Ninjurin-1 Negatively Regulates Humoral and Cellular Immune Responses Induced by the Saponin-Based Adjuvant Quil-A in Mice

Abstract

Adjuvants are co-administered with antigens to enhance vaccine-induced protection. Saponins are plant-derived compounds with adjuvant properties, some of which are used in licensed vaccines. Macrophages and dendritic cells (DCs) exposed to saponin-based adjuvants have been reported to exhibit NLRP3-dependent interleukin-1 beta (IL-1β) release, and NLRP3 signaling has been shown to limit their adjuvant activity. Saponin-based adjuvants also induce plasma membrane rupture (PMR) and the release of high-molecular-weight intracellular molecules; however, the molecular mechanisms that mediate PMR and its impact on adjuvant-induced immune responses remain unclear. Here, we investigated the involvement of Ninjurin-1 (NINJ1), a key executor of PMR, in Quil-A-induced PMR and its immunological consequences. Upon stimulation with the saponin mixture Quil-A, peritoneal macrophages and bone marrow-derived dendritic cells (BMDCs) showed NLRP3-independent PMR but NLRP3-dependent IL-1β release. Quil-A-induced PMR was almost completely suppressed in Ninj1^−/− peritoneal macrophages and BMDCs compared with wild-type cells, whereas IL-1β release remained unaffected by NINJ1 deficiency. Immunization with Quil-A and ovalbumin (OVA) increased OVA-specific serum immunoglobulin G (IgG), IgG2b, and IgG2c levels in Ninj1^−/− mice compared with wild-type mice. Splenocytes from Ninj1^−/− mice produced higher levels of interferon-gamma upon stimulation with class I- and class II-restricted OVA peptides than those from wild-type mice. Ninj1^−/− mice also showed a higher frequency of OVA-bearing cells, particularly monocyte-derived DCs, in the draining lymph nodes. These results demonstrate that NINJ1 is critical for Quil-A-induced PMR and that NINJ1-mediated PMR negatively regulates Quil-A-induced humoral and cellular immune activation by restricting antigen delivery via antigen-presenting cells.

INTRODUCTION

Vaccination remains the most effective strategy for reducing the morbidity and mortality associated with infectious diseases. Many current vaccines use purified protein, polysaccharide, or recombinant antigens instead of whole pathogens. Although these vaccines are generally safe, they have weak intrinsic immunogenicity and often require additional immune stimulation to provide effective protection. This is typically accomplished by including adjuvants, which are substances that enhance immune responses to antigens.¹⁾ However, the mechanisms of many clinically approved adjuvants remain incompletely understood, limiting their rational design and optimized use in vaccination. A better understanding of their mechanisms could guide evidence-based vaccine development strategies.

Activation of antigen-presenting cells (APCs), such as dendritic cells (DCs) and macrophages, is particularly critical for promoting adaptive immunity. Adjuvants achieve this through multiple modes of action. Some mimic pathogen-associated molecular patterns and directly engage pattern recognition receptors, thereby activating APCs.²⁾ Other adjuvants function through a distinct mechanism that involves the induction of cellular perturbations, leading to the release of intracellular molecules that modulate the activity of surrounding immune cells, including APCs.³⁾ Several clinically approved adjuvants belong to this latter category. For example, aluminum-based adjuvants induce the release of host DNA at the injection site, which enhances the recruitment of APCs.⁴⁾ The release of intracellular molecules induced by such adjuvants is an active process mediated by molecular mechanisms.^5,6) A detailed understanding of the molecules responsible for releasing intracellular molecules, as well as the effector molecules released during this process that modulate immune responses, could enable the rational design and control of adjuvants that act through pathways promoting the release of intracellular mediators.

Saponins are plant-derived compounds, some of which have been incorporated into clinically approved adjuvants.^7,8) Compared with traditional aluminum-based adjuvants, which primarily induce humoral immunity, saponin-based adjuvants can induce both strong cellular and humoral immunity and thus have attracted increasing attention.⁹⁾ Several mechanisms of action have been reported for saponin-based adjuvants, including the promotion of antigen endocytosis and the induction of Syk activation in APCs, as well as the direct delivery of costimulatory signals to T cells.^10,11) Importantly, some saponin-based adjuvants have been shown to activate the NLRP3 inflammasome and trigger plasma membrane rupture (PMR) at high concentrations.^12,13) Activation of the NLRP3 inflammasome is considered to induce pore formation in the plasma membrane, leading to the release of low-molecular-weight intracellular molecules such as interleukin-1 beta (IL-1β).¹⁴⁾ Notably, Nlrp3^−/− mice immunized with QS-21, a saponin-based adjuvant used in multiple vaccines, showed higher serum levels of antigen-specific immunoglobulin G (IgG) and increased production of interferon-gamma (IFN-γ) by T cells compared with wild-type controls, suggesting that molecules released from cells in an NLRP3-dependent manner may restrict the adjuvant effects of QS-21.¹²⁾ However, NLRP3-deficient cells still exhibited a loss of Calcein fluorescence upon stimulation with QS-21, suggesting that PMR occurs via an NLRP3-independent pathway.¹²⁾ PMR leads to the release of high-molecular-weight intracellular molecules that can mediate inflammation.¹⁵⁾ Elucidating the molecular mechanisms of PMR induced by saponin-based adjuvants would help to better understand and modulate their adjuvant effects.

