SEARCH

SEARCH BY CITATION

Keywords:

  • IL-12;
  • Infection;
  • Memory T cells;
  • Signal 3;
  • Type I IFN

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

The relevance of direct inflammatory signals (signal 3) for the activation of memory CD8+ T cells during recall responses is so far unknown. We therefore investigated the direct impact of IL-12 and type I IFN on the formation, recall potential and protective capacity of memory T cells. Using CD8+ T cells deficient for IL-12 or type I IFN receptors in an adoptive transfer system, we generated memory populations after infection with vaccinia virus, lymphocytic choriomeningitis virus or Listeria monocytogenes. The results demonstrate that in the absence of signal 3 cytokines during primary infection, functional memory T cells were formed. After retransfer into naïve mice, signal 3-deficient memory T cells were able to specifically lyse target cells in vivo under non-infectious conditions. However, after reinfection, secondary effector CD8+ T cells lacking signal 3 were impaired in expansion and protective capacity dependent on the nature of the pathogen. We conclude that memory CD8+ T cells depend on a signal 3 for expansion, independent of signals obtained during priming, thereby being influenced by the pathogen-induced inflammatory milieu during secondary infection. In summary, our results reveal an essential role for direct inflammatory cytokine signaling in secondary T-cell responses.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

The ability of the adaptive immune system to remember pathogens via the generation of memory cells improves the host immune response in the case of reinfection with the same pathogen and is the basis for protective immunity after vaccination. The relationship between effector and memory T cells is still a matter of debate although it is widely accepted that memory T cells transit through an effector phase after antigen contact 1–3. Multiple signals determining which effector T cells survive to become memory cells have been proposed: asymmetric cell division, TCR signaling strength, duration of antigenic stimulation, metabolic fitness, and availability of proinflammatory cytokines and growth or survival factors 4–8. Investigating the requirements for primary activation of naïve CD8+ T cells is therefore crucial to understanding the effector-to-memory transition. In recent years, it became clear that, during acute infection, CD8+ T-cell fate is regulated by three signals: TCR engagement (signal 1), costimulation (signal 2), and an inflammatory stimulus (signal 3) via cytokines such as IL-12 or type I interferons (type I IFN) 9–11. Direct signaling via IL-12 and type I IFN receptors thereby favors terminal differentiation of effector CD8+ T cells, which demonstrate high cytolytic activity, produce IFN-γ but are limited in survival. This subpopulation of terminal effector CD8+ T cells expresses killer cell lectin-like receptor G1 (KLRG1) downregulates CD127 and is not able to establish memory after clearance of the infection. In contrast, memory precursor effector CD8+ T cells demonstrating comparable cytolytic activity and IFN-γ expression but producing IL-2 and exhibiting a phenotypically more naïve status (low levels of KLRG1 and high levels of CD127 expression) are supposed to survive and establish memory 3, 12. We and others demonstrated that CD8+ T cells with deficiencies in receptors for IL-12 or type I IFN mainly differentiate into KLRG1lo/CD127hi memory precursor effector T cells 13–16. The importance of type I IFN versus IL-12 in a given T-cell response is thereby dictated by the pathogen-induced cytokine milieu. Interestingly, a recent study could demonstrate that direct type I IFN signaling to CD8+ T cells substituted for T-cell help during vaccinia virus (VV) infection. In this experimental setting, type I IFN replaces T-cell help by directly supporting survival and differentiation of virus-specific CD8+ T cells independent of IL-2 and IL-15 signals 17.

It was suggested that the preferential differentiation of CD8+ T cells into memory precursors in the absence of signal 3 cytokines during priming leads to higher numbers and more active memory T cells. In line with this, it was demonstrated that missing the IL-12 signal during priming favors memory CD8+ T-cell formation after infection 18. However, this finding is so far controversial, since another study suggested that programming of memory T-cell development strictly requires signal 3 in the context of VV or Listeria infections and thus, in the absence of signal 3, no memory T-cell formation occurs 19. In contrast, we showed recently that CD8+ T cells lacking IL-12 or type I IFN signals were able to form stable memory T-cell populations 13, 15. Altogether, a conclusive study addressing the influence of signal 3 availability during priming on memory formation of T cells is missing.

Memory CD8+ T cells have been classified into CD44hi/CD62Llo/CCR7lo effector memory T cells (TEM), that exhibit high granzyme B expression levels, direct ex vivo cytolytic activity and an increased capacity to migrate to peripheral tissues, and CD44hi/CD62Lhi/CCR7hi central memory T (TCM) cells, that have strong proliferative capacities, produce IL-2, and reside mainly in lymph nodes 20. Memory CD8+ T cells were shown to be more efficient in response to antigens compared with naïve CD8+ T cells, thereby demonstrating rapid proliferation, an accelerated cytolytic activity and cytokine production. Based on this faster recall potential and on in vitro activation and in vivo peptide immunization studies, it was proposed that memory T cells are independent of a third, inflammatory signal for recall responses 10, 21. However, the direct influence of the proinflammatory cytokines IL-12 and type I IFN on the activation and recall capacity of memory/secondary effector CD8+ T cells during infections has not been addressed so far.

In this study, we analyzed the influence of IL-12 and type I IFN on formation, recall potential and protective capacity of memory/secondary effector T cells. Using CD8+ T cells with defined antigen-specificity lacking receptors for IL-12 or type I IFN in an adoptive transfer (AdTf) system, we generated memory populations after infection with VV, lymphocytic choriomeningitis virus (LCMV), or Listeria. We first analyzed the formation of memory cell populations, demonstrating that the presence or absence of signal 3 cytokines does not prevent memory formation and function of CD8+ T cells, even in the situations where primary T-cell expansion was limited. Second, we investigated signal 3 dependency for recall and protective capacity of memory/secondary effector CD8+ T cells. We found that reactivation of memory CD8+ T cells is controlled by signal 3 cytokines for expansion and protective capacity, dependent on the nature of the challenge infection.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

CD8+ T cells form functional memory cells independent of type I IFN after VV infection

