Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Sep 1.
Published in final edited form as: Eur J Immunol. 2013 Jul 3;43(9):10.1002/eji.201243124. doi: 10.1002/eji.201243124

Cooperative intracellular interactions between MyD88 and TRIF are required for CD4 T cell TH1 polarization with a synthetic TLR4 agonist adjuvant

Mark T Orr 1, Malcolm S Duthie 1, Hillarie Plessner Windish 1, Elyse A Lucas 1, Jeff Guderian 1, Thomas E Hudson 1, Narek Shaverdian 1, Joanne O’Donnell 1, Anthony L Desbien 1, Steven G Reed 1, Rhea N Coler 1
PMCID: PMC3803998  NIHMSID: NIHMS510993  PMID: 23716300

Summary

GLA-SE is a synthetic adjuvant agonist of TLR4 that promotes potent poly-functional TH1 responses. Different TLR4 agonists may preferentially signal via MyD88 or TRIF to exert adjuvant effects. However the contribution of MyD88 and TRIF signaling to the induction of polyclonal TH1 responses by TLR4 agonist adjuvants has not been studied in vivo. To determine whether GLA-SE preferentially signals through MyD88 or TRIF, we evaluated the immune responses against a candidate tuberculosis vaccine antigen following immunization of mice lacking either signaling adapter compared to intact mice. We find that both MyD88 and TRIF are necessary for GLA-SE to induce a poly-functional TH1 immune response characterized by CD4 T cells producing IFN-γ, TNF and IL-2, as well as IgG2c class switching, when paired with the tuberculosis vaccine antigen ID93. Accordingly, the protective efficacy of ID93/GLA-SE immunization against aerosolized Mycobacterium tuberculosis was lost when either signaling molecule was ablated. We demonstrate that MyD88 and TRIF must be expressed in the same cell for the in vivo TH1-skewing adjuvant activity indicating that these two signaling pathways cooperate on an intracellular level. Thus engagement of both the MyD88 and TRIF signaling pathways are essential for the effective adjuvant activity of this TLR4 agonist.

Keywords for submission: adjuvant, TLR4, vaccine, CD4 T cells, GLA

1. Introduction

Subunit vaccines consisting of purified or recombinant antigens represent a significant advancement in the development of safe vaccines. However protein antigens alone are often poorly immunogenic. The addition of adjuvants can significantly enhance the immune response to the target antigen. The most well established adjuvants alum and oil-in-water based emulsions such as MF59 largely increase the antibody response to the target antigen (reviewed in [1]). In many cases this is sufficient to provide protective efficacy, however for pathogens that are controlled by cellular rather than humoral immunity, new adjuvants are needed. Toll-like receptor (TLR) agonists provide a means to inducing cellular responses and are attractive candidates for adjuvant development [2, 3]. The TLR4 agonist monophosphoryl lipid A (MPLA) is the first approved adjuvant that can drive a robust TH1 response when properly formulated [4, 5].

We have developed a novel synthetic TLR4 agonist, glucopyranosyl lipid adjuvant (GLA) that when formulated in a MF59-like stable oil-in-water emulsion (SE) induces IL-12 production by antigen presenting cells (APCs) and drives TH1 responses to a variety of antigens that are protective against intracellular infections [6-14]. Similar to MPLA, GLA is monophosphorylated and is thus safer than the diphosphorylated parental Salmonella lipopolysaccharide (LPS), which shows excellent adjuvant activity but unacceptable toxicity. GLA-SE is currently being tested as an adjuvant in clinical trials of vaccines against tuberculosis (TB), leishmaniasis, hookworm, malaria, HIV and other diseases that may require TH1 responses for protection. Using one of our candidate TB vaccine antigens, we have shown that that the GLA-SE adjuvant drives protective immune responses that can control TB in a mouse model, whereas the SE adjuvant alone does not provide protection [15].

TLR4 is unique among the TLRs as it activates two distinct signaling pathways downstream of the signaling adapters MyD88 and TRIF (also known as TICAM1) [16]. MyD88 and TRIF signaling downstream of TLR4 activation induce partially overlapping gene expression profiles [17]. Based on these expression profile differences and proliferation of adoptively transferred TCR transgenic CD4 and CD8 T cells, it has been proposed that preferential TRIF activation is necessary and sufficient for effective adjuvant activity of TLR4 agonists [18-20]. Others have found both MyD88 and TRIF contribute to dendritic cells activation of T cells in vitro following LPS activation with MyD88-deficiency having a more pronounced impairment of this activity [21]. In vivo, GLA induces activation of both MyD88 and TRIF-associated gene expression [22]. Additionally, induction of CD8 T cell responses by an adenovirus vector vaccine was enhanced by addition of LPS as an adjuvant and the adjuvant effect was dependent on both MyD88 and TRIF [23]. However the requirement of MyD88 and/or TRIF for TLR4 agonist adjuvant activity to drive a polyclonal TH1 response in vivo following immunization has not been rigorously tested. The requirement for MyD88 or TRIF is not necessarily an either/or proposition as signaling through both pathways may activate different cell types, elicit different cytokine responses or combine to produce unique cytokine responses necessary for effective adjuvant activity. In the present study we determine the requirements for MyD88 and TRIF signaling for the induction of TH1 responses to our candidate TB vaccine antigen, ID93 when adjuvanted with GLA-SE. We find that MyD88 and TRIF signaling are both required and must collaborate in the same cell for ID93/GLA-SE to induce a TH1 response necessary for an effective vaccine against TB.

