ABSTRACT
In a previous study, we have identified MTBK_24820, the complete protein form of PPE39 in the hypervirulent Mycobacterium tuberculosis (Mtb) strain Beijing/K by using comparative genomic analysis. PPE39 exhibited vaccine potential against Mtb challenge in a murine model. Thus, in this present study, we characterize PPE39-induced immunological features by investigating the interaction of PPE39 with dendritic cells (DCs). PPE39-treated DCs display reduced dextran uptake and enhanced MHC-I, MHC-II, CD80 and CD86 expression, indicating that this PPE protein induces phenotypic DC maturation. In addition, PPE39-treated DCs produce TNF-α, IL-6 and IL-12p70 to a similar and/or greater extent than lipopolysaccharide-treated DCs in a dose-dependent manner. The activating effect of PPE39 on DCs was mediated by TLR4 through downstream MAPK and NF-κB signaling pathways. Moreover, PPE39-treated DCs promoted naïve CD4+ T-cell proliferation accompanied by remarkable increases of IFN-γ and IL-2 secretion levels, and an increase in the Th1-related transcription factor T-bet but not in Th2-associated expression of GATA-3, suggesting that PPE39 induces Th1-type T-cell responses through DC activation. Collectively, the results indicate that the complete form of PPE39 is a so-far-unknown TLR4 agonist that induces Th1-cell biased immune responses by interacting with DCs.
This article has an associated First Person interview with the first author of the paper.
INTRODUCTION
Tuberculosis (TB), which is caused by the bacterium Mycobacterium tuberculosis (Mtb), is one of the deadliest infectious diseases. Mtb was reported to have caused ∼10.4 million cases of tuberculosis and ∼1.7 million tuberculosis-related deaths in 2016 (World Health Organization, 2017). Moreover, the high risk of infection is worsened by the emergence of HIV co-infection and multidrug-resistant strains (Gandhi et al., 2010; Reece and Kaufmann, 2008). Approximately a century after its introduction, Bacille Calmette-Guerin (BCG) still remains the only licensed vaccine that protects against TB (Trunz et al., 2006). However, a major limitation of the BCG vaccine is that it confers protection against only a severe disseminated form of TB in childhood, and that it inconsistently protects adolescents and adults from pulmonary TB (Orme et al., 2015). To tackle this issue, targeting of Mtb antigens is regarded to be a promising strategy to elicit both preventative and protective immune responses against Mtb infection, suggesting that the identification of novel antigens is urgently required.
In recent years, the Beijing family of Mtb, the predominant Mtb strain genotype, has been related to major TB outbreaks worldwide; Mtb Beijing is prevalent in East Asia, as Beijing genotype infections are estimated to account for 92 % of Mtb infections in China, 72 % in Korea and 40 % in Vietnam (Glynn et al., 2002; Jeon et al., 2008, 2012; Parwati et al., 2010). Moreover, this genotype has been highly associated with drug resistance phenotypes, such as single-drug resistance and multidrug resistance, and exhibits high relapse rates from the latent state (Burman et al., 2009). Therefore, the genotypic, pathogenic and phenotypic variation among Mtb strains need to be considered because each Mtb strain displays a different level of virulence (Jeon et al., 2008).
Given this context, we have previously performed a comparative genomic analysis of Mtb K, a highly transmissible Mtb strain belonging to the Beijing genotype, and the laboratory-adapted Mtb H37Rv strain (Kim et al., 2017). Through this analysis, we did identify MTBK_24820, a Pro-Pro-Glu (PPE) gene family (GenBank accession no. AIB49026.1) inside the 5.7 kb inserted region of Mtb K (Han et al., 2015; Kim et al., 2017). MTBK_24820 belongs to the subfamily of PPE-major polymorphic tandem repeats (MPTRs), which harbors an Asn-X-Gly-X-Gly-Asn-X-Gly sequence in the C-terminus and is orthologous to Rv2353c (PPE39) of Mtb H37Rv. Moreover, MTBK_24820 expression in Mtb K was ∼8.1-fold higher than that in Mtb H37Rv in the previous microarray study (Kim et al., 2017). MTBK_24820 exists independently without Pro-Glu (PE) proteins, unlike other PE-PPE proteins, which are located in a row (Bitter et al., 2009). Notably, the H37Rv PPE39 protein is truncated at the N-terminus, which is consistent with the numerous genetic variations of PPE39, caused by single-nucleotide polymorphisms (SNPs) and IS6110 integration, observed among several Mtb strains, whereas the PPE39 of Mtb K harbors 259 extra amino acids in its N-terminus and is defined as the complete form of PPE39 (Kim et al., 2017; McEvoy et al., 2009).
Mtb is an intracellular pathogen armed with many strategies to subvert and evade the host immune response in order to better its survival chances. In particular, the majority of PE/PPE proteins have been reported to contribute to immune evasion and are described as mycobacterial virulence factors (Bhat et al., 2012; Li et al., 2005; Nair et al., 2011, 2009; Sampson, 2011). The PE/PPE family occupies 10% of coding capacity in the genome and harbors PE and PPE multiple repeats near the N-terminus; family members have been demonstrated to interact with host factors in innate immune pathways (Goulding et al., 2000; Sampson, 2011). For example, PPE18 suppresses inflammation by inhibiting macrophage effector functions (Nair et al., 2009). PPE24 and PPE53 have been demonstrated to promote Mtb survival inside macrophages (Mehta et al., 2006). Additionally, PPE57 and PPE26 induce the secretion of proinflammatory cytokines to establish fine-tuned Th1-biased environments (Su et al., 2015; Xu et al., 2015). Furthermore, some PE/PPE proteins, PPE68 (Okkels et al., 2003), PE-PGRS33 (Basu et al., 2007) and PPE36 (Le Moigne et al., 2005), have been reported to function in host-pathogen interactions, as these proteins are located in the cell wall. Thus, PE/PPE proteins are associated with TB pathogenesis, and a deeper understanding of these proteins might provide new protective strategies against Mtb infection.