Recently, Ninjurin-1 (NINJ1) has been identified as a key executor of PMR in response to stimuli such as lipopolysaccharide.^16–18) NINJ1 is a small double-pass transmembrane protein broadly expressed across multiple tissues and hematopoietic lineages, with relatively high transcript levels in innate immune cells such as macrophages.¹⁹⁾ It is primarily localized to the plasma membrane, where it can oligomerize in response to cellular stress to form filamentous assemblies that disrupt the lipid bilayer and induce PMR.²⁰⁾ Given that saponin-based adjuvants also induce PMR, it is plausible that NINJ1 mediates PMR induced by these adjuvants as well. To date, the role of NINJ1 in modulating adjuvant-induced immune responses has not been investigated. In this study, we aimed to determine whether NINJ1 mediates PMR induced by the saponin-based adjuvant Quil-A and to evaluate its impact on Quil-A-induced humoral and cellular immune responses.

MATERIALS AND METHODS

Reagents

Quil-A adjuvant, monophosphoryl lipid A (MPLA), and OVA_323–339 peptide were purchased from InvivoGen (San Diego, CA, U.S.A.). OVA_257–264 peptide was purchased from Selleck Chemicals (Houston, TX, U.S.A.). Ovalbumin, Low Endotoxin was purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan). Ovalbumin, Fluorescein Conjugate and recombinant mouse granulocyte-macrophage colony-stimulating factor (GM-CSF) were purchased from Thermo Fisher Scientific (Waltham, MA, U.S.A.). The Cytotoxicity LDH Assay Kit-WST was purchased from Dojindo Laboratories (Kumamoto, Japan). CellTiter-Glo 2.0 Assay was purchased from Promega (Madison, WI, U.S.A.). PE-conjugated anti-mouse F4/80 antibody (BM8), PE-Cy7-conjugated anti-mouse CD11b antibody (M1/70), APC/Cy7-conjugated anti-mouse CD11c antibody (N418), and enzyme-linked immunosorbent assay (ELISA) kits for mouse IL-1β and IFN-γ were purchased from BioLegend (San Diego, CA, U.S.A.). Block Ace was purchased from KAC (Kyoto, Japan). Goat Anti-Mouse IgG, Human ads-HRP was purchased from SouthernBiotech (Birmingham, AL, U.S.A.). Goat Anti-Mouse IgG1 (HRP), Goat Anti-Mouse IgG2b heavy chain (HRP), and Goat Anti-Mouse IgG2c heavy chain (HRP) were purchased from Abcam (Cambridge, U.K.). Red Blood Cell Lysing Buffer Hybri-Max was purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). ELISA POD Substrate TMB Kit, RPMI 1640 medium, penicillin-streptomycin mixed solution, MEM non-essential amino acids solution, Dulbecco’s Phosphate Buffered Saline, and 2-mercaptoethanol were purchased from Nacalai Tesque (Kyoto, Japan).

Mice

Wild-type C57BL/6N and C57BL/6J mice were purchased from CLEA Japan (Shizuoka, Japan) and Japan SLC, Inc. (Shizuoka, Japan). Nlrp3^−/− mice (JAX stock #021302) were obtained from Jackson Laboratory (Bar Harbor, ME, U.S.A.).²¹⁾ To generate Ninj1^−/− mice, C57BL/6N-Ninj1^{tm1a(KOMP)Wtsi}/MbpMmucd sperm were imported from the KOMP repository (MMRRC:048810-UCD, http://www.komp.org) and in vitro fertilized with eggs from wild-type C57BL/6N mice. Homozygous mice were generated by breeding the heterozygote mice. The mice were housed in standard cages in a temperature-controlled room under a 12-h light/dark cycle at the Animal Care Facility of the Graduate School of Pharmaceutical Sciences, The University of Osaka, during the experimental period. The mice were provided with standard laboratory mouse chow and drinking water ad libitum. All animal experiments were approved by the Animal Care and Use Committee of the Graduate School of Pharmaceutical Sciences at The University of Osaka.

Cell Preparation and Stimulation

Peritoneal exudate cells (PECs) were isolated from mice 3 d after intraperitoneal injection with 2 mL of 4% thioglycolate and used as macrophages. To prepare DCs, mouse bone marrow cells were cultured in RPMI 1640 supplemented with 8% fetal calf serum, 10 ng/mL GM-CSF, and 50 μM 2-mercaptoethanol. The culture medium was replaced every 2 d. On day 9, cells were harvested and used as bone marrow-derived DCs (BMDCs). PECs and BMDCs were seeded at a density of 1 × 10⁵ cells per well in flat-bottom 96-well cell culture plates and cultured in RPMI 1640 supplemented with 8% fetal calf serum overnight. On the following day, cells were primed with MPLA (500 ng/mL) or left unprimed for 6 h, followed by stimulation with Quil-A (30 μg/mL) for 2 h. Culture supernatants were collected after centrifugation at 440 × g for 5 min at 4°C. Supernatants and cell samples were analyzed as described below.