As previously described, CD8+ T cells lacking signal 3 cytokine receptors were able to establish memory populations after AdTf and infection with different pathogens even in the situation when expansion of T cells in the primary response was limited. First, we analyzed the contribution of signal 3 cytokines to memory formation of P14 T cells competent (P14.WT) or deficient for type I IFN signaling (P14.IFNARKO) after infection with recombinant vaccinia virus expressing LCMV glycoprotein (rVVGP) (expressing the LCMV-GP33 epitope). This infection model was chosen since CD8+ T cells are largely independent of type I IFN signals for expansion after primary VV infection (Fig. 1A, 13) and thus, WT and IFNARKO memory T cells are supposed to display similar characteristics and functions. Forty days after primary infection P14.WT or P14.IFNARKO memory T cells constituted ∼7–8% of total CD8+ T cells in the blood (Fig. 1A). To distinguish different memory T-cell subsets, we examined expression of KLRG1 and CD127, as well as CD62L and CD44 (Fig. 1B and C). Only CD8+ T cells expressing high levels of CD127 survived and formed memory cells (90% of P14.WT and P14.IFNARKO T cells), with 25% of P14.WT and 19% of P14.IFNARKO T cells still expressing KLRG1 (Fig. 1B and C), reflecting the slightly impaired KLRG1 expression on P14.IFNARKO T cells during primary infection (our unpublished results). No differences in the development into central or effector memory T cells were observed, as ∼30% of P14.WT and P14.IFNARKO T cells exhibited a TCM and ∼70% a TEM phenotype. Furthermore, the expression of the transcription factors T-box transcription factor (T-bet) and Eomes was analyzed. About 95% of both P14 memory T-cell populations expressed T-bet after VV infection, with comparable expression levels on a per cell basis (see MFI, Fig. 1C). After VV infection, ∼30% of P14.WT and ∼25% of P14.IFNARKO memory T cells expressed Eomes (Fig. 1C). The comparison of cytokine production after short-term peptide restimulation in vitro revealed that ∼80% of both memory T-cell populations produced IFN-γ and TNF-α, about 40% of both populations expressed IL-2 and almost all T cells degranulated (CD107a surface translocation). Interestingly, on a per cell basis, P14.IFNARKO memory T cells produced less IFN-γ, TNF-α, and IL-2 (but not statistically significant) and were somewhat reduced in degranulation compared with P14.WT memory T cells (see MFI, Fig. 1D).

thumbnail image

Figure 1. Analysis of memory T cells day 40 after VV infection. (A) 105 naïve P14.WT (•) or P14.IFNARKO (○) T cells were adoptively transferred into B6 mice which were infected with 2×106 PFU rVVGP. Expansion of P14 T-cell populations was analyzed in the acute (day 4 p.i.; left) and in the memory phase (day 40 p.i.; right). Plots show percentages of P14 T cells of total CD8+ T cells in the blood. (B) Analysis of KLRG1 and CD127 expression on P14.WT (left) and P14.IFNARKO (right) memory T cells in the spleen. Plots are gated on CD8+, Thy1.1+ (P14) cells. (C) Analysis of differentiation of splenic memory P14.WT (black bars and symbols) and P14.IFNARKO (white bars and symbols) T cells. Percentages of KLRG1 and CD127 positive as well as percentages of central (CD44hi/CD62Lhi) and effector (CD44hi/CD62Llo) memory P14 T cells of total P14 T cells are depicted (top, left to right). Percentages of T-bet and Eomes-positive P14 T cells of total P14 memory T cells, as well as mean fluorescence intensity (MFI) of T-bet and Eomes of P14 T cells are indicated (bottom). (D) Effector functions of P14 memory T cells were analyzed after short-time restimulation ex vivo. Percentages of cytokine-positive P14 memory T cells of total P14 T cells, as well as MFI of cytokine expression are indicated. (A, C, D) Values are expressed as mean (+SEM), n=5. Results are representative of three independent experiments. n.s., not significant (Student's unpaired t-test).

Download figure to PowerPoint

These data indicate that CD8+ T-cell memory formation was possible in the absence of type I IFN signaling after VV infection.

Memory CD8+ T cells depend on type I IFN as a signal 3 after LCMV recall infection

In a first approach, we studied in vivo recall capacities and effector functions of memory T cells, lacking type I IFN signaling, in a retransfer system. We used memory P14.WT and P14.IFNARKO T cells generated in the context of VV infection since T cells were largely independent of type I IFN signals during primary infection (comparable expansion, differentiation, and effector functions) and therefore leading to memory cells with the same experience.

To examine the contribution of type I IFN signaling to memory CD8+ T-cell recall responses, equal numbers of purified P14.WT and P14.IFNARKO memory T cells, generated by VV infections (Fig. 1), were transferred into naïve recipients (experimental setup, Fig. 2A). The transfer of equal numbers of memory T cells guaranteed a direct comparability of the two T-cell populations. To analyze in vivo cytolytic activity under non-infectious conditions, we performed an “in vivo killer” assay. After AdTf of P14.WT and P14.IFNARKO memory T cells, target cells were transferred into recipient mice. Splenocytes of H8 mice that ubiquitously express GP33 and are recognized by P14 T cells were labeled with a low amount of CFSE and used as target cells. Spleen cells of C57BL/6 (B6) mice were marked with a high concentration of CFSE and used as a control. In total, 28 h after transfer of a 1:1 mixture of the two target cell populations, the specific cytolytic activity of P14.WT and P14.IFNARKO memory T cells reduced the number of H8 target cells in the blood to ∼32% (from originally 50%, Fig. 2B). Thereafter, H8 target cells were gradually reduced, whereas control target cells were still detectable in the blood 66 h after transfer. From these results, we concluded that both memory T-cell populations exhibited comparable, cytolytic activity in vivo under non-infectious conditions independent of type I IFN-signaling.