2. Results

2.1 Both MyD88 and TRIF signaling are required for TH1 priming

We have previously found that human and mouse dendritic cells stimulated with GLA activate both MyD88 and TRIF associated genes [12]. Immunization with ID93/GLA-SE induces a strong TH1 immune response against ID93 [7]. Further, in the absence of GLA, ID93 formulated in SE drove a TH2 response that was not protective against aerosolized Mtb challenge [15]. Thus GLA is necessary for induction of a protective TH1 response by ID93/GLA-SE. To assess whether MyD88 and/or TRIF were required for effective vaccination, wild type (WT), Ticam1-/- (lacking expression of the TRIF protein), and Myd88-/- mice were immunized with ID93 either alone or adjuvanted with either SE or GLA-SE. Immunization of WT mice with ID93/GLA-SE produced ID93-specific CD4 T cells that produced IFN-γ, TNF and IL-2, upon ex-vivo re-stimulation (Figure 1A). GLA was required for this TH1 skewing as CD4 T cells from animals immunized with ID93 alone or ID93-SE failed to produce these cytokines upon re-stimulation with ID93. Genetic ablation of MyD88 or TRIF abolished the TH1 response to immunization with ID93/GLA-SE, indicating that both signaling pathways are necessary for GLA-SE to drive a TH1 response (Figure 1A). Of note IL-17 was not produced by CD4 T cells from any of the immunized groups upon ex-vivo restimulation (data not shown).

Figure 1. TRIF and MyD88 are required for generation of TH1 cells in vivo following immunization with ID93/ GLA-SE.

Figure 1

Open in a new tab

WT, Ticam1-/-, and Myd88-/- mice were immunized with ID93 alone ID93-SE, or ID93/GLA-SE or unimmunized. One month after the final immunization spleen cells were isolated and re-stimulated with ID93. CD4 T cells were analyzed for the production of (A) CD154, IFN-γ, TNF, IL-2, and IL-5 or co-expression of CD154, IFN-γ, TNF and IL-2. N= 3-5 animals/group. Results are representative of two similar experiments.

In the Myd88-/- and Ticam1-/- mice immunized with ID93/GLA-SE there was a residual population of CD4 T cells that responded to ID93 restimulation by expressing CD154 (Figure 1A). These cells also made IL-5, indicating a small level of priming of antigen specific cells in the absence of either of these signaling adapters (Figure 1A). The frequency of IL-5 producing TH2 cells did not vary significantly among Myd88-/- and Ticam1-/- mice immunized with ID93, ID93/SE and ID93/GLA-SE suggesting that protein alone is sufficient to drive TH2 responses. Importantly immunization of WT mice with ID93/GLA-SE impaired the induction of TH2 cells compared to ID93 or ID93/SE immunization. The frequency of poly-functional TH1 cells (CD4 T cells making combinations of IFN-γ, TNF, and/or IL-2) have been proposed to be a correlate of vaccine efficacy against M.tb. in mice. The majority of TH1 cells elicited by immunization of WT mice with ID93/GLA-SE co-expressed CD154, IFN-γ and TNF upon restimulation with approximately half of these cells also expressing IL-2 (Figure 1B).

The MyD88 and TRIF contribution to TH1 skewing with ID93/GLA-SE immunization was also evident in ID93-specific antibody class switching. Immunization of WT mice with ID93/GLA-SE produced ID93-specific IgG1 and IgG2c antibodies, whereas immunization with ID93 alone did not elicit significant antibody responses (Figure 2). IgG1 titers were similar between ID93-SE and ID93/GLA-SE immunized WT mice, indicating that GLA does not play a role in induction of IgG1 responses in the presence of the SE adjuvant. The IgG1 response to either ID93-SE or ID93/GLA-SE was not substantially impaired by TRIF-deficiency, but was reduced in MyD88-deficient immunized animals (P<0.001)(Figure 2B and C). Similar to the IFN-γ response, GLA was required for efficient switching to IgG2c as ID93-SE elicited only low titers of IgG2c responses in WT mice (P<0.001) (Figure 2E). The IgG2c response to ID93/GLA-SE immunization was partially dependent on both TRIF (P<0.001) and MyD88 (P<0.001) (Figure 2F). Together these data indicate that both TRIF and MyD88 are necessary for efficient induction of TH1 responses by the GLA-SE adjuvant. Additionally MyD88 appears to be necessary for efficient induction of IgG1 responses by ID93-SE.

Figure 2. TRIF and MyD88 are required for optimal generation of an IgG2c response following immunization with ID93/ GLA-SE.

Figure 2

Open in a new tab

WT, Ticam1-/-, and Myd88-/- mice were immunized with ID93, ID93-SE or ID93/GLA-SE. Sera were collected three weeks after the first immunization and serially diluted to assess levels of anti-ID93 (A - C) IgG1 and (D - F) IgG2c isotype antibody titers from WT,Ticam1-/, and Myd88-/- mice immunized with (A and D) ID93, (B and E) ID93-SE or (C and F) ID93/GLA-SE. N= 5 animals/group. Results are representative of two similar experiments.