Measurement of Cytokine Levels in Culture Supernatants

The levels of IL-1β and IFN-γ in the culture supernatants were measured using ELISA, in accordance with the manufacturer’s instructions.

Measurement of LDH Release

Lactate dehydrogenase (LDH) release from mouse PECs and BMDCs was measured as a marker of PMR using the Cytotoxicity LDH Assay Kit-WST according to the manufacturer’s instructions.

Measurement of Cellular Activity

The activity of mouse PECs and BMDCs was measured using CellTiter-Glo 2.0 Cell Viability Assay according to the manufacturer’s instructions.

Immunization

Mice were immunized subcutaneously at the base of the tail with 10 μg Quil-A plus 20 μg OVA on days 0, 28, and 42. Blood samples were collected from the retro-orbital vein before the first immunization and on days 7 and 28 after immunization. Serum OVA-specific antibody levels in serum were measured as described below. Ten days after the final immunization, spleens were collected and subjected to T cell stimulation assay as described below.

Measurement of OVA-Specific Antibody Levels in the Serum

Clear Flat-Bottom Immuno Nonsterile 96-Well Plates (Thermo Fisher Scientific) were coated overnight with 10 μg/mL of OVA in 100 mM bicarbonate buffer, pH 9.4. The plates were blocked with 2× Block Ace buffer for 1 h at room temperature. The plates were incubated with mouse serum (diluted 1 : 256 in 1× Block Ace buffer) for 2.5 h. Each HRP-conjugated secondary antibody (goat anti-mouse IgG, IgG1, IgG2b heavy chain, and IgG2c heavy chain) was diluted at 1 : 2000 in 1× Block Ace and added to the plates, which were then incubated at room temperature for 2 h. The wells were washed four times with phosphate-buffered saline plus 0.05% (v/v) Triton X-100 between each step. After adding TMB, the plates were developed at room temperature in the dark. Finally, 1 M HCl was added, and the absorbance of the solution was measured at 450 nm.

T Cell Stimulation Assay

Spleens were homogenized to obtain single-cell suspensions. Splenocytes were filtered through a 40-μm cell strainer, and red blood cells were lysed using Red Blood Cell Lysing Buffer Hybri-Max. Cells were seeded into flat-bottom 96-well cell culture plates at a density of 1 × 10⁶ cells per well in 200 μL of RPMI 1640 supplemented with 8% fetal bovine serum, 1% penicillin-streptomycin, 1% non-essential amino acids, and 50 μM 2-mercaptoethanol. Antigen-specific stimulation was performed by adding OVA_323-339 and OVA_257-264 peptides (10 μg/mL each), while control wells contained medium alone. Cells were cultured at 37°C for 3 d, and the supernatants were analyzed by ELISA.

Flow Cytometric Analysis

Cells collected from the inguinal lymph nodes of mice injected subcutaneously at the base of the tail with 10 μg Quil-A plus 100 μg fluorescein isothiocyanate (FITC)-labeled OVA were stained with antibodies against mouse F4/80, CD11b, and CD11c in accordance with the manufacturer’s instructions. FITC-positive cells were identified as OVA-bearing cells. F4/80⁺ CD11c⁺ CD11b⁺ cells, F4/80⁻ CD11c⁺ cells, and F4/80⁺ CD11c⁻ CD11b⁺ cells were identified as monocyte-derived DCs (MoDCs), DCs, and macrophages, respectively. Data were acquired using a flow cytometer (CytoFLEX; Beckman Coulter, Brea, CA, U.S.A.) and analyzed using FlowJo software (Tree Star, Ashland, OR, U.S.A.).

BMDCs were seeded into 96-well plates at a density of 2 × 10⁵ cells per well in 200 μL of RPMI 1640 medium supplemented with 8% fetal bovine serum and 1% penicillin-streptomycin, containing FITC-labeled OVA (100 μg/mL). The cells were incubated at 37°C for 1 h. After incubation, cells were washed and analyzed by flow cytometry using antibodies against mouse CD11b and CD11c, as described above.

Statistical Analysis

Statistical analysis was performed with GraphPad Prism 8.0 (Boston, MA, U.S.A.). Unpaired two-tailed Student’s t-tests were used for comparisons between two groups. A one-way ANOVA, followed by Tukey–Kramer post hoc test, was performed to compare multiple groups. Statistical significance was set at p < 0.05.

RESULTS

Quil-A-Induced PMR in Mouse Macrophages and DCs Depends on NINJ1

Quil-A, a crude saponin extract from Quillaja saponaria, is widely used as a model saponin-based adjuvant in experimental studies.²²⁾ It contains QS-21, a more highly purified fraction.¹²⁾ We first examined whether NINJ1 mediates PMR in PECs and BMDCs induced by Quil-A. To adequately induce NLRP3 inflammasome activation and subsequent IL-1β release, cells were primed with the Toll-like receptor 4 (TLR4) agonist MPLA. MPLA is an attenuated derivative of lipopolysaccharide and is used as an immunostimulatory component in clinically approved saponin-based adjuvant formulations.⁷⁾ In wild-type PECs and BMDCs, Quil-A stimulation induced LDH release both with and without MPLA priming, indicating that PMR occurs independently of TLR4 activation (Figs. 1A, 1B). IL-1β was markedly released only in MPLA-primed cells after Quil-A stimulation. Quil-A induced minimal IL-1β release from Nlrp3^−/− cells, whereas LDH release was comparable to that in wild-type cells. Importantly, Ninj1^−/− cells exhibited a reduction in LDH release upon Quil-A stimulation, whereas IL-1β release remained unaffected (Figs. 1C, 1D). These results suggest that Quil-A-induced PMR is dependent on NINJ1 but not on NLRP3, while NLRP3-dependent IL-1β release occurs independently of NINJ1.