Although protective immune correlates, i.e. measurable indicators of immunity, against TB in humans have not yet been identified, the protective role of Th1 cellular immunity is crucial to control Mtb infection in experimental animal models (Coler et al., 2018; Lindenstrøm et al., 2009). Efficient control of microbial infection not only requires immune activation upon pathogen invasion but also demands the generation of an appropriate immune response. In our previous study, mice that had been vaccinated with the complete form of PPE39 displayed enhanced antigen-specific multifunctional CD4+ T cells coexpressing IFN-γ and TNF-α, resulting in durable protection against the highly virulent Mtb K (Kim et al., 2017). These T cells were primed and fine-tuned in the draining lymph nodes by dendritic cells (DCs). Therefore, DCs play important roles as pivotal intermediaries of innate and adaptive immunity by processing and presenting antigens to naïve T cells (Trumpfheller et al., 2012). Furthermore, well-known Mtb antigens, such as ESAT-6, CFP10, Ag85 and HspX, which have been regarded as potential vaccine candidates, have been demonstrated to interact with DCs and elicit antigen-specific Th1-polarized responses during Mtb infection (Jung et al., 2014; Kim et al., 2016b; Lowrie, 2012; Palma et al., 2013; Wang et al., 2012; Yuk and Jo, 2014). In this context, antigens that drive Th1 polarization through activation of DCs can contribute to enhanced protection against Mtb infection. In this study, we investigate the immunological mechanisms mediated by complete PPE39 in order to determine how this antigen mediates Th1 polarization and study DC functions strengthened by the vaccine potential of PPE39 against a highly virulent Mtb infection.
RESULTS
Purification of recombinant PPE39 and its recognition during Mtb infections
Recombinant PPE39 was purified as previously described (Kim et al., 2017). PPE39 consists of 622 amino acids, and its molecular mass is ∼59 kDa. SDS-PAGE analysis of the purified protein showed the expected molecular mass. Any contaminating endotoxin in the protein preparations was removed (Fig. 1A). Endotoxin content was measured by a limulus amebocyte lysate (LAL) assay and was <0.1 EU/ml in the recombinant PPE39 preparations (data not shown). Then, we investigated whether the vaccine potential of PPE39 allows it to be recognized by the host immune system during in vivo Mtb infections (Kim et al., 2017). Thus, the ability to elicit antigen-specific increases of IFN-γ levels in response to stimulation with PPE39 or ESAT-6 (positive control) was analyzed in single-cell suspensions from the spleen and lung of Mtb K- and H37Rv-infected mice at 4 weeks (early) and 8 weeks (late) post infection (Fig. 1B). The analyzed cells isolated from Mtb K- and H37Rv-infected mice displayed similar levels of IFN-γ in response to ESAT-6 stimulation at both 4 and 8 weeks post infection. Previously, we have identified potential epitopes of PPE39 by measuring IFN-γ secretion levels in spleen cells from Mtb K-infected mice at 4 and 9 weeks post infection (Kim et al., 2017). Furthermore, synthetic peptides that overlap within the N-terminus of PPE39 (Table S1) were applied ex vivo to lung cells from Mtb H37Rv- and K-infected mice at 4 weeks post infection, where Mtb was delivered to the lungs by using aerosol. Notably, elevated levels of IFN-γ were observed upon stimulation with peptides comprising amino acids 85–102, 205–222 or 217–234 in lung cells from Mtb K-infected mice compared to those from Mtb H37Rv-infected mice (Fig. S1). This demonstrated that the N-terminus of PPE39, which is conserved in the Mtb K strain, possesses potent T-cell epitopes. Considering these epitopes, the complete form of PPE39 significantly increased levels of IFN-γ in both lung and spleen cells from Mtb K-infected mice at 4 and 8 weeks compared with those from Mtb H37Rv-infected mice (Fig. 1C,D). Notably, this complete form of PPE39 also exhibited immunogenicity during the Mtb H37Rv infection, as it elicited antigen-specific secretion of IFN-γ, suggesting that complete PPE39 is well recognized during both Mtb K and H37Rv infection.
PPE39 induces DC maturation and activation
Prior to studying the immunological functions in DCs, we examined whether PPE39 affects the viability of bone marrow-derived dendritic cells (BMDCs). Recombinant PPE39 did not influence cell viability when used at concentrations of <5 µg/ml, indicating that PPE39 does not interfere with our subsequent analyses (Fig. 2A). To investigate whether PPE39 affects BMDC maturation, BMDCs were stimulated for 24 h with 1 or 5 µg/ml PPE39, or 100 ng/ml lipopolysaccharide (LPS). LPS was used as a positive control for BMDC maturation. DC maturation is crucial for DC functions in antigen-presenting cells (APCs), represented by reduced endocytic capacity and increased cytokine levels. We examined whether PPE39 regulates BMDC endocytic activity. Compared with untreated BMDCs, PPE39-treated BMDCs exhibited a significantly decreased percentage of dextran+CD11c+ cells (Fig. 2B). Furthermore, PPE39 induced BMDC activation, as evidenced by the marked dose-dependent increase in the expression of the co-stimulatory molecules CD80 and CD86, as well as antigen-presenting molecules, including MHC-I and MHC-II (Fig. 2C). PPE39 induced not only BMDC but also splenic DC activation by significantly upregulating the expression of co-stimulatory molecules, including CD40, CD80, CD86, MHC-I, and MHC-II, to levels similar to those induced by ESAT-6 antigen that was used as another positive control (Fig. S2A and B). In addition, PPE39 induced BMDCs to secrete high levels of TNF-α, IL-6, IL-10 and IL-12p70 in a dose-dependent manner (Fig. 2D). Collectively, these results demonstrated that PPE39 promotes DC activation through phenotypic and functional maturation.