2.2 MyD88 and TRIF are required for the protective efficacy of ID93/GLA-SE against Mycobacterium tuberculosis infection

We have previously found that immunization with ID93/GLA-SE limits Mtb burden following aerosol challenge and that inclusion of GLA is necessary for the protective efficacy [15]. To determine whether MyD88 and/or TRIF signaling contribute to the protective efficacy of ID93/GLA-SE, we immunized WT, Myd88-/-, and Ticam1-/- mice with ID93 either alone or adjuvanted with SE or GLA-SE or immunized them with saline. Mice were then challenged with a low dose of aerosolized Mtb. Four weeks later WT mice immunized with ID93/GLA-SE had significantly lower bacterial burdens in both the spleen and lungs compared to unimmunized animals (P<0.001 and P<0.001 respectively) (Figure 3). As we found previously, GLA was necessary for this protective response as immunization with ID93 or ID93-SE did not significantly reduce the bacterial burden in either organ. The protective efficacy of immunization with ID93/GLA-SE was ablated in both MyD88 and TRIF-deficient mice as immunized and unimmunized mice of the same genotype had similar levels of bacterial burden (Figure 3). In agreement with previous reports ablation of MyD88 enhanced lung bacterial loads ~1000-fold compared to WT mice regardless of immunization, whereas TRIF-deficiency did not alter bacterial burden at this time (P<0.001 and not significant for saline immunized WT vs Myd88-/- and Ticam1-/-, respectively) [24, 25]. Compared to ID93/GLA-SE immunized Myd88-/- or Ticam1-/- mice, ID93/GLA-SE immunized WT mice showed significantly lower bacterial burden (P<0.001) (Figure 3). These data indicate that signaling through both MyD88 and TRIF during ID93/GLA-SE immunization is critical for the generation of a protective TH1 immune response that limits Mtb infection as loss of either signaling adapter was sufficient to ablate vaccine efficacy.

Figure 3. ID93/GLA-SE-induced protection against M. tuberculosis infection requires TRIF and MyD88.

Figure 3

Open in a new tab

WT, Myd88-/-, and Ticam1-/- mice were immunized with ID93, ID93-SE, or ID93/GLA-SE or left unimmunized. One month after the final immunization, animals were infected with aerosolized M. tuberculosis. Four weeks after infection, (A) lungs and (B) spleens were harvested and bacterial burden determined. N=7 animals per group. Results are representative of two similar experiments. Statistically significant differences between unimmunized and immunized groups within the same genotype are indicated, n.s. indicates not significant.

2.3 MyD88 and TRIF must be expressed in the same cell for GLA adjuvant activity

There are several non-mutually exclusive scenarios for the cooperation of MyD88 and TRIF signaling for the adjuvant activity of GLA-SE including: (1) activation of multiple cell types through either MyD88 or TRIF, (2) secretion of multiple MyD88 and TRIF-dependent cytokines or chemokines that collaborate to drive TH1 programming or (3) cooperative signaling within the same cell to produce cytokines and/or chemokines that drive TH1 programming. To distinguish among these possibilities we generated bone marrow chimeras with WT, Ticam1-/-, Myd88-/-, or a 1:1 mixture of Ticam1-/- and Myd88-/- bone marrow in lethally irradiated WT. In the latter animals, both MyD88 and TRIF are expressed, but are in mutually exclusive cell populations. Three months after reconstitution, the chimeras were immunized with ID93 adjuvanted with GLA-SE. Similar to the intact WT mice immunized with ID93/GLA-SE in Figure 1, mice reconstituted with WT bone marrow produced antigen-specific CD4 T cells that produced IFN-γ, TNF, and IL-2 and expressed CD154 upon antigen restimulation (Figure 4). Similar to intact Myd88-/-or Ticam1-/- mice, CD4 T cells from chimeras reconstituted with Myd88-/- or Ticam1-/- bone marrow failed to express IFN-γ, TNF, or IL-2 upon antigen re-stimulation (Figure 4A). CD154 was still expressed on a small population of CD4 T cells, indicating antigen specific cells were still produced in these animals. Mice expressing both MyD88 and TRIF, but in different cell populations (reconstituted with a mixture of Myd88-/- and Ticam1-/- bone marrow), failed to make a TH1 response following immunization (Figure 4A). As with intact WT mice, the CD4 T cell responses in mice reconstituted with WT bone marrow were multifunctional, with the vast majority co-expressing CD154, IFN-γ, and TNF with or without co-expression of IL-2 (Figure 4B). These data indicate that TRIF and MyD88 must be expressed in the same cell for effective TH1 adjuvant activity of GLA. Further these data suggest that both MyD88 and TRIF must be expressed on a radiation sensitive cell population for the ID93/GLA-SE vaccine to drive a TH1 immune response.

Figure 4. GLA-SE induced TH1 responses require MyD88 and TRIF in the same cell.

Figure 4

Open in a new tab

WT, Myd88-/-, Ticam1-/-, or a 1:1 mix of Myd88-/- and Ticam1-/- was used to reconstitute lethally irradiated WT mice. After reconstitution mice were immunized with ID93/GLA-SE. One week after the final immunization, splenocytes were stimulated with antigen or media and analyzed for CD4 T cell production of CD154, IFN-γ, TNF and IL-2 by intracellular cytokine staining, either for (A) individual responses or (B) multi-functionality. N=3-5 animals/group. Data represent antigen – saline responses. Statistical differences compared to the WT group are shown.

2.4 IL-12 production is dependent on both MyD88 and TRIF

Dendritic cells (DCs) are necessary for the GLA-SE adjuvant to drive TH1 responses during vaccination [13]. To determine the GLA-specific responses controlled by both MyD88 and TRIF in DCs, we measured changes in gene expression in WT, Ticam1-/-, or Myd88-/- bone marrow derived dendritic cells (BMDCs) stimulated with GLA. We also stimulated cells with LPS to evaluate canonical TL4 signaling as well as R848 or polyI:C to evaluate exclusive MyD88 or TRIF- dependent signaling downstream of TLR7 and TLR3, respectively. Compared to GLA, LPS, or R848, polyI:C at the dose used induced a more muted, but detectable change in gene expression. As expected, changes in gene expression in response to R848 and polyI:C were ablated in MyD88 and TRIF-deficient BMDCs, respectively. The responses to GLA and LPS followed similar patterns, with many of the genes evaluated being partially controlled by both MyD88 and TRIF, including: IL1α, Il1β, Cxcl1, Ccl4, Il2b, Ccl3, Gcsf, and Ccl5 (Figure 5A). Because IL-12 is necessary for the induction and stabilization of TH1 programming, we measured IL-12p40 and IL-12p35 protein production from these cells 24 hours after stimulation. Similar to the changes in Il12b expression, production of IL-12 in response to either GLA or LPS stimulation was dependent on both MyD88 and TRIF (Figure 5B). IL-12p40 can associate with either IL-12p35 to drive TH1 commitment or IL-23p19 to drive TH17 commitment. We detected low levels of IL-12p70 elicited by GLA, LPS and R848 (Figure 5C). Both TRIF and MyD88 were necessary for maximal induction of IL-12p70 by both GLA and LPS. Because IL-17-producing CD4 T cells were not produced by in vivo immunization with ID93/GLA-SE (data not shown) we did not test for IL-23p19 production by stimulated DCs. CCL5 (aka RANTES) protein production, which is important for the recruitment of CD4 T cells, was also dependent on both TRIF and MyD88 (Figure 5D).