Fig. 1. Antigen-Presenting Cells Stimulated with Quil-A Exhibit NINJ1-Dependent LDH Release and NLRP3-Dependent IL-1β Release

(A) Peritoneal exudate cells (PECs) and (B) bone marrow-derived dendritic cells (BMDCs) were isolated from wild-type and Nlrp3^−/− mice. Cells were primed with 500 ng/mL MPLA or left unprimed for 6 h, followed by stimulation with 30 μg/mL Quil-A for 2 h. Supernatants were collected and analyzed for interleukin-1 beta (IL-1β) and lactate dehydrogenase (LDH) release. (C) PECs and (D) BMDCs from wild-type and Ninj1^−/− mice were primed with MPLA or left unprimed, followed by stimulation with Quil-A. Supernatants were collected and analyzed for LDH and IL-1β release. Data are shown as means ± standard deviation (S.D.) of values from triplicate wells. ^****p < 0.0001; ns: not significant. The experiment was independently repeated twice with similar results.

NINJ1 Negatively Regulates Quil-A-Induced Antigen-Specific Humoral and Cellular Immune Responses

To assess whether NINJ1 influences humoral immunity induced by saponin-based adjuvants, wild-type and Ninj1^−/− mice were immunized subcutaneously with Quil-A-adjuvanted OVA. On day 7 after immunization, an early increase in OVA-specific total IgG was observed in Ninj1^−/− mice, whereas the levels of IgG subclasses, including IgG1, IgG2b, and IgG2c, were comparable between wild-type mice and Ninj1^−/− mice (Fig. 2). By day 28, higher serum levels of OVA-specific total IgG, IgG2b, and IgG2c were observed in Ninj1^−/− mice compared with wild-type mice, whereas IgG1 levels were almost comparable. Because saponin-based adjuvants also elicit potent cellular immunity, we further examined whether NINJ1 modulates Quil-A-induced OVA-specific T cell responses. Splenocytes from Ninj1^−/− mice produced higher levels of IFN-γ upon stimulation with OVA_323-339 and OVA_257-264 than those from wild-type mice (Figs. 3A, 3B), suggesting that both the Th1 and cytotoxic T lymphocyte responses were enhanced in the absence of NINJ1. IL-4 and IL-17 were also measured but were undetectable under the experimental conditions used (data not shown). Collectively, these results indicate that NINJ1 restrains Quil-A-induced antigen-specific humoral and cellular immune responses.

Fig. 2. Quil-A-Immunized Ninj1^−/− Mice Exhibit Enhanced Antigen-Specific IgG Production

Wild-type and Ninj1^−/− mice were immunized subcutaneously with 10 μg Quil-A plus 20 μg ovalbumin (OVA). Blood samples were collected before the first immunization and on days 7 and 28 after immunization. OVA-specific total IgG, IgG1, IgG2b, and IgG2c antibodies were measured by ELISA. Data are presented as means ± S.D. (n = 4). *p < 0.05; **p < 0.01; ****p < 0.0001; ns: not significant. The experiment was independently repeated twice with similar results.

Fig. 3. Quil-A-Immunized Ninj1^−/− Mice Exhibit Enhanced Antigen-Specific IFN-γ Production from Splenocytes

Wild-type and Ninj1^−/− mice were immunized subcutaneously on days 0, 28, and 42 with 10 μg Quil-A plus 20 μg ovalbumin (OVA), and splenocytes were isolated on day 52. Splenocytes were stimulated with (A) OVA_323-339 or (B) OVA_257-264 peptides (10 μg/mL each) for 3 d. Supernatants were collected and analyzed for IFN-γ production. Data are presented as means ± S.D. (n = 4). ****p < 0.0001. The experiment was independently repeated twice with similar results.

NINJ1 Restricts Antigen Delivery by MoDCs to Draining Lymph Nodes after Quil-A-Adjuvanted Immunization

Quil-A-induced PMR depends on NINJ1, and PMR is accompanied by the release of intracellular molecules, which may modulate the activity of the surrounding APCs. We then investigated whether antigen delivery to the draining lymph nodes by APCs is affected by NINJ1 deficiency in mice following Quil-A-adjuvanted immunization. Mice were immunized with FITC-labeled OVA and Quil-A, and OVA-bearing (FITC⁺) cells were analyzed among the total cells as well as within MoDCs (F4/80⁺ CD11c⁺ CD11b⁺), conventional DCs (F4/80⁻ CD11c⁺), and macrophages (F4/80⁺ CD11c⁻ CD11b⁺) (Fig. 4A). Compared with wild-type mice, Ninj1^−/− mice exhibited a higher frequency of FITC⁺ cells in the draining lymph nodes (Fig. 4B). The frequency of FITC⁺ cells among MoDCs was increased in Ninj1^−/−mice, whereas that in conventional DCs and macrophages was comparable (Figs. 4C–4E). The frequency of whole MoDCs, DCs, and macrophages among total cells was comparable between Ninj1^−/− and wild-type mice (data not shown). Thus, NINJ1 negatively regulates APC-mediated antigen delivery to the draining lymph nodes after Quil-A-adjuvanted immunization, particularly in MoDCs.

