profile. Data presented in Figure 6a show that whilst the functional profile of pp65-specific CD4 þ T cells remained unaltered following in vitro stimulation, a dramatic change in IE-1specific CD4 þ T cells was observed. Following in vitro stimulation with IE-1 protein in combination with IL-2, IL-12, IL-2 þ IL-4 or Immunology and Cell Biology
IL-2 þ IL-12, the majority of T cells specific for this protein switched from TNF-a only to both IFN-g þ and TNF-a þ . No major alteration in the functional profile was observed following stimulation with IE-1 antigens and IL-4 (Figure 6a). Consistent with this altered functional profile, we also observed a dramatic change in the expression of T-box transcription factors. The majority of the IE-1-specific T cells switched from T-bet Eomes to a dominant T-betHiEomesHi/Lo expressing population (Figure 6b). This impact was particularly evident when these cells were cultured in the presence of IL-12 or IL-2 þ IL-12. Upon IL-12 or IL-2 þ IL-12 treatment, a statistical increase in the proportion of TbetLoEomesHi and TbetHiEomesHi/Lo cells was observed, with a concomitant significant decrease in Tbet Eomes cells. Furthermore, coculture of IE-1-specific T cells with IL-12 or IL-2 þ IL-12 increased the proportion of perforin-positive cells indicating that these cells were also capable of acquiring cytolytic function under appropriate stimulation conditions (Figure 6c). Finally, we also observed a change in the differentiation status following stimulation with viral antigen in the presence of IL-12 or IL-2 þ IL-12. The majority of the T cells acquired a late/intermediate phenotypic profile. These observations further emphasize an intrinsic link between cytokine expression, transcriptional regulation and cellular differentiation and their impact on functional plasticity of antigen-specific CD4 þ T cells. DISCUSSION The CD4 þ T-cell repertoire to an individual infectious agent such as CMV, is directed towards a diverse array of antigenic peptides and it has been shown that a large frequency of antigen-specific T cells is
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 185
Frequency of cytokine prodycing CD4+ T-cells (%)
a
Granzyme B ns
100
Perforin
IL-2
**
**
80 60 pp65 40 20 0 TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+
Frequency of cytokine producing CD4+ T-cells (%)
b
GranzymeB
100
ns
Perforin
IL-2
**
**
80 60 IE-1 40 20 0 TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+
c Frequency of cytokine producing CD4+ T-cells (%)
Granzyme B 100
Perforin
ns
IL-2 ns
*
80 60 gB 40 20 0 TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+ TNF-+ only TNF-+ IFN-+
d
e 4000
**
**
**
3000
IFN-+TNF-+
**
**
ns
pp65
IE-1 Protein
gB
3000 IL-2 MFI
Perforin MFI
TNF-+ alone 4000
2000
2000 1000
1000 0 pp65
IE-1 Protein
gB
0
Figure 3 Cytolytic potentiality and IL-2 production by CMV-specific CD4 þ T cells. PBMCs from CMV-seropositive donors (n ¼ 10) were incubated overnight with recombinant proteins to pp65, IE-1 or gB, in the presence of brefaldin A. Cells were then labelled with anti-CD4 and assessed for the intracellular expression of IFN-g, TNF-a, granzyme B, perforin and IL-2. Panels a–c show frequency of granzyme B þ , perforin þ and IL-2 þ CD4 þ T cells directed towards pp65, IE-1 and gB proteins. The mean value is represented by a horizontal line. Panels d and e show mean±SEM of the mean fluorescence intensity of perforin and IL-2 expression in CMV-specific CD4 þ T-cell populations with distinct cytokine profile (TNF-a þ or IFN-g þ TNF-a þ ). *Po0.05; **Po0.001; ns, not significant.