PPE39 activates DCs by interacting with TLR4
Pattern recognition receptors, such as Toll-like receptors (TLRs), recognize pathogen-associated molecular patterns from mycobacterial cell wall components. The role of TLRs in APCs is particularly important (Harding and Boom, 2010). Thus, we examined whether PPE39 is recognized by TLRs in DCs. To identify the TLRs on DCs that interact with PPE39, DCs of wild-type (WT), TLR2 knockout (TLR2−/−) or TLR4 knockout (TLR4−/−) mice were treated with PPE39. We measured the levels of surface molecule, and pro- and anti-inflammatory cytokines in PPE39-treated DCs from WT, TLR2−/− and TLR4−/− mice. Levels of cell-surface molecules (Fig. 3A) and pro- and anti-inflammatory cytokine secretion (Fig. 3B) were enhanced in WT and TLR2−/− DCs after PPE39 treatment. In contrast, these effects were strongly diminished in TLR4−/− DCs, indicating that PPE39 functions as a TLR4 agonist in DCs. These data clearly suggested that PPE39 induces DC maturation in a TLR4-dependent manner, causing increased expression of cell-surface molecules and pro- and anti-inflammatory cytokines.
PPE39 activates DCs via the MAPK and NF-κB signaling pathways
The mitogen-activated protein kinases (MAPKs) and nuclear factor (NF)-κB signaling pathways are essential for maturation of DCs induced by mycobacterial antigens (Bansal et al., 2010a; Byun et al., 2012). Therefore, we examined whether PPE39 activates MAPKs and NF-κB in DCs. After 4 h of serum starvation, DCs were stimulated with 5 μg/ml PPE39 for different durations (0, 5, 15, 30 or 60 min), followed by analysis of levels of phosphorylated p38 (p-p38), ERK1/2 (p-ERK1/2) and JNK (p-JNK), of phosphorylated IκB-α (p-IκB-α) as well as of IκB-α degradation and nuclear translocation of NF-κB p65. We found that PPE39 triggered the phosphorylation of DC MAPKs (including p38, ERK1/2 and JNK) (Fig. 4A). Additionally, we found that PPE39 induced phosphorylation and degradation of IκB-α and substantial translocation of NF-κB p65 from the cytosol to the nucleus (Fig. 4B,C). Because PPE39 promoted NF-κB activation by inducing its nuclear translocation, we next determined whether TLR2 or TLR4 are essential for NF-κB activation by using confocal microscopy. Upon stimulation, NF-κB p65 was found to translocate into the nucleus in DCs from WT and TLR2−/− mice but not in DCs from TLR4−/− mice, indicating that the immunological activation mediated by PPE39 depends on TLR4 (Fig. 4D). In addition, the PPE39-induced levels of TNF-α, IL-6, IL-12p70 and IL-10 by DCs were markedly reduced after pretreatment with pharmacological MAPK or NF-kB inhibitors (Fig. 4E). Although there were some exceptions, such as expression of IL-10 in response to the ERK1/2 inhibitor and expression of IL-12p70 in response to the ERK1/2 and JNK inhibitors, we were able to collectively demonstrate that these signaling pathways have a crucial role in DC activation mediated by PPE39.
PPE39 promotes naïve T-cell proliferation
The roles of mature DCs are characterized by interactions with T cells and antigen presentation followed by T-cell activation (Breedveld et al., 2017). Thus, we investigated whether PPE39-mediated DC maturation promotes naïve T-cell proliferation. CFSE-labeled ovalbumin (OVA)-specific CD4+ T cells were co-cultured with PPE39-treated BMDCs that were pulsed with OVA323-339. These T cells displayed increased proliferation compared with T cells co-cultured with OVA-pulsed BMDCs or with ESAT-6-treated BMDCs (Fig. 5A). In addition, naïve CD4+ T cells primed with PPE39-treated DCs produced more IFN-γ and IL-2 compares to those primed with untreated DCs. Compared to the CD4+ T cells co-cultured with ESAT-6-treated BMDCs, increased levels of IL-2 but not IFN-γ were observed as a consequence of the priming of naïve CD4+ T cells by PPE39-treated BMDCs, which correlated with a high level of T-cell proliferation (Fig. 5B). Furthermore, PPE39-treated splenic DCs also exhibited the ability to promote naïve CD4+ T-cell proliferation. This was accompanied by increased secretion levels of both IFN-γ and IL-2 compared with levels in untreated splenic DCs, and increased levels of IFN-γ – but not IL-2 – compared with splenic DCs treated with ESAT-6 as a control (Fig. 5C,D). These observations demonstrated that PPE-39 elicits optimal DC activation, resulting in the proliferation of naïve CD4+ T cells.