Fig. 4. Quil-A-Immunized Ninj1^−/− Mice Exhibit Increased Frequencies of Antigen-Bearing Cells in the Draining Lymph Nodes

Wild-type and Ninj1^−/− mice were immunized subcutaneously with 100 μg of FITC-conjugated OVA plus 10 μg of Quil-A. Leukocytes from the inguinal lymph nodes were harvested 18 h post-immunization. Single-cell suspensions were stained with mouse anti-F4/80, anti-CD11c, and anti-CD11b antibodies and analyzed by flow cytometry. (A) Gating strategy used to define monocyte-derived dendritic cells (MoDCs), dendritic cells (DCs), and macrophages. Leukocytes were gated based on forward scatter (FSC-A) versus side scatter (SSC-A) and then selected for singlets using FSC-A versus FSC-H. Within singlets, MoDCs (F4/80⁺ CD11c⁺ CD11b⁺), DCs (F4/80⁻ CD11c⁺), and macrophages (F4/80⁺ CD11c⁻ CD11b⁺) were identified. The frequency of OVA-bearing (FITC⁺) (B) leukocytes, (C) MoDCs, (D) DCs, and (E) macrophages among total leukocytes was analyzed. Data are presented as means ± S.D. (n = 7). ^*p < 0.05; ns: not significant. The experiment was independently repeated twice with similar results.

To investigate the mechanism underlying the increased antigen delivery in Ninj1^−/− mice, we assessed intracellular ATP levels in PECs and BMDCs after Quil-A stimulation. ATP levels may influence numerous cellular functions, including migration, antigen processing, and other metabolic processes, all of which could potentially contribute to the frequency of antigen-bearing APCs in the draining lymph nodes. Quil-A reduced intracellular ATP levels in both wild-type and Ninj1^–/– cells to a comparable extent (Figs. 5A, 5B). Additionally, we examined whether NINJ1 deficiency affects the antigen uptake capacity of BMDCs. Among BMDCs, the proportion of the major DC subset that expresses CD11c and CD11b did not differ between Ninj1^−/− and wild-type mice (Fig. 6A). These DCs showed comparable uptake of FITC-labeled OVA between Ninj1^−/− and wild-type mice (Fig. 6B). Therefore, the increased frequency of antigen-bearing APCs observed in Ninj1^−/− mice may not result from differences in ATP availability or their antigen uptake capacity.

Fig. 5. ATP Decrease in Quil-A-Stimulated Antigen-Presenting Cells Is Unaffected by NINJ1 Deficiency

(A) Peritoneal exudate cells and (B) bone marrow-derived dendritic cells from wild-type and Ninj1^−/− mice were stimulated with 30 μg/mL Quil-A for 6 h. Cells were collected and analyzed for intracellular ATP levels. Data are shown as means ± S.D. of values from triplicate wells. The experiment was independently repeated twice with similar results. ns: not significant.

Fig. 6. Antigen Uptake Capacity of BMDCs Is Unaffected by NINJ1 Deficiency

Bone marrow-derived dendritic cells (BMDCs) were isolated from naïve wild-type and Ninj1^−/− mice. BMDCs were incubated with fluorescein isothiocyanate (FITC)-labeled ovalbumin (100 μg/mL) at 37°C for 1 h. Single-cell suspensions were stained with mouse anti-CD11c and anti-CD11b antibodies and analyzed by flow cytometry. (A) The frequency of CD11c⁺ CD11b⁺ cells (DCs) among total cells was analyzed. (B) The frequency of ovalbumin-bearing (FITC⁺) DCs among total cells was analyzed. Data are presented as means ± S.D. (n = 3). ns: not significant.

DISCUSSION

This study demonstrated that Quil-A induces NINJ1-dependent PMR and that both humoral and cellular immune responses are enhanced in Ninj1^−/− mice compared with wild-type controls following Quil-A-adjuvanted immunization. These findings suggest that NINJ1-mediated PMR negatively regulates the immune-potentiating effects of saponin-based adjuvants. Since saponin-based adjuvants other than Quil-A are also known to induce LDH release,^12,13,23) it is likely that saponin-based adjuvants generally trigger PMR. Therefore, NINJ1 may negatively regulate the activity of other saponin-based adjuvants as well, through its involvement in PMR induction. Interestingly, a previous study reported that activation of the NLRP3-dependent pathway by QS-21 also attenuated the ability of QS-21 to induce antigen-specific humoral and cellular immune responses.¹²⁾ Our findings suggest that saponin-based adjuvants activate two independent negative regulatory pathways: NINJ1-dependent PMR that releases high-molecular-weight inhibitory factors, and NLRP3-dependent pathway that releases low-molecular-weight ones. Although NINJ1 and NLRP3 are primarily recognized as pro-inflammatory mediators, our findings reveal a paradoxical regulatory role in the context of saponin-based adjuvants, where both molecules function as negative regulators that limit vaccine-induced immunity. This represents a previously unrecognized regulatory mechanism that may have broader implications for vaccine design.

Importantly, our results imply that inhibition of NINJ1 activity could be an effective strategy to enhance the efficacy of vaccines formulated with saponin-based adjuvants. The activity of NINJ1 can be suppressed by neutralizing antibodies or by pharmacological agents such as glycine.^24,25) Furthermore, specific inhibitors of NLRP3 have already been developed for clinical use.²⁶⁾ Therefore, appropriately suppressing the activities of both NINJ1 and NLRP3 during administration of vaccines formulated with saponin-based adjuvants would effectively enhance the induction of antigen-specific humoral and cellular immune responses.

The enhanced adjuvant effect of Quil-A observed in Ninj1^−/− mice can be mechanistically attributed to increased accumulation of antigen-bearing APCs, particularly MoDCs, in the draining lymph nodes. MoDCs are known to be highly efficient at priming naive T cells and can bridge innate and adaptive immunity. The increased frequency of antigen-bearing MoDCs in the draining lymph nodes likely promoted more efficient T cell priming, leading to enhanced antigen-specific cellular immune responses in the spleen and increased antibody production by plasma cells in the secondary lymphoid organs. The comparable intracellular ATP levels after Quil-A stimulation in wild-type and Ninj1^−/− cells indicate that Quil-A-induced metabolic perturbation occurs independently of NINJ1-mediated PMR. This finding is consistent with a previous report showing that ATP release through small membrane pores formed by Gasdermin D does not require NINJ1, and suggests that the differential immune responses in Ninj1^−/− mice are not due to the bioavailability of intracellular ATP.²⁷⁾ Additionally, we showed that NINJ1 deficiency does not affect the development of BMDCs or their antigen uptake capacity. GM-CSF-induced BMDCs include cells derived from monocytes that are present in the BM. Moreover, a previous report also showed that the proportions of various leukocyte populations in the circulation of Ninj1^−/− mice, including monocytes that serve as precursors of MoDCs, are comparable to those in wild-type mice.²⁸⁾ The same study also reported that NINJ1 deficiency does not affect M-CSF-induced differentiation of BM cells into macrophages. Thus, it seems unlikely that NINJ1 intrinsically affects the development of APCs, including monocytes, DCs, and macrophages, or the antigen uptake capacity of MoDCs. Hence, we hypothesize that Quil-A-induced PMR in wild-type cells promotes the release of intracellular molecules that exert paracrine inhibitory effects on the trafficking or accumulation of antigen-bearing APCs in the draining lymph nodes. In APCs in which PMR is not induced, Quil-A may still enhance antigen endocytosis and activation. However, the inhibitory molecules released upon PMR may act to suppress these functions. In Ninj1^−/− cells, where PMR is substantially reduced, this negative regulatory mechanism is attenuated, resulting in increased antigen delivery in the draining lymph nodes.

Quil-A induces the release of both low-molecular-weight molecules, such as mature IL-1β (approx. 17 kDa), and high-molecular-weight molecules, such as LDH (approx. 140 kDa), by DCs and macrophages. The former are released in an NLRP3-dependent but NINJ1-independent manner, and likely exit through small pores approximately 10–20 nm in diameter formed by Gasdermin D in the plasma membrane following NLRP3 activation.^16,29) In contrast, the latter are released in a NINJ1-dependent but NLRP3-independent manner.^16,30) High-mobility group box 1 (HMGB1) is one of the molecules released extracellularly in a NINJ1-dependent manner during PMR.¹⁶⁾ HMGB1 is a nuclear protein ubiquitously expressed across tissues and released into the extracellular space during various forms of sterile inflammatory responses or tissue damage.^31,32) HMGB1 mediates inhibitory signaling via the CD24-siglec-10 axis in DCs.³³⁾ However, extracellular HMGB1 has also been reported to act as a pro-inflammatory mediator in some contexts.³¹⁾ In mice immunized with QS-21, HMGB1 was reported to promote rather than suppress the adjuvant activity of QS-21.³⁴⁾ Therefore, comprehensive proteomic analysis is needed to identify the full spectrum of high-molecular-weight molecules released through NINJ1-dependent PMR following saponin stimulation. Such studies should focus particularly on molecules with known immunosuppressive properties that could act on APCs to limit their migration, antigen presentation capacity, or T cell priming ability. Identifying these key inhibitory factors could reveal novel therapeutic targets for enhancing saponin-based vaccine efficacy.

In conclusion, this study reveals a previously unrecognized negative regulatory mechanism in vaccine adjuvant responses, wherein NINJ1-mediated PMR limits immune activation by suppressing antigen delivery to the draining lymph nodes. These findings have several important implications. First, they identify NINJ1 as a potential therapeutic target for enhancing saponin-based vaccine efficacy. Second, because many clinically approved adjuvants, including aluminum-based formulations and emulsion-based adjuvants, also induce the release of intracellular molecules, NINJ1-dependent negative regulation may represent a general mechanism limiting adjuvant activity. Finally, understanding how NINJ1-mediated PMR suppresses APC function could enable the rational design of next-generation adjuvants that maximize immune activation while avoiding unwanted negative regulatory pathways. Future studies should focus on identifying the specific inhibitory molecules released through PMR and developing strategies to selectively block their activity during vaccination.

Acknowledgments

The authors thank A. Sato for secretarial assistance. The authors also thank the MMRRC for providing the C57BL/6N-Ninj1^{tm1a(KOMP)Wtsi}/MbpMmucd sperm. This work was, in part, supported by the Japan Society for the Promotion of Science KAKENHI (Grant Numbers: 25K22759 to T.S.; 24K02687 to N.T.; 25K03064 to M.T.); the Platform Project for Supporting Drug Discovery and Life Science Research (Basis for Supporting Innovative Drug Discovery and Life Science Research [BINDS]) from AMED under Grant Number: JP25ama121052; the Support for Pioneering Research Initiated by the Next Generation (SPRING) program (Grant Number: JPMJSP2138 to M.S.); and the Takeda Science Foundation (to T.S. and N.T.).

DECLARATION

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1)  Reed SG, Orr MT, Fox CB. Key roles of adjuvants in modern vaccines. Nat. Med., 19, 1597–1608 (2013).
• 2)  Zhao T, Cai Y, Jiang Y, He X, Wei Y, Yu Y, Tian X. Vaccine adjuvants: mechanisms and platforms. Signal Transduct. Target. Ther., 8, 283 (2023).
• 3)  Lee SM, Kim P, You J, Kim EH. Role of damage-associated molecular pattern/cell death pathways in vaccine-induced immunity. Viruses, 13, 2340 (2021).
• 4)  Marichal T, Ohata K, Bedoret D, Mesnil C, Sabatel C, Kobiyama K, Lekeux P, Coban C, Akira S, Ishii KJ, Bureau F, Desmet CJ. DNA released from dying host cells mediates aluminum adjuvant activity. Nat. Med., 17, 996–1002 (2011).
• 5)  Li H, Willingham SB, Ting JP-Y, Re F. Cutting edge: inflammasome activation by alum and alum’s adjuvant effect are mediated by NLRP3. J. Immunol., 181, 17–21 (2008).
• 6)  Kim EH, Woodruff MC, Grigoryan L, Maier B, Lee SH, Mandal P, Cortese M, Natrajan MS, Ravindran R, Ma H, Merad M, Gitlin AD, Mocarski ES, Jacob J, Pulendran B. Squalene emulsion-based vaccine adjuvants stimulate CD8 T cell, but not antibody responses, through a RIPK3-dependent pathway. eLife, 9, e52687 (2020).
• 7)  Alving CR, Rao M, Matyas GR. Similarities and differences of chemical compositions and physical and functional properties of adjuvant system 01 and army liposome formulation with QS21. Front. Immunol., 14, 1102524 (2023).
• 8)  Zarnegar B, Carow B, Eriksson J, Spennare E, Öhlund P, Akpinar E, Bringeland E, Osterman IL, Lundqvist L, Antti J, Handin N, Helgesson PH, Bankefors J, Lövgren Bengtsson K, Sellin ME, Palm AKE, Stertman L, Lunderius Andersson C. Matrix-M adjuvant triggers inflammasome activation and enables antigen cross-presentation through induction of lysosomal membrane permeabilization. NPJ Vaccines, 10, 184 (2025).
• 9)  Rajput ZI, Hu SH, Xiao CW, Arijo AG. Adjuvant effects of saponins on animal immune responses. J. Zhejiang Univ. Sci. B, 8, 153–161 (2007).
• 10)  Marciani DJ. Elucidating the mechanisms of action of saponin-derived adjuvants. Trends Pharmacol. Sci., 39, 573–585 (2018).
• 11)  Wang P. Natural and synthetic saponins as vaccine adjuvants. Vaccines, 9, 222 (2021).
• 12)  Marty-Roix R, Vladimer GI, Pouliot K, Weng D, Buglione-Corbett R, West K, MacMicking JD, Chee JD, Wang S, Lu S, Lien E. Identification of QS-21 as an inflammasome-activating molecular component of saponin adjuvants. J. Biol. Chem., 291, 1123–1136 (2016).
• 13)  Wilson NS, Duewell P, Yang B, Li Y, Marsters S, Koernig S, Latz E, Maraskovsky E, Morelli AB, Schnurr M, Ashkenazi A. Inflammasome-dependent and -independent IL-18 production mediates immunity to the ISCOMATRIX adjuvant. J. Immunol., 192, 3259–3268 (2014).
• 14)  Blevins HM, Xu Y, Biby S, Zhang S. The NLRP3 inflammasome pathway: a review of mechanisms and inhibitors for the treatment of inflammatory diseases. Front. Aging Neurosci., 14, 879021 (2022).
• 15)  Ramos S, Hartenian E, Broz P. Programmed cell death: NINJ1 and mechanisms of plasma membrane rupture. Trends Biochem. Sci., 49, 717–728 (2024).
• 16)  Kayagaki N, Kornfeld OS, Lee BL, et al. NINJ1 mediates plasma membrane rupture during lytic cell death. Nature, 591, 131–136 (2021).
• 17)  Ramos S, Hartenian E, Santos JC, Walch P, Broz P. NINJ1 induces plasma membrane rupture and release of damage-associated molecular pattern molecules during ferroptosis. EMBO J., 43, 1164–1186 (2024).
• 18)  Dondelinger Y, Priem D, Huyghe J, Delanghe T, Vandenabeele P, Bertrand MJM. NINJ1 is activated by cell swelling to regulate plasma membrane permeabilization during regulated necrosis. Cell Death Dis., 14, 755 (2023).
• 19)  Zhu L, Xu Y. Multifaceted roles of ninjurin1 in immunity, cell death, and disease. Front. Immunol., 16, 1519519 (2025).
• 20)  David L, Borges JP, Hollingsworth LR, Volchuk A, Jansen I, Garlick E, Steinberg BE, Wu H. NINJ1 mediates plasma membrane rupture by cutting and releasing membrane disks. Cell, 187, 2224–2235.e16 (2024).
• 21)  Kovarova M, Hesker PR, Jania L, Nguyen M, Snouwaert JN, Xiang Z, Lommatzsch SE, Huang MT, Ting JP-Y, Koller BH. NLRP1-dependent pyroptosis leads to acute lung injury and morbidity in mice. J. Immunol., 189, 2006–2016 (2012).
• 22)  Kamstrup S, San Martin R, Doberti A, Grande H, Dalsgaard K. Preparation and characterisation of quillaja saponin with less heterogeneity than Quil-A. Vaccine, 18, 2244–2249 (2000).
• 23)  Wu CA, Yang YW. Induction of cell death by saponin and antigen delivery. Pharm. Res., 21, 271–277 (2004).
• 24)  Borges JP, Sætra RSR, Volchuk A, Bugge M, Devant P, Sporsheim B, Kilburn BR, Evavold CL, Kagan JC, Goldenberg NM, Flo TH, Steinberg BE. Glycine inhibits NINJ1 membrane clustering to suppress plasma membrane rupture in cell death. eLife, 11, e78609 (2022).
• 25)  Kayagaki N, Stowe IB, Alegre K, et al. Inhibiting membrane rupture with NINJ1 antibodies limits tissue injury. Nature, 618, 1072–1077 (2023).
• 26)  Zahid A, Li B, Kombe AJK, Jin T, Tao J. Pharmacological inhibitors of the NLRP3 inflammasome. Front. Immunol., 10, 2538 (2019).
• 27)  Schachter J, Guijarro A, Angosto-Bazarra D, Pinilla M, Hurtado-Navarro L, Meunier E, Pérez-Oliva AB, Schwarzbaum PJ, Pelegrin P. Gasdermin D mediates a fast transient release of ATP after NLRP3 inflammasome activation before ninjurin 1-induced lytic cell death. Cell Rep., 44, 115233 (2025).
• 28)  Ahn BJ, Le H, Shin MW, Bae SJ, Lee EJ, Wee HJ, Cha JH, Lee HJ, Lee HS, Kim JH, Kim CY, Seo JH, Lo EH, Jeon S, Lee MN, Oh GT, Yin GN, Ryu JK, Suh JK, Kim KW. Ninjurin1 deficiency attenuates susceptibility of experimental autoimmune encephalomyelitis in mice. J. Biol. Chem., 289, 3328–3338 (2014).
• 29)  Ramos-Junior ES, Morandini AC. Gasdermin: a new player to the inflammasome game. Biomed. J., 40, 313–316 (2017).
• 30)  Degen M, Santos JC, Pluhackova K, Cebrero G, Ramos S, Jankevicius G, Hartenian E, Guillerm U, Mari SA, Kohl B, Müller DJ, Schanda P, Maier T, Perez C, Sieben C, Broz P, Hiller S. Structural basis of NINJ1-mediated plasma membrane rupture in cell death. Nature, 618, 1065–1071 (2023).
• 31)  Kwak MS, Kim HS, Lee B, Kim YH, Son M, Shin JS. Immunological significance of HMGB1 post-translational modification and redox biology. Front. Immunol., 11, 1189 (2020).
• 32)  Andersson U, Tracey KJ. HMGB1 is a therapeutic target for sterile inflammation and infection. Annu. Rev. Immunol., 29, 139–162 (2011).
• 33)  Chen GY, Tang J, Zheng P, Liu Y. CD24 and siglec-10 selectively repress tissue damage-induced immune responses. Science, 323, 1722–1725 (2009).
• 34)  Detienne S, Welsby I, Collignon C, Wouters S, Coccia M, Delhaye S, Van Maele L, Thomas S, Swertvaegher M, Detavernier A, Elouahabi A, Goriely S, Didierlaurent AM. Central role of CD169⁺ lymph node resident macrophages in the adjuvanticity of the QS-21 component of AS01. Sci. Rep., 6, 39475 (2016).