thumbnail image

Figure 2. Functional analysis of VV-experienced P14.WT and P14.IFNARKO memory T cells. (A) Experimental setup. Forty days after AdTf and primary VV infection, MACS-purified P14.WT and P14.IFNARKO memory T cells (Thy1.1) were retransferred into B6 mice (Thy1.2) and analyzed for functionality. (B) Cytolytic activity of P14 memory T cells was determined by an in vivo cytotoxicity assay. In total, 2×105 purified P14.WT and P14.IFNARKO memory T cells were transferred into naïve B6. Spleen cells of H8 mice (CFSE low) and B6 mice (CFSE high) were used as target cells. Cytolytic activity was measured 28, 42, and 66 h after transfer of target cells by analyzing PBLs. Histograms show target cell populations, with numbers indicating percentages of CFSE-positive cells. (C, D) Analysis of re-expansion of VV-experienced P14 memory T cells. Briefly, 2×104 Thy1.1+ P14.WT or P14.IFNARKO memory T cells were transferred into B6 mice (Thy1.2+) followed by secondary infection with (C) rVVGP or (D) LCMV. Expansion of P14 T cells was analyzed in the spleen (C) at day 5 after rVVGP infection or (D) at day 7 after infection with LCMV. Percentages and absolute numbers of P14.WT (•) and P14.IFNARKO (○) T cells are shown (top). Percentages of KLRG1 and CD127+ memory P14.WT (black bars) and P14.IFNARKO (white bars) T cells of total P14 T cells are shown (middle). Percentages of IFN-γ-positive P14 T cells of total P14 memory T cells, as well as MFI of IFN-γ of P14.WT (black bars and symbols) and P14.IFNARKO (white bars and symbols) memory T cells are indicated (bottom). (E) Protective capacity of P14 memory T cells was tested after AdTf of 2×105 P14.WT or P14.IFNARKO memory T cells in naïve B6, followed by infection with LCMV-WE. Viral titers in the spleen at day 4 after challenge are shown. Values are expressed as mean (+SEM), n=3. *p<0.05, **p<0.01, ***p<0.001, n.s., not significant (Student's unpaired t-test). Results are representative of two independent experiments.

Download figure to PowerPoint

As memory CD8+ T cells were thought to be independent of a third signal for re-expansion and survival, we next tested the recall capacity of P14.WT and P14.IFNARKO memory T cells in vivo. After AdTf of low numbers (2×104) of memory T cells, recipient mice were infected with either rVVGP or LCMV (Fig. 2C and D). We decided to use these two infection models because CD8+ T cells were largely independent of type I IFN during primary VV infection, whereas fully depended on type I IFN signals for expansion during primary LCMV infection 13.

On day 5 after VV infection, analysis of the P14 memory T cells in the spleen revealed that P14.IFNARKO secondary effector T cells were slightly reduced in expansion compared with P14.WT T cells, constituting only 2% of total CD8+ T cells (compared with 5% of P14.WT T cells). This two to three-fold lower expansion was also seen in absolute cell numbers in the spleen. It is worth noting that both secondary effector T-cell populations demonstrated strongly diminished expansion compared with primary infection with VV, where both T-cell populations expanded to ∼45% of CD8+ T cells in the blood (Fig. 1A). This impaired secondary expansion of VV-experienced memory T cells was also observed when higher numbers of P14.WT and P14.IFNARKO T cells were transferred and the challenging dose was increased (Supporting Information Fig. 1). One may speculate that the VV-induced inflammatory milieu during T-cell priming negatively influences the fitness of the memory T cells or that antigen availability during VV reinfection is limited leading to diminished memory T-cell triggering and expansion.

Despite this difference in expansion, analysis of surface markers KLRG1 and CD127, as well as CD44 and CD62L revealed no differences between the two secondary effector T-cell populations after VV infection. About 70–80% of P14 secondary effector T cells expressed KLRG1 and 90% demonstrated a TEM phenotype (CD44hi/CD62Llo, Fig. 2C and Supporting Information Fig. 2A), suggesting highly activated T cells after challenge infection. Additionally, P14.WT and P14.IFNARKO secondary effector T cells showed comparable cytokine production after short-time restimulation in vitro, with 70% of P14 T cells producing IFN-γ, 20% of P14 T cells producing TNF-α and only 2–3% demonstrated IL-2 production (Fig. 2C and Supporting Information Fig. 2A). No differences were detectable in the degranulation capacity according to CD107a staining of both secondary effector T-cell populations (Supporting Information Fig. 2A). Thus, after secondary VV infection, CD8+ T cells were dependent on type I IFN as a signal 3 for expansion, but were independent for differentiation into highly activated, cytokine-producing cells.

After retransfer of equal numbers of memory T cells and LCMV challenge infection, P14.WT secondary effector T cells expanded to ∼60% of CD8+ T cells, whereas P14.IFNARKO secondary effector T cells were severely impaired in their re-expansion, as seen in relative and total cell numbers on day 7 postinfection in the spleen (Fig. 2D). The drastically reduced expansion correlated with diminished KLRG1 expression, as only 40% of P14.IFNARKO secondary effector T cells expressed KLRG1 compared with 90% of P14.WT T cells (Fig. 2D). In addition, only 80% of P14.IFNARKO secondary effector T cells demonstrated a TEM phenotype (compared with 95% of P14.WT T cells) and were reduced in cytokine production. Less P14.IFNARKO T cells produced IFN-γ, TNF-α, and IL-2 compared with P14.WT cells after short-time restimulation in vitro, demonstrating a less activated phenotype (Fig. 2D and Supporting Information Fig. 2B). Interestingly, on a per-cell basis, expression levels of all cytokines analyzed were higher with P14.IFNARKO secondary effector T cells, probably due to a compensatory mechanism of the few P14.IFNARKO T cells left for in vitro restimulation.

To test the protective capacity of P14 secondary effector T cells in vivo, we adoptively transferred equal numbers of P14.WT and P14.IFNARKO memory T cells in B6 mice and challenged the recipient mice with low-dose LCMV-WE. Four days later, viral titers in the spleen of recipient mice were analyzed. P14.WT secondary effector T cells were able to clear the virus close to detection limit until day 4, whereas P14.IFNARKO T cells were markedly impaired in virus control (Fig. 2E). In summary, our results show that P14.IFNARKO secondary effector T cells are severely impaired in expansion and protective capacity during LCMV infection, although fully functional under non-infectious conditions or during VV recall infection. The reverse approach, transfer of memory P14.IFNARKO T cells that were initially activated with LCMV and then challenged with rVVGP is an obvious complementary experimental setting. However, such an experiment is technically not possible, since P14.IFNARKO T cells are drastically limited in expansion after primary LCMV infection (there are virtually no effector T cells present) and the frequencies of P14.IFNARKO memory T cells are too low for retransfer experiments.