induced ready for secondary exposure to antigen.5,27–29 CD4 þ T-cell populations specific for certain proteins are preferentially expanded and comprises heterogeneous memory T cells.30,31 To date, the factors responsible for the generation of the functional and phenotypic
heterogeneity of T cells have yet to be elucidated. However, studies have suggested that antigen load and persistence,31–33 precursor frequency, pMHCII/TCR avidity and cytokine milieu have an impact on Th commitment and memory T cell generation. Immunology and Cell Biology
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 186
10
10
2
2
2
5
3
10
10
4
10
5
Frequency of cytokine producing CD4+ T-cells (%)
4
10
10
10
3
3
10
10
2
100
10
4
10
5
CCR7
3
4
10
5
10
5
0
3
10
4
10
0 0 10
5
10
5.8%
4
3
10
4
10
5
2
10
0 10 5
10
9.9%
4
3
4
10
10
5
10
0.72%
5
10
3
3
10
4
10
3
10
10
2
2
10
10
0
0 2
0 10
3
4
10
10
5
10
CD27
CD45RA
**
**
**
2
10 2
10
10 0
0 3
10
2
0
2
10 2
10
10 2
10
48.3%
3
10
0 10
50.0%
4
10
3
3
5
4
0 10
2
0 10
3
10
4
10
5
10
CD57
2
0 10
3
4
10
10
5
10
CD62L
**
**
80 60
pp65
40 20 0 TNF-+ alone
Frequency of cytokine producing CD4+ T-cells (%)
3
10
2
c
10
5
10 0
b
0
10
6.9%
4
10
2
10
3.3%
4
10
12.3%
4
10 0
0
CD4
10
10
10
0 10
10
10
3
3
10 0
10
10
58.3%
4
10
CD62L 5
+
10
4
CD57 5
TNF- alone
10
CD45RA 5
+
10
CD27 5
9.0%
+
CCR7 5
IFN- TNF-
a
100
IFN-+ TNF-+
CCR7
**
TNF-+ alone
IFN-+ TNF-+
TNF-α+ alone
IFN-+ TNF-+
CD27
CD45RA
**
**
TNF-α+ alone
IFN-+ TNF-+
CD57 ns
TNF-+ alone
IFN-+ TNF-+
CD62L
**
80 60 IE-1 40 20 0 TNF-+ alone
IFN-+ TNF-+
TNF-+ alone
IFN-+ TNF-+
TNF-+ alone
IFN-+ TNF-+
TNF-+ alone
IFN-+ TNF-+
TNF-+ alone
IFN-+ TNF-+
Figure 4 Differential phenotypic profile of CMV-specific CD4 þ T cells expressing TNF-a or IFN-g and TNF-a. PBMCs from CMV-seropositive donors (n ¼ 10) were incubated overnight with recombinant proteins to pp65, IE-1 or gB, stained with fluorescently labelled anti-CD4, CCR7, CD27, CD45RA, CD57 and CD62L followed by staining for the intracellular expression of IFN-g and/or TNF-a. Panel a shows representative analysis of cell surface marker expression by TNF-a þ and IFN-g þ TNF-a þ CMV-specific CD4 þ T cells. Cytokine-producing cells are represented as an overlay of total CD4 þ T cells. Panels b and c show comprehensive analysis of individual surface marker on pp65 and IE-1-specific CD4 þ T cells expressing TNF-a or IFN-g and TNF-a. Horizontal bars represent the mean frequency of CD4 þ CMV-specific cells expressing each individual marker. **Po0.001; ns, not significant.
Recently, the transcriptional regulation of Th cells has received unprecedented attention. In particular, the impact of T-box transcription factors (for example, T-bet and Eomes) on the commitment of cells to the Th1 lineage and cytokine expression patterns has been identified. T-bet expression in CD4 þ T cells has been shown to be required for commitment of these cells to the Th1 lineage, optimal IFN-g production1,8,34,35 and proinflammatory trafficking.36 Eomes expression has also been shown to invoke IFNg production even in the absence of T-bet. Thus, as for CD8 þ T cells, Immunology and Cell Biology
cooperative T-bet and Eomes expression may be required for full effector function.13 However, the full impact of Eomes expression on the functional and phenotypic profile of T cells, as well as the impact of cooperative transcription factor expression has yet to be elucidated. The finding of a CMV-specific CD4 þ T-cell population which exhibited predominant TNF-a only expression was unexpected, as previous studies by Widmann and colleagues had indicated that simultaneous induction of Th1 cytokines occurs in CMV-specific CD4 þ T cells.21 It is important to note that Widmann and colleagues
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 187 250K
150K
CD4
SSC-A
200K
100K
105
105
104
104 CD4
a
103
103
102 0
50K
0K
0 102
25
0K
0K
20
15
0K 10
50
0
K
0
102 0 103
104
105
0 102
CD8
103 104 CD95
105
105
105
104
104
104
103 102 0
IFN-γ
105
CCR7
CD27
FSC-A
103
102 0
102 0 0 102
103
104
105
0 102
CD45RA
Frequency of cytokine producing CD4+ T-cells (%)
b
103
104
CD45RA
**
103
105
0 102
103
104
105
TNF-
**
100
Late (CD95+CD45RA+/-CCR7-CD27-) Intermediate (CD95+CD45RA+CCR7-CD27+) Early (CD95+CD45RA-CCR7+CD27+) TSCM (CD95+CD45RA+CCR7+CD27+)
50
0
TNF-+ only
TNF-+ IFN-+
pp65
TNF-+ only
TNF-+ IFN-+ IE-1
CD4 þ
