ARTICLE
Received 3 Dec 2014 | Accepted 1 Apr 2015 | Published 8 May 2015
Valrie Martinet1,*, Sandrine Tonon1,*, David Torres1, Abdulkader Azouz1, Muriel Nguyen1, Arnaud Kohler1, Vronique Flamand1, Chai-An Mao2, William H. Klein2, Oberdan Leo1 & Stanislas Goriely1
CD8 T-cell memory phenotype and function are acquired after antigen-driven activation. Memory-like cells may also arise in absence of antigenic exposure in the thymus or in the periphery. Eomesodermin (Eomes) is a key transcription factor for the development of these unconventional memory cells. Herein, we show that type I interferon signalling in CD8
T cells directly activates Eomes gene expression. Consistent with this observation, the phenotype, function and age-dependent expansion of virtual memory CD8 T cells are strongly affected in absence of type I interferon signalling. In addition, type I interferons induce a sustained expansion of virtual memory CD8 T cells in an Eomes-dependent fashion. We further show that the development of innate thymic CD8 T cells is dependent on the same pathway. In conclusion, we demonstrate that type I interferon signalling in
CD8 Tcells drives Eomes expression and thereby regulates the function and homeostasis of memory-like CD8 T cells.
DOI: 10.1038/ncomms8089 OPEN
Type I interferons regulate eomesodermin expression and the development of unconventional memory CD8 T cells
1 WELBIO and Institute for Medical Immunology (IMI), Universit Libre de Bruxelles, Gosselies 6041, Belgium. 2 Department of Systems Biology, University of Texas, MD Anderson Cancer Center, Houston, Texas 77030, USA. * These authors contributed equally to this work. Correspondence and requests for materials should be addressed to S.G. (email: mailto:[email protected]
Web End [email protected] ).
NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 1
& 2015 Macmillan Publishers Limited. All rights reserved.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089
CD8 T cells are important effectors of the immune response against tumours, viruses and other intracellular pathogens. During infection or vaccination, CD8 T cells
undergo antigen-specic activation and expansion to give rise to cellular progeny, acquiring effector functions for pathogen clearance. The pool of activated CD8 T cells then undergoes a contraction phase, leaving behind a small fraction of memory cells that contributes to antigen-specic life-long protection1,2. In absence of antigen exposure, CD8 T cells may also acquire a memory phenotype in the thymus (innate-like CD8 T cells)3,4 or in the periphery (virtual memory (VM) cells)5,6. Recent evidences indicate that conventional and unconventional memory CD8 T-cell subsets promptly secrete large amounts of cytokines in response to inammatory cues in the context of infection7,8.
This non-cognate activation of memory CD8 T cells that leads to rapid interferon (IFN)g production and acquisition of cytolytic functions contributes to the rst line of defence and favours a Th1-prone environment6,7,911.
The transcriptional networks implicated in the alternative differentiation of memory-phenotype CD8 T cells are poorly understood. In these subpopulations, Eomesodermin (Eomes), a transcription factor closely related to T-bet, appears to play a central role in the acquisition of memory phenotype and function1214. In conventional memory cells, Eomes favours the development of central memory cells (TCM) characterized by longer survival and an important potential for homeostatic proliferation15,16. However, in the context of chronic viral infection, Eomes is also important for the terminal differentiation of virus-specic CD8 T cells in response to persisting antigen17. In different mice models that give rise to innate-like CD8 T cells, interleukin (IL)-4-dependent Eomes induction within CD8 single-positive (SP) thymocytes is required for their differentiation12,14,18,19. The development of VM CD8
T cells in the periphery also relies on high Eomes expression that mediates CD122 expression and responsiveness to IL-15 trans-presentation by CD8a dendritic cells13. Despite the important role of Eomes in these contexts, the signalling pathways responsible for its sustained expression in memory CD8 T cells are still ill-dened.
Type I IFNs display important direct and indirect immunomodulatory effects on CD8 T cells20,21. They promote the expression of specic cytokines by antigen-presenting cells (APCs) such as IL-15 or IL-27, which play a critical role in CD8 T-cell activation or differentiation2225. Similar to IL-12, they act as a third signal that promotes full activation, proliferation and survival of CD8 T cells activated by T cell receptor and costimulatory molecules21,26. In contrast, several studies showed that type I IFNs generally inhibit CD8 T-cell proliferation by increasing their sensitivity to apoptosis2729.
These mediators also induce the rapid acquisition of effector functions in absence of antigenic stimulation both in naive and memory cells30,31. Type I IFNs activate multiple signal transducer and activator of transcription (STAT) molecules, including STAT1, STAT3 homo/heterodimers and the IFN-stimulated gene factor 3 (ISGF3) complex composed of STAT1, STAT2 and IFN regulatory factor (IRF) 9 (ref. 21).
In the present work, we demonstrate that type I IFNs induce direct Eomes gene expression through activation of the ISGF3 complex within CD8 T cells. We further show that this pathway contributes to the homeostasis and innate functions of memory-like CD8 T cells both in the periphery and in the thymus.
ResultsReduced pool of VM CD8 T cells in IFNAR / mice. Type I
IFNs are known to regulate immune cell homeostasis through their ability to affect cellular proliferation and survival20. In an
initial set of experiments, we analysed the relative frequency of CD8 T-cell subpopulations in naive mice lacking type I IFN receptor (IFNAR / mice). We observed that the pool of memory CD44 CD62LCD8 T cells was reduced in these mice as compared with age-matched wild-type (WT) animals (Fig. 1a). Recent data indicate that a signicant proportion of memory CD8 T cells in the periphery of naive animals are not antigen-experienced true memory (TM) cells but so-called VM
cells that express low levels of CD49d, an integrin alpha subunit5,13,32. We observed that the reduced proportion of memory cells in IFNAR / mice was due to lower frequency and absolute counts of VM cells (Fig. 1a,b). The pool of TM cells was not consistently altered in IFNAR / mice (Fig. 1b). As
Eomes plays a critical role in the development of this particular memory subset13,33, we evaluated its expression in VM cells from both strains. We observed a signicant decrease in Eomes expression in absence of type I IFN signalling (Fig. 1c).
Type I IFNs directly activate Eomes expression in CD8 T cells. On the basis of these initial results, we postulated that type I IFN signalling could play a role in the regulation of VM CD8 T-cell homeostasis possibly by regulating Eomes expression. To test this hypothesis, we incubated isolated CD8 T cells with recombinant IFNb. As a positive control, we used rIL-4 that activates Eomes expression in a STAT6-dependent fashion18,34,35. In these experimental conditions, rIFNb alone or in combination with rIL-4 was a potent inducer of Eomes expression both at the messenger RNA (mRNA) and the protein levels (Fig. 2a). This effect was observed both in naive and CD44 CD62LCD49dlow (VM) or CD49dhigh (central memory) subsets. As previously shown15, Eomes expression in CD44 CD62L effector cells was lower than in the other memory subsets and was not upregulated in response to rIFNb (Fig. 2a).
Type I IFNs signal through the formation of the ISGF3 complex (STAT1/2/IRF9). We assessed the specic role of this pathway by using CD8 T cells from IRF9 / and STAT1 /
mice. Induction of Eomes by rIFNb was found to be IRF9- and STAT1-dependent (Fig. 2b). T-bet expression was not signicantly modulated in these conditions, indicating that type I IFN signalling differentially regulates these two related transcription factors. We further conrmed the role of IRF9 in type I IFN-induced Eomes mRNA upregulation both in naive and CD44 CD62L memory subsets (Fig. 2c).
Previous studies indicated that type I IFNs prime CD8 T-cell effector function by indirectly targeting APCs31. To assess a possible CD8 T-cell extrinsic role of IRF9 signalling on Eomes expression, we performed co-culture in vitro experiments. For this purpose, Thy1.2 CD8 T cells puried from WT or
IRF9 / mice were cultured with congenic WT Thy1.1 splenocytes for 16 h with or without rIFNb. Eomes expression was analysed by ow cytometry (Fig. 2d). We observed that IRF9 / CD8 T cells exhibited impaired expression of Eomes in response to rIFNb even in presence of WT splenocytes. Next, we analysed Eomes expression upon in vivo induction of type I IFNs by injection of polyI:C. To study the CD8 T-cell intrinsic role of IRF9, we performed adoptive transfers of WT or IRF9 /
Thy1.2 CD8 T cells into congenic Thy1.1 WT mice before polyI:C injection. Sixteen hours after injection, Eomes expression was analysed by ow cytometry in spleen cells. We observed a clear upregulation of Eomes upon polyI:C injection in WT but not in IRF9-decient CD8 T cells (Fig. 2e). These experiments indicate that type I IFN signalling within CD8 T cells leads to the activation of Eomes both in vitro and in vivo.
Next, we looked at classical histone modications in the proximal promoter region of Eomes (Fig. 2f). We found an
2 NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
& 2015 Macmillan Publishers Limited. All rights reserved.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089 ARTICLE
IRF9-dependent increase in H3K4me3 and H3K9Ac in response to rIFNb, highly suggestive of transcriptional activation. We identied a putative IFN-stimulated response element site located 150 base pairs upstream of the transcriptional start site. We observed direct recruitment of STAT1 to this region in chromatin immunoprecipitation experiments (Fig. 2g). In contrast, STAT1 recruitment to a control region located downstream of the gene body did not increase above the background levels. Taken together, these experiments indicate that type I IFNs directly activate Eomes gene expression in CD8 T cells in a STAT1/
IRF9-dependent fashion.
IRF9 regulates the homeostasis of memory CD8 T cells. Our results indicate that IRF9 is critical for type I IFN-induced Eomes expression. We therefore analysed CD8 T-cell memory subsets in IRF9 / mice. As observed in IFNAR / mice (Fig. 1), we found that the frequency and absolute counts of VM cells was signicantly lower in naive IRF9 / mice as compared with WT animals (Fig. 3a).
In agreement with recent reports32,36, we observed an important expansion of circulating memory CD44 CD49dlo
CD8 T cells in old (410 months) WT mice as compared with younger (68 weeks) mice (Fig. 3b). Interestingly, the
WT
18.3%
0.8%
IFNAR/
104
103
104
103
102
101
0
104
103
102
101
0
7.5%
0.5%
1%
102
CD44highCD49dlo (VM) CD8+ T cells
CD62L
CD49d
20 Blood **
101
0
0 103
102
101
102
103
0
101
CD44
CD44
104
103
102
101
0
2%
16%
6.6%
WT IFNAR/
0
101
102
103
0
101
102
103
Naive CD8+ T cells CD44highCD49dhighCD8+
T cells (TM)
CD44highCD49dloCD8+ T cells (VM)
15 1.0
0.5
0.0 0
Spleen Spleen Spleen
4
NS
***
% Of max Abs. numbers (10 6 cells)
Abs. numbers (106cells)
Abs. numbers (106 cells)
NS
10
5
0
200
2
WT
IFNAR/
Eomes in VM CD8+ T cells Eomes
*
Eomes MFI
150
100
50
0
100
80
60
40
20
0
0 101 102 103
WT IFNAR/
WT IFNAR/ WT IFNAR/
Figure 1 | The pool of virtual memory CD8 T cells is decreased in IFNAR / mice. (a) Total circulating CD8 T cells were analysed ex vivo for the frequency of memory CD44CD62LCD8 T cells, virtual memory CD44 CD49dlo CD8 T cells (VM) and true memory CD44 CD49dhi CD8
Tcells (TM) in WTand IFNAR / mice. (b) Absolute numbers of naive, VM and TM CD8 Tcells in the spleens of WT and IFNAR / mice. (c) Mean uorescence intensity (MFI) of Eomes in VM CD8 T cells. Representative histograms of Eomes expression in WT (blue line) and IFNAR / (red line)
VM CD8 T cells are shown. As a control, we included staining in CD8 T cells from Eomes/CD4cre mice (plain histogram). *Po0.05, **Po0.01 and ***Po0.001, NS: not signicant (non-parametric MannWhitney). In graphs, each dot represents an individual mouse and bars represent means.e.m.
NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 3
& 2015 Macmillan Publishers Limited. All rights reserved.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089
age-associated expansion of the VM compartment was totally abrogated in absence of IRF9 (Fig. 3b). In contrast, the frequency and age-dependent expansion of CD44 CD49dhigh TM CD8
T cells was not affected by the lack of IRF9 (Fig. 3b).
Eomes is strongly implicated in the development of VM cells13,14. As compared with antigen-experienced (CD49dhigh) memory cells, Eomes was found to be highly expressed in circulating VM cells from WT old animals (Fig. 3c).
Eomes mRNA
Eomes expression Eomes expression
Medium
Copies per 106 actin mRNA
3.0105
2.0105
* *
*
MFI
150
100
50
0
MFI
150
100
50
0
IFN
1.0105
NS
0 IFN
IL-4
IFN
IL-4
Naive
CD8+ T-cell subpopulations
TCM TEM/Eff VM
+ +
+
+
+
+
+
+
Eomes mRNA
CD44CD62L+CD8+ T cells Naive
CD44CD62L+CD8+ T cells Memory
4.0107
3.0107
2.0107
1.0107
8.0107
6.0107
4.0107
2.0107
WT IRF9/
0.2% 0.3% 0%
0.4%
11% 10%
13.5%
103
103
102
101
0
Copies per 106 actin mRNA
Copies per 106 actin mRNA
WT IRF9/ STAT1/
10%
103
102
101
0
103
102
101
0
102
Medium
101
0
0
49%
101
102
103
101
0 103
102
101
0
103
102
103
0 Medium 3 h 6 h 16 h
0 Medium 3 h 6h 16 h
0.8%
19%
2.5%
103
102
101
0
103
102
101
0
IFN
Eomes
0
101
102
103
0
101
102
103
0
101
102
Tbet
IFN
Eomes expression
Eomes mRNA
100
80
60
40
20
150
100
MFI
*
MFI
50
0.5
0 Medium Medium
IFN
IFN
WT CD8+ T cells +WT splenocytes
IRF9/ CD8+ T cells
+ WT splenocytes
0 PBS PBS
PolyI:C PolyI:C
CD8+Thy1.2+ WT T cells
CD8+Thy1.2+ IRF9/ T cells
H3K4me3 (Eomes promoter) H3K9ac (Eomes promoter)
H3K4me3 (Gapdh promoter) H3K9ac (Gapdh promoter)
+14.6 kb region
0.5
0.4
0.3
0.2
WT
IRF9/
WT IRF9/
0.3
0.2
WT
IRF9/
0.3
0.2
WT IRF9/
% Input
% Input
0.4
0.3
0.2
% Input
0.1
0.0
% Input
0.1
0.0
% Input
0.1
0.0
0.1
0.0
+ +
IFN (4 h)
+ +
IFN (4 h)
IFN (4 h)
Promoter region
0.05 *
*
0.04
0.03
0.02
0.01
0.00 Medium 45 90
0.05 IgG
STAT1
% Input
0.04
0.03
0.02
0.01
0.00
Medium 45 90
IFN
+ +
IFN (4 h)
+ +
IFN
4 NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
& 2015 Macmillan Publishers Limited. All rights reserved.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089 ARTICLE
Its expression was signicantly decreased in age-matched IRF9 / animals (Fig. 3c) both in VM and TM CD8
T cells. VM cells are characterized by high levels of CXCR3 and CD122, which are direct or indirect targets of Eomes3740. We found that expression of these two surface molecules was signicantly decreased in IRF9 / VM cells (Fig. 3c). We observed that CD127 was highly expressed in VM as compared with TM CD8 T cells. CD127 expression was also signicantly decreased in VM cells from IRF9 / old animals (Fig. 3c).
To gain further insight into the mechanisms responsible for decreased frequency of VM cells in IRF9 / mice, we analysed
Ki67 and Bcl2 expression within these subsets. Interestingly, we found that the expression of these molecules also strongly differs between VM and TM CD8 T cells. While Bcl2 was highly expressed in VM CD8 T cells, Ki67 levels were higher in TM cells. We observed that expression of both molecules was reduced in IRF9 / cells, indicating lower proliferation rate and an increase in their sensitivity to apoptosis (Fig. 3c). Taken together, these results indicate that VM cells display a unique phenotype that favours their responsiveness towards homeostatic cytokines such as IL-7 and IL-15. Our results further suggest that IRF9-dependent signals are critical for their general tness.
Type I IFNs regulate innate IFNc production by CD8 T cells. To determine the innate functions of these memory CD8
T cells in steady-state conditions, we stimulated splenocytes from naive 68-week-old WT and IRF9 / animals with rIL12
rIL18, a combination known to trigger rapid IFNg production by memory CD8 T cells in a non-cognate fashion9,22. IFNg production in these conditions was restricted to Eomeshigh memory CD8 T cells (Fig. 4a) and was found to be decreased in IRF9 / CD8 T cells (Fig. 4b). This was not only the consequence of the reduced memory pool, as IRF9 / CD44
CD8 T cells displayed lower IFNg production as compared with their WT counterparts (Fig. 4b).
To further assess the functional implications of our observations in the acquisition of innate effector functions in the context of acute Listeria monocytogenes (LM) infection, we analysed IFNg expression in WT and IRF9 / mice 16 h post infection.
Listeria-elicited IFNg production by CD8 T cells was strongly decreased in IRF9 / mice (Fig. 4c). In contrast, IFNg production by natural killer (NK) cells was not affected in
Listeria-infected IRF9 / mice, indicating that the inamma-tory cues that induce IFNg production were globally maintained in these mice (Fig. 4c).
To dene whether IRF9 acts in a T-cell intrinsic fashion, we performed mixed bone marrow chimera by transferring WT Thy1.1 and IRF9 / Thy1.2 myeloid progenitors into an irradiated RAG2 / host. Ten weeks after reconstitution, we observed that the proportion of memory CD8 T cells was reduced among the IRF9 / CD8 T cells as compared with WT CD8 T cells. This was reected by a decrease in the VM subset and reduced Eomes expression within memory cells, demonstrating the T-cell intrinsic role of IRF9 in our observations (Fig. 4d). Upon ex vivo rIL-12 rIL-18 stimulation,
we observed that IRF9 / cells produced lower IFNg levels. As IRF9 could play a T-cell intrinsic role independently of type I IFN signalling, we performed similar mixed bone marrow chimera with WT and IFNAR / cells (Fig. 4e). We conrmed our ndings in this experimental setting, showing that type I IFN signalling within CD8 T cells contributes to the homeostasis of VM CD8 T cells and their innate functions.
PolyI:C injection increases the pool of VM CD8 T cells. Our results clearly demonstrate a role of IRF9 in the homeostasis of
VM cells in steady state and upon aging. Injection of polyI:C induced an important upregulation of Eomes expression in CD8 T cells 16 h after injection (Fig. 2e). PolyI:C was previously shown to induce the expansion of CD8 memory T cells in a type I IFN- and IL-15-dependent manner25,41. Three days after injection, both VM and TM CD8 T-cell subsets expanded signicantly (Fig. 5a). This was reected by a strong increase in their proliferation rate (Fig. 5a). In contrast to in vitro mRNA data (Fig. 2c), showing transient induction of Eomes by IFNb, we observed that in vivo, increased Eomes protein expression in naive or memory subsets was maintained 3 days after polyI:C injection (Fig. 5a). To study the long-term effects of this stimulation, we analysed VM and TM cell subsets 1 month after polyI:C injection. The expansion of memory CD8 T cells was still observed at this later time point in the VM compartment but not in the TM compartment (Fig. 5b). These short- and long-term effects on VM cells were abrogated in IFNAR /
mice conrming the role of type I IFNs in this process (Fig. 5c). To address the role of Eomes in the effects of type I IFNs on the homeostasis of memory cells, we next used Eomes/CD4Cre mice. PolyI:C-induced expansion of VM CD8 T cells did not occur in these mice (Fig. 5c). Taken together, these experiments support the involvement of a type I IFN/Eomes axis in the homeostasis of VM cells.
Figure 2 | Type I IFNs directly activate Eomes gene expression in CD8 T cells. (a) Puried CD8 Tcells were incubated in medium alone or stimulated with rIFNb (100 U ml 1) and/or rIL-4 (20 ng ml 1). After 6 h of stimulation, Eomes mRNA levels were quantied and normalized against b-actin mRNA levels. Histograms represent means.e.m. of triplicates (representative of three independent experiments). Eomes expression (mean uorescence intensity (MFI)) in total or the indicated subpopulation of CD8 Tcells was assessed by ow cytometry after 16 h of stimulation. (b) Splenocytes from WT,
IRF9 / or STAT1 / mice were cultured for 16 h with or without rIFNb (100 U ml 1) in presence of rIL-2 (100 U ml 1). Eomes expression in total CD8 T cells was assessed by ow cytometry. (c) Naive CD44 CD8 T cells and CD44 CD62L memory CD8 T cells were sorted and cultured for the indicated time with or without rIFNb (100 U ml 1). Eomes mRNA levels were normalized against b-actin mRNA levels (experiment performed in triplicates). (d) WT or IRF9 / Thy1.2 CD8 T cells were cultured with WT Thy1.1 splenocytes for 16 h with or without rIFNb (100 U ml 1).
Eomes expression in WT or IRF9 / Thy1.2 CD8 T cells is shown. Means.e.m. of two independent experiments is shown. (e) WT or IRF9 / Thy1.2 CD8 T cells were adoptively transferred i.v. into WT Thy1.1 mice 24 h before PBS or polyI:C (200 mg per mouse) injection (intraperitoneally).
Mice were killed 16 h post injection. Eomes expression within the Thy1.2 CD8 T-cell population was analysed by ow cytometry. Histograms represent means.e.m. (ve mice per group). (f) Total CD8 T cells from WT and IRF9 / mice were stimulated for 4 h with IFNb or medium alone. H3K4me3 and H3K9Ac modications in the proximal promoter region of the Eomes or Gapdh loci were quantied by quantitative (q)PCR after chromatin immunoprecipitation (ChIP) experiments (normalization against input controls). Values are expressed as means.e.m. of duplicates and are representative of two independent experiments. (g) Total CD8 T cells from WT mice were incubated with rIFNb or medium alone for the indicated time. STAT1 recruitment to the proximal region or to a control region located 14.6 kb downstream of the transcriptional start site of the Eomes gene was quantied
by qPCR after ChIP experiments (normalization against input controls). Values represent means.e.m. of three independent experiments. *Po0.05 (non-parametric MannWhitney).
NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 5
& 2015 Macmillan Publishers Limited. All rights reserved.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089
CD44highCD49dlo (VM) CD8+ T cells
CD44highCD49dlo (VM) CD8+ T cells
Blood Blood
***
*** *** NS
CD44highCD49dlo (VM) CD8+ T cells
CD44highCD49dhigh (TM) CD8+ T cells
2.5
2.0
1.5
1.0
0.5
0.0
Spleen
10 Spleen **
VM CD8+ T cells (106 )
**
%
WT
IRF9/
IRF9/
WT
IRF9/
WT
IRF9/
103
0
103
0
60
40
20
60
40
20
**
102
102
Young
Old
101
101
*
0
10%
32.8%
0
6.9%
0 101 102 103
0
101 102 103
%
%
103
103
102
CD49d
102
101
101
0 Young Old Young Old
IRF9/
CD44
0 Young Old Young Old
7.3%
WT
WT
0 101 102 103 0 101 102 103
Eomes
CXCR3
Bcl2 Ki67+ cells
CD122
400
300
200
100
250
200
150
100
50
***
*
***
*** NS NS
MFI
MFI
MFI
MFI
30
20
10
0 WT WT IRF9/
IRF9/
0 WT WT IRF9/
IRF9/
VM TM
0 WT WT IRF9/
IRF9/
VM TM
VM TM
VM TM
VM TM
50
30
20
10
150
100
50
CD127
*** *** **
*
NS NS
MFI
50
30
20
10
40
40
%
IRF9/
Figure 3 | IRF9 regulates the homeostasis of memory CD8 T cells. (a) Frequency and absolute numbers of splenic VM CD8 T cells in 68-week-old WTand IRF9 / mice (representative groups of three independent experiments). (b) Frequency of circulating CD44 CD49dlo VM and CD44 CD49dhi
TM CD8 Tcells in young (610 weeks) and aged (410 month) WTand IRF9 / mice. Representative dot plots (out of at least seven mice) and graphs are shown. (c) Expression of Eomes, CD122, CD62L, CxCR3, CD127, Bcl2 (mean uorescence intensity (MFI)) and Ki67 (%) among CD44 CD49dlo and CD44CD49dhi CD8 T cells from age-matched WT and IRF9 / old mice. For all graphs, each dot represents an individual mouse. Bars represent means.e.m. *Po0.05, **Po0.01 and ***Po0.001, NS: not signicant.
0 WT WT IRF9/
IRF9/
0 WT WT IRF9/
IRF9/
0 WT WT IRF9/
VM TM
6 NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
& 2015 Macmillan Publishers Limited. All rights reserved.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089 ARTICLE
WT IRF9/
0%
103
0%
103
IFN (CD8+T cells)
IFN (CD44+CD8+ T cells)
102
102
10
50
IL12+IL18
Control
101
101
*
***
***
40
0
0
8
6
30
0
101
102
103 0 101 102 103
%
%
103
5,7% 103
1,5%
*
20
4
2
0 0
10
102
102
IL12+IL18
IL12+IL18
IFN
101
101
0
0
+ + WT WT
IRF9/
IRF9/
0
101
102
103 0 101 102 103
Eomes
IFN (CD8+ T cells)
IFN (CD44+CD8+ T cells)
10
40
IFN (NK cells)
ns
***
8
***
30
80
60
40
6
%
%
20
%
4
2
10
20
0
0
Infection
WT IRF9/
+ +
Infection
WT IRF9/
+ +
0 Infection
WT IRF9/
+ +
CD44highCD49dlo CD8+ T cells
(VM) Eomes
250 60
40
20
200
150
MFI
IFN (CD8+ T cells)
**
40
**
*
30
%
20
%
10
100
50
0 Thy1.1 (WT)
Thy1.2 (IRF9/)
0 Thy1.1 (WT)
Thy1.2 (IRF9/)
0 Thy1.1 (WT)
Thy1.2 (IRF9/)
CD44highCD49dlo CD8+ T cells (VM)
5
Eomes
IFN (CD8+ T cells)
20 **
300 ** **
20
15
15
200
%
10
MFI
%
10
100
5
0 0
CD45.1 (WT)
CD45.2 (IFNAR/)
CD45.1 (WT)
CD45.2 (IFNAR/)
0 CD45.1 (WT)
CD45.2 (IFNAR/)
Figure 4 | Type I IFNs regulate innate IFNc production by CD8 T cells. (a,b) Splenocytes from WT and IRF9 / mice were cultured for 16 h with or without IL-12 IL-18. Brefeldin A (5 mg ml 1) was added for the last 3 h of culture. (a) Dot plots of IFNg and Eomes expression among total CD8 T cells
from WT and IRF9 / mice are illustrated. (b) Frequency of IFNg cells among total and memory CD44 CD8 T cells in WT and IRF9 / mice. Data shown are representative of at least three independent experiments. (c) WT and IRF9 / mice were infected with Listeria monocytogenes 1 day before killing. Splenocytes were isolated and cultured for 4 h in presence of Brefeldin A before intranuclear staining and analysis by ow cytometry.
Total CD8 T cells, CD44CD8 T cells and NK cells were analysed for IFNg expression. (d,e) Mixed bone marrow chimera experiments. WT Thy1.1 and IRF9 / Thy1.2 (d) or WT CD45.1 and IFNAR / CD45.2 (e) myeloid progenitors were transferred into the irradiated RAG2 / host.
After 3 months, the frequency of CD44 CD49dlo CD8 T cells among WT and IRF9 / or IFNAR / CD8 T cells was analysed by ow cytometry. Eomes expression levels are represented among WT and IRF9 / or IFNAR / VM CD8 T cells. Right panel: splenocytes from the chimeras were cultured for 16 h with IL-12 IL-18. Brefeldin A (5 mg ml 1) was added for the last 3 h of culture. The frequency of IFNg cells among WT and IRF9 /
or IFNAR / total CD8 T cells are shown. For all graphs, each dot represents an individual mouse. Bars represent means.e.m. *Po0.05, **Po0.01 and ***Po0.001, NS: not signicant (non-parametric MannWhitney or paired Wilcoxon test for mixed bone marrow chimera experiments).
Effects of type I IFN signalling on Ag-specic CD8 T cells. Data obtained in steady state or polyI:C-treated mice suggest that type I IFN signalling contributes to the homeostasis of VM but not TM cells. To specically address the role of IRF9 in antigen-specic CD8 T cells, we rst assessed the effect of type I IFN stimulation in the context of an in vitro polyclonal activation assay. Consistent with previous reports42, we observed that type I IFNs potentiated polyclonal-induced Eomes expression (Fig. 6a).
We next evaluated the role of this signalling pathway upon antigenic challenge in vivo. For this purpose, we infected WT or IRF9 / animals with LM deleted for ActA that expresses the ovalbumin (OVA) antigen (DactA rLmOVA)43, and we analysed SIINFEKL-specic CD8 T cells in the course of infection (Fig. 6b). In absence of IRF9, the frequency of OVA-specic
CD8 T cells in the effector phase was slightly increased (Fig. 6b). This could be related to the deleterious effects of type I
NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 7
& 2015 Macmillan Publishers Limited. All rights reserved.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089
CD44highCD49dlo (VM)
CD8+ T cells
CD44highCD49dhigh (TM)
CD8+ T cells Ki67
+CD8+ T cells
Eomes expression
Abs. numbers (106 cells)
5
4
Abs. numbers (106 cells)
1.5
1.0
0.5
0.0 0
Spleen (d3)
CD44highCD49dlo (VM)
CD8+ T cells
CD44highCD49dlo (VM)
CD8+ T cells WT mice
Spleen (d3) Spleen (d3)
100
** **
**
* *
MFI eomes
150
3
60
%
80
100
2
40
1
20
50
0
0 PBS Polyl:C
PBS Polyl:C
Naive VM TM
Polyl:C
+ + +
200 Spleen (d3)
Polyl:C
Naive VM TM
+ + +
CD44highCD49dhigh (TM)
CD8+ T cells Spleen (d30)
NS
Abs. numbers (106 cells)
2
Abs. numbers (106 cells)
2.0
1.5
1.0
0.5
0.0
4 Spleen (d30)
*
3
1
0 PBS
Polyl:C
PBS Polyl:C
CD44highCD49dlo (VM)
CD8+T cells IFNAR/ mice
CD44highCD49dlo (VM)
CD8+ T cells
CD44highCD49dlo (VM)
CD8+ T cells
CD44highCD49dlo (VM) CD8+ T cells Eomesfl/flCD4cre mice
Blood
Blood
30 Blood PBS
Polyl:C
30
20
30
20
VM/CD8+ T cells (%)
VM/CD8+ T cells (%)
VM/CD8+ T cells (%)
10
10
10
0 d0 d5 d30
0 d0 d14 d30
0 d0 d5 d30
Abs. numbers (106 cells)
Abs. numbers (106 cells)
6
Polyl:C
WT IFNAR/
Spleen Spleen
NS
4
* *
4
3
2
NS
2
1
+
0 +
0 +
Polyl:C
WT EomesT
+
Figure 5 | PolyI:C injection increases the pool of VM CD8 Tcells. (ac) Mice were injected with 200 mg of polyI:C intraperitoneally (a) Three days after injection, mice were analysed for the absolute numbers of VM and TM CD8 T-cell subsets in the spleen. The frequency of Ki67 cells and Eomes expression among naive and memory subpopulations are depicted in the right panels. Each dot represents an individual mouse. (b) Thirty days after injection, mice were analysed for the absolute number of VM and TM CD8 T-cell subsets in the spleen. Each dot represents an individual mouse.
Data are representative of three independent experiments. (c) WT, IFNAR / and Eomes/CD4cre (EomesDT) mice were injected with polyI:C and analysed at the indicated time points for the frequency of VM CD8 Tcells in blood samples. One month after injection, mice were killed and the absolute numbers of VM CD8 T cells in indicated strains or WT littermates are illustrated. For the upper panels, each dot represents the mean of 57 mice per group. For the bottom panels, each dot represents an individual mouse. For all graphs, bars represent means.e.m. *Po0.05 **Po0.01 (non-parametric MannWhitney).
IFN signalling in the course of Listeria infection27. The frequency of memory cells, their capacity to expand upon secondary challenge or the proportion of central memory CD8 T cells were not affected in absence of IRF9 (Fig. 6c). We could demonstrate that Eomes expression was decreased in antigen-specic CD8 T cells from IRF9 / mice during the effector phase (day 7 post infection) and the memory phase (days 3570)
(Fig. 6d). These results conrm the data obtained in total CD44 CD49dhi cells (TM) under steady-state conditions (Fig. 3b). In contrast, T-bet expression in IRF9 / mice was slightly increased (effector phase) or unaffected (memory phase)
(Fig. 6d). WT/IRF9 mixed bone marrow chimera experiments conrmed the T-cell intrinsic role of IRF9 in the induction of
Eomes expression in OVA-specic CD8 T cells upon DactA rLmOVA infection (Fig. 6e).
These results indicate that IRF9 contributes to Eomes expression in the course of antigenic activation and differentiation into antigen-specic memory CD8 T cells. However, in sharp contrast to VM cells, the maintenance and function of conventional memory CD8 T cells were not compromised in absence of IRF9 signalling, a nding that could be related to the partial contribution of Eomes to their maintenance or phenotype15.
IRF9 regulates the homeostasis of innate CD8 thymocytes. Our results highlight the role of type I IFN/IRF9/Eomes pathway
8 NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
& 2015 Macmillan Publishers Limited. All rights reserved.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089 ARTICLE
Eomes
4.0106
3.0106
2.0106
1.0106
0
IFN Medium
Copies per 106
actin mRNA
800
MFI
400
0
CD3/CD28 0 h 3 h 6 h 20 h40 h 0 h 4 h 24 h 48 h
WT
Eomes mRNA
1,200
IRF9/
103
103
102
102
Effector (D7)
Memory (D70)
101
101
30
Penramer+ (%)
0
20
0
0 101
102
103
0
101
102
103
104
104
10
103
103
2
102
102
1
CD8
101
101
0
0 D0
D7
D35
0
0 101
102
103
104
0
101
102
103
104
Pentamer H-2Kb/SIINFEKL (OVA)
SIINFEKL-specific CD8+ T cells
WT IRF9/
D70
Recall SIINFEKL-specific CD8+ T cells
TCM SIINFEKL-specific
CD8+ T cells
Eomes SIINFEKL-specific CD8+ T cells
Tbet SIINFEKL-specific CD8+ T cells
WT IRF9/
60
40
D5 post II
150
100
50
NS
50
D70
NS
***
800
**
Pentamer+ (%)
20
CD62Lhigh (%)
10
40
20
600
400
200
**
30
**
MFI
MFI
NS
NS
Pre
WT IRF9/
0 Post
Pre
Post
0 WT
IRF9/
0 d7 d35 d70
0 d7 d35 d70
Eomes SIINFEKL-specific CD8+ T cells
Tbet SIINFEKL-specific CD8+ T cells
150 D7 D7
MFI
*
600
400
200
***
100
50
MFI
0 Thy1.1
(WT)
Thy1.2 (IRF9/)
0 Thy1.1
(WT)
Thy1.2 (IRF9/)
Figure 6 | Effects of type I IFN signalling on Ag-specic CD8 T cells. (a) Puried CD8 T cells were activated by polyclonal stimulation (aCD3/28) in vitro with or without rIFNb. Eomes expression is shown at the mRNA level by qRTPCR (mean uorescence intensity (MFI), left panel) and at the protein level by uorescence-activated cell sorting (FACS) (right panel). (bd) WT and IRF9 / mice were infected by LM deleted for ActA that expresses the ovalbumin antigen (DactA rLmOVA). (b) Blood samples were harvested at the indicated time after infection to monitor SIINFEKL-specic CD8 T-cell responses. For this purpose, blood cells were stained with H-2Kb/SIINFEKL pentamers. Representative dot plots illustrate the frequency of Ag-specic
CD8 T cells in WT and IRF9 / mice. Graphs represent means.e.m. of at least seven mice per group and are representative of three independent experiments (c). Mice were killed following 5 days of rLmOVA challenge at day 18 (recall phase) and at day 70 post infection (memory phase) and analysed, respectively, for the frequency of SIINFEKL-specic CD8 T cells and the proportion of CD44 CD62L (central memory) cells. (d) At day 7 (effector phase) and days 3570 (memory phase) of infection, the mice were killed and splenocytes were collected. Eomes and T-bet expression (MFI)
within SIINFEKL-specic CD8 T cells were determined by FACS staining. (e) Mixed bone marrow chimera were generated by transferring WT Thy1.1 and IRF9 / Thy1.2 myeloid progenitors into the irradiated RAG2 / host. These mice were inoculated with DactA rLmOVA and killed at day 7 post infection to analyse the expression of Eomes and T-bet among SIINFEKL-specic CD8 T cells. Each dot represents an individual mouse. Means.e.m. is shown. *Po0.05, **Po0.01 and ***Po0.001, NS: not signicant (non-parametric MannWhitney or paired Wilcoxon test for mixed bone marrow chimera experiments).
in the homeostasis of unconventional rather than conventional memory CD8 T cells. Another memory-like CD8 T-cell population, characterized by high expression of CD44, CXCR3,
CD122 and Eomes, has been described in the thymus4,12. IL-4 is known to facilitate Eomes expression in CD8 SP thymocytes and the acquisition of this particular phenotype19,35,44. Indeed, we conrmed that incubation with rIL-4 induced Eomes expression
in CD8 SP thymocytes (Fig. 7a). Consistent with our results in peripheral CD8 T cells (Fig. 2a), rIFNb alone or in combination with rIL-4 also induced Eomes expression in these experimental conditions (Fig. 7a). These results suggest that type I IFN signalling could participate in the development of innate thymic CD8 T cells. Despite the low abundance of these cells in the thymi of C57BL/6 mice, we observed a signicant reduction in
NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 9
& 2015 Macmillan Publishers Limited. All rights reserved.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089
EomeshiCD8 SP cells CD44+CXCR3+CD8 SP cells
Thymus C57BL/6
Thymus
**
1.5
C57BL/6
1.0
%
25
20
15
10
5
0
%
0.5
0.0
IFN WT
+ +
+
+
IRF9/
IL-4
CD44+CXCR3+CD8 SP cells
CD44+CXCR3+CD8 SP cells
10 Thymus C57BL/6
C57BL/6 PBS C57BL/6 polyI:C
10.8%
10 Thymus C57BL/6
** **
102
CD8
3.2% 103
103
102
101
0
%
%
101
0
4
2
0
0 103
102
101
0
101
102
103
CD41.2%
CD44
Polyl:C
4.3%
103
CD44+Eomes+CD8 SP cells Ki67+ cells
PBS PBS
Polyl:C
Thymus C57BL/6
100 Thymus C57BL/6
CXCR3
102
103
102
101
0
10
8
6
%
**
101
*
0
**
0
101
102
103
0
101
102
103
%
40
20
0Polyl:C +
+
Conventional CD44CXCR3
Innate CD44+CXCR3+
4
2
0
PBS Polyl:C
BALB/c WT BALB/c IRF9/
7%
5.2%
CD44+Eomes+CD8 SP cells
CD44+CD122+CD8 SP cells
100 Thymus BALB/c
103
103
102
102
100 Thymus BALB/c
CD8
101
101
** **
0
0
%
%
40
20
0 0
WT
40
20
0
101
102
103
0
101
102
103
CD4
44% 24.2%
103
103
WT IRF9/
IRF9/ CD44+Eomes+CD8
SP cells
100 Thymus BALB/c
**
102
102
Eomes
101
101
0
0
101
102
103
0
101
102
103
0 CD44
42.8% 21.4%
103
103
%
CXCR3
102
102
101
101
40
20
0
0
0
101
102
103
0
101
102
103
0 WT IRF9/
CD44
Figure 7 | Type I IFNs regulate the homeostasis of innate CD8 thymocytes. (a) Thymocytes from C57BL/6 mice were incubated in medium alone or stimulated with rIFNb (100 U ml 1) and/or rIL-4 (20 ng ml 1). After 16 h, Eomes expression was assessed by ow cytometry among CD8 SP thymocytes.
(b) The frequency of innate thymic CD8 T cells (CD44 CXCR3 CD8 SP) was analysed among CD8 SP thymocytes from C57BL/6 WT and IRF9 / mice. (c) PBS or polyI:C (200 mg per mouse) was injected intraperitoneally into C57BL/6 mice. After 3 days, mice were killed and thymocytes were isolated for ow cytometry analysis. The frequency of innate thymic CD8 T cells (CD44CXCR3, CD44CD122 and CD44Eomes CD8 SP) was analysed among CD8 SP thymocytes. The frequency of Ki67 was assessed in conventional or innate CD8 SP thymocytes. (d) The frequency of innate thymic CD8 T cells (CD44CXCR3, CD44 CD122 and CD44Eomes CD8 SP) was analysed among CD8 SP thymocytes of BALB/c WT and
IRF9 / mice. For all the graphs, each dot represents an individual mouse. Bars represent means.e.m. *Po0.05 and **Po0.01 (non-parametric MannWhitney).
10 NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
& 2015 Macmillan Publishers Limited. All rights reserved.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089 ARTICLE
the frequency of this population in IRF9 / mice under steady-state conditions as compared with WT animals (Fig. 7b). Type I
IFNs are known to affect the survival of thymocytes45. However, 3 days after polyI:C injection, despite a strong decrease in the cellularity of the thymus, we observed a global rise in the proportion of CD8 SP cells in the thymus, suggesting that these cells might be relatively spared. Moreover, as observed in the periphery, we found an increase in the proportion of thymic innate cells among CD8 SP thymocytes in polyI:C-injected animals (Fig. 7c). PolyI:C induced a decrease in the proportion of Ki67 classical CD44 CD8 SP cells, which could reect reduced proliferation rate or overall loss in CD4 CD8 double-positive precursors. In sharp contrast, the proportion of Ki67 cells among innate memory CD8 SP increased, indicating that along with Eomes expression, type I IFNs promote their expansion (Fig. 7c).
While innate CD8 T cells are rare in the thymi of C57BL/6 mice, they were found to be abundant in BALB/c mice under steady-state conditions, as this population is dependent on IL-4-expressing NK T cells that are highly represented in this strain4,19. We therefore evaluated the proportion of innate thymic CD8 SP thymocytes in WT or IRF9 / BALB/c mice.
This population was signicantly reduced in IRF9 / mice as compared with their WT counterparts (Fig. 7d). Taken together, these results indicate that together with IL-4, type I IFNs contribute to the development of innate thymic CD8 T cells.
DiscussionUpon sensing microbial signals, APCs such as inammatory monocytes, subcapsular macrophages or conventional dendritic cells produce cytokines that may directly or indirectly inuence the function of CD8 T cells9,22. Among these mediators, type I
IFNs are known to regulate many aspects of T-cell functions20,21,46. Herein, we clearly demonstrate a direct action of type I IFNs and IRF9-dependent signals on Eomes gene activation in CD8 T cells. This effect was observed in peripheral naive and memory CD8 T cells and in CD8 SP thymocytes.
We show that type I IFNs favour the expansion of unconventional VM CD8 T cells in an Eomes-dependent manner. This was found to be the case upon induction of type I IFNs in the context of polyI:C injection, and we observed that IRF9 was also required for the accumulation of this subset with age. How Eomes drives the acquisition of a memory-like phenotype is not clear. An important direct target of Eomes is CD122, a subunit shared by the IL-2 and IL-15 receptor38. Indeed, expression of T-bet and/or Eomes is required and sufcient to confer cellular responsiveness to IL-15 (ref. 38). This cytokine is critical for the maintenance of memory cells in general and the development of VM cells in particular13,23,4752. Importantly, type I IFNs signalling within dendritic cells and/or inammatory monocytes is also required for IL-15 production and efcient trans-presentation to CD8 T cells23,52. Hence, both reduced responsiveness to homeostatic cytokines and inefcient IL-15 trans-presentation could be responsible for a lower proliferation rate and Bcl2 expression by IRF9 / VM cells50. Bone marrow chimera experiments clearly indicate that type I IFN signalling in CD8 T cells inuences VM cell homeostasis. Taken together with previous works, we propose that type I IFNs directly upregulate Eomes expression in naive T cells leading to the induction of CD122 that confers responsiveness to IL-15 and subsequent conversion into VM cells. Type I IFNs might also be required to maintain high Eomes expression on VM cells, thereby favouring their survival and IL-15-driven expansion. Further work will be necessary to dene the global transcriptional program imposed by Eomes in these cells. There is no direct evidence that upregulation of Eomes
gene expression by type I IFNs are directly responsible for our observations. It is important to note that the effects of type I IFNs on VM CD8 T cells probably involve other pathways that have not been investigated in this work. For example, type I IFNs were shown to promote IL-15Ra expression in human T cells, resulting in enhanced IL-15 signalling53.
The function of these memory-like cells is still a matter of debate6,9,11,33,36. Like classical memory CD8 T cells, VM cells are able to rapidly produce IFNg during the rst stage of an infection, in absence of antigenic recognition9. They respond to specic antigens with a better proliferation rate than true naive CD8 T cells6. These cells could therefore contribute both to innate and adaptive immune responses. Furthermore, memory-like cells are present in large amount in the neonatal period and also accumulate with age, so they could play an important role in these periods of high susceptibility to infections32,33,36,54,55. It is tempting to speculate that high Eomes (and/or T-bet) expression is responsible for these innate properties. Indeed, in response to IL-12/IL-18 stimulation, IFNg production was restricted to
Eomes-positive CD8 T cells. Furthermore, in absence of IRF9 within CD8 T cells, the proportion of VM cells, their Eomes expression and their capacity to produce IFNg were decreased.
IRF9 was also required for optimal Eomes expression in the context of antigen-driven differentiation into effector and memory cells in vivo. Despite these ndings, the persistence or the phenotype of these TM cells under steady-state conditions, upon polyI:C injection or in the context of Listeria infection, was not consistently affected in absence of type I IFN signalling. We cannot exclude that IRF9-dependent induction of Eomes in antigen-experienced cells could contribute to their long-term maintenance in other contexts. However, this hypothesis is difcult to test in viral models such as lymphocytic choriomeningitis virus infection. Indeed, in contrast to Listeria infection, type I IFNs are critical for the induction of effector CD8 T cells in these models, which precludes the analysis of their role in the memory phase26,46,56.
Another unconventional memory CD8 T-cell population arises in the thymus. They differ from VM cells as this innate-like population is strongly dependent on IL-4 produced by PLZF-expressing NK T cells that are highly represented in BALB/c mice4,19. Of note, the thymus was shown to be an important source of type I IFN under physiological conditions57,58. Furthermore, we observed that rIFNb strongly potentiated IL-4 driven Eomes expression, suggesting that type I IFNs could also inuence the homeostasis of thymocytes. Indeed, we show that the proportion of Eomes innate CD8 SP thymocytes was signicantly decreased in IRF9 / BALB/c mice. This result indicates that, in addition to IL-4, type I IFN signalling contributes to the differentiation of these memory-like cells.
In conclusion, we demonstrate that type I IFN signalling within CD8 T cells directly regulates Eomes expression, a transcription factor associated with the acquisition of a memory-like phenotype and innate functions. Together with the role of type I IFNs on IL-15/IL-15R expression, we show that this mechanism contributes to the long-term homeostasis, tness and function of VM CD8 T cells in the periphery and the differentiation of innate memory cells in the thymus.
Methods
Mice. IRF9-decient (IRF9 / ) mice on C57BL/6 or BALB/c backgrounds were obtained from the Riken BioResource Center (Ibaraki, Japan) with the approval of
T. Taniguchi (University of Tokyo, Tokyo, Japan). RAG2 / , CD3e / , Thy1.1 and CD45.1 congenic mice on C57BL/6 background were obtained from the
Jackson Laboratory. STAT1 / and IFNAR1 / mice on C57BL/6 background were kindly provided by D.E. Levy (New York University School of Medicine,
NYC, USA) and Claude Libert (Department for Molecular Biomedical Research, VIB, Ghent, Belgium), respectively. These mice were housed and bred in our
NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 11
& 2015 Macmillan Publishers Limited. All rights reserved.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089
specic pathogen-free animal facility. Eomesox/ox mice were previously described59. These mice were further back-crossed onto C57BL/6 background for eight generations. To inactivate Eomes in T lymphocytes, mice were crossed with CD4-Cre animals. All experimental groups were matched for sex and age (610 weeks or as indicated in the legend of the gures). All animal studies were approved by the institutional Animal Care and local committee for animal welfare.
Cell preparation and cultures. Total CD8 T cells were puried by magnetic-activated cell sorting (CD8 microbeads, mice; MACS Miltenyi Biotec or
Dynabeads Untouched Mouse CD8 Cells Kit from Life Technologies). All cultures were performed in RPMI with 10% (vol/vol) FCS, 2 mM l-glutamine, 1 mM sodium pyruvate, 0.1 mM non-essential amino acids, 40 mM b-mercaptoethanol,100 U ml 1 of penicillin and 100 U ml 1 of streptomycin (all from Lonza). Total
CD8 T cells were cultured with or without rIFNb at 100 U ml 1 (PBL Interferon Source) and rIL-4 at 20 ng ml 1 (R&D Systems). In addition, we performed cocultures with rIFNb and plate-bound anti-CD3 (5 mg ml 1; 145-2C11, BD biosciences) and soluble anti-CD28 (1 mg ml 1; 37.51, BD biosciences). In another set of experiments, we performed Thy1.1 /Thy1.2 cocultures; Thy1.2 (CD90.2 ) CD8 T cells puried from WT or IRF9 / mice were cultured with
WT Thy1.1 (CD90.1 ) splenocytes for 16 h in the presence or absence of rIFNb. In another panel of experiments, splenocytes were cultured for 16 h with a cocktail of rIL-12 (5 ng ml 1) and rIL-18 (10 ng ml 1, MBL) or medium alone.
In vivo experiments. For adoptive transfer experiments using polyI:C, 3.106 WT or IRF9 / CD8 T cells (CD90.2 ) were injected intravenously (i.v.) into congenic C57BL/6 recipient mice (CD90.1 ). After 24 h, recipient mice were injected intraperitoneally with 200 mg of polyI:C (GE Health Care). After 24 h, recipient mice were killed and donor-derived T cells were identied by surface staining with anti-CD90.2 Pe-Cy7 (Biolegend, clone 30-H12m, 1/100) and anti-CD8 PerCP (BD biosciences, clone 536.7, 1/50).
For in vivo infections, WT and IRF9 / mice were inoculated i.v. with a recombinant attenuated strain of LM deleted for ActA that expresses the OVA antigen43 (DactA rLmOVA) or a recombinant strain of LM expressing ovalbumin (LM-OVA)43,60 (kindly provided by Dr Hao Shen, Department of Microbiology, University of Pennsylvania School of Medicine, Philadelphia).
Antigen-specic T-cell responses were monitored in blood and spleen using pentamer stainings (H-2Kb/SIINFEKL (OVA), Proimmune). To study innate CD8 T-cell response in blood and spleen, animals were killed 16 h post injection of DactA rLmOVA i.v. (5 105 colony-forming units per mouse). To evaluate the
recall of the CD8 T-cell response, mice were killed at day 5 following an injection of LM-OVA (1.106 colony-forming units per mouse).
We also performed mixed bone marrow chimera by injecting either WT Thy1.1 and IRF9 / Thy1.2 or WT CD45.1 and IFNAR / CD45.2 myeloid progenitors into lethally irradiated RAG2 / host recipient mice. When indicated, 8 weeks post engraftment, animals were inoculated with DactA rLmOVA and followed the same protocol as described previously.
Cell surface and intracellular staining. Puried CD8 T cells, splenocytes and thymocytes were rst stained for surface antigens and then treated with Foxp3 staining buffer set according to the manufacturers directions (eBioscience). Anti-Eomes AlexaFluor 647 or eFluor 660 (Dan11mag, 1/75), anti-T-bet PE (eBio4B10, 1/100) and anti-CD49d FITC or PE (R1-2, 1/50) antibodies were purchased from eBioscience. Anti-CD8 PercP (536.7, 1/50), anti-CXCR3 APC (CXCR3-173, 1/50), anti-CD4 Pe-Cy7 (RM4-5, 1/100), anti-CD62L PE (1/100)or V450 (1/50) (MEL-14), anti-Bcl2 PE (3F11, 1/25), anti-Ki67 FITC (B56, 1/25), anti-CD44 FITC or V450 (IM7, 1/50), anti-CD127 Pe-Cy7 (SB/199, 1/50), anti-CD122 FITC (TM-BETA1, 1/50), anti-NK1.1 FITC (PK136, 1/50), anti-CD90.2 Pe-Cy7 (53-2.1, 1/100) and anti-IFNg APC or PB or PE (XMG1.21/50) were purchased from BD biosciences. Anti-CD3 Pe-Cy7 (2C11, 1/100) was purchased from Biolegend.
In some experiments, brefeldin A (5 mg ml 1, Sigma) was added in samples for 3 h at 37 C before intracytoplasmic staining. Blood samples were directly stained for surface antigens and then treated with FACS lysing buffer (BD biosciences) as described in the product data sheet. All samples were xed with 1% paraformaldehyde in PBS prior to their processing using a Cyan ow cytometer (Dako Cytomation).
RNA purication and real-time RTPCR. Total RNA from cells was extracted using a MagnaPure LC RNA-High Performance Isolation Kit (Roche Diagnostics). Reverse transcription (RT) and real-time PCR reactions were carried out using an RNA amplication kit (one-step procedure) on a Lightcycler 480 Real-Time PCR system (Roche Diagnostics). The sequences of primers and probes were: b-actin: 50-TCCTGAGCGCAAGTACTCTGT-30, 50-CTGATCCACATCTGCTGGAAG-30 and probe 50-ATCGGTGGCTCCATCCTGGC-30, Eomes: 50-CCTTCACCTTCT CAGAGACACAGTT-30, 50-TCGATCTTTAGCTGGGTGATATCC-30 and probe 50-TCGCTGTGACGGCCTACCAAAACA-30, T-bet: 50-CAAGTTCAACCAGCA CCAGA-30, 50-CCACATCCACAAACATCCTG-30 and probe 50-TCATCACTA AGCAAGGACGGCGA-30.
Chromatin immunoprecipitation. Puried CD8 T cells were stimulatedwith rIFNb (107 cells per condition). Cells were then xed for 10 min at room temperature with 1% formaldehyde, and glycine was added to a nal concentration of 0.125 M. Cells were washed twice with ice-cold PBS, resuspended in lysis buffer and sonicated to obtain chromatin fragments that were 200500 base pairs in length using a bioruptor device (Diagenode). Chromatin was then incubated overnight at 4 C with monoclonal rabbit anti-H3K4me3 (Millipore), polyclonal rabbit anti-H3K9ac (Millipore), polyclonal rabbit anti-STAT1 (M-22, sc-592, Santa Cruz) or with rabbit polyclonal IgG (CS200581; Millipore) and protein G magnetic-activated beads (Active Motif). Beads were washed ve times. Eluted samples were incubated with NaCl (nal concentration: 200 mM) for 4 h at 65 C. Samples were treated with RNAse and Proteinase K for 1 h at 45 C and DNA was then puried using the QIAquick kit according to the manufacturers instructions (Qiagen). Quantitative PCR was performed with primers encompassing the proximal promoter region of Eomes. The sequences of primers for Eomes were: 50-AAAGAAACACCAAACCAGCA-30, 50-GGGACTTTGCTATTGGCTGT-30 and probe 50-CGCAGGCGACCCGATCCAATTA-30. As a positive control for histone modications, we used primers located in the Gapdh proximal promoter region: 50-CCACCATCCGGGTTCCTAT-30, 50-GCGATTTTCACCTGGCACT-30 and probe: 50-CTCCTCCCTGTTCCAGAGACGGC-30. As a negative control, we used a region located 14.6 kb downstream of Eomes transcriptional start site:
50-GCGCACACACACACACATAC-30, 50-AAATGGCAGGTTTCTTTACCC-30 and probe: 50-CATCTCACTAGACCTTGAGTCAGTCCTCTCTCTC-30.
Statistical analysis. Statistical analysis was performed using a non-parametric MannWhitney or paired Wilcoxon test when appropriate.
References
1. Serbina, N. & Pamer, E. G. Quantitative studies of CD8 T-cell responses
during microbial infection. Curr. Opin. Immunol. 15, 436442 (2003).2. Sallusto, F., Geginat, J. & Lanzavecchia, A. Central memory and effector memory T cell subsets: function, generation, and maintenance. Annu. Rev. Immunol. 22, 745763 (2004).
3. Prince, A. L. et al. Development of innate CD4 and CD8 T cells in
Itk-decient mice is regulated by distinct pathways. J. Immunol. 193, 688699 (2014).4. Weinreich, M. A., Odumade, O. A., Jameson, S. C. & Hogquist, K. A. T cells expressing the transcription factor PLZF regulate the development of memory-like CD8 T cells. Nat. Immunol. 11, 709716 (2010).
5. Haluszczak, C. et al. The antigen-specic CD8 T cell repertoire in
unimmunized mice includes memory phenotype cells bearing markers of homeostatic expansion. J. Exp. Med. 206, 435448 (2009).6. Lee, J. Y., Hamilton, S. E., Akue, A. D., Hogquist, K. A. & Jameson, S. C. Virtual memory CD8 T cells display unique functional properties. Proc. Natl Acad. Sci. USA 110, 1349813503 (2013).
7. Berg, R. E., Crossley, E., Murray, S. & Forman, J. Relative contributions of NK and CD8 T cells to IFN-gamma mediated innate immune protection against Listeria monocytogenes. J. Immunol. 175, 17511757 (2005).
8. Kambayashi, T., Assarsson, E., Lukacher, A. E., Ljunggren, H. G. & Jensen, P. E. Memory CD8 T cells provide an early source of IFN-gamma. J. Immunol.
170, 23992408 (2003).9. Berg, R. E., Crossley, E., Murray, S. & Forman, J. Memory CD8 T cells
provide innate immune protection against Listeria monocytogenes in the absence of cognate antigen. J. Exp. Med. 198, 15831593 (2003).10. Bou Ghanem, E. N., Nelson, C. C. & DOrazio, S. E. T cell-intrinsic factors contribute to the differential ability of CD8 T cells to rapidly secrete
IFN-gamma in the absence of antigen. J. Immunol. 186, 17031712 (2011).11. Chu, T. et al. Bystander-activated memory CD8 T cells control early pathogen load in an innate-like, NKG2D-dependent manner. Cell Rep. 3, 701708 (2013).
12. Lee, Y. J., Jameson, S. C. & Hogquist, K. A. Alternative memory in the CD8 T cell lineage. Trends Immunol. 32, 5056 (2011).
13. Sosinowski, T. et al. CD8alpha dendritic cell trans presentation of IL-15 to
naive CD8 T cells produces antigen-inexperienced T cells in the periphery
with memory phenotype and function. J. Immunol. 190, 19361947 (2013).14. Gordon, S. M. et al. Requirements for eomesodermin and promyelocytic leukemia zinc nger in the development of innate-like CD8 T cells. J.
Immunol. 186, 45734578 (2011).15. Banerjee, A. et al. Cutting edge: the transcription factor eomesodermin enables CD8 T cells to compete for the memory cell niche. J. Immunol. 185,
49884992 (2010).16. Rao, R. R., Li, Q., Odunsi, K. & Shrikant, P. A. The mTOR kinase determines effector versus memory CD8 T cell fate by regulating the expression of
transcription factors T-bet and Eomesodermin. Immunity 32, 6778 (2010).17. Paley, M. A. et al. Progenitor and terminal subsets of CD8 T cells cooperate
to contain chronic viral infection. Science 338, 12201225 (2012).18. Nayar, R. et al. TCR signaling via Tec kinase ITK and interferon regulatory factor 4 (IRF4) regulates CD8 T-cell differentiation. Proc. Natl Acad. Sci.
USA 109, E2794E2802 (2012).
12 NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
& 2015 Macmillan Publishers Limited. All rights reserved.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8089 ARTICLE
19. Lai, D. et al. KLF13 sustains thymic memory-like CD8( ) T cells in BALB/c
mice by regulating IL-4-generating invariant natural killer T cells. J. Exp. Med. 208, 10931103 (2011).20. Mattei, F., Schiavoni, G. & Tough, D. F. Regulation of immune cell homeostasis by type I interferons. Cytokine Growth Factor Rev. 21, 227236 (2010).
21. Welsh, R. M., Bahl, K., Marshall, H. D. & Urban, S. L. Type 1 interferons and antiviral CD8 T-cell responses. PLoS Pathog. 8, e1002352 (2012).
22. Soudja, S. M., Ruiz, A. L., Marie, J. C. & Lauvau, G. Inammatory monocytes activate memory CD8( ) T and innate NK lymphocytes independent of
cognate antigen during microbial pathogen invasion. Immunity 37, 549562 (2012).23. Colpitts, S. L. et al. Cutting edge: the role of IFN-alpha receptor and MyD88 signaling in induction of IL-15 expression in vivo. J. Immunol. 188, 24832487 (2012).
24. Remoli, M. E. et al. IFN-beta modulates the response to TLR stimulation in human DC: involvement of IFN regulatory factor-1 (IRF-1) in IL-27 gene expression. Eur. J. Immunol. 37, 34993508 (2007).
25. Tough, D. F., Borrow, P. & Sprent, J. Induction of bystander T cell proliferation by viruses and type I interferon in vivo. Science 272, 19471950 (1996).
26. Curtsinger, J. M., Valenzuela, J. O., Agarwal, P., Lins, D. & Mescher, M. F. Type I IFNs provide a third signal to CD8 T cells to stimulate clonal expansion and differentiation. J. Immunol. 174, 44654469 (2005).
27. Carrero, J. A., Calderon, B. & Unanue, E. R. Type I interferon sensitizes lymphocytes to apoptosis and reduces resistance to Listeria infection. J. Exp. Med. 200, 535540 (2004).
28. Kaser, A., Nagata, S. & Tilg, H. Interferon alpha augments activation-induced T cell death by upregulation of Fas (CD95/APO-1) and Fas ligand expression. Cytokine 11, 736743 (1999).
29. McNally, J. M. et al. Attrition of bystander CD8 T cells during virus-induced T-cell and interferon responses. J. Virol. 75, 59655976 (2001).
30. Kohlmeier, J. E., Cookenham, T., Roberts, A. D., Miller, S. C. & Woodland, D.L. Type I interferons regulate cytolytic activity of memory CD8( ) T cells in
the lung airways during respiratory virus challenge. Immunity 33, 96105 (2010).31. Marshall, H. D., Prince, A. L., Berg, L. J. & Welsh, R. M. IFN-alpha beta and self-MHC divert CD8 T cells into a distinct differentiation pathway characterized by rapid acquisition of effector functions. J. Immunol. 185, 14191428 (2010).
32. Chiu, B. C., Martin, B. E., Stolberg, V. R. & Chensue, S. W. Cutting edge: central memory CD8 T cells in aged mice are virtual memory cells. J. Immunol. 191, 57935796 (2013).
33. Akue, A. D., Lee, J. Y. & Jameson, S. C. Derivation and maintenance of virtual memory CD8 T cells. J. Immunol. 188, 25162523 (2012).
34. Oliver, J. A., Stolberg, V. R., Chensue, S. W. & King, P. D. IL-4 acts as a potent stimulator of IFN-gamma expression in CD8 T cells through STAT6-
dependent and independent induction of Eomesodermin and T-bet. Cytokine 57, 191199 (2012).35. Carty, S. A., Koretzky, G. A. & Jordan, M. S. Interleukin-4 regulates eomesodermin in CD8 T cell development and differentiation. PLoS ONE 9,
e106659 (2014).36. Renkema, K. R., Li, G., Wu, A., Smithey, M. J. & Nikolich-Zugich, J. Two separate defects affecting true naive or virtual memory T cell precursors combine to reduce naive T cell responses with aging. J. Immunol. 192, 151159 (2014).
37. Cruz-Guilloty, F. et al. Runx3 and T-box proteins cooperate to establish the transcriptional program of effector CTLs. J. Exp. Med. 206, 5159 (2009).
38. Intlekofer, A. M. et al. Effector and memory CD8 T cell fate coupled by
T-bet and eomesodermin. Nat. Immunol. 6, 12361244 (2005).39. Intlekofer, A. M. et al. Anomalous type 17 response to viral infection by CD8
T cells lacking T-bet and eomesodermin. Science 321, 408411 (2008).40. Pearce, E. L. et al. Control of effector CD8 T cell function by the
transcription factor Eomesodermin. Science 302, 10411043 (2003).41. Judge, A. D., Zhang, X., Fujii, H., Surh, C. D. & Sprent, J. Interleukin 15 controls both proliferation and survival of a subset of memory-phenotype CD8( ) T cells. J. Exp. Med. 196, 935946 (2002).
42. Agarwal, P. et al. Gene regulation and chromatin remodeling by IL-12 and type I IFN in programming for CD8 T cell effector function and memory.J. Immunol. 183, 16951704 (2009).43. Pearce, E. L. & Shen, H. Generation of CD8 T cell memory is regulated by IL-12. J. Immunol. 179, 20742081 (2007).
44. Nayar, R. et al. Graded levels of IRF4 regulate CD8 T cell differentiation and
expansion, but not attrition, in response to acute virus infection. J. Immunol. 192, 58815893 (2014).45. Baron, M. L. et al. TLR ligand-induced type I IFNs affect thymopoiesis.J. Immunol. 180, 71347146 (2008).
46. Keppler, S. J., Rosenits, K., Koegl, T., Vucikuja, S. & Aichele, P. Signal 3 cytokines as modulators of primary immune responses during infections: the interplay of type I IFN and IL-12 in CD8 T cell responses. PLoS ONE 7, e40865 (2012).
47. Cui, G. et al. Characterization of the IL-15 niche in primary and secondary lymphoid organs in vivo. Proc. Natl Acad. Sci. USA 111, 19151920 (2014).
48. Dubois, S., Waldmann, T. A. & Muller, J. R. ITK and IL-15 support two distinct subsets of CD8 T cells. Proc. Natl Acad. Sci. USA 103, 1207512080 (2006).
49. Itsumi, M., Yoshikai, Y. & Yamada, H. IL-15 is critical for the maintenance and innate functions of self-specic CD8( ) T cells. Eur. J. Immunol. 39,
17841793 (2009).50. Lai, Y. G. et al. IL-15 modulates the balance between Bcl-2 and Bim via a Jak3/1-PI3K-Akt-ERK pathway to promote CD8alphaalpha intestinal
intraepithelial lymphocyte survival. Eur. J. Immunol. 43, 23052316 (2013).51. Stonier, S. W., Ma, L. J., Castillo, E. F. & Schluns, K. S. Dendritic cells drive memory CD8 T-cell homeostasis via IL-15 transpresentation. Blood 112, 45464554 (2008).
52. Zhang, X., Sun, S., Hwang, I., Tough, D. F. & Sprent, J. Potent and selective stimulation of memory-phenotype CD8 T cells in vivo by IL-15. Immunity 8,
591599 (1998).53. Hansen, M. L. et al. IFN-alpha primes T- and NK-cells for IL-15-mediated signaling and cytotoxicity. Mol. Immunol. 48, 20872093 (2011).
54. Sprent, J. & Surh, C. D. Normal T cell homeostasis: the conversion of naive cells into memory-phenotype cells. Nat. Immunol. 12, 478484 (2011).
55. Szabolcs, P. et al. Coexistent naive phenotype and higher cycling rate of cord blood T cells as compared to adult peripheral blood. Exp. Hematol. 31, 708714 (2003).
56. Xiao, Z., Casey, K. A., Jameson, S. C., Curtsinger, J. M. & Mescher, M. F. Programming for CD8 T cell memory development requires IL-12 or type I IFN. J. Immunol. 182, 27862794 (2009).
57. Lienenklaus, S. et al. Novel reporter mouse reveals constitutive and inammatory expression of IFN-beta in vivo. J. Immunol. 183, 32293236 (2009).
58. Colantonio, A. D. et al. IFN-alpha is constitutively expressed in the human thymus, but not in peripheral lymphoid organs. PLoS ONE 6, e24252 (2011).
59. Mao, C. A. et al. Eomesodermin, a target gene of Pou4f2, is required for retinal ganglion cell and optic nerve development in the mouse. Development 135, 271280 (2008).
60. Foulds, K. E. et al. Cutting edge: CD4 and CD8 T cells are intrinsically different in their proliferative responses. J. Immunol. 168, 15281532 (2002).
Acknowledgements
This study was supported by the Fonds National de la Recherche Scientique (FRS-FNRS, Belgium) and an Interuniversity Attraction Pole of the Belgian Federal Science Policy. V.M. and S.G. are supported by the FRS-FNRS. We thank the David and Alice Van Buuren and Rose and Jean Hoguet foundations for their support. We also thank Dr Hao Shen, Department of Microbiology, University of Pennsylvania School of Medicine, Philadelphia, for the recombinant strain of Listeria monocytogenes expressing OVA134387. The conditional Eomes-null mouse line was generated by Mao and Klein with support from NIH/NEI grant EY011930 and from the Welch Foundation grant G-0010 both to W.H.K.
Author contributions
V.M., S.T. and M.N. conducted most of the experiments. D.T., A.A. and A.K. contributed to some experiments. C.-A.M. and W.H.K. provided critical reagents. V.M. and S.T. analysed the data. V.F. and O.L. provided input for research design and interpretation.V.M. and S.G. designed the experiments and wrote the manuscript.
Additional information
Competing nancial interests: The authors declare no competing nancial interests.
Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions
Web End =http://npg.nature.com/ http://npg.nature.com/reprintsandpermissions
Web End =reprintsandpermissions/
How to cite this article: Martinet, V. et al. Type I interferons regulate eomesodermin expression and the development of unconventional memory CD8 T cells. Nat.
Commun. 6:7089 doi: 10.1038/ncomms8089 (2015).
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articles Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
Web End =http://creativecommons.org/licenses/by/4.0/
NATURE COMMUNICATIONS | 6:7089 | DOI: 10.1038/ncomms8089 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 13
& 2015 Macmillan Publishers Limited. All rights reserved.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Copyright Nature Publishing Group May 2015
Abstract
CD8+ T-cell memory phenotype and function are acquired after antigen-driven activation. Memory-like cells may also arise in absence of antigenic exposure in the thymus or in the periphery. Eomesodermin (Eomes) is a key transcription factor for the development of these unconventional memory cells. Herein, we show that type I interferon signalling in CD8+ T cells directly activates Eomes gene expression. Consistent with this observation, the phenotype, function and age-dependent expansion of 'virtual memory' CD8+ T cells are strongly affected in absence of type I interferon signalling. In addition, type I interferons induce a sustained expansion of 'virtual memory' CD8+ T cells in an Eomes-dependent fashion. We further show that the development of 'innate thymic' CD8+ T cells is dependent on the same pathway. In conclusion, we demonstrate that type I interferon signalling in CD8+ T cells drives Eomes expression and thereby regulates the function and homeostasis of memory-like CD8+ T cells.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer