IFN-β modulates the response to TLR stimulation in human DC: Involvement of IFN regulatory factor-1 (IRF-1) in IL-27 gene expression


  • Maria Elena Remoli,

    1. Department of Infectious, Parasitic and Immune-mediated Diseases, Istituto Superiore di Sanità, Rome, Italy
    Search for more papers by this author
    • M. E. R. and V. G. equally contributed to this work.

  • Valérie Gafa,

    1. Department of Infectious, Parasitic and Immune-mediated Diseases, Istituto Superiore di Sanità, Rome, Italy
    Search for more papers by this author
    • M. E. R. and V. G. equally contributed to this work.

  • Elena Giacomini,

    1. Department of Infectious, Parasitic and Immune-mediated Diseases, Istituto Superiore di Sanità, Rome, Italy
    Search for more papers by this author
  • Martina Severa,

    1. Department of Infectious, Parasitic and Immune-mediated Diseases, Istituto Superiore di Sanità, Rome, Italy
    Search for more papers by this author
  • Roberto Lande,

    1. Department of Infectious, Parasitic and Immune-mediated Diseases, Istituto Superiore di Sanità, Rome, Italy
    Search for more papers by this author
  • Eliana M. Coccia

    Corresponding author
    1. Department of Infectious, Parasitic and Immune-mediated Diseases, Istituto Superiore di Sanità, Rome, Italy
    • Department of Infectious, Parasitic and Immune-mediated Diseases, Istituto Superiore di Sanità, 00161 Rome, Italy, Fax: +39-06-49387115
    Search for more papers by this author


Type I IFN are cytokines which play a central role in host resistance to viral or microbial infections and are important components linking innate and adaptive immunity. We and others have previously demonstrated that the production of IFN-β by DC following bacterial infections or TLR triggering influences, in an autocrine manner, their maturation. In this study, we investigated whether IFN-β release modulates the phenotype of the immature DC and their response to a subsequent TLR stimulation. The induction of CD86, HLA-DR, CD38 and B7H1 and the absence of CCR7 and CD83 expression upon IFN-β treatment suggest that IFN-β-primed DC remain at the site of infection acquiring an activated phenotype. These results prompted us to investigate the response of IFN-β-primed DC to TLR stimulation. While IFN-β pretreatment increases slightly the expression of maturation markers in TLR2- or TLR4-stimulated DC, it is able to modulate selectively the secretion of inflammatory and immuno-regulating cytokines. Interestingly, IL-27p28 subunit was induced by IFN-β alone or during LPS-induced maturation of DC in a type I IFN-dependent manner through IFN regulatory factor-1 (IRF-1) activation. Taken together, our results shed light on the capacity of IFN-β to finely tune DC response to invading pathogens.


cytometric bead assay


EBV-induced gene 3


IFN regulatory factor-1


IFN-stimulated response element


monocyte-derived DC




small interfering RNA


DC are a central component of the immune system for their extraordinary capacity to initiate and modulate the immune responses elicited upon recognition of infectious agents. Indeed, immature DC play a crucial role in the surveillance of peripheral sites by migrating through all the tissues and actively taking up foreign antigens 1, 2. Once in contact with the antigen, immature DC utilize the pattern recognition receptors to specifically recognize pathogen-related molecules. TLR are the best characterized class of pattern recognition receptors in the mammalian species 3, 4. Immediately after contact with and recognition of the pathogens through TLR, DC undergo a process termed maturation modifying their phenotypical features and leading to production of cytokines that regulate the immune responses acting sequentially in different micro-environments and on different leukocyte populations 1.

Among the released cytokines, type I IFN play a crucial immuno-regulatory role for their ability to modulate several DC functions 5. Besides representing one of the major producer of type I IFN, DC are also key targets of their actions. By binding to specific transmembrane receptors, they trigger a response that culminates in the induction of a large number of genes modulating and linking the innate and the adaptive immune responses 6. Several reports have shown that the autocrine production of type I IFN is required for both activation and maturation of DC induced by TLR3 and TLR4 triggering or by viral infection 79. Moreover, type I IFN-induced phenotypic changes are associated with an increased ability of DC to stimulate T cell proliferation 5, 9, 10.

In the past, we and other groups have demonstrated that the activation of certain TLR or the infection with different pathogens can lead to the transcriptional regulation of type I IFN subtypes in APC 1116. Since TLR4, triggered by its ligand LPS, induces IFN-β as a ‘signature’ cytokine 11, 17, 18, in recent studies we have investigated the IFN-β-related transcriptional and subsequent functional consequences that occur in human monocyte-derived DC (MoDC) along LPS-induced maturation 19, 20. These data showed that TLR4-mediated type I IFN release activates specific transcription programs in DC amplifying the expression of pathogen sensors to correctly and promptly respond to a bacterial – as well as viral – infection.

In the present study we have investigated the ability of IFN-β to influence DC phenotype and their response to TLR stimulation. We first investigated whether IFN-β alone or in combination with TLR2 and TLR4 agonists differentially modulates maturation of human MoDC. Having found that IFN-β alone activates immature MoDC and the pretreatment with IFN-β slightly increases the expression of maturation markers in MoDC stimulated with LPS or the bacterial triacylated lipopeptide N-palmitoyl-S-[2,3-bis(palmitoyloxy)-(2RS)- propyl]-(R)-cysteine (Pam3Cys), we investigated its effect on the release of inflammatory and immuno-regulating cytokines. An enhancement of pro-inflammatory cytokines, such as TNF and IL-6, was observed in maturing MoDC pretreated with IFN-β. Conversely, IFN-β treatment does not affect the capacity of MoDC to release IL-10 and IL-12 upon TLR2 stimulation, while it significantly reinforces the production of both cytokines from LPS maturing cells.

Given the important role of the novel IL-12 family members in the development of cell-mediated immunity 21, the effect of IFN-β was investigated on the expression of IL-23, a heterodimer consisting of the p19 and p40 subunit, and IL-27, composed of the subunits p28 and EBV-induced gene 3 (EBI3). No significant modification in the expression of IL-23p19 subunit suggested that IL-23 production was not strengthened by IFN-β treatment of maturing MoDC. Conversely, IL-27p28 subunit was induced by IFN-β alone or during LPS-induced maturation of MoDC in a type I IFN-dependent manner. The induction of IL-27p28 subunit in LPS-maturing MoDC was mainly a consequence of the transcriptional activity of IFN regulatory factor-1 (IRF-1), whose binding site was located within p28 subunit promoter. Our results, therefore, highlight the ability of IFN-β alone or in combination with TLR to finely tune DC response to invading pathogens.


Activation of human MoDC following IFN-β treatment

We previously described that DC could release IFN-β following Mycobacterium tuberculosis infection or TLR4 agonist stimulation 11, 12. To understand whether IFN-β released by infected DC could modify the status of immature DC, we evaluated by flow cytometry analysis the phenotype of MoDC following IFN-β treatment. We compared the effect of IFN-β with those induced by TNF and IL-1β, two pro-inflammatory cytokines released following microbial infection of DC (Fig. 1).

Figure 1.

Analysis of maturation markers and CCR7 expression on DC treated with TNF/IL-1β and IFN-β. MoDC were treated with IFN-β and TNF/IL-1β for 24 h at 37°C. DC from a total of six different blood donors were used. (A) The expression of the CD83, CD86, HLA-DR, B7H1 and CD38 was evaluated by flow cytometry. The results are the mean of MFI ± SE values. (B) Percentages of CCR7-positive cells ± SE were also evaluated by flow cytometry.

Both IFN-β and TNF/IL-1β treatment induced a significant increase of CD86 and HLA-DR expression while no effect was observed on CD83 expression (Fig. 1A). Moreover, IFN-β-treated MoDC displayed a selective increased expression of CD38 and B7H1, two IFN-inducible markers (Fig. 1A) 20, 22, whereas, as expected, their expression was not observed upon TNF/IL-1β treatment. Interestingly, IFN-β was unable to promote the expression of CCR7, partly induced on TNF/IL-1β-treated MoDC (14% of the examined cells) (Fig. 1B). Altogether, these results indicate that IFN-β-conditioned DC display an activated phenotype although are not able to leave the site of infection because of the lack of CCR7, the chemokine receptor required for the migration of mature MoDC into secondary lymphoid organs 23.

Effects of IFN-β pretreatment on MoDC maturation induced by TLR4 and TLR2 triggering

Since it is conceivable that IFN-β-conditioned DC, which do not migrate but reside in the infected tissues, may encounter the invading pathogen or its products soon after, we investigated their response to TLR triggering. Among the TLR, we decided to investigate the effects induced by TLR4 and TLR2 agonists, LPS and Pam3Cys respectively, since they exhibit a differential capacity to induce type I IFN production 18. MoDC were cultured for 4 h in the presence or absence of IFN-β and then stimulated with LPS and Pam3Cys for 24 h. The expression of CD83, CD86, CD38 and B7H1 surface markers was evaluated by flow cytometry analysis (Fig. 2).

Figure 2.

Analysis of IFN-β pretreatment effect on the maturation of DC stimulated with LPS and Pam3Cys. MoDC were pretreated for 4 h with IFN-β before the addition of LPS or Pam3Cys. After 24 h, cells were stained and analyzed by flow cytometry for the expression of CD83, CD86, B7H1, and CD38. The results are the means of MFI ± SE values from six different blood donors.

As expected, LPS and Pam3Cys treatment induced CD83, CD86 and B7H1 expression in MoDC. As previously demonstrated 20, CD38 expression was up-regulated only during LPS-induced maturation of MoDC in a type I IFN-dependent manner. IFN-β did not impair the capacity of DC to undergo full maturation; conversely, it slightly increased the LPS-induced stimulation of CD86, B7H1 and CD38 expression, while it did not modulate CD83 expression. Interestingly, a more pronounced effect of IFN-β pretreatment was observed on CD83 and CD86 expression in Pam3Cys-stimulated MoDC compared to LPS-stimulated ones. However, the Pam3Cys-mediated induction of B7H1 was only poorly increased in MoDC pre-exposed to IFN-β, while the IFN-β-induced CD38 expression was not further stimulated in IFN-β-primed DC upon TLR2 triggering. These results suggest that, although able to activate DC, IFN-β poorly increases the expression of maturation markers in TLR2- or TLR4-stimulated MoDC.

Effects of IFN-β pretreatment on cytokine production from TLR4- and TLR2-stimulated MoDC

We then studied whether IFN-β could modulate the profile of inflammatory and immune-regulating cytokines released by TLR4- and TLR2-stimulated MoDC. Supernatants were harvested from MoDC stimulated for 24 h with LPS and Pam3Cys with or without 4 h IFN-β pretreatment and analyzed by cytometric bead assay (CBA) (Fig. 3). The presence of IFN-β further increased the secretion of IL-6 by LPS- and Pam3Cys-stimulated DC, while it exerted only a modest effect on TNF production. Moreover, a significant induction of IL-10 (2.5-fold) and IL-12 (5-fold) was found in IFN-β-pretreated DC upon LPS stimulation compared to cells that were not pretreated with IFN-β. Nevertheless, this enhanced production of IL-10 was not observed in DC stimulated with Pam3Cys. Furthermore, the presence of IFN-β did not rescue the capacity of Pam3Cys-stimulated DC to secrete IL-12. Overall, these results indicate that IFN-β pretreatment does not confer to DC the ability to produce a different profile of cytokines following stimulation with specific TLR agonists but it rather reinforces the TLR-induced cytokine production.

Figure 3.

Cytokine production by IFN-β-pretreated DC stimulated with LPS or Pam3Cys. MoDC were stimulated as described in Fig. 2 and cell culture supernatants were collected. The amounts of IL-12, IL-10, TNF and IL-6 were measured in the supernatants by CBA. The results represent the means ± SE of six separate experiments.

Effects of IFN-β on the expression of IL-12 family members in mature MoDC

Having found that IFN-β treatment is able to strengthen IL-12p70 secretion, we asked whether this effect was dependent on transcriptional or translational events. Moreover, we also asked the question whether the expression of the recently characterized members of the IL-12 family, such as IL-23 and IL-27, could be modulated by IFN-β in MoDC upon TLR stimulation. Therefore, the expression of IL-12p35, IL-12p40, IL-23p19, IL-27EBI3 and IL-27p28 subunits was evaluated by real-time RT-PCR (Fig. 4A–E). Total RNA was extracted at 4 h after LPS or Pam3Cys treatment with or without a 4-h IFN-β priming. As shown in Fig. 4A, B, TLR4 triggering induced the expression of both IL-12p35 and IL-12p40 subunits while only the induction of this latter subunit was observed in TLR2-stimulated DC. Interestingly, IFN-β pretreatment was able to further induce the expression of IL-12p35 subunit in LPS-stimulated DC while it did not rescue the expression in the Pam3Cys-treated cells. This result confirms the CBA data shown in Fig. 3 and also indicates that the increased expression of bioactive IL-12p70 upon TLR4 stimulation of IFN-β-primed DC was mainly the result of an enhanced IL-12p35 gene expression.