Figure 5. GLA stimulation of DC induces both MyD88 and TRIF-associated genes.

Figure 5

Open in a new tab

WT, Ticam1-/-, and Myd88-/- bone marrow derived DC were stimulated with GLA, LPS, polyI:C or R848 (A) mRNA expression for MyD88 associated genes (top) and TRIF associated genes (bottom) was assessed after two hours of stimulation. Fold increases over media stimulation are indicated by the heat map. (B) IL-12p40, (C) IL-12p70 and (C) CCL5 production in the culture supernatant was measured after 24 hours of stimulation. Data are representative of 2 experiments with 2 replicates per condition. *, **,*** and **** indicates P< 0.05, 0.01, 0.001 and 0.0001 relative to WT respectively.

To determine if GLA activates different signaling pathways in primary macrophages and dendritic cells in the spleen and bone marrow we stimulated whole bone marrow and splenocytes from WT, Myd88-/- and Ticam1-/- mice directly ex vivo with GLA, LPS, or R848 in the presence of Brefeldin A. All three agonists evoked production of TNF and IL-12p40 from splenic and bone marrow macrophages and DCs (Figure 6). The R848 response was completely lost in Myd88-/- cells, but similar to WT in Ticam1-/- cells. GLA and LPS showed similar dependencies on MyD88 and TRIF with MyD88 being essential for IL-12p40 production by all four cell types whereas TNF production by bone marrow resident macrophages and DCs was only partially reduced in Myd88-/- cells. TRIF-deficiency led to a partial reduction in TNF and IL-12p40 production by both splenic and bone marrow resident dendritic cells and macrophages (Figure 6). We did not detect IL-12p35 or IL-23p19 in of these samples. Together, these data indicate that at the dose tested GLA signals through TRIF and MyD88 to activate BMDCs to produce the TH1 driving cytokines IL-12 and RANTES as well as the pro-inflammatory cytokine TNF.

Figure 6. MyD88 and TRIF are required for GLA induction of IL-12p40 and TNF in primary macrophages and dendritic cells.

Figure 6

Open in a new tab

WT, Ticam1-/-, and Myd88-/- splenocytes and bone marrow were stimulated ex vivo with GLA, LPS, or R848 and stained for intracellular IL-12p40 and TNF production (A) Representative staining of splenic macrophages. Quantitation of IL-12p40 and TNF production by (B) splenic macrophages, (C) splenic DCs (C) bone marrow macrophages and (D) bone marrow dendritic cells. Data are representative of 2 experiments with 3 replicates per condition.

3. Discussion

We demonstrate that the TLR4 agonist adjuvant GLA-SE signals through both MyD88 and TRIF to drive a polyclonal TH1 response in vivo characterized by IFN-γ, TNF and IL-2 producing cells and IgG2c isotype switching. Importantly we find that MyD88 and TRIF must cooperate within the same cell to drive this in vivo TH1 response. Further we find that MyD88 and TRIF are both required for the induction of IL-12p40, IL-12p35, RANTES and TNF by GLA in several different APCs including splenic and bone marrow resident macrophages and DCs. We hypothesize that cooperative IL-12 production may be a necessary point of intersection for the MyD88 and TRIF-dependent induction of TH1 responses by ID93/GLA-SE immunization.

TLR agonists in general and TLR4 agonists such as MPLA and GLA in particular are safe and effective adjuvants for the induction of TH1 responses to recombinant vaccine antigens. Among the TLRs, TLR4 is unique in its ability to activate both the MyD88 and TRIF signaling pathways. Based on gene expression analysis it has been proposed that preferential activation of the TRIF pathway is sufficient for effective adjuvant activity of TLR4 agonists and that this proposed TRIF bias is necessary for decreased toxicity compared to the parental LPS molecule. However, at least for GLA, we find that both the TRIF and MyD88 pathway are important for the TH1 skewing capacity of TLR4 agonist adjuvants. In this aspect the adjuvant activity of GLA may be more similar to LPS than MPLA. Similar to LPS we find that GLA-SE induces memory CD4 T cells that persist for at least four months following immunization (manuscript in preparation). These findings largely agree with the finding that the ability of LPS activated DCs to drive to prime a T cell response in vitro depends on both MyD88 and TRIF [21]. However the present study uses a much more rigorous in vivo immunization protocol to test this hypothesis. Importantly GLA is safe and well tolerated in multiple clinical trials involving hundreds of doses of GLA-containing vaccines, suggesting that, at least for GLA, preferential TRIF usage is not a requirement for an acceptable safety profile. These data provide important insights into the requirements of such TLR4 agonist adjuvants and may shape the further development of TLR4 agonist adjuvants [26].