Figure 5 Frequencies of CMV-specific T cells displaying differential maturation status expressing TNF-a alone or IFN-g and TNF-a. PBMCs from CMV-seropositive donors were incubated overnight with recombinant pp65 or IE-1, then stained with fluorescently labelled anti-CD4, CD95, CD27, CCR7 and CD4RA and assessed for the intracellular expression of IFN-g and TNF-a. Panel a shows gating strategy for the identification of CMV-specific CD4 þ T cells expressing different maturation status that is, late (CD95 þ CD45RA þ / CCR7 CD27 ), intermediate (CD95 þ CD45RA þ CCR7 CD27 þ ), early (CD95 þ CD45RA CCR7 þ CD27 þ ) or TSCM (CD95 þ CD45RA þ CCR7 þ CD27 þ ). Panel b shows relative proportions of TNF-a þ and IFN-g þ TNF-a þ CMV-specific CD4 þ T cells directed towards pp65 or IE-1 exhibiting a late, intermediate, early or TSCM phenotype. TSCM frequency was compared between TNF-a þ only and TNF-a þ IFN-g þ . **Po0.001.
stimulated T cells with whole CMV lysate which would primarily activate T cells specific for immunodominant antigens including pp65 and gB with minimal activation of IE-1-specific T cells. In our study we have used pp65, gB and IE-1 proteins individually which allows dissection of cytokine expression pattern in T cells specifically directed to each of these antigens. Data presented in our study also show that the predominant expression of either TNF-a only or IFN-g and TNFa by CMV-specific CD4 þ T cells was coincident with dominant expansion of either T-bet Eomes or T-betHiEomesHi/Lo antigenspecific T cells, respectively. More importantly, the T-bet Eomes phenotype of antigen-specific CD4 þ T cells correlated with poor functional capacity (specifically, the production of perforin), as well as immature status when compared with antigen-specific T cells with T-betHiEomesHiLo profile. Previous studies have shown that T-bet and Eomes expression has a concomitant impact upon the generation of memory and effector CD8 þ T cells in both humans and mice.13,37–39 A recent study by Gattinoni et al., has also shown the ability of earlydifferentiated T cells known as TSCM to express cytokines including
TNF-a. Apart from the upregulation of CD95, there is no evidence of phenotypic differences in the expression of differentiation markers compared with naı¨ve T cells.25 Our results have shown that the presence of less-differentiated populations endows T cells with functional flexibility. This indicates remarkable heterogeneity of CMV-specific CD4 þ T cells which are amenable to modulation in antiviral immune responses. In CMV infection, periodic viral reactivation occurs, which makes T cells susceptible to functional exhaustion. These T cells lose the ability to control active virus replication and thus viral control wanes.40,41 However, data presented here show that heterogeneity exists with respect to cellular differentiation of CMV-specific T cells. Thus when moredifferentiated T cells become exhausted, the less-differentiated T cells have the propensity to further differentiate and acquire the ability to upregulate effector functions such as perforin production. This heterogeneity allows less-differentiated T cells to reinforce the immune responses control of viral infection, upon exhaustion of their more-differentiated counterparts. Immunology and Cell Biology
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 188
a
IFN-+ alone IL-2
IL-4
IFN-+TNF-+ alone
IL-12
IL-2+IL-4
IL-2+IL-12
IE-1
pp65
No cytokine
TNF-+ alone
b
T-bet-Eomes-
% Cytokine Positive
100
T-betLoEomesHi
ns
ns
ns
**
**
T-betHiEomesHi/Lo
*
**
**
**
80 60 40 20 0 No cytokine
IL-2
IL-2+IL-12
50
ns
*
**
40 30 20 10 no IL-2 cytokine
IL-12 IL-2+IL-12
Frequeccy of cytokine producing cells CD4+ T-cells (%)
d
c % Perforin+
IL-12
*
100
*
80 60 40 20 0 No IL-2 cytokine
IL-12 IL-2+IL-12
Late (CD95+CD45RA+/-CCR7-CD27-) Intermediate (CD95+CD45RA+CCR7-CD27+) Early (CD95+CD45RA-CCR7+CD27+) Figure 6 Impact of antigenic stimulation and inflammatory cytokines on the functional and transcriptional profile of CMV-specific T cells. PBMCs from CMVseropositive individuals were cultured with or without pp65 and IE-1 protein in the presence of IL-2, IL-4 and/or IL-12 or with no cytokine for 7 days. Following incubations, these cells were assessed for the expression of IFN-g, TNF-a, perforin, IL-2, T-bet and Eomes as outlined in the ‘Material and Methods’ section. Panel a shows relative proportion of pp65 and IE-1-specific CD4 þ T cells exhibiting an IFN-g þ alone, TNF-a þ alone or IFN-g þ TNF-a þ expression profile following antigenic stimulation in the absence or presence of IL-2, IL-4 and/or IL-12. Panels b and c show impact of antigenic stimulation in combination with IL-2 and/or IL-12 on the transcriptional profile (T-bet and Eomes) and perforin expression in IE-1-specific CD4 þ T cells. T-bet and Eomes profiles were compared between T cells treated with no cytokine and those treated with IL-2 and/or IL-12. *Po0.05; **Po0.001; ns, not significant. Panel d shows the alteration in the maturation status (early, intermediate or late) of IE-1-specific CD4 þ T cells following incubation with IL-2 and/or IL-12 and recombinant viral protein. For maturation status analysis, T cells following incubation with viral antigens and IL-2 and/or IL-12 were stained with fluorescently labelled anti-CD4, CD95, CD27, CCR7 and CD4RA antibodies and then analysed using FACS Canto II.
Previous studies have shown that cytokines (for example, IL-12) can modulate the expression of T-bet and Eomes, which could have a concomitant impact upon cellular differentiation status. It is important to note that newly activated mature CD8a þ dendritic cells secrete significant amounts of IL-12 upon murine CMV infection.42,43 Thus IL-12 may have a role in modulating expression of T-box transcription factors post CMV infection. In this study, in vitro exposure to IE-1 antigen in combination with IL-12 and/or IL-2 induced T-bet upregulation and graded expression of Eomes, which Immunology and Cell Biology
indicates that these cytokines have a role in Th1 programming of cells. IL-12 in particular directs upregulation of both IFN-g and perforin expression and transition to a more-differentiated phenotype. Thus production of IL-12 by infected dendritic cells, particularly during the priming phase, has a role in reinforcement of T-bet expression, differentiation and perforin acquisition. It is unclear whether this change in cytokine expression pattern is as a result of phenotypic changes in a select population of IE-1 cells or expansion of the IE-1 population as a whole. Further investigations will be required to
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 189
dissect the cellular basis of this observation. Nevertheless, the discordance of phenotypic and functional programming in different CMV-specific T-cell populations suggests that antigen load of infected cells has a pivotal role in generation of T cells with different levels of functional flexibility. Recent studies have shown that to maintain the effector T-cell pool for MCMV, continuous stimulation is required after latency has been established, thereby providing protection against viral reactivation.44,45 This would account for the expansion of intermediate to late memory T-cell populations with higher perforin expression concurrent with higher T-bet expression. Antigen density upon priming has been shown to influence generation of memory CD8 þ T cells.33 Therefore, we postulate that the different expression kinetics of the CMV proteins yield an explanation for why T cells specific for the different proteins are heterogeneous in their phenotypic and transcriptional programming. For example, IE-1 is expressed intermittently compared with pp65 and gB, which are expressed throughout the viral cycle.24 This intermittent antigen expression may suggest that antigen load is lower for this protein and as such is less available for encounter with T cells. Antigen density has already been shown to influence memory generation of CD8 þ T cells at priming, thereby explaining the relative immaturity of these cells compared with those specific for other antigens. Thus, persistent virus and cytokine signals cooperate to determine the phenotype and functionality of CMV-specific T cells. These observations have important implications for the modulation of immune responses in therapeutic interventions where a protective response to virus is required. Further investigation of the precise differences in antigen processing and presentation of different CMV proteins will provide an insight into why different antigens induce phenotypic and transcriptional variation in T-cell populations. In addition, the elucidation of the precise mechanism by which cytokines modulate T-cell phenotype will improve our understanding of how cytokine milieu variation can be used to modify antiviral immune responses.