Figure 4.

Expression of IL-12 family members in IFN-β-pretreated DC upon stimulation with LPS or Pam3Cys. (A–E) Total RNA was extracted from MoDC treated for 4 h with the TLR4 and TLR2 agonists with or without 4-h pretreatment with IFN-β, or stimulated with IFN-β alone. (F) Total RNA was isolated from MoDC following 4 h LPS treatment in presence or absence of anti-IFN Ab. (G) Total RNA was extracted from MoDC incubated for 4 h with the indicated concentrations of IFN-β and LPS, or pretreated for 4 h with IFN-β (200 pM) before the addition of different doses of LPS. Real-time RT-PCR was performed to measure the expression of IL-12 family members as indicated. The results shown are from one of three experiments that yielded similar results.

Next, we evaluated the effect of IFN-β treatment on the TLR-stimulated expression of IL-23, a heterodimer consisting of the p19 and p40 subunits. Since the expression of IL-23p19 subunit was not modulated by IFN-β pretreatment, we inferred that IL-23 expression is not a target of IFN-β effect (Fig. 4C). Conversely, the analysis of IL-27 subunits, p28 and EBI3, suggested that IFN-β profoundly stimulates the expression of this cytokine even in immature DC (Fig. 4D, E). Indeed, while IFN-β did not increase the constitutive expression of EBI3, it induced IL-27p28 expression by approximately 10-fold. Moreover, we also observed that LPS was able to induce IL-27p28 expression mainly through the release of autocrine IFN-β, since its neutralization by the addition of anti-IFN Ab clearly reduced IL-27p28 expression (Fig. 4F). On the contrary, Pam3Cys treatment, which fails to induce type I IFN production, is not able to stimulate IL-27p28 expression. Overall, the analysis of the IL-12 family members reveals the ability of IFN-β to differentially regulate their expression in both immature and TLR-matured DC.

Dose-response experiments were performed to further characterize the IL-27p28 inducibility by IFN-β and LPS. As shown in Fig. 4G, when using higher dose of IFN-β (1000 pM), the IL-27p28 expression was induced at the same extent to that observed with the dose generally used (200 pM), while a less pronounced expression was found with lower doses, such as 20 and 2 pM. Similarly, a modest induction of IL-27p28 expression was obtained with 1 ng/mL of LPS, a dose of LPS which is not able to promote DC maturation. However, a strengthening of IFN-β effect was obtained when 10 and 1 ng/mL of LPS were used to stimulate DC. Moreover, the level of IL-27p28 found in IFN-β-primed DC is reinforced by Pam3Cys treatment, suggesting the existence of cooperative signals acting when sub-optimal stimuli are used to induce the expression of IL-27p28.

Role of IRF-1 in IL-27p28 expression

To understand the molecular mechanisms involved in the transcriptional regulation of IL-27p28 mediated by IFN-β, a region spanning 450 bp upstream of the transcription starting site of the human p28 promoter was analyzed by the Transcription Factor Binding Site prediction program MatInspector. Among the multiple predicted transcription factor binding sites, NF-κB (–393/–381) and IRF-1/IFN-stimulated response element (ISRE) (–61/–45) binding sites were found (Fig. 5A).

Figure 5.

Transcriptional regulation of IL-27p28 gene. (A) Schematic map of ∼450 bp of the human p28 promoter. (B–C) DC were treated with IFN-β, Pam3Cys or LPS for 45 min to detect NF-κB and for 4 h to analyze IRF-1 binding activity, respectively. Nuclear extracts (15 μg) were used in EMSA with the indicated radiolabeled oligonucleotides. Supershift assay (SS) was performed using Ab specific for IRF-1, p50 and p65 NF-κB subunits, where indicated. (D) Nuclear cell lysates (15 μg) obtained from cells treated for 4 h as described in (B) were analyzed by Western blotting using a specific anti-IRF-1 and anti-TFIIH Ab. This analysis was repeated two additional times using DC from a total of three different blood donors.