PPE39 promotes Th1 cell polarization through DC activation, resulting in enhanced control of intracellular mycobacterial growth
Accumulating evidence has demonstrated the protective contribution of the Th1-mediated immune response because Th1 responses are reduced during Mtb active infections (Derrick et al., 2011; Nandakumar et al., 2014; Qiu et al., 2012). As shown in Fig. 5, CD4+ T cells co-cultured with PPE39-treated DCs that had been pulsed with OVA323-339 showed a significant increase in T-cell proliferation compared with control cells. First, we investigated whether PPE39 antigen or PPE39-stimulated BMDCs induces the development of the CD4+ effector/memory T-cell subpopulation, in order to characterize PPE39 as a potent T-cell antigen during Mtb infection. We analyzed the expansion of PPE39-specific effector/memory T cells through the ex vivo treatment of PPE39 in lung and spleen cells from Mtb-infected mice at 4 weeks post infection. As shown in Fig. S3A,B, PPE39 stimulation significantly induced effector/memory T cells in both lung and spleen cells, except in spleen cells from Mtb H37Rv-infected mice. Additionally, PPE39-treated BMDCs also markedly induced the expansion of spleen effector/memory T cells from Mtb-infected mice compared with untreated and ESAT-6-treated BMDCs (Fig. S3C). Because PPE39 might function as a potent recall antigen, we next analyzed T-cell subsets associated with T-cell polarization. When BMDCs or splenic DCs stimulated with PPE39 were co-cultured with T-cells, both DC types exhibited elevated expression of a T-cell-associated T-box transcription factor, the Th1-specific regulator T-bet, to levels similar to those observed in BMDCs and splenic DCs stimulated with ESAT-6. However, in PPE39-treated DCs the increased expression of GATA-binding protein 3 (GATA-3), which is essential for Th2 development, was not observed as compared with either non-treated DCs or ESAT-6-treated DCs (Fig. 6A–C). Based on the PPE39-induced Th1 phenotype, we further evaluated whether T cells activated by PPE39-treated BMDCs can suppress mycobacterial growth inside bone marrow-derived macrophages (BMDMs). Importantly, T cells activated by PPE39-treated BMDCs specifically suppress bacterial growth compared to untreated BMDCs or ESAT-6-treated BMDCs accompanied by elevated levels of both IFN-γ and nitric oxide, both of which are important for bacterial control (Chan et al., 1992; Flynn et al., 1993) (Fig. 6D,E). These results suggest that PPE39 induces the expansion of effector/memory T cells and controls mycobacterial growth through polarization of Th1 cells.
DISCUSSION
In this current study, we first investigated any functional roles in interactions with DCs mediated by the complete form of PPE39. We found that the complete form of PPE39 is well recognized by the host immune system, as represented by antigen-specific increase of IFN-γ levels during both Mtb H37Rv and K infections (Fig. 1). Moreover, PPE39 acted as a TLR4 agonist, inducing DC maturation (Figs 2 and 3) and, ultimately, promoting proliferation of naïve CD4+ T-cells and polarization of Th1 cells (Figs 5 and 6). To the best of our knowledge, this is the first time that the complete form of PPE39, derived from the highly virulent Mtb K strain, has been described as a TLR4 agonist.
Although the immunological and pathophysiological roles of PE/PPE proteins are not yet fully understood, mounting evidence suggests that these proteins play key roles in antigenic diversity because of their immune evasion functions and genetic variability, as evidenced by their involvement in pathogenesis and host-pathogen interactions (Jiang et al., 2016; Sayes et al., 2012; Shah et al., 2015). Several secreted or cell-surface PE/PPE proteins could also have access to the extracellular compartment of infected host cells by potentiating interactions with host cells (Brennan and Delogu, 2002; Mukhopadhyay and Balaji, 2011). Interestingly, PE/PPE proteins appear to elicit either pro- or anti-inflammatory immune responses. For example, Rv1917c (PPE34) strongly induces secretion of IL-10 via the p38 MAPK signaling pathway by favoring a Th2-polarizing environment that is beneficial for Mtb (Bansal et al., 2010b; Nair et al., 2009). In contrast, Rv2769c (PE27) activates dendritic cells by inducing the secretion of proinflammatory cytokines and polarization of Th1 cells. Moreover, on the one hand, a single PE35-DNA vaccination – not in combination with PPE68 DNA– significantly induces antigen-specific Th1 responses in mice, with PE35 stimulation eliciting the secretion of IFN-γ but not of Th2-related cytokines, such as IL-5 and IL-10. On the other hand, however, PE35 has been reported to have an immunoregulatory role together with its cooperonic gene product PPE68 (Rv3873), by stimulating secretion of IL-10 in human macrophages in a TLR2-dependent manner, leading to an Mtb-favoring environment (Hanif et al., 2011; Tiwari et al., 2014). Nevertheless, PE/PPE proteins are still of great interest as they elicit potent CD4+ T-cell-derived cellular immune responses (Kim et al., 2016b).
Thus, we first analyzed the entire genome of the Mtb Beijing K strain and identified the complete form of PPE39, which possesses additional amino acids in the N-terminus compared to that of the Mtb H37Rv strain (Han et al., 2015; Kim et al., 2017). Considering that PPE39 from Mtb H37Rv has previously been reported to localize to the cell membrane (Målen et al., 2011), the complete form of PPE39 from Mtb K is highly likely to be immunogenic and to regulate the host immune system. We observed markedly increased levels of IFN-γ in response to the complete form of PPE39 within lung and spleen cells against both Mtb strains at 4 and 8 weeks post infection (Fig. 1). Although PPE39-specific secretion of IFN-γ in response to Mtb H37Rv infection could be elicited because PPE39 is also present in Mtb H37Rv, the increased levels of antigen-specific IFN-γ in response to Mtb K infection can be explained by additional amino acid epitopes in the N-terminus of full-length Mtb K PPE39. At 4 and 9 weeks, respectively, amino acids 85–102 and 217–234 within the overlapping complete form of PPE39 specifically induce secretion of IFN-γ in response to Mtb K – but not H37Rv – infection (Kim et al., 2017). Moreover, in our current study, we observed increased levels of IFN-γ in lung cells from Mtb K-infected mice in response to stimulation with amino acids 85–102, 205–222 or 217–234 compared to those detected in Mtb H37Rv-infected mice (Fig. S1). The difference in IFN-γ levels in response to PPE39 against Mtb H37Rv and K infection can, therefore, be associated with potent T-cell epitopes within the N-terminus of PPE39 from Mtb K.
DCs play a crucial role in priming and modulating T-cell-mediated immunity against Mtb infection (Steinman and Inaba, 1999). Mtb subverts CD4+ T-cell-derived cellular immunity by hampering the initiation of optimal T-cell responses through the regulation of DC functions (Choi et al., 2017). Thus, correct DC activation followed by T-cell stimulation is crucial to confer protective immunity against Mtb infection. In this study, we attempted to elucidate the immunological functions of the complete form of PPE39, by investigating PPE39 interaction with BMDCs or splenic DCs on the basis of its previously demonstrated vaccine potential against Mtb infection. PPE39 induced an increase increased expression of surface molecules on DCs and the secretion of immunoregulatory cytokines (Fig. 2 and Fig. S2). In this context, the complete form of PPE39 is capable of eliciting DC activation through phenotypic and functional maturation.
It appears that different members of the TLR family can interact with mycobacterial components. We have previously reported that several Mtb antigens, such as Rv0577 and Rv3628, can regulate innate and adaptive immune responses against Mtb infection via TLR2 engagement (Byun et al., 2012; Kim et al., 2016a). Although TLR2-mediated signaling pathways have been highlighted more than TLR4-mediated signaling pathways in Mtb infection, TLR4 signaling seems to be involved in apoptosis and Mtb growth (Grover et al., 2018; Kim et al., 2013; Tiwari et al., 2015). However, we have also demonstrated that TLR4 engagement by distinct mycobacterial antigens can direct Th1-biased immune responses; thus these Mtb antigens display potential as future vaccine antigens (Kim et al., 2013, 2018). In this context, cell membrane-associated PPE39 seems to be a key modulator of innate and adaptive immunity by optimally activating DCs in a TLR4-dependent manner in response to activated downstream signaling pathways (Figs 3 and 4).
CD4+ T cells play a vital role in the induction of adaptive immune responses and in directing the responses toward particular effector phenotypes (Diebold, 2008). The initiation of the Th1 immune response is crucial to effectively control Mtb replication within APCs. On the basis of these findings, maintenance of Th1-polarized immune responses is thought to be crucial in protecting the host against Mtb infection (Diebold, 2008). Our results clearly demonstrate that PPE39-treated DCs specifically promote proliferation of naïve CD4+ T cells accompanied by significantly elevated secretion levels of IFN-γ and IL-2, which are markedly related to Th1-biased responses (Fig. 5). In addition, PPE39-treated DCs induced the development of T cells with an effector phenotype (Fig. S3) as well as the enhanced expression of the Th1-related transcription factor T-bet in CD4+ cells but did not induce expression of the Th2-associated transcription factor GATA-3 (Fig. 6). Furthermore, T cells activated through co-culture with PPE39-treated DCs exhibited upregulated levels of nitric oxide and elevated levels of IFN-γ in Mtb-infected BMDMs. This activity contributed to an enhanced control of intracellular bacterial growth (Fig. 6) and indicates that PPE39 participates in both an enforced innate response and adaptive immunity by directing Th1 polarization.
Collectively, our data demonstrated that immunogenic MTBK_24820, the complete form of PPE39, plays a pivotal role in the activation of DCs in a TLR4-dependent manner and prompts naïve CD4+ T-cell proliferation accompanied by Th1 polarization. Importantly, the future vaccine potential of PPE39 as a novel antigen candidate is strengthened by this investigation, describing the interaction of PPE39 with DCs at the molecular level.
MATERIALS AND METHODS
Ethics statement
All animal experiments were performed according to the guidelines of the Korean Food and Drug Administration. Protocols for the animal studies used in this study were approved by the Ethics Committee and Institutional Animal Care and Use Committee (Permit Number: 2014-0197) of Yonsei University Health System (Seoul, Korea).
Animals
Six- to seven-week-old specific pathogen-free (SPF) female C57BL/6 (H-2Kb and I-Ab), TLR2 knockout (TLR2−/−), TLR4 knockout (TLR4−/−) and C57BL/6 OT-II T-cell receptor (TCR) transgenic mice at were purchased from Japan SLC, Inc. (Shijuoka, Japan) or the Jackson Laboratory (Bar Harbor, ME). Mice were maintained under barrier conditions in an SPF animal facility at Yonsei University Medical Research Center.
Antibodies and reagents
Recombinant mouse granulocyte-macrophage colony stimulating factor (GM-CSF) and recombinant mouse IL-4 were purchased from JW CreaGene (Gyeonggi, Korea). A fluorescein isothiocyanate (FITC)-annexin V/propidium iodide (PI) kit was purchased from R&D Systems (Minneapolis, MN). Dextran-FITC (molecular mass, 40,000 Da) was obtained from Sigma-Aldrich (St Louis, MO). LPS (from E. coli O111:B4) and palmitoyl-3-CysSer-(Lys)4 (Pam3CSK4) were purchased from InvivoGen (San Diego, CA). An endotoxin filter (END-X) and an endotoxin removal resin (END-X B15) were acquired from Associates of Cape Cod (East Falmouth, MA). AbFrontier (Seoul, Korea) synthesized the OT-II peptide (OVA323-339). Antibodies against the phosphorylated forms of p38 (p-p38; #9215, 1:1000), ERK1/2 (p-ERK1/2; #4377, 1:1000), JNK (p-JNK; #9251, 1:1000), and IκB-α (p-IκB-α; #2859, 1:1000) were obtained from Cell Signaling Technology (Danvers, MA). Antibodies against IκB-α, (#sc-371, 1:1000) NF-κB (p65; #sc-372, 1:1000) and histone H3 (#sc-10809, 1:1000) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Horseradish peroxidase (HRP)-conjugated anti-mouse IgG (#71045, 1:2000) and HRP-conjugated anti-rabbit IgG (#12-348, 1:2000) antibodies were obtained from Calbiochem (San Diego, CA). Anti-β-actin monoclonal antibody (AC-15; #A5441, 1:10,000) was purchased from Sigma-Aldrich (St Louis, MO). Anti-Thy1.2 (53-2.1, BV605; #140317, 1:300), anti-CD8 (53-6.7, BV711; #100747, 1:300), and anti-CD40 (3/23, FITC; #124607, 1:300) antibodies were purchased from BioLegend (San Diego, CA). Anti-CD44 (IM7, V450; #560451, 1:300), anti-IL-2 (JES6-5H4, PE-Cy7; #560538, 1:300), and anti-CD4 (RM4-5, PerCP-Cy5.5; #550954, 1:300) antibodies were obtained from BD Biosciences (San Diego, CA). The following antibodies were obtained from eBioscience (San Diego, CA): anti-CD80 (16-10A1, APC; #17-0801-81, 1:300), anti-CD86 (GL1, APC; #17-0862-82, 1:300), anti-MHC-I (34-1-2S, PE; #12-5998-82, 1:300), anti-MHC-II (M5/114.15.2, PE; #12-5321-82, 1:300), anti-CD11c (N418, PE-Cy7; #25-0114-82, 1:300), anti-CD62L (MEL-14, FITC; #11-0621-82, 1:300), anti-CD44 (IM7, PE; #12-0441-82, 1:300), anti-CD127 (A7R34, APC; #17-1271-82, 1:300) anti-T-bet (4B10, PE-Cy7; #25-5825-82, 1:200), and anti-GATA-3 (eBM2a, PE; #12-9966-42, 1:200). TNF-α, IL-6, IL-12p70, and IL-2 ELISA kits were obtained from eBioscience. IL-10, IL-5, and IFN-γ ELISA kits were obtained from BD Biosciences. Ultrapure CD11c Microbeads, Thy1.2 Microbeads, and a CD4+ T-cell isolation kit for cell isolation were obtained from Miltenyi Biotec (Miltenyi Biotec, Bergisch Gladbach, Germany).
Expression and purification of recombinant PPE39 protein
Transformed E. coli (pET22b-mtbk_24820) and vector control E. coli (pET22b in E. coli BL21 cells) were incubated in Luria-Bertani (LB) broth supplemented with 50 μg/ml ampicillin and 1 mM isopropyl-β-D-thio-galactoside (IPTG) at 37°C for 12 h. The recombinant E. coli were harvested through centrifugation (8000 g) and washed with lysis buffer (20 mM Tris-Cl, 0.5 M NaCl, 5 mM imidazole and 1 mM phenylmethylsulfonyl fluoride; Sigma-Aldrich). The obtained cells were lysed with lysis buffer (1 M DTT, lysozyme and 1 mM phenylmethylsulfonyl fluoride) and sonicated. To confirm PPE39 expression, the products from the recombinant E. coli were separated on a 12% SDS-PAGE gel. The separated proteins were electrophoretically transferred to a nitrocellulose (NC) membrane. The protein in the membrane was detected by using a mouse anti-histidine primary antibody (1:1000; 12631977, Lab Vision) at 4°C overnight. Additionally, the membrane was blocked with 5% skimmed milk and washed with PBST, followed by immediate addition of goat anti-mouse IgG secondary antibody (Sigma-Aldrich; #12-349, 1:5000). Immunoreactive bands were visualized with 3,3′-diaminobenzidine tetrahydrochloride (DAB HCl). Finally, the 6× His-tagged proteins attached to a nickel-nitrilotriacetic acid (Ni-NTA) column were purified in 50–100 mM elution buffer and assessed in a Limulus Amebocyte Lysate (LAL) test (Lonza) according to the manufacturer's instructions. Protein purity was confirmed by Coomassie Brilliant Blue R-250 staining.
Preparation of Mtb strains
Mtb H37Rv (ATCC 27294) was purchased from the American Type Culture Collection (ATCC, Manassas, VA). The Mtb K strain was obtained from the strain collection at the Korean Institute of Tuberculosis (Osong, Chungchungbuk-do, Korea). All of the mycobacterial strains used in this study were cultured and prepared as described previously (Rachman et al., 2006).
Analysis of immunological recognition during the infection with Mtb
Mice were infected with the Mtb K or H37Rv strain by using aerosol as previously described (Kwon et al., 2017). Briefly, mice were infected with the Mtb K or H37Rv strain in a calibrated inhalation chamber by using an airborne infection system (Glas-Col, Terre Haute, IN) for 60 min bearing ∼200 viable bacteria. At 4 and 8 weeks post infection, single cells prepared from the spleen and lung of Mtb-infected mice were stimulated with PPE39 (5 µg/ml) and ESAT-6 (1 µg/ml) for 24 h at 37°C. Subsequently, the level of IFN-γ in the cell culture supernatant was measured by using ELISA kits according to the manufacturers' instructions.
Cell culture
Murine bone marrow-derived DCs (BMDCs) were prepared and cultured as previously described (Kim et al., 2013). Briefly, bone marrow cells were plated in Petri dishes and cultured at 37°C in the presence of 5% CO2 using RPMI 1640 medium supplemented with 100 units/ml penicillin/streptomycin (Lonza, Basel, Switzerland), 10% fetal bovine serum (FBS) (Lonza, Basel, Switzerland), 50 μM mercaptoethanol (Lonza), 20 ng/ml GM-CSF and 5 ng/ml IL-4. On day 6, over 80% of the non-adherent cells expressed CD11c. In certain experiments, to obtain highly purified populations (>95% cell purity), DCs were labeled with bead-conjugated anti-CD11c monoclonal antibodies (Miltenyi Biotec; #130-052-001, 100 μl/108 cells), followed by positive selection on paramagnetic columns (LS columns; Miltenyi Biotec) according to the manufacturer's instructions. To isolate splenic DCs, spleens were put through 40-µm cell strainers (Corning, NY) to generate single-cell suspensions and red blood cells were lysed with ACK lysis buffer (Lonza) for 2 min at room temperature. After washing the cells, prepared single-cell suspensions were labeled with bead-conjugated monoclonal antibodies against CD11c (Miltenyi Biotec), followed by selection on paramagnetic columns (LS columns; Miltenyi Biotec) according to the manufacturer's instructions. Isolated mouse bone marrow cells were differentiated into BMDMs using L929-conditioned medium and maintained in L929-conditioned DMEM supplemented with 10% FBS and 100 units/ml of penicillin/streptomycin at 37°C in the presence of 5% CO2 for 6 days.
Cell viability analysis
PPE39 (5 μg/ml) was added to cultures of isolated BMDCs in 12-well plates (0.5×106 cells/ml). To investigate the effect of PPE39 on cells, the cell death pattern of DCs was analyzed after treatment with PPE39 (1 μg/ml and 5 μg/ml). After 24 h of treatment, the harvested DCs were stained with FITC-Annexin V/PI (BD Biosciences) according to the manufacturer's instructions. Then, DC cell death was measured by using a FACSVerse flow cytometer. The data were analyzed using FlowJo version 10 software (TreeStar, Ashland, OR).
Analysis of surface-molecule expression
After 24 h of treating DCs with PPE39, cells were washed with 2% FBS in phosphate-buffered saline (PBS) and stained with PE-Cy7-conjugated anti-CD11c, PE-conjugated anti-MHC-I, PE-conjugated anti-MHC-II, FITC-conjugated anti-CD40, APC-conjugated anti-CD80 and APC-conjugated anti-CD86 antibodies for 30 min at 4°C. Cells were washed three times with PBS containing 2% FBS and were then resuspended in PBS containing 2% FBS. Fluorescence was measured with a flow cytometer and FlowJo software. LPS, Pam3CSK4, and ESAT-6 were used as positive controls for DC maturation.
Antigen uptake by DCs induced by PPE39
DCs (0.5×106 cells) were equilibrated at 37 or 4°C for 45 min and then pulsed with fluorescein-conjugated dextran at a concentration of 50 μg/ml. Cold staining buffer was added to stop the reaction. The cells were washed three times, stained with monoclonal antibody against CD11c, and analyzed using a FACSVerse flow cytometer. Nonspecific binding of dextran to the DCs was determined by incubating DCs with dextran at 4°C, and subtracting the resulting background value from the specific binding values.
Quantification of cytokine levels
Cell culture supernatants were collected and stored at −80°C until use. A sandwich ELISA was used to detect TNF-α, IL-6, IL-10, IL-12p70, IFN-γ and IL-2 in culture supernatants of DCs and DCs co-cultured with T cells. In addition, the level of IFN-γ was measured in culture supernatants of Mtb-infected BMDMs that had been cultured together with T cells or with single cells isolated from lungs of Mtb-infected mice upon stimulation with antigens or synthesized peptides. The cytokine levels in the culture medium were analyzed using ELISA kits according to the manufacturers’ protocols. TNF-α, IL-6, IL-12p70 and IL-2 ELISA kits were obtained from eBioscience. IL-10, IL-5, and IFN-γ ELISA kits were obtained from BD Biosciences.
Immunoblot analysis
After stimulation with 5 μg/ml PPE39, DCs were lysed in 100 μl lysis buffer containing 50 mM Tris-HCl pH 7.5, 150 mM NaCl, 1% Triton X-100, 1 mM EDTA, 50 mM NaF, 30 mM Na4PO7, 1 mM phenylmethanesulfonyl fluoride, 2 μg/ml aprotinin, and 1 mM pervanadate. Immunoblotting was carried out as previously described (Kim et al., 2013). Epitopes in target proteins, including MAPKs and NF-κB, recognized by specific antibodies were visualized using an ECL Advance kit (GE Healthcare, Little Chalfont, UK).
Confocal laser scanning microscopy
DCs were plated overnight on poly-L-lysine-coated glass coverslips. After treatment with PPE39 (5 μg/ml), cells were fixed in 4% (w/v) paraformaldehyde/PBS, permeabilized in 0.1% (v/v) Triton X-100/PBS, and then blocked with 2% bovine serum albumin (BSA) in PBS containing 0.1% Tween-20 (PBS/T) for 2 h. Cells were then incubated at room temperature with a mouse anti-NF-κB/p65 polyclonal antibody for 2 h. After washing with PBS/T, cells were incubated again with a FITC-conjugated secondary antibody (Molecular Probes; A11008, 1:200) in a dark room for 1 h, and were then stained with DAPI for 10 min at room temperature. Cell morphology and fluorescence intensity were observed by using a confocal laser scanning microscope (Zeiss LSM510 Meta). Images were acquired using LSM510 Meta software and processed using LSM image examiner.
Pharmacological inhibitors of signaling pathways
All of the pharmacological inhibitors were purchased from Calbiochem (San Diego, CA). Dimethyl sulfoxide (Sigma-Aldrich) was added to cultures at 0.1% (vol/vol) as a solvent control. DCs were washed with PBS and pretreated with inhibitors in RPMI 1640 medium for 1 h prior to the treatment with PPE39 (5 μg/ml) for 24 h. The following inhibitors were used: SB203580 (inhibiting p38; 20 μM), U0126 (inhibiting ERK1/2; 10 μM), SP600125 (inhibiting JNK; 10 μM), and Bay11-7082 (inhibiting NF-κB; 20 μM). In all experiments in which inhibitors were used, a tested concentration was used after careful titration experiments by MTT assay, assessing the viability of DCs.
Proliferation and polarization of CD4+ T cells
CD4+ T cells were isolated from spleen cell suspensions prepared from OT-II mice by using a MACS column. CD4+ T cells were stained with 1 μM CFSE (Invitrogen) as previously described (Kim et al., 2013). Subsequently, CFSE-labeled CD4+ T cells were cultured together with stimulated BMDCs or splenic DCs (LPS; 100 ng/ml, ESAT-6; 1 µg/ml, PPE39; 5 µg/ml, 2×105 cells per well) pulsed with OVA peptide (OVA323–339) at a DC:T cell ratio of 1:10. On day 3 of the co-culture, CD4+ T cells were stained with monoclonal antibodies against Thy1.2 and CD4 for 30 min at 37°C, and the cell proliferation rate was analyzed using a CytoFLEX S Flow cytometer (Beckman Coulter, Indianapolis, IN). Supernatants were harvested and ELISA was used to measure the protein levels of IFN-γ and IL-2. For the T-cell polarization assays, labeled T cells were cultured together with stimulated BMDCs or splenic DCs in the presence of OVA peptide for 24 h. T cells were first stained with monoclonal antibodies against Thy1.2 and CD4 for 30 min in the dark, after which they were fixed and permeabilized using a fixation/permeabilization buffer (eBioscience) according to the manufacturer's instructions. The transcription factors T-bet and GATA-3 were detected with fluorescein-conjugated antibodies in a permeabilization buffer. Cells were subsequently analyzed by using a CytoFLEX S Flow cytometer.
Analysis of effector/memory T cells
To analyze memory responses, C57BL/6 mice were aerogenically infected with Mtb K and H37Rv as described above. The isolated lung and spleen cells at 4 weeks post infection were stimulated with ESAT-6 (1 µg/ml) or PPE39 (5 µg/ml) and incubated for 12 h at 37°C. Then, the cells were stained with BV605-conjugated anti-Thy1.2, PerCP-Cy5.5-conjugated anti-CD4, FITC-conjugated anti-CD62L, APC-conjugated anti-CD127 and PE-conjugated anti-CD44, and activation of effector/memory T cells was analyzed by CytoFLEX S Flow cytometer. For ex vivo co-culture with BMDCs, splenocytes from Mtb-infected mice were labeled with bead-conjugated monoclonal antibody against Thy1 (Thy1.2, Miltenyi Biotec; #130-049-101, 10 μl/107 cells), which was followed by positive selection on paramagnetic columns (LS columns; Miltenyi Biotec). BMDCs (2×105 cells) were stimulated with LPS, ESAT-6 or PPE39 for 24 h. After washing with PBS, the BMDCs were cultured together with Thy1.2-enriched splenocytes (2×106 cells) from Mtb K-infected mice at a DC:T-cell ratio=1:10. On day 3 of the co-culture, activation of effector/memory T cells was analyzed.
Measurement of intracellular bacterial growth in macrophages
Intracellular bacterial growth was evaluated as previously described with slight modifications (Choi et al., 2018). Briefly, BMDMs (1.5×105 cells/well in 48 well plates) were infected with Mtb K at one bacterium per cell in antibiotic-free DMEM and incubated for 4 h. Amikacin (200 µg/ml) was then used to treat Mtb-infected BMDMs to remove extracellular mycobacteria. Next, the T-cell mixture, which was co-cultured in advance with agonist- or antigen-activated BMDCs for 3 days (DC:T-cell ratio=1:10), was added to Mtb-infected BMDMs and incubated for 3 days. Three days after infection, the cells were lysed with 0.05% Triton X-100 (Sigma-Aldrich) prepared in sterile water. The lysates were then serially diluted in sterile PBS and spotted onto Middlebrook 7H10 agar (Difco) supplemented with 10% OADC. After incubation for 4 weeks at 37°C, the bacterial colony-forming units (CFU) were counted.
Nitric oxide detection
Nitric oxide (NO) was detected through the enzymatic reduction of NO3 by nitrate reductase followed by spectrophotometric analysis of total nitrite using Griess reagent. Analysis was performed using a commercial kit according to the manufacturer's instructions (iNtRON Biotechnology, Gyeonggi, Korea).
Statistical analysis
All experiments were repeated at least three times and showed consistent results. Statistical significance between two groups was determined using unpaired Student's t-test; more than three groups were assessed usng one-way ANOVA followed by Tukey's multiple comparison test and statistical software (GraphPad Prism Software, version 5.01; GraphPad Software, San Diego, CA). Data in graphs are expressed as mean value±standard deviation (s.d.). *P<0.05, **P<0.01 and ***P<0.001 were considered statistically significant.
Acknowledgements
Footnotes
Author contributions
Conceptualization: S.-N.C., S.J.S.; Methodology: H.-H.C., K.W.K., S.M.K., E.C., A.K.; Formal analysis: S.J.H., S.M.K., E.C.; Investigation: H.-H.C., K.W.K.; Resources: S.J.H., A.K., S.-N.C., S.J.S.; Writing - original draft: H.-H.C., K.W.K.; Writing - review & editing: S.J.S.; Supervision: S.J.S.; Project administration: S.J.S.
Funding
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (NRF-2016R1A2A1A05005263 and NRF-2019R1A2C2003204) and by a grant from the Korean Health Technology R&D Project of the Ministry of Health and Welfare (HI17C0175), Republic of Korea.
References
Competing interests
The authors declare no competing or financial interests.