We concluded from these experiments that memory CD8+ T cells are regulated by type I IFN as signal 3 for recall expansion dependent on the inflammatory milieu induced by different pathogens, probably irrespective of signals obtained originally during primary infection.

Memory CD8+ T cells depend on IL-12 as a signal 3 during Listeria recall infection

To further support our findings that signal 3 cytokines modulate memory CD8+ T-cell recall responses dependent on the infection, we analyzed, in a second approach, memory CD8+ T cells lacking IL-12 as signal 3. For the generation of memory T cells, naïve P14.WT and P14.IL-12RKO T cells were adoptively transferred and recipient mice were infected with LCMV (experimental setup, Fig. 3A). We chose LCMV as primary infection since P14.IL-12RKO T cells expanded and gained effector functions like P14.WT cells and therefore memory T cells with comparable experience are generated 15. About 50 days after primary infection, memory T cells were purified and equal numbers of P14.WT and P14.IL-12RKO memory T cells transferred into B6 mice.

thumbnail image

Figure 3. Functional analysis of LCMV-experienced P14.WT and P14.IL-12RKO memory T cells. (A) Experimental setup. Forty days after AdTf and primary LCMV infection, MACS-purified P14.WT and P14.IL-12RKO memory T cells (Thy1.1) were retransferred into B6 mice (Thy1.2) and analyzed for functionality. (B) Cytolytic activity of P14 memory T cells was determined by an in vivo cytotoxicity assay. In total, 5×105 purified P14.WT or P14.IL-12RKO memory T cells were transferred into naïve B6. Spleen cells of H8 mice (CFSE low) and B6 mice (CFSE high) were used as target cells. Cytolytic activity was measured 17 and 40 h after transfer of target cells by analyzing PBLs. Histograms show target cell populations, with numbers indicating percentages of CFSE-positive cells. (C) Analysis of long-term protective capacity using B16GP melanoma cells. Two wks after AdTf of 2×104 P14 memory cells and 5×105 B16GP melanoma cells, lungs of recipient mice were analyzed for metastases. (D, E) Analysis of re-expansion of LCMV-experienced P14 memory T cells. In total, 2×104 Thy1.1+ P14.WT (black symbols) or P14.IL-12RKO (white symbols) memory T cells were transferred into B6 mice (Thy1.2+) followed by infection with (D) LCMV or (E) rListeriaGP33. Expansion of P14 T cells was analyzed in the spleen (D) at day 7 after LCMV infection or (E) at day 6 after infection with rListeriaGP33. Dot plots demonstrate expansion of P14.WT (top, upper row) and P14.IL-12RKO LCMV memory cells (top, lower row). Percentages and absolute numbers of P14.WT (•, black bars) and P14.IL-12RKO (○, white bars) T cells are shown (top). Percentages of KLRG1 and CD127+ memory P14 T cells of total P14 T cells are shown (middle). Percentages of IFN-γ-positive P14 T cells of total P14 memory T cells, as well as MFI of IFN-γ of P14.WT (black bars and symbols) and P14.IL-12RKO (white bars and symbols) memory T cells are indicated (bottom). Values are expressed as mean (+SEM), n=3. *p<0.05, **p<0.01, ***p<0.001, n.s., not significant (Student's unpaired t-test). Results are representative of two independent experiments.

Download figure to PowerPoint

First, in vivo cytolytic activity of transferred P14.WT and P14.IL-12RKO memory T cells was tested in an “in vivo killer” assay with CFSE-labeled H8 and B6 spleen cells as targets. Both memory T-cell populations lysed the specific target cells with comparable kinetics, which is indicated as a reduction in H8 splenocytes in the blood (Fig. 3B).

In a second approach, to investigate cytolytic activity of memory T cells under non-infectious conditions over a longer period of time, control or B16 melanoma cells expressing the GP33-epitope were used as read-out system. After AdTf of P14.WT or P14.IL-12RKO memory T cells, B16GP33 melanoma cells were injected i.v. into recipient mice. Two weeks after transfer of tumor cells, a high number of B16 metastases were found in the lungs of mice, which did not receive memory T cells. In contrast, lungs of recipient mice which received P14.WT or P14.IL-12RKO memory T cells were almost free of metastases (Fig. 3C). Thus, P14.IL-12RKO and P14.WT memory T cells exhibited comparable short-term and long-term in vivo cytolytic activities under non-infectious conditions. We conclude that memory T cells are independent of IL-12 (or type I IFN) as signal 3 for the development of cytolytic effector functions under non-infectious conditions.

To analyze expansion and effector functions under infectious conditions, recipient mice of memory T cells were challenged with either LCMV (no dependency on IL-12 signals during primary infection) or recombinant ListeriaGP33 (dependency on IL-12 signals during primary infection). On day 7 after secondary LCMV infection, P14.WT and P14.IL-12RKO secondary effector T cells showed vigorous expansion in the spleen with about 50% P14 T cells in the CD8+ compartment. Although equally well in expansion, P14.IL-12RKO secondary effector T cells were slightly but significantly impaired in KLRG1 expression (only 60% expressing KLRG1 compared with 80% of P14.WT T cells) but not in T-bet expression on a per cell basis (Fig. 3D, Supporting Information Fig. 3A). Despite the lower expression of KLRG1, no differences were observed between P14.WT and P14.IL-12RKO secondary effector T cells concerning expression of other surface markers (CD127, CD44, and CD62L) or cytokine production. About 80% of both secondary effector T-cell populations produced IFN-γ, 50% TNF-α, and 3–4% showed IL-2 production after short-term in vitro restimulation, with no significant differences in the MFI (Fig. 3D, Supporting Information Fig. 3A).

After secondary Listeria infection, however, P14.IL-12RKO secondary effector T cells exhibited severely reduced expansion in the spleen compared with P14.WT cells (5 versus 22% P14 T cells of total CD8+). Results were confirmed by absolute cell numbers (Fig. 3E). In addition, P14.IL-12RKO secondary effector T cells were impaired in KLRG1 expression and downregulated CD127 to a lesser extend than P14.WT T cells. The diminished expression of T-bet on a per cell basis correlated with the impaired KLRG1 expression on P14.IL-12RKO secondary effector T cells. In addition, more P14.IL-12RKO secondary effector T cells expressed IL-2 after restimulation, altogether indicating a less activated secondary effector phenotype. However, no differences were observed in differentiation into TCM and TEM subpopulations, as almost 100% of P14.WT and P14.IL-12RKO secondary effector T cells showed a TEM phenotype. Again, expression of IFN-γ and TNF-α as well as degranulation capacity was comparable in both secondary effector T-cell populations (Fig. 3E and Supporting Information Fig. 3B).

In line with our findings of memory CD8+ T cells lacking type I IFN signals, these results indicate that memory CD8+ T cells depend on a signal 3 for expansion and differentiation into highly activated secondary effector cells under defined infectious conditions.

Memory T cells with different experience exhibit equal proliferative capacity after LCMV rechallenge

So far, we examined memory T cells competent or deficient for signal 3 cytokines that were established during primary infection conditions leading to the same experience (comparable expansion, differentiation, and effector functions). Therefore, we analyzed in a third approach the impact of signal 3 cytokines on memory T-cell re-expansion in situations where already the primary T-cell response was determined by signal 3 and thus memory T cells with different experience were generated. To this end, P14.WT and P14.IL-12RKO memory T cells were retransferred into naïve recipient mice 50 days after primary Listeria infection (experimental setup, Supporting Information Fig. 4A). During primary Listeria infection, P14.IL-12RKO T cells are diminished in their expansion (Fig. 4A), show severely reduced KLRG1 expression, limited downregulation of CD127, and produce less IFN-γ on a per cell basis compared with P14.WT T cells. Nevertheless, P14.IL-12RKO T cells form memory populations of comparable size to P14.WT T cells, but are slightly reduced in IFN-γ and TNF-α production after ex vivo restimulation 15. Thus, after primary Listeria infection, P14.WT and P14.IL-12RKO develop into memory T cells with different experience. To investigate re-expansion capacities of P14.WT and P14.IL-12RKO memory T cells generated after primary Listeria infection, equal numbers of memory T cells were retransferred into B6 mice and recipients challenged with either LCMV or rListeriaGP33. In accordance to our previous findings with naïve T cells, P14.WT and P14.IL-12RKO secondary effector T cells expanded to the same extent after LCMV infection (50–60% of CD8+ in the spleen) (Fig. 4B). In contrast after Listeria infection, P14.IL-12RKO secondary effector T cells demonstrated a five-fold reduced expansion compared with P14.WT T cells (Fig. 4C). P14.IL-12RKO secondary effector T cells showed lower KLRG1 expression and diminished production of IFN-γ during both LCMV and Listeria infections. In addition, during Listeria infection, more P14.IL-12RKO secondary effector T cells produced IL-2, thereby indicating a less activated phenotype compared with P14.WT T cells (Fig. 4C). However, production of TNF-α and degranulation were comparable in P14.WT and P14.IL-12RKO secondary effector T cells after LCMV and Listeria infections (Supporting Information Fig. 4B and C).

thumbnail image

Figure 4. Functional analysis of Listeria-experienced P14.WT and P14.IL-12RKO memory T cells. (A) Expansion of naïve P14.WT (top) and P14.IL-12RKO (bottom) T cells day 6 after AdTf and primary rListeriaGP33 infection. Numbers in brackets indicate percentages of P14 T cells of total CD8+ T cells in the spleen. Forty days after primary AdTf, 2×104 MACS-purified P14.WT and P14.IL-12RKO memory T cells (Thy1.1) were retransferred into B6 mice (Thy1.2) and analyzed for recall capacity. Expansion of P14 T cells in the spleen (B) at day 7 after LCMV infection or (C) at day 6 after infection with rListeriaGP33. Dot plots demonstrate expansion of P14.WT (upper row) and P14.IL-12RKO memory cells (lower row). Percentages and absolute numbers of P14.WT (•, black bars) and P14.IL-12RKO (○, white bars) T cells are shown (top, right of dot plots). Percentages of KLRG1 and CD127+ memory P14 T cells of total P14 T cells are depicted (middle). Percentages of IFN-γ and IL-2-positive P14 T cells of total P14 memory T cells, as well as MFI of IFN-γ and IL-2 of P14.WT (black symbols and bars) and P14.IL-12RKO (white symbols and bars) memory T cells are indicated (lower rows). Values are expressed as mean (+SEM), n=4. *p<0.05, **p<0.01, ***p<0.001, n.s., not significant (Student's unpaired t-test). Results are representative of three independent experiments.

Download figure to PowerPoint

In summary, while Listeria- and LCMV-experienced P14.IL-12RKO memory T cells developed robust secondary immune responses following LCMV infection, both rListeria- and LCMV-experienced P14.IL-12RKO memory T cells were limited in expansion after rListeria infection. Thus, these results indicate that the availability of signal 3 cytokines during priming does not significantly influence the recall capacity of memory T cells and that memory T cells are modulated by signal 3 cytokines for expansion dependent on the inflammatory milieu induced by the pathogens used for challenge.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

The ability to develop and sustain memory T-cell populations after infection or immunization is a hallmark of the adaptive immune response. The aim is to manipulate the primary activation of CD8+ T cells in a way to improve formation and efficiency of memory responses to challenge. Here, we analyzed the direct influence of the proinflammatory cytokines IL-12 and type I IFN on T-cell memory formation as well as the recall responses and protective capacity of memory CD8+ T cells after antigen re-encounter. We therefore followed two different approaches: first, we applied equal priming conditions for WT CD8+ T cells and T cells lacking the receptor for IL-12 or type I IFN, using infectious conditions under which CD8+ T cells were largely independent of the respective signal 3 cytokine (LCMV infection for IL-12RKO T cells and VV infection for IFNARKO T cells). We thus generated wild-type and signal 3-deficient memory T cells with the same experience and comparable effector to memory transition.

Second, we analyzed memory T cells from WT and IL-12R-deficient T cells that developed differently during primary Listeria infection. In this setting, during acute phase of infection, IL-12R-deficient CD8+ T cells are diminished in their expansion, show severely reduced KLRG1 expression and downregulation of CD127, and produce less IFN-γ on a per cell basis compared with WT T cells. Additionally, CD8+ T cells that lack IL-12 signals during priming with Listeria differentiated preferentially into a memory precursor phenotype. Thus, memory cells with different experience and different history of effector to memory transition were generated.

The analysis of memory T cells at day 40 after infection generated from the two described settings revealed that CD8+ T cells formed memory populations of comparable size independent of signal 3 availability during priming. Wild-type and signal 3-deficient memory T cells were able to produce cytokines after ex vivo restimulation (Fig. 1, 15).

Our finding that memory T cells formed independently of signal 3 contrasts previously published results which demonstrated that OT-1 T cells lacking type I IFN receptor or IL-12 receptor failed to form memory populations in response to VV or Listeria infections 19. In this study, OT-1 T cells deficient in the IL-12Rβ1 chain were used, which lack both IL-12 and IL-23 signals, whereas in the current article we worked with P14 T cells deficient in the IL-12Rβ2 chain which exclusively lack IL-12 signals. Furthermore, the use of different TCR transgenic T cells (varying in TCR avidity), infection with different recombinant VV and different application routes may have an impact on signal 3 dependency of T-cell responses. In our hands, as demonstrated in the current and our previously published article, the generation of functional P14 memory T cells lacking IL-12 or type I IFN signals was observed after infections with LCMV, recombinant vesicular stomatitis virus expressing LCMV glycoprotein (rVSVGP), rVVGP and rListeriaGP33 13, 15 (our unpublished results). Even under conditions, in which severely impaired expansion occurred during primary infection (IFNAR-deficient T cells during LCMV infection) a small number of memory T cells were detectable at late time points 13, 14.

It was suggested that altering the ratio of effector subpopulations during primary infections toward less activated KLRG1lo effector cells (memory precursors) by reducing the inflammatory milieu might lead to higher numbers of efficient memory T cells 3, 7, 22, 23. Findings of one study thus indicated that CD8+ T cells in IL-12p35-deficient mice, and therefore in the absence of IL-12, differentiated more into a memory precursor T-cell phenotype and generated a higher frequency of long-term memory T cells with an enhanced protective capacity after rechallenge infection 22. However, IL-12p35-deficient mice exhibit an altered cytokine pattern after infections as other cytokines are produced to compensate the general lack of IL-12. Thus, CD8+ T cells in IL-12p35-deficient mice might be influenced by other cytokines and therefore conclusions of direct effects of IL-12 on memory formation of CD8+ T cells are difficult to draw. In contrast, we used IL-12Rβ2-deficient T cells adoptively transferred into B6 mice, thereby depriving only CD8+ T cells from IL-12 signals during primary Listeria infection. Thus, we were able to study the transition of KLRG1lo memory precursor effector cells to memory cells in an intact environment concerning the ability to produce an inflammatory milieu. In the context of primary Listeria infection, IL-12R-deficient T cells were limited in their expansion but demonstrated preferentially a memory precursor phenotype when compared with WT T cells. Thereafter, the stronger contraction of WT T cells mainly correlated with the disappearance of the terminal effector cell population over time. As the total number of memory precursor T cells was largely the same in WT and IL-12R-deficient populations during the peak of infection, it is not surprising that both cell populations formed memory pools of comparable size and the same ratio of TCM and TEM subpopulations until day 40 postinfection 15.

A similar memory precursor phenotype of CD8+ T cells during primary infection was described in studies, in which antibiotic treatment limited inflammation after Listeria infection. In this setting, the availability of inflammatory cytokines was reduced due to a shortened duration of infection and memory-like CD8+ T cells were able to rapidly expand in response to a booster challenge at day 7 post primary infection. In contrast, CD8+ T cells primed during Listeria infection without antibiotic treatment were not able to re-expand at this early time point after primary infection 7. These and our findings clearly demonstrate that under reduced inflammatory conditions the generation of memory precursor T cells is favored. It was suggested that the rapid transition to memory phenotype of cells lacking inflammatory signals during priming might lead to a faster recall capacity compared with terminally differentiated effector cells 7, 24. However, conclusive results are still lacking. In our model, the rapid transition into the memory precursor phenotype seems to have no influence on the size of the memory T-cell population formed at later time points.

Altogether, these findings suggest that the quantity and quality of proinflammatory cytokines during priming determine (i) the expansion capacity of T cells and (ii) whether CD8+ T cells exhibit a terminal effector phenotype to successfully combat an acute infection or rapidly progress to a memory status.

To investigate the recall potential and protective capacity of memory T cells at day 40 after primary infection, we decided to retransfer normalized numbers of memory CD8+ T cells into B6 mice, as the direct challenge of memory mice does not allow discrimination between individual variations in precursor frequencies and direct impact of signal 3 on reactivation of memory T cells. Thus being able to analyze the functionality of memory cells on a per cell basis, we found that signal 3-deficient memory CD8+ T cells showed comparable cytolytic activity to wild-type cells under non-infectious conditions. This is in line with in vivo data from peptide immunization experiments revealing comparable expansion of memory cells in the presence or absence of IL-12 signals during secondary antigen contact 21.

Under infectious conditions during recall responses, we observed comparable effector functions and differentiation into secondary effector T-cell subpopulations independent of IL-12 or type I IFN signaling. However, memory CD8+ T cells lacking signal 3 cytokine receptors were impaired in their re-expansion during secondary infections, dependent on the nature of the pathogen. In addition, secondary effector CD8+ T cells lacking the type I IFN receptor were not only reduced in their proliferative, but also in their protective capacity after LCMV challenge infection. From this, we concluded that memory T cells exert immediate effector functions independent of signal 3 cytokines, and thus mount more efficient immune responses than naïve T cells under non-infectious conditions. However, protective immunity against rechallenge infection requires expansion of memory T cells, dictated by sensing inflammatory signals. Thus, memory/secondary effector T cells during secondary immune responses demonstrated a signal 3 dependency for expansion similar to naïve T cells during priming. Signal 3 dependency is thereby determined by the pathogen induced inflammatory milieu.

Although inflammatory signals obtained during priming did not affect formation and re-expansion capacity of memory T cells, results of IL-12R-deficient memory T cells suggest that IL-12 signals during priming are able to imprint CD8+ T cells for KLRG1 upregulation and T-bet expression. Lower KLRG1 expression on IL-12RKO secondary effector T cells after secondary infection correlated with increased IL-2 production and therefore points at less activated secondary effector CD8+ T cells. These findings were independent of the experience of memory T cells during priming and therefore the effector to memory transition in both settings analyzed.

In summary, our data clearly show that even in the absence of signal 3 cytokines during priming, functional memory T cells are formed, which exhibit cytokine production and in vivo cytolytic activity. However, depending on the rechallenge infection, signal 3 cytokines were crucial to support expansion and protective capacity of secondary effector CD8+ T cells after pathogen re-encounter. To our knowledge, this is the first study to demonstrate a signal 3 dependency of secondary effector T cells. Our results therefore reveal an essential role for inflammatory cytokines in secondary immune responses, thereby contributing to the understanding of the developmental fate of naïve and memory T cells.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

Mice

C57BL/6 (B6) mice were obtained from Harlan Winkelmann, IL-12Rβ2-deficient (IL-12RKO) (Il12rb2tm1Jm) mice and IFN-α-deficient IFNARKO (B6.129S7-Ifnar1tm1Agt) from the Jackson Laboratory 25, 26. P14.WT TCR transgenic (B6.D2-Tg(TcrLCMV)327Sdz/JDvsJ), P14.IFNARKO, and P14.IL-12RKO mice on a B6.Thy1.1 background were generated by breeding. Mice were kept under specific pathogen-free conditions and used at 8–16 wk of age. Animal care and use was approved by the Regierungspräsidium Freiburg.

Infections

LCMV-WE and rVVGP 27 expressing the LCMV-GP were grown on L929 or BSC-1 cells. rListeriaGP33 28, expressing the LCMV-GP epitope gp33-41, were grown in TSB medium. Mice were infected with 200 PFU of LCMV-WE, 2×106 PFU of rVVGP, and 2×104 CFU of rListeriaGP33 i.v.

Protection was determined 4 days after infection with 200 PFU LCMV-WE in the spleen of recipient mice using a focus forming assay as described previously 29.

AdTf experiments

P14 T cells express a transgenic TCR specific for the LCMV-GP epitope gp33-41 in the context of H-2 Db. Splenic P14 T cells (P14.WT, P14.IFNARKO, or P14.IL-12RKO) from donor mice were purified and equal numbers were transferred i.v. into sex-matched B6 mice as described previously 15. After AdTf of P14 T cells, recipient mice were infected with LCMV, rVVGP, or rListeriaGP33. For AdTf of memory cells, P14 memory T cells were isolated 40–50 days after primary infections. After MACS purification, 2×104 P14 memory T cells were retransferred into B6 mice for expansion and B16GP protection experiments; 2–5×105 P14 memory T cells were retransferred for the analysis of cytolytic activity (in vivo killer) and protection (LCMV-WE challenge). For tumor protection experiments, 5×105 B16GP melanoma cells were injected i.v. and 2 wk later lungs of recipient mice were analyzed.

Flow cytometry

All antibodies were purchased from eBioscience. For the analysis of intracellular cytokines, 106 lymphocytes per well were stimulated with 10−7 M LCMV-GP33-41 peptide in the presence of Brefeldin A for 4 h, followed by surface staining for CD8 and Thy1.1 and intracellular staining for IFN-γ and TNF-α using the Cytofix/Cytoperm kit (BD Bioscience). Intracellular staining for T-bet and Eomes was done directly ex vivo. All flow cytometry was analyzed on a FACSCalibur or a FACSCanto (BD).

In vivo cytolytic activity

Cytolytic activity of memory P14 T cells was tested in vivo. Memory P14 T cells were MACS-purified and 2–5×105 P14 T cells (effector cells) transferred into B6. CFSE-labeled spleen cells from H8 mice (CFSE low), which ubiquitously express the LCMV-GP and are therefore recognized by P14 T cells, and B6 cells (CFSE high, as control) were used as targets. Target cells were mixed at a ratio of 1:1 and 2×107 cells injected i.v. into recipient mice. Cytolytic activity was detected at indicated time points in the blood 30.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

The authors thank Dr. H. Pircher for comments on the manuscript. This work was supported by Deutsche Forschungsgemeinschaft grant AI 34/2-1 (P. A.) and the Boehringer Ingelheim Fonds (S. J. K.).

Conflict of interest: The authors declare no financial or commercial conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information
  • 1
    Kaech, S. M., Hemby, S., Kersh, E. and Ahmed, R., Molecular and functional profiling of memory CD8 T cell differentiation. Cell 2002. 111: 837851.
  • 2
    Opferman, J. T., Ober, B. T. and Ashton-Rickardt, P. G., Linear differentiation of cytotoxic effectors into memory T lymphocytes. Science 1999. 283: 17451748.
  • 3
    Sarkar, S., Kalia, V., Haining, W. N., Konieczny, B. T., Subramaniam, S. and Ahmed, R., Functional and genomic profiling of effector CD8 T cell subsets with distinct memory fates. J. Exp. Med. 2008. 205: 625640.
  • 4
    D'Souza, W. N. and Hedrick, S. M., Cutting edge: latecomer CD8 T cells are imprinted with a unique differentiation program. J. Immunol. 2006. 177: 777781.
  • 5
    Rao, R. R., Li, Q., Odunsi, K. and 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.
  • 6
    Araki, K., Turner, A. P., Shaffer, V. O., Gangappa, S., Keller, S. A., Bachmann, M. F., Larsen, C. P. et al., mTOR regulates memory CD8 T-cell differentiation. Nature 2009. 460: 108112.
  • 7
    Badovinac, V. P. and Harty, J. T., Manipulating the rate of memory CD8+ T cell generation after acute infection. J. Immunol. 2007. 179: 5363.
  • 8
    Chang, J. T., Palanivel, V. R., Kinjyo, I., Schambach, F., Intlekofer, A. M., Banerjee, A., Longworth, S. A. et al., Asymmetric T lymphocyte division in the initiation of adaptive immune responses. Science 2007. 315: 16871691.
  • 9
    Curtsinger, J. M., Johnson, C. M. and Mescher, M. F., CD8 T cell clonal expansion and development of effector function require prolonged exposure to antigen, costimulation, and signal 3 cytokine. J. Immunol. 2003. 171: 51655171.
  • 10
    Curtsinger, J. M., Schmidt, C. S., Mondino, A., Lins, D. C., Kedl, R. M., Jenkins, M. K. and Mescher, M. F., Inflammatory cytokines provide a third signal for activation of naive CD4+ and CD8+ T cells. J. Immunol. 1999. 162: 32563262.
  • 11
    Curtsinger, J. M., Valenzuela, J. O., Agarwal, P., Lins, D. and Mescher, M. F., Type I IFNs provide a third signal to CD8 T cells to stimulate clonal expansion and differentiation. J. Immunol. 2005. 174: 44654469.
  • 12
    Joshi, N. S., Cui, W., Chandele, A., Lee, H. K., Urso, D. R., Hagman, J., Gapin, L. et al., Inflammation directs memory precursor and short-lived effector CD8(+) T cell fates via the graded expression of T-bet transcription factor. Immunity 2007. 27: 281295.
  • 13
    Aichele, P., Unsoeld, H., Koschella, M., Schweier, O., Kalinke, U. and Vucikuja, S., CD8 T cells specific for lymphocytic choriomeningitis virus require type I IFN receptor for clonal expansion. J. Immunol. 2006. 176: 45254529.
  • 14
    Kolumam, G. A., Thomas, S., Thompson, L. J., Sprent, J. and Murali-Krishna, K., Type I interferons act directly on CD8 T cells to allow clonal expansion and memory formation in response to viral infection. J. Exp. Med. 2005. 202: 637650.
  • 15
    Keppler, S. J., Theil, K., Vucikuja, S. and Aichele, P., Effector T-cell differentiation during viral and bacterial infections: Role of direct IL-12 signals for cell fate decision of CD8(+) T cells. Eur. J. Immunol. 2009. 39: 17741783.
  • 16
    Wilson, D. C., Matthews, S. and Yap, G. S., IL-12 signaling drives CD8+ T cell IFN-gamma production and differentiation of KLRG1+effector subpopulations during Toxoplasma gondii infection. J. Immunol. 2008. 180: 59355945.
  • 17
    Wiesel, M., Kratky, W. and Oxenius, A., Type I IFN substitutes for T cell help during viral infections. J. Immunol. 186: 754763.
  • 18
    Pearce, E. L. and Shen, H., Generation of CD8 T cell memory is regulated by IL-12. J. Immunol. 2007. 179: 20742081.
  • 19
    Xiao, Z., Casey, K. A., Jameson, S. C., Curtsinger, J. M. and Mescher, M. F., Programming for CD8 T cell memory development requires IL-12 or type I IFN. J. Immunol. 2009. 182: 27862794.
  • 20
    Sallusto, F., Lenig, D., Forster, R., Lipp, M. and Lanzavecchia, A., Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 1999. 401: 708712.
  • 21
    Schmidt, C. S. and Mescher, M. F., Peptide antigen priming of naive, but not memory, CD8 T cells requires a third signal that can be provided by IL-12. J. Immunol. 2002. 168: 55215529.
  • 22
    Pearce, E. L., Mullen, A. C., Martins, G. A., Krawczyk, C. M., Hutchins, A. S., Zediak, V. P., Banica, M. et al., Control of effector CD8+T cell function by the transcription factor Eomesodermin. Science 2003. 302: 10411043.
  • 23
    Prlic, M. and Bevan, M. J., Exploring regulatory mechanisms of CD8+ T cell contraction. Proc. Natl. Acad. Sci. USA 2008. 105: 1668916694.
  • 24
    Harty, J. T. and Badovinac, V. P., Shaping and reshaping CD8+ T-cell memory. Nat. Rev. Immunol. 2008. 8: 107119.
  • 25
    Magram, J., Connaughton, S. E., Warrier, R. R., Carvajal, D. M., Wu, C. Y., Ferrante, J., Stewart, C. et al., IL-12-deficient mice are defective in IFN gamma production and type 1 cytokine responses. Immunity 1996. 4: 471481.
  • 26
    Muller, U., Steinhoff, U., Reis, L. F., Hemmi, S., Pavlovic, J., Zinkernagel, R. M. and Aguet, M., Functional role of type I and type II interferons in antiviral defense. Science 1994. 264: 19181921.
  • 27
    Pinschewer, D. D., Perez, M., Jeetendra, E., Bachi, T., Horvath, E., Hengartner, H., Whitt, M. A. et al., Kinetics of protective antibodies are determined by the viral surface antigen. J. Clin. Invest. 2004. 114: 988993.
  • 28
    Zenewicz, L. A., Foulds, K. E., Jiang, J., Fan, X. and Shen, H., Nonsecreted bacterial proteins induce recall CD8 T cell responses but do not serve as protective antigens. J. Immunol. 2002. 169: 58055812.
  • 29
    Battegay, M., Cooper, S., Althage, A., Banziger, J., Hengartner, H. and Zinkernagel, R. M., Quantification of lymphocytic choriomeningitis virus with an immunological focus assay in 24- or 96-well plates. J. Virol. Methods 1991. 33: 191198.
  • 30
    Aichele, P., Brduscha-Riem, K., Oehen, S., Odermatt, B., Zinkernagel, R. M., Hengartner, H. and Pircher, H., Peptide antigen treatment of naive and virus-immune mice: antigen-specific tolerance versus immunopathology. Immunity 1997. 6: 519529.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

Detailed facts of importance to specialist readers are published as ”Supporting Information”. Such documents are peer-reviewed, but not copy-edited or typeset. They are made available as submitted by the authors.

FilenameFormatSizeDescription
eji_201141537_sm_SupplInfo.pdf197KSupplInfo

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.