Therefore, to investigate whether these transcription factors might play a role in the transcriptional regulation of IL-27p28, EMSA was performed using double-stranded oligonucleotides corresponding to the binding regions for NF-κB (–393/–381) and IRF-1/ISRE (–61/–45). Nuclear extracts were prepared from DC stimulated with IFN-β, LPS and Pam3Cys for 45 min to study NF-κB activity and for 4 h to analyze IRF-1 activity. A complex binding the p28 κB site was detected in extracts from DC stimulated with LPS and Pam3Cys (Fig. 5B). The identity of the induced complex was assessed using Ab raised against NF-κB p50 and p65 subunits. Moreover, when the oligonucleotides corresponding to the putative IRF-1/ISRE were used, a clear induction of DNA-binding activity was observed in LPS- and IFN-β-treated cells (Fig. 5C). Supershift experiment demonstrated that this complex contained IRF-1 since it was retarded by the addition of anti-IRF-1 Ab but not by Ab directed towards IRF-7 (data not shown). Conversely, no IRF-1 binding was detected in cell lysates from Pam3Cys-treated DC, indicating that the release of type I IFN is required for IRF-1 expression. The expression of IRF-1 in LPS-, Pam3Cys- and IFN-β-treated cells was also confirmed by Western blot analysis (Fig. 5D).

Taken together, these results indicate that the IFN-β-mediated activation of IRF-1 is required for expression of IL-27p28, which is likely strengthened by the simultaneous NF-κB activity. However, the lack of IL-27p28 subunit expression in Pam3Cys-treated DC indicates that the mere activation of NF-κB is not sufficient for IL-27p28 expression.

To confirm the essential role of IRF-1 in the regulation of IL-27p28 gene expression in a more physiologically relevant setting, we knocked down IRF-1 endogenous expression in LPS-treated DC using the small interfering RNA (siRNA) silencing technology. We were able to achieve substantial reduction in IRF-1 mRNA in LPS-treated DC transfected with an IRF-1 siRNA while IRF-1 expression was not affected in DC transfected with a scrambled control siRNA (Fig. 6A). Accordingly, the LPS-induced IL-27p28 expression was clearly reduced when IRF-1 was silenced by specific siRNA (Fig. 6B). The expression of IL-12p40 mRNA was evaluated as control of the specificity of siRNA (Fig. 6C). No effect on the expression of IL-12p40 was indeed observed when both IRF-1 siRNA or control siRNA were used.

Figure 6.

Effect of IRF-1 silencing on IL-27p28 gene expression. Total RNA was extracted from DC nucleofected with 1 μM of siIRF-1 or scrambled control siRNA oligonucleotides and treated for 16 h with LPS. IRF-1 (A), IL-27p28 (B) and IL-12p40 (C) expression was evaluated by real-time RT-PCR. (D) Total RNA was extracted from MoDC treated for 4 h with LPS and Pam3Cys with or without 4 h pretreatment with IFN-γ or stimulated with IFN-γ alone. The results shown are from a representative experiment out of three performed that yielded similar results.

Since IRF-1 is a transcription factor induced by IFN-γ 24, we asked whether also IFN-γ alone is able to stimulate the expression of IL-27p28 in DC and whether it can also synergize with TLR2 and TLR4 signaling in the induction of this subunit. Therefore, total RNA was extracted at 4 h after LPS or Pam3Cys treatment with or without a 4-h IFN-γ priming and the expression of IL-27p28 subunit was evaluated by real-time RT-PCR (Fig. 6D). As expected, IFN-γ per se induces a robust expression of IL-27p28 subunit and strengths the expression induced by LPS. Interestingly, the level of IL-27p28 expression induced by IFN-γ was reinforced following Pam3Cys treatment as it was observed in IFN-β-primed DC. These data provide further evidence that IRF-1 plays a key role in the regulation of IL-27p28 induced by the treatment of human DC with IFN-β and IFN-γ.


The key function of DC in driving a specific immune response relies also on their potential to produce different cytokines in response to specific microbial signals 2. We and others 12, 1416 have previously demonstrated a selective production of IFN-β by myeloid DC following microbial infections or TLR triggering as opposed to induction of all subtypes upon virus infection. Although the induction of multiple IFN-α subtypes plays a role in the antiviral response, it is not clear yet whether in the context of microbial infections the selective IFN-β expression exerts a direct antimicrobial effect or whether it indirectly influences the innate and adaptive immune response 15, 25. In this context, we were interested in studying the effects of IFN-β in the modulation of DC response subsequently to a stimulation with TLR agonists. Indeed, following pathogen infection, the rapid release of IFN-β might act in an autocrine fashion on the producing DC but also in paracrine manner on the surrounding DC, which have not interacted yet with the pathogen, thus influencing the infection outcome.

It has previously been reported that a single stimulation with type I IFN induces in DC a state of partial maturation characterized by the up-regulation of costimulatory and MHC class I and II molecules 5, 9, 10, 26. In line with these data, we found that IFN-β per se induces the activation of MoDC characterized by an increased expression of CD86, HLA-DR and B7H1 but not of CD83. Moreover, we also found that IFN-β does not induce the expression of CCR7, indicating that IFN-β-conditioned DC acquire an activated phenotype although they do not migrate to the lymph nodes. Based on these data, we asked whether IFN-β could influence the response of MoDC to a subsequent TLR stimulation. A modest up-regulation of the maturation markers was observed in DC pretreated with IFN-β upon stimulation with TLR2 and TLR4 agonists compared to the expression found in DC stimulated with the only TLR agonists. These results confirmed the data by Walker and colleagues 26 showing that the pretreatment with IFN-β exerts a poor effect on TLR-stimulated DC.

DC not only present microbial antigens to T cells but also provide them information about the features of the pathogen by the release of a specific panel of cytokines directing the development of an appropriate immune response 2. Thus, we investigated the ability of IFN-β to modulate the release of inflammatory (IL-6 and TNF) and immuno-regulating cytokines (IL-10 and IL-12 family) in LPS- or Pam3Cys-stimulated MoDC. An enhancement in the release of pro-inflammatory cytokines was observed in IFN-β-conditioned DC following TLR2 and TLR4 stimulation, indicating that IFN-β promotes an inflammatory milieu into the site of pathogen entry. Conversely, a differential response to the IFN-β priming was observed in the case of the expression of immuno-regulatory cytokines. In particular, IFN-β treatment acted synergistically with TLR4 in the induction of IL-12p70 and IL-10 production while it was not able to modulate the synthesis of these cytokines in TLR2-stimulated DC. This differential response of maturing DC to IFN-β well correlates with previous results showing the requirement of an autocrine loop of type I IFN for bioactive IL-12p70 and IL-10 secretion by myeloid DC stimulated by TLR4 27, 28. The enhanced production of IL-12p70 from IFN-β-primed DC upon TLR4 stimulation is mainly dependent on an increased transcription of IL-12p35 gene. Thus, it is conceivable that IFN-β-mediated induction of IRF-1 may contribute to the expression of IL-12p35 gene which contains within its promoter region an IRF-E binding site. Keeping with this hypothesis, it has recently been demonstrated that following TLR stimulation both IRF-1 and IRF-3 may regulate IL-12p35 expression through their recruitment to this IRF-E binding element 29, 30.

Other members of the IL-12 family, IL-23 and IL-27, have recently been described to play important roles in the regulation of T cell response 21, 31, 32. Since the effects of type I IFN on the expression of IL-23 and IL-27 have been investigated to a limited extent so far 33, we decided to extend our analysis also to the release of these molecules in TLR-stimulated DC. Whilst no modification in the expression of IL-23p19 subunit was found in maturing MoDC pretreated with IFN-β, IL-27p28 gene expression was found to be highly induced in a type I IFN-dependent manner during LPS-induced maturation of MoDC. Moreover, a clear expression of IL-27p28 subunit was also found in DC treated with IFN-β alone. Thus, since the IL-27EBI3 subunit is constitutively expressed, we can suggest that IFN-β per se might stimulate the production of IL-27 in activated DC by only p28 induction. Since a commercial ELISA assay is not yet available, the amount of bioactive IL-27 released by IFN-β-treated DC remains to be determined. Interestingly, the IL-27p28 induction by IFN-β can be further increased by the addition of sub-optimal stimuli, such as low dose of LPS or TLR2 ligand, suggesting the existence of a cross-talk between the IFN- and TLR-stimulated intracellular pathways leading to the induction of IL-27p28 gene expression.

The induction of IL-27p28 subunit in LPS-maturing MoDC was mainly a consequence of the transcriptional activity of IRF-1, whose binding site was located within p28 subunit promoter 34. Indeed, a dramatic decrease in LPS-induced IL-27p28 subunit expression was caused by depletion of IRF-1 by siRNA or by the neutralization of autocrine IFN-β production by LPS-matured DC. Consistent with these findings TLR2 triggering failed to do so, likely for its inability to stimulate IFN-β production 18, while the treatment with IFN-γ, a primary inducer of IRF-1 24, turns on IL-27p28 gene expression. Our results are in line with recent data by J. Liu and colleagues 34, who reported that IFN-γ alone can stimulate IL-27p28 synthesis through IRF-1 activity in murine macrophages although a synergism with NF-κB c-Rel induced by LPS was also observed.

However, we should also consider that other IRF members may contribute to IL-27p28 gene expression as previously demonstrated for the p35 subunit of IL-12. Indeed, it has recently been described that an IRF-E site within IL-27p28 promoter is also recognized by IRF-3 and is important for IL-27p28 gene regulation in DC stimulated by TLR4 agonist 35. Thus, it is possible that while both IRF-1 and IRF-3 may cooperate in the regulation of IL-27p28 expression following TLR4 stimulation, the ability of IFN-β to stimulate the expression of this subunit is primarily dependent on the activation of IRF-1. Taken together these data shed light on a novel target of IRF-1 activity along the maturation of human DC and further provide evidence for a close relationship between IRF transcription factor activity and the expression of IL-12 family members in DC.

What is the role of IFN-β-stimulated release of IL-27 observed in activated or maturing DC? IL-27 is still an enigmatic cytokine that has been described to exert both pro- and anti-inflammatory effects on the immune response 21, 31. Indeed, it can promote the early phases of the Th1 response and suppress Th2 and Th17 differentiation although there is growing evidence that, upon exposure to microbes, IL-27 plays a crucial role in limiting the excessive inflammation deleterious to the host 3640. Thus, under these conditions the anti-inflammatory properties of IL-27 appear to dominate its ability to skew T cell polarization.

Consistent with this notion, we can envisage the following scenario. Upon the interaction with certain microbes, DC release IFN-β, which may influence the immune response acting at different levels. At early stage of infection, IFN-β may represent an alarm signal sent from infected cells to local DC in order to promote their activation, which enables them to promptly and strongly respond to a subsequent microbial infection. Indeed, once reached the lymph node, IFN-β-primed DC may strongly promote Th1 cell response by virtue of their enhanced IL-12 and IL-27 secretion. However, at later stage of infection when the immune response has to be switched off since the pathogen has been eliminated, the release of IFN-β may induce the expression of IL-27 in the DC present in the infected tissue. The local release of IL-27 may, in turn, reduce the inflammatory response acting on T cell activity to prevent a deleterious excessive inflammation. Taken together our findings suggest a model in which IFN-β may play a role in orchestrating the positive and negative signals required to generate and shape a protective inflammatory response occurring after the interaction of DC with an invading pathogen.

Materials and methods

Generation of MoDC

MoDC were prepared as previously described 12. Briefly, monocytes were purified by positive sorting using anti-CD14-conjugated magnetic microbeads (Miltenyi, Bergisch Gladbach, Germany). DC were generated by culturing monocytes with 25 ng/mL GM-CSF and 1000 U/mL IL-4 (R&D Systems, Abingdon, UK) for 5 days. MoDC were starved from IL-4 and GM-CSF for 20 h before their stimulation. Cytokine deprivation did not affect MoDC survival rate, which was >90% as evaluated by phase-contrast light microscopic examination and by the Trypan blue dye exclusion method.


DC were generally stimulated with 200 pM of IFN-β (Avonex, Biogen Inc., Cambridge, MA) or 100 ng/mL of LPS from Escherichia coli 0111:B4 (Sigma-Aldrich, St. Louis, MO) unless otherwise specified, 10 ng/mL IFN-γ, 100 ng/mL of TNF and 10 ng/mL of IL-1β (PeproTech EC Ltd., London, UK) and 2 μg/mL of Pam3Cys (Alexis Corporation, San Diego, CA). Sheep antiserum raised against human leukocyte IFN, a kind gift from Gilles Uzé (CNRS UMR5124, Montpellier, France), was used at a 1:100 dilution.

Flow cytometry analysis

mAb specific for CD1a, CD14, CD86, CD83, HLA-DR, B7H1, CD38 and CCR7 as well as IgG1 (BD Bioscience PharMingen, San Diego, CA) were used as pure Ab or direct conjugates to FITC or PE. Goat anti-mouse IgG F(ab’)2-FITC was used as secondary Ab for B7H1 and CCR7. After incubation with mAb, 105 cells were washed and fixed with 2% paraformaldehyde before analysis on a FACScan using CellQuest software (Becton Dickinson, Mountain View, CA). A total of 5000 cells were analyzed per sample. The expression of the cell surface molecules was evaluated using the MFI after subtraction of the values of the isotype controls or by percentage of positive cells.

Cytokine determinations

Supernatants from MoDC cultures were harvested and stored at –80°C. IL-6, TNF, IL-10 and IL-12p70 were measured with the human inflammation CBA (BD Bioscience, Pharmingen). The assay was conducted according to manufacturer's instructions.

RNA isolation and real-time RT-PCR quantification

RNA was extracted with RNeasy Mini kit (Qiagen Inc., Valencia, CA) according to the manufacturer's instructions. Reverse transcriptions and quantitative PCR assays were performed as previously described 11. The sequences of the primer pairs used for the quantification of GAPDH have previously been described 11. Primer pairs for IL-p40, IL-12p35, IL-23p19, IL-27p28 and EBI3 were described by Nagai 41. The primers used for mRNA quantification of IRF-1 were: IRF-1 forward: 5′-GATGCTTCCACCTCTCACCAAG-3′; IRF-1 reverse: 5′-CCTGCTCCACCTCCAAGTCC-3′.

Western blot analysis

Western blots were performed as previously described 12. Briefly, 15 μg of nuclear cell extracts were separated by 10% SDS-PAGE gel and blotted onto nitrocellulose membranes. Blots were incubated with IRF-1 Ab (Santa Cruz Biotechnology, Santa Cruz, CA) and reacted with anti-rabbit HRP-coupled secondary Ab (Amersham Pharmacia Biotech, Little Chalfont, UK) using an ECL system. Blots, after stripping, were incubated with TFIIH Ab (Santa Cruz) as loading control.


Nuclear cell extracts were prepared as previously described 12. To measure the association of DNA-binding proteins with different DNA sequences, synthetic double-stranded oligonucleotides were end-labeled with [γ-32P]ATP by T4 polynucleotide kinase. For the analysis of IRF-1 and NF-κB complexes, nuclear cell lysates (15 μg) were used in EMSA experiments as previously described 12. The sequence of oligonucleotides were as follows: κB p28 forward: 5′-GGTGGGCAATCCCCTTACCT-3′, κB p28 reverse: 5′-AGGTAAGGGGATTGCCCACC-3′; IRF-1/ISRE p28 forward: 5′-GGACGGAAAGTGAAACCGGGC-3′; IRF-1/ISRE p28 reverse: 5′-GCCCGGTTT-CACTTTCCGTCC-3′. For supershift analysis, 1 μg of anti-IRF-1, anti-p50 and anti-p65 Ab (Santa Cruz Biotechnology) were added to the reaction.

In silico analysis of human p28 promoter

The analysis of the human promoter of p28 gene was done using the Transcription Factor Binding Site prediction program MatInspector (online at http://www.genomatix.de/matinspector.html) as previously described 12. The analyzed sequence was a region spanning 450 bp upstream the transcription starting site.

siRNA nucleofection

MoDC were nucleofected with 1 μM of siIRF-1 (catalog number L-011704-00-0005; Dharmacon, Lafayette, CO) or scrambled control siRNA oligonucleotides (catalog number D-001810-10-05; Dharmacon) using the Human Dendritic Cell Nucleofector Kit according to the manufacturer's instruction (Amaxa, Cologne, Germany). Two hours after transfection DC were treated with LPS. After 16 h, RNA was extracted and analyzed for IRF-1, IL-27p28 and IL12-p40 expression by real-time RT-PCR.

Statistical analysis

Statistical analysis was calculated using the Wilcoxon matched-pairs test (p<0.05 was considered significant).


This work was supported by grants of the Istituto Superiore di Sanità (5AD/F2) and Italian Multiple Sclerosis Foundation (2005/R/7) to E.M.C. We are grateful to Sandra Pellegrini (Pasteur Institute, Paris, France) for helpful discussion and critical reading of the manuscript. We would like also to thank Gilles Uzé (CNRS UMR5124 Montpellier, France) and Federico Giannoni (Department of Infectious, Parasitic and Immune-mediated Diseases, Istituto Superiore di Sanità, Rome, Italy) for providing reagents, and Eugenio Morassi for preparing drawings.


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


  1. 1


  2. 2


  3. 3


  4. 4


  5. 5


  6. 6