GLA-SE is also being used as an adjuvant for vaccines against other pathogens including influenza A virus where antibody responses are critical for vaccine efficacy [27]. Surprisingly although TRIF is necessary for induction of IFN-γ producing CD4 T cells by ID93/GLA-SE, in the same mice there is residual, but diminished induction of IgG2c class switching. This suggests that the residual signaling via MyD88 in these animals may be sufficient to promote production of T follicular helper cells.

The inclusion of GLA in our candidate TB vaccines is essential for the induction of TH1 responses and reduction of bacterial burden [11, 15]. This activity depends on both MyD88 and TRIF as deletion of either results in ablation of TH1 induction and loss of vaccine efficacy. The ability of ID93/GLA-SE to elicit protective responses against Mtb is due to the induction of durable memory CD4 T cells, rather than acute changes in the innate immune system in response to GLA as there is a four-week gap between the time of last exposure to GLA and Mtb challenge. Additionally we have found that CD4 T cells are both necessary and sufficient for the protective efficacy of GLA-SE adjuvanted vaccines against aerosolized Mtb challenge (manuscript in preparation). Immunization with ID93 formulated in the MF59-like adjuvant SE elicited a modest TH2 response characterized by IL-5 production and substantial IgG1 antibody responses. Addition of GLA to the formulation eliminated the IL-5 response. Interestingly based on IgG1 titers MyD88 appears to contribute to the TH2 skewing of the SE adjuvant, likely through TLR independent mechanisms such as caspase 1 activity, although testing this hypothesis is outside the scope of the current study.

Loss of MyD88 or TRIF signaling impaired a number of GLA dependent responses in BMDCs and splenic and bone marrow resident DCs and macrophages including IL-12 and RANTES, both of which may contribute to effective induction of TH1 responses as well as the pro-inflammatory cytokine TNF. These results are in agreement with the findings of others [21]. IL-12 in particular is critical for stable TH1 programming and IL-12 production in response to GLA was lost in both Myd88-/- and Ticam1-/- BMDCs. This fits with our finding that MyD88 and TRIF must be expressed in the same cell for GLA-SE to be an effective TH1-biasing adjuvant. That other TLR agonists such as R848 can efficiently induce IL-12 production from BMDCs without activating the TRIF pathway suggests that TLR4 engagement of MyD88 is fundamentally different than that of TLR7. This may be due to the difference in cellular localization of the two receptors or recruitment of different signaling molecules in addition to MyD88 such as MAL which is important for TLR4 but not TLR7 signaling [28].

Previous studies concluded that the TLR4 agonist adjuvant MPLA induces antigen-specific T cell proliferation in MyD88-independent, TRIF-dependent manner [18, 19]. This would appear to contradict our present results. However, there are many important differences between these studies. Foremost, although both signal via TLR4, MPLA and GLA are very different adjuvants. Whereas MPL is a complex mixture with different acyl chain numbers and lengths, GLA is a synthetic hexa-acylated mono-species with a fixed acyl chain length. In the present study we analyze polyclonal TH1 responses in WT, Myd88-/-, and Ticam1-/- mice to recombinant protein immunization as read out by cytokine production and protective efficacy. Conversely the earlier studies relied on acute proliferation of adoptively transferred TCR transgenic cells, either OT-1 or OT-II, into Myd88-/- and Ticam1-/- recipients, following immunization with targeted peptides. Further the present study analyzes the contribution of MyD88 and TRIF following a homologous prime, boost, boost strategy compared to the acute response following priming of transgenic cells with peptide and MPLA. Although the cells transferred into Myd88-/- recipients were competent to proliferate in response to peptide and MPLA immunization, it is unclear whether Myd88 signaling contributed to TH1 skewing in this study [18]. The latter approach is only feasible using TCR transgenic cells, but these types of studies may be affected by the lack of a polyclonal repertoire and artifacts resulting from the adoptive transfer of large number of cells [29]. In light of these caveats, we believe that the conclusions of the present study and the previous study are not necessarily contradictory. Indeed the same group has subsequently reported that synthetic monophosphoryl lipid A signals through both MyD88 and TRIF to activate p38 upstream of several important cytokines [19].

In conclusion we find that collaboration between MyD88 and TRIF signaling pathways at an intracellular level is necessary for induction of dendritic cell produced IL-12 by the GLA-SE adjuvant. This in turn facilitates poly-functional ID93-specific CD4 T cell differentiation into multi-functional TH1 cells in vivo that promote IgG2c class-switching and provide protection against aerosolized Mtb challenge. Further, this usage of both MyD88 and TRIF may be important for the induction of long-lived memory CD4 T cells [30]. These findings provide important insight into the mechanistic activity of an adjuvant that is in clinical trials for use with a number of vaccines against both intra- and extracellular pathogens.

4. Materials and Methods

4. 1 Mice and immunizations

WT C57BL/6 mice and Ticam1-/- mice on the C57Bl/6 background were purchased from Jackson Laboratories (Bar Harbor, ME). Myd88-/- mice on the C57Bl/6 background were obtained from Institute for Systems Biology, Seattle. Mice were immunized with recombinant protein ID93 alone or formulated with stable emulsion (SE) or stable emulsion plus GLA(also known as PHAD, Avanti Polar Lipids) to provide a final vaccine dose of 0.5μg ID93 and 5 μg GLA-SE [7]. Mice were immunized 3 times at 3 week intervals. All mice were maintained in specific pathogen-free condition. All procedures were approved by the IDRI institutional animal care and use committee.