METHODS Human volunteer recruitment A panel of healthy CMV-seropositive donors (range ¼ 28–59 years) were recruited for this study. This study was approved by the QIMR Human Ethics Committee. Informed consent was obtained from each donor according to QIMR Human Ethics Committee guidelines.
Intracellular Cytokine analysis of CMV-specific
CD4 þ
T cells
To assess cytokine expression by CMV-specific T-cells, PBMCs from CMVseropositive individuals were incubated for 2 h with 25 mg ml 1 CMV proteins pp65, gB or IE-1 and then incubated overnight in the presence of Brefaldin A. After overnight incubation, cells were incubated with live/dead fixable Aqua (Invitrogen, Mulgrave, VIC, Australia), Pacific Blue-labelled anti-CD4 (BD Biosciences, Franklin Lakes, NJ, USA) and Peridinin chlorophyll cyanin-Cy5.5 labelled anti-CD8 (BD Biosciences). Cells were then fixed and permeabilised using cytofix/cytoperm kit (BD Biosciences) and incubated with R-Phycoerythrin (PE)-Cy7 anti- IFN-g (BD Biosciences), Alexa Fluor-700 or Allophycocyanin-labelled anti-TNF-a (BD Biosciences), and combinations of the following antibodies, PE-labelled anti-IL-2 (BD Biosciences), fluorescein isothiocyanate or Alexa Fluor-700-labelled anti-granzyme B (BD Biosciences), anti-perforin PE (BD Biosciences). Acquisition of data from these cells was carried out using a FACSCanto with FACSDiva software (BD Biosciences, Palo Alto, CA, USA) and post acquisition analysis was performed using FlowJo software (Treestar, Ashland, OR, USA).
In vitro expansion of CMV-specific CD4 þ T cells To expand CMV-specific CD4 þ T cells, PBMCs from CMV-seropositive donors were initially incubated with human CD8 microbeads (Miltenyi Biotech GmbH, Bergisch Gladbach Germany), and separated on an AutoMACS column to deplete CD8 þ T cells according to the manufacturer’s guidelines (Miltenyi Biotech GmbH). These CD8 þ T cells depleted PBMCs were then incubated with 25 mg ml 1 recombinant CMV proteins (pp65, IE-1 or gB) for 2 h. Following incubation, these cells were cultured in RPMI-1640 medium supplemented with 10% fetal calf serum, recombinant IL-2 (10 IU ml 1) (R&D Systems, Minneapolis, MN, USA) and 30% MLA-144 supernatant. On day 14, antigen-specific CD4 þ T cells were assessed for their production of IFN-g and/or TNF-a using intracellular cytokine assay described above. In some experiments, CMV-specific CD4 þ T cells were also expanded as outlined above in the presence or absence of recombinant IL-2 (10 IU ml 1), recombinant IL-4 (10 ng ml 1) (R&D systems) and/or recombinant IL-12 (10 ng ml 1) (R&D systems).
Phenotypic analysis of CMV-specific CD4 þ T cells To assess ex vivo phenotypic profile of CMV-specific CD4 þ T cells directed towards IE-1, pp65 or gB protein, PBMCs were stimulated for 2 h with 25 mg ml 1 CMV protein alone, followed by addition of Brefaldin A for a further 14 h. Cells were then incubated with live/dead fixable Aqua, Pacific Blue-labelled anti-CD4, Peridinin chlorophyll cyanin-Cy5.5-labelled anti-CD8 and a combination of fluorescein isothiocyanate-labelled anti-CD45RA, Alexa Fluor-700-labelled antiCCR7, fluorescein isothiocyanate-labelled anti-CD57, PE-labelled anti-CD27 and Allophycocyanin-labelled anti-CD62L (all antibodies purchased from BD Biosciences). Cells were then fixed and permeabilised using the Cytofix/Cytoperm kit (BD Biosciences), and incubated with PE-Cy7-labelled IFN-g and Allophycocyanin or AF700-labelled TNF-a. Acquisition of data from these cells was carried out using a FACSCanto with FACSDiva software (BD Biosciences) and post acquisition analysis was performed using FlowJo software (Treestar).
Intracellular T-bet and Eomes analysis in CMV-specific CD4 þ T cells To assess the expression of the transcription factors T-bet and Eomes in CMVspecific CD4 þ T-cell populations, PBMCs from CMV-seropositive individuals were incubated for 2 h with 25 mg ml 1 CMV proteins (IE-1, pp65 or gB), followed by overnight incubation in the presence of Brefaldin A. After overnight incubation, cells were incubated with live/dead fixable Aqua (Invitrogen), Pacific Blue-labelled anti-CD4 and Peridinin chlorophyll cyanin-Cy5.5-labelled anti-CD8 (BD Biosciences), then fixed and permeabilised using the FoxP3 staining kit (eBioscience, San Diego, CA, USA). Cells were then incubated with PE-Cy7 anti- IFN-g (BD Biosciences), AF700-labelled anti-TNF-a (BD Bioscience), PE-labelled anti-T-bet and Allophycocyaninlabelled anti-Eomes (eBioscience). Acquisition of data from these cells was carried out using a FACSCanto with FACSDiva software (BD Biosciences) and post acquisition analysis was performed using FlowJo software (Treestar).
Statistical analysis All statistical analysis was undertaken using GraphPad Prism 5 Software (GraphPad Software Inc., La Jolla, CA, USA). A One-way analysis of variance followed by a post-hoc Tukey test was used to determine the significant difference observed between the mean cytokine expressions of experimental samples in Figure 1 and the T-bet and Eomes profiles in Figure 6. In addition, a non-parametric Wilcoxon test was used to determine the significance of differences between experimental samples for Figure 2 onwards. In all cases, differences where Po0.05 were considered to be significant.
CONFLICT OF INTEREST The authors declare no conflict of interest.
ACKNOWLEDGEMENTS We acknowledge the help of Linda Jones and Leone Beagley who assisted with sample processing. Thanks to Leesa Wockner for her statistical advice. This work was supported by the National Health and Medical Research Council (NH&MRC). RK is supported by NH&MRC Senior Principal Research Fellowship. Immunology and Cell Biology
Transcriptional programming of CD4 þ T cells ESJ Edwards et al 190 1 Szabo SJ, Sullivan BM, Peng SL, Glimcher LH. Molecular mechanisms regulating Th1 immune responses. Annu Rev Immunol 2003; 21: 713–758. 2 O’Shea JJ, Paul WE. Regulation of T(H)1 differentiation–controlling the controllers. Nat Immunol 2002; 3: 506–508. 3 Murphy KM, Reiner SL. The lineage decisions of helper T cells. Nat Rev Immunol 2002; 2: 933–944. 4 Marshall NB, Swain SL. Cytotoxic CD4 T cells in antiviral immunity. J Biomed Biotechnol 2011; 2011: 954602. 5 Crompton L, Khan N, Khanna R, Nayak L, Moss PA. CD4 þ T cells specific for glycoprotein B from cytomegalovirus exhibit extreme conservation of T-cell receptor usage between different individuals. Blood 2008; 111: 2053–2061. 6 Aslan N, Yurdaydin C, Wiegand J, Greten T, Ciner A, Meyer MF et al. Cytotoxic CD4 T cells in viral hepatitis. J Viral Hepat 2006; 13: 505–514. 7 Schulz EG, Mariani L, Radbruch A, Hofer T. Sequential polarization and imprinting of type 1 T helper lymphocytes by interferon-gamma and interleukin-12. Immunity 2009; 30: 673–683. 8 Szabo SJ, Kim ST, Costa GL, Zhang X, Fathman CG, Glimcher LH. A novel transcription factor, T-bet, directs Th1 lineage commitment. Cell 2000; 100: 655–669. 9 Szabo SJ, Sullivan BM, Stemmann C, Satoskar AR, Sleckman BP, Glimcher LH. Distinct effects of T-bet in TH1 lineage commitment and IFN-gamma production in CD4 and CD8 T cells. Science 2002; 295: 338–342. 10 Mullen AC, High FA, Hutchins AS, Lee HW, Villarino AV, Livingston DM et al. Role of T-bet in commitment of TH1 cells before IL-12-dependent selection. Science 2001; 292: 1907–1910. 11 Mullen AC, Hutchins AS, High FA, Lee HW, Sykes KJ, Chodosh LA et al. Hlx is induced by and genetically interacts with T-bet to promote heritable TH1 gene induction. Nat Immunol 2002; 3: 652–658. 12 Long M, Slaiby AM, Hagymasi AT, Mihalyo MA, Lichtler AC, Reiner SL et al. T-bet down-modulation in tolerized Th1 effector CD4 cells confers a TCR-distal signaling defect that selectively impairs IFN-gamma expression. J Immunol 2006; 176: 1036–1045. 13 Pearce EL, Mullen AC, Martins GA, Krawczyk CM, Hutchins AS, Zediak VP et al. Control of effector CD8 þ T cell function by the transcription factor Eomesodermin. Science 2003; 302: 1041–1043. 14 Yang Y, Xu J, Niu Y, Bromberg JS, Ding Y. T-bet and eomesodermin play critical roles in directing T cell differentiation to Th1 versus Th17. J Immunol 2008; 181: 8700–8710. 15 Elkington R, Shoukry NH, Walker S, Crough T, Fazou C, Kaur A et al. Cross-reactive recognition of human and primate cytomegalovirus sequences by human CD4 cytotoxic T lymphocytes specific for glycoprotein B and H. Eur J Immunol 2004; 34: 3216–3226. 16 Gamadia LE, Remmerswaal EB, Weel JF, Bemelman F, van Lier RA, Ten Berge IJ. Primary immune responses to human CMV: a critical role for IFN-gamma-producing CD4 þ T cells in protection against CMV disease. Blood 2003; 101: 2686–2692. 17 Gamadia LE, Rentenaar RJ, van Lier RA, Ten Berge IJ. Properties of CD4( þ ) T cells in human cytomegalovirus infection. Hum Immunol 2004; 65: 486–492. 18 Villacres MC, Longmate J, Auge C, Diamond DJ. Predominant type 1 CMV-specific memory T-helper response in humans: evidence for gender differences in cytokine secretion. Hum Immunol 2004; 65: 476–485. 19 Rentenaar RJ, Gamadia LE, van DerHoek N, van Diepen FN, Boom R, Weel JF et al. Development of virus-specific CD4( þ ) T cells during primary cytomegalovirus infection. J Clin Invest 2000; 105: 541–548. 20 Zhu J, Shearer GM, Marincola FM, Norman JE, Rott D, Zou JP et al. Discordant cellular and humoral immune responses to cytomegalovirus infection in healthy blood donors: existence of a Th1-type dominant response. Int Immunol 2001; 13: 785–790. 21 Widmann T, Sester U, Gartner BC, Schubert J, Pfreundschuh M, Kohler H et al. Levels of CMV specific CD4 T cells are dynamic and correlate with CMV viremia after allogeneic stem cell transplantation. PLoS One 2008; 3: e3634. 22 Sylwester AW, Mitchell BL, Edgar JB, Taormina C, Pelte C, Ruchti F et al. Broadly targeted human cytomegalovirus-specific CD4 þ and CD8 þ T cells dominate the memory compartments of exposed subjects. J Exp Med 2005; 202: 673–685. 23 Slezak SL, Bettinotti M, Selleri S, Adams S, Marincola FM, Stroncek DF. CMV pp65 and IE-1 T cell epitopes recognized by healthy subjects. J Transl Med 2007; 5: 17.
Immunology and Cell Biology
24 Stinski MF. Sequence of protein synthesis in cells infected by human cytomegalovirus: early and late virus-induced polypeptides. J Virol 1978; 26: 686–701. 25 Gattinoni L, Lugli E, Ji Y, Pos Z, Paulos CM, Quigley MF et al. A human memory T cell subset with stem cell-like properties. Nat Med 2011; 17: 1290–1297. 26 Lugli E, Gattinoni L, Roberto A, Mavilio D, Price DA, Restifo NP et al. Identification, isolation and in vitro expansion of human and nonhuman primate T stem cell memory cells. Nat Protoc 2013; 8: 33–42. 27 Crough T, Beagley L, Smith C, Jones L, Walker DG, Khanna R. Ex vivo functional analysis, expansion and adoptive transfer of cytomegalovirus-specific T-cells in patients with glioblastoma multiforme. Immunol Cell Biol 2012; 90: 872–880. 28 Burrows SR, Moss DJ, Khanna R. Understanding human T-cell-mediated immunoregulation through herpesviruses. Immunol Cell Biol 2011; 89: 352–358. 29 Khan N, Hislop A, Gudgeon N, Cobbold M, Khanna R, Nayak L et al. Herpesvirusspecific CD8 T cell immunity in old age: cytomegalovirus impairs the response to a coresident EBV infection. J Immunol 2004; 173: 7481–7489. 30 Verhoeven D, Teijaro JR, Farber DL. Heterogeneous memory T cells in antiviral immunity and immunopathology. Viral Immunol 2008; 21: 99–113. 31 Harari A, Vallelian F, Meylan PR, Pantaleo G. Functional heterogeneity of memory CD4 T cell responses in different conditions of antigen exposure and persistence. J Immunol 2005; 174: 1037–1045. 32 Vanguri V, Govern CC, Smith R, Huseby ES. Viral antigen density and confinement time regulate the reactivity pattern of CD4 T-cell responses to vaccinia virus infection. Proc Natl Acad Sci 2013; 110: 288–293. 33 Leignadier J, Labrecque N. Epitope density influences CD8 þ memory T cell differentiation. PLoS One 2010; 5: e13740. 34 Afkarian M, Sedy JR, Yang J, Jacobson NG, Cereb N, Yang SY et al. T-bet is a STAT1-induced regulator of IL-12R expression in naive CD4 þ T cells. Nat Immunol 2002; 3: 549–557. 35 Szabo SJ, Sullivan BM, Stemmann C, Satoskar AR, Sleckman BP, Glimcher LH. Distinct effects of T-bet in TH1 lineage commitment and IFN-g production in CD4 and CD8 T cells. Science 2002; 295: 338–342. 36 Lord GM, Rao RM, Choe H, Sullivan BM, Lichtman AH, Luscinskas FW et al. T-bet is required for optimal proinflammatory CD4 þ T-cell trafficking. Blood 2005; 106: 3432–3439. 37 Smith C, Elhassen D, Gras S, Wynn KK, Dasari V, Tellam J et al. Endogenous antigen presentation impacts on T-box transcription factor expression and functional maturation of CD8 þ T cells. Blood 2012; 120: 3237–3245. 38 Intlekofer AM, Takemoto N, Wherry EJ, Longworth SA, Northrup JT, Palanivel VR et al. Effector and memory CD8 þ T cell fate coupled by T-bet and eomesodermin. Nat Immunol 2005; 6: 1236–1244. 39 Intlekofer AM, Takemoto N, Kao C, Banerjee A, Schambach F, Northrop JK et al. Requirement for T-bet in the aberrant differentiation of unhelped memory CD8 þ T cells. J Exp Med 2007; 204: 2015–2021. 40 Jelley-Gibbs DM, Dibble JP, Filipson S, Haynes L, Kemp RA, Swain SL. Repeated stimulation of CD4 effector T cells can limit their protective function. J Exp Med 2005; 201: 1101–1112. 41 Antoine P, Olislagers V, Huygens A, Lecomte S, Liesnard C, Donner C et al. Functional exhaustion of CD4 þ T lymphocytes during primary cytomegalovirus infection. J Immunol 2012; 189: 2665–2672. 42 Orange JS, Biron CA. Characterization of early IL-12, IFN-alphabeta, and TNF effects on antiviral state and NK cell responses during murine cytomegalovirus infection. J Immunol 1996; 156: 4746–4756. 43 Ruzek MC, Miller AH, Opal SM, Pearce BD, Biron CA. Characterization of early cytokine responses and an interleukin (IL)-6-dependent pathway of endogenous glucocorticoid induction during murine cytomegalovirus infection. J Exp Med 1997; 185: 1185–1192. 44 Seckert CK, Schader SI, Ebert S, Thomas D, Freitag K, Renzaho A et al. Antigenpresenting cells of haematopoietic origin prime cytomegalovirus-specific CD8 T-cells but are not sufficient for driving memory inflation during viral latency. J Gen Virol 2011; 92: 1994–2005. 45 Torti N, Walton SM, Brocker T, Rulicke T, Oxenius A. Non-hematopoietic cells in lymph nodes drive memory CD8 T cell inflation during murine cytomegalovirus infection. PLoS Pathog 2011; 7: e1002313.