4. 2 Antigen stimulation and cytokine responses

Splenocytes were restimulated with 10 μg/mL protein or media in the presence of Brefeldin A for eight hours. Cells were washed with PBS and stained with LIVE/DEAD Fixable Stain (Invitrogen) for 30 minutes at 4°C. Cells were washed and surface stained with fluorochrome labeled antibodies to CD4 (clone GK1.5) and CD8 (clone 53-6. 7) (BioLegend and eBioscience) in the presence of 20% normal mouse serum for 20 minutes at 4°C. Cells were washed and permeabilized with Cytofix/Cytoperm (BD Biosciences) for 20 minutes at room temperature. Cells were washed twice with Perm/Wash (BD Biosciences) and stained intracellularly with fluorochrome labeled antibodies to IFN-γ (clone XMG-1.2), IL-2 (JES6-5H4), TNF (MP6-XT22), IL-17A (tc11-18h10.1) and CD154 (MR1) (BioLegend and eBioscience) for 20 minutes at room temperature. Cells were washed and resuspended in PBS. Up to 106 events were collected on a four laser LSRFortessa flow cytometer (BD Biosciences). Data were analyzed with FlowJo. Cells were gated as singlets > live > lymphocytes > CD4+ CD8- > response positive. Analysis and presentation of distributions was performed using SPICE version 5.2, downloaded from <http://exon.niaid.nih.gov/spice.

4.3 Antibody responses

Mouse sera were prepared by collection of retro-orbital blood into microtainer serum collection tubes (VWR International, West Chester, PA), followed by centrifugation at 1200rpm for 5 minutes. Each serum sample was then analyzed by antibody capture ELISA. Briefly, ELISA plates (Nunc, Rochester, NY) were coated with 1 μg/ml recombinant antigen in 0.1 M bicarbonate buffer and blocked with 1% BSA-PBS. Then, in consecutive order and following washes in PBS/Tween20, serially diluted serum samples, anti-mouse IgG, IgG1 or IgG2c-HRP (all Southern Biotech, Birmingham, AL) and ABTS-H2O2 (Kirkegaard and Perry Laboratories, Gaithersburg, MD) were added to the plates. Plates were analyzed at 405nm (ELX808, Bio-Tek Instruments Inc, Winooski, VT).

4.4 Aerosol M. tuberculosis infection of mice

Mice were infected with Mycobacterium tuberculosis strain H37Rv (ATCC) in an aerosol exposure chamber designed by the University of Wisconsin, College of Engineering Shops (Madison, WI). A low dose of aerosol M. tuberculosis (50-100 colony forming units; CFU) was delivered to the lungs of each animal. Four weeks after M. tuberculosis infection the lungs and spleens were collected and homogenized. An aliquot of organ homogenate was serially diluted in 0.1% PBS-Tween 80 and plated on 7H10 plates (Molecular Toxicology, Boone NC) to allow bacterial growth. Plates were incubated for 14-21 days at 37°C with 5% CO2 before manual counting of colony forming units (CFUs).

4.5 Cytokine production by dendritic cells and dendritic cells

Mouse bone marrow cells were seeded at 2 × 105/ml in 20ml dishes and cultured for 5-7 days in 10% heat inactivated FBS, 1% penicillin/streptomycin, 1% L-glutamine, 1% non-essential amino acids in the presence of 400ng/mL GMCSF and 20ng/mL IL-4 (Peprotech, Rocky Hill, NJ). Following culture, BMDC were collected, 5 × 105 cells seeded in each well of 24-well plates and stimulated with GLA (1μg/mL) (also known as PHAD Avanti Polar Lipids, Alabaster AL), Salmonella Minnesota R595 ultrapure LPS (10 ng/mL) (InvivoGen, San Diego, CA), R848 (1μg/mL) (InvivoGen) or polyI:C (20 μg/mL) (Sigma-Aldrich). Two hours later cells were lysed in QuantiGene buffer. mRNA for multiple targets (CCL3, CCL4, CCL5, CXCL1, CXCL10, GAPDH, G-CSF, ICAM-1, IFIT-1, IFNβ, IL-1α, IL-1β, IL-6, IL-12p40, TNF) was extracted and captured within the same sample using the QuantiGene multiplex assay according to the manufacturer’s instructions (Panomics). Transcript levels were assessed using a Luminex 200 Multi-Analyte system and analyzed using MasterPlex EX software (MiraiBio, San Francisco, CA). Data were normalized against the housekeeping gene GAPDH and expressed as fold induction versus stimulated BMDC with media alone. After 24 hours of stimulation IL-12p40, IL-12p70 and CCL5 protein concentration were determined MILLIPLEX MAP magnetic bead analysis according to manufacturer’s instructions (Millipore).

Alternatively single cell suspensions of mouse splenocytes and bone marrow were stimulated with GLA, LPS, R848 (1 μg/mL each) or media in the presences of Brefeldin A for eight hours using a method adopted from Hajjar et.al. [31]. Cells were washed with PBS and treated with anti-CD16/CD32 blocking antibody (clone 93). Cells were washed and surface stained with fluorochrome labeled antibodies to CD3 (17A2), CD19 (1D3), CD11b (m1/70), CD11c (n418), and Ly6G (1A8) (BioLegend and eBioscience) for 20 minutes at 4°C. Cells were washed and permeabilized with Cytofix/Cytoperm (BD Biosciences) for 20 minutes at room temperature. Cells were washed twice with Perm/Wash (BD Biosciences) and stained intracellularly with fluorochrome labeled antibodies to IL-12p40 (C17.8), TNF, IL-12p35 (4D10p35), and IL-23p19 (fc23cpg) (BioLegend and eBioscience) for 20 minutes at room temperature. Cells were washed and resuspended in PBS. Up to 106 events were collected on a four laser LSRFortessa flow cytometer (BD Biosciences). Data were analyzed with FlowJo. Cells were gated as singlets > live > CD3-CD19-> not neutrophils (Ly6Ghi CD11bhi) > DCs (CD11c+) or macrophages (CD11c- CD11b+) > response positive.

4.6 Statistics

Statistical analyses were performed using MS Excel (Microsoft Corporation, Redmond, WA) or Prism software (GraphPad Software, Inc., La Jolla, CA). Assays resulting in normally distributed data including two groups were analyzed using the Student’s t-test. ANOVA analyses with Bonferroni correction for multiple tests were used when more than two groups were compared. Data were log-transformed for non-normal data sets prior to analysis.

Acknowledgments

We thank Lynn Hajjar for experimental advice. We thank David Argilla, Valerie Reese, Charles Davis, and John Laurance for technical assistance. This work was conducted with support from National Institutes of Health grant AI-078054 to R. Coler and contract HHSN272200800045C to R. Coler, funding from the American Leprosy Missions to M. Duthie, and grant 42387 from the Bill and Melinda Gates Foundation to S. Reed.

Abbreviations used in this paper

GLA

Glucopyranosyl Lipid Adjuvant

SE

stable emulsion

Mtb

Mycobacterium tuberculosis

TB

tuberculosis

Footnotes

The authors declare no financial or commercial conflicts of interest.

References

  • 1.Lambrecht BN, Kool M, Willart MA, Hammad H. Mechanism of action of clinically approved adjuvants. Current opinion in immunology. 2009;21:23–29. doi: 10.1016/j.coi.2009.01.004. [DOI] [PubMed] [Google Scholar]
  • 2.Guy B. The perfect mix: recent progress in adjuvant research. Nature reviews Microbiology. 2007;5:505–517. doi: 10.1038/nrmicro1681. [DOI] [PubMed] [Google Scholar]
  • 3.Kwissa M, Kasturi SP, Pulendran B. The science of adjuvants. Expert review of vaccines. 2007;6:673–684. doi: 10.1586/14760584.6.5.673. [DOI] [PubMed] [Google Scholar]
  • 4.Casella CR, Mitchell TC. Putting endotoxin to work for us: monophosphoryl lipid A as a safe and effective vaccine adjuvant. Cellular and molecular life sciences : CMLS. 2008;65:3231–3240. doi: 10.1007/s00018-008-8228-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Cluff CW. Monophosphoryl lipid A (MPL) as an adjuvant for anti-cancer vaccines: clinical results. Advances in experimental medicine and biology. 2010;667:111–123. doi: 10.1007/978-1-4419-1603-7_10. [DOI] [PubMed] [Google Scholar]
  • 6.Anderson RC, Fox CB, Dutill TS, Shaverdian N, Evers TL, Poshusta GR, Chesko J, Coler RN, Friede M, Reed SG, Vedvick TS. Physicochemical characterization and biological activity of synthetic TLR4 agonist formulations. Colloids and surfaces B, Biointerfaces. 2010;75:123–132. doi: 10.1016/j.colsurfb.2009.08.022. [DOI] [PubMed] [Google Scholar]
  • 7.Baldwin SL, Bertholet S, Kahn M, Zharkikh I, Ireton GC, Vedvick TS, Reed SG, Coler RN. Intradermal immunization improves protective efficacy of a novel TB vaccine candidate. Vaccine. 2009;27:3063–3071. doi: 10.1016/j.vaccine.2009.03.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Baldwin SL, Shaverdian N, Goto Y, Duthie MS, Raman VS, Evers T, Mompoint F, Vedvick TS, Bertholet S, Coler RN, Reed SG. Enhanced humoral and Type 1 cellular immune responses with Fluzone adjuvanted with a synthetic TLR4 agonist formulated in an emulsion. Vaccine. 2009;27:5956–5963. doi: 10.1016/j.vaccine.2009.07.081. [DOI] [PubMed] [Google Scholar]
  • 9.Raman VS, Bhatia A, Picone A, Whittle J, Bailor HR, O’Donnell J, Pattabhi S, Guderian JA, Mohamath R, Duthie MS, Reed SG. Applying TLR synergy in immunotherapy: implications in cutaneous leishmaniasis. Journal of immunology. 2010;185:1701–1710. doi: 10.4049/jimmunol.1000238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Raman VS, O’Donnell J, Bailor HR, Goto W, Lahiri R, Gillis TP, Reed SG, Duthie MS. Vaccination with the ML0276 antigen reduces local inflammation but not bacterial burden during experimental Mycobacterium leprae infection. Infection and immunity. 2009;77:5623–5630. doi: 10.1128/IAI.00508-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Bertholet S, Ireton GC, Ordway DJ, Windish HP, Pine SO, Kahn M, Phan T, Orme IM, Vedvick TS, Baldwin SL, Coler RN, Reed SG. A defined tuberculosis vaccine candidate boosts BCG and protects against multidrug-resistant Mycobacterium tuberculosis. Science translational medicine. 2010;2:53ra74. doi: 10.1126/scitranslmed.3001094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Coler RN, Bertholet S, Moutaftsi M, Guderian JA, Windish HP, Baldwin SL, Laughlin EM, Duthie MS, Fox CB, Carter D, Friede M, Vedvick TS, Reed SG. Development and characterization of synthetic glucopyranosyl lipid adjuvant system as a vaccine adjuvant. PloS one. 2011;6:e16333. doi: 10.1371/journal.pone.0016333. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Pantel A, Cheong C, Dandamudi D, Shrestha E, Mehandru S, Brane L, Ruane D, Teixeira A, Bozzacco L, Steinman RM, Longhi MP. A new synthetic TLR4 agonist, GLA, allows dendritic cells targeted with antigen to elicit Th1 T-cell immunity in vivo. European journal of immunology. 2012;42:101–109. doi: 10.1002/eji.201141855. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Windish HP, Duthie MS, Ireton G, Lucas E, Laurance JD, Bailor RH, Coler RN, Reed SG. Protection of mice from Mycobacterium tuberculosis by ID87/GLA-SE, a novel tuberculosis subunit vaccine candidate. Vaccine. 2011 doi: 10.1016/j.vaccine.2011.07.094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Baldwin SL, Bertholet S, Reese VA, Ching LK, Reed SG, Coler RN. The importance of adjuvant formulation in the development of a tuberculosis vaccine. Journal of immunology. 2012;188:2189–2197. doi: 10.4049/jimmunol.1102696. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Oshiumi H, Matsumoto M, Funami K, Akazawa T, Seya T. TICAM-1, an adaptor molecule that participates in Toll-like receptor 3-mediated interferon-beta induction. Nature immunology. 2003;4:161–167. doi: 10.1038/ni886. [DOI] [PubMed] [Google Scholar]
  • 17.Coler RN, Baldwin SL, Shaverdian N, Bertholet S, Reed SJ, Raman VS, Lu X, DeVos J, Hancock K, Katz JM, Vedvick TS, Duthie MS, Clegg CH, Van Hoeven N, Reed SG. A synthetic adjuvant to enhance and expand immune responses to influenza vaccines. PloS one. 2010;5:e13677. doi: 10.1371/journal.pone.0013677. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Mata-Haro V, Cekic C, Martin M, Chilton PM, Casella CR, Mitchell TC. The vaccine adjuvant monophosphoryl lipid A as a TRIF-biased agonist of TLR4. Science. 2007;316:1628–1632. doi: 10.1126/science.1138963. [DOI] [PubMed] [Google Scholar]
  • 19.Cekic C, Casella CR, Eaves CA, Matsuzawa A, Ichijo H, Mitchell TC. Selective activation of the p38 MAPK pathway by synthetic monophosphoryl lipid A. The Journal of biological chemistry. 2009;284:31982–31991. doi: 10.1074/jbc.M109.046383. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Persing DH, Coler RN, Lacy MJ, Johnson DA, Baldridge JR, Hershberg RM, Reed SG. Taking toll: lipid A mimetics as adjuvants and immunomodulators. Trends in microbiology. 2002;10:S32–37. doi: 10.1016/s0966-842x(02)02426-5. [DOI] [PubMed] [Google Scholar]
  • 21.Shen H, Tesar BM, Walker WE, Goldstein DR. Dual signaling of MyD88 and TRIF is critical for maximal TLR4-induced dendritic cell maturation. Journal of immunology. 2008;181:1849–1858. doi: 10.4049/jimmunol.181.3.1849. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Lambert SL, Yang CF, Liu Z, Sweetwood R, Zhao J, Cheng L, Jin H, Woo J. Molecular and cellular response profiles induced by the TLR4 agonist-based adjuvant Glucopyranosyl Lipid A. PloS one. 2012;7:e51618. doi: 10.1371/journal.pone.0051618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Rhee EG, Kelley RP, Agarwal I, Lynch DM, La Porte A, Simmons NL, Clark SL, Barouch DH. TLR4 ligands augment antigen-specific CD8+ T lymphocyte responses elicited by a viral vaccine vector. Journal of virology. 2010;84:10413–10419. doi: 10.1128/JVI.00928-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Fremond CM, Yeremeev V, Nicolle DM, Jacobs M, Quesniaux VF, Ryffel B. Fatal Mycobacterium tuberculosis infection despite adaptive immune response in the absence of MyD88. The Journal of clinical investigation. 2004;114:1790–1799. doi: 10.1172/JCI21027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Scanga CA, Bafica A, Feng CG, Cheever AW, Hieny S, Sher A. MyD88-deficient mice display a profound loss in resistance to Mycobacterium tuberculosis associated with partially impaired Th1 cytokine and nitric oxide synthase 2 expression. Infection and immunity. 2004;72:2400–2404. doi: 10.1128/IAI.72.4.2400-2404.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Bowen WS, Minns LA, Johnson DA, Mitchell TC, Hutton MM, Evans JT. Selective TRIF-dependent signaling by a synthetic toll-like receptor 4 agonist. Science signaling. 2012;5:ra13. doi: 10.1126/scisignal.2001963. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Clegg CH, Roque R, Van Hoeven N, Perrone L, Baldwin SL, Rininger JA, Bowen RA, Reed SG. Adjuvant solution for pandemic influenza vaccine production. Proceedings of the National Academy of Sciences of the United States of America. 2012;109:17585–17590. doi: 10.1073/pnas.1207308109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.O’Neill LA, Bowie AG. The family of five: TIR-domain-containing adaptors in Toll-like receptor signalling. Nature reviews Immunology. 2007;7:353–364. doi: 10.1038/nri2079. [DOI] [PubMed] [Google Scholar]
  • 29.Badovinac VP, Haring JS, Harty JT. Initial T cell receptor transgenic cell precursor frequency dictates critical aspects of the CD8(+) T cell response to infection. Immunity. 2007;26:827–841. doi: 10.1016/j.immuni.2007.04.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Thompson BS, Chilton PM, Ward JR, Evans JT, Mitchell TC. The low-toxicity versions of LPS, MPL adjuvant and RC529, are efficient adjuvants for CD4+ T cells. Journal of leukocyte biology. 2005;78:1273–1280. doi: 10.1189/jlb.0305172. [DOI] [PubMed] [Google Scholar]
  • 31.Hajjar AM, Ernst RK, Fortuno ES, 3rd, Brasfield AS, Yam CS, Newlon LA, Kollmann TR, Miller SI, Wilson CB. Humanized TLR4/MD-2 mice reveal LPS recognition differentially impacts susceptibility to Yersinia pestis and Salmonella enterica. PLoS pathogens. 2012;8:e1002963. doi: 10.1371/journal.ppat.1002963. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES