Identification of Aim2 as a sensor for DNA vaccines ¹

Animals

C57BL/6 mice were obtained from Taconic Laboratories. Aim2^−/− mice were generated in house by K. Fitzgerald’s group at the University of Massachusetts Medical School (UMMS) as previously described (27). Aim2^−/− mice were on a mixed B6/129 background and therefore B6×129 mice were utilized as controls. B6.129 (hereafter referred to as Aim2^+/+) mice were obtained from Jackson Laboratories (Bar Harbor, ME) and were bred at UMMS. IL-1 receptor (Il-1r), IL-18 receptor (Il-18r), and Asc-deficient mice were produced in house. All mice were maintained in the Department of Animal Medicine at UMMS according to IACUC-approved protocols. Mice received 100 μg of codon optimized H1HA DNA vaccine expressing the full-length wild type HA protein from A/Texas/04/09 (pH1HA), divided between quadriceps, at 2 and 4 weeks. A third boosting immunization was delivered 1-week prior to sacrifice.

Gene Induction Analysis

C57BL/6 mice were shaved and immunized with 100 μg H1-TX04-09.tPA DNA vaccine plasmid intramuscularly injected into the hind quad muscle. Punch biopsies were harvested from the site of immunization at 6 hours, 12 hours, and 24 hours post immunization and snap frozen. RNA was isolated from tissue biopsies using TRIzol Reagent (Life Technologies). We analyzed gene expression using the Nanostring nCounter Analysis system (Nanostring Technologies). Each reaction contained 100 ng RNA in a 5 μl aliquot, plus reporter and capture probes. We also included 6 pairs of positive control and 8 pairs of negative control probes. Gene induction analysis and normalization was conducted using nSolver Analysis Software v1.1. Raw counts were normalized to naïve mice using 3 reference genes: Gapdh, Gusb, and Hprt1.

Cell Culture and Cytokine ELISA

Mouse BMDCs were generated from Aim2^+/+ or Aim2^−/− mice by culturing fresh bone marrow in R10 medium containing GM-CSF for 8 days at 37 ° C. Aim2^+/+ and Aim2^−/− immortalized BMDM were produced in house. Cells were first primed with 200 ng/ml LPS (Sigma Aldrich) for 4–5 hrs. prior to treatment with appropriate stimuli. All media was removed from the cells, and the appropriate stimulus was added. Poly (dA:dT) (Sigma Aldrich) DNA and H1-TX04-09.tPA DNA vaccine plasmid were transfected using Lipofectamine 2000 at a concentration of 1.5 μg/ml. ATP was added at a concentration of 1.25 μg/ml. Cultures were incubated 16–18 hours at 37 ° C, and supernatants were harvested. Cell culture supernatants were assayed for IL-1β (BD Biosciences, Franklin Lakes, NJ) by ELISA. Lactate dehydrogenase (LDH) release was used to measure pyroptotic cell death. LDH assays were performed using the Promega CytoTox96 Non-radioactive Cytotoxicity Kit according to manufacturer’s directions (Promega, Madison, WI)

Enzyme Linked Immunosorbent Assay (ELISA)

Microtiter plates were coated with transiently expressed H1HA antigen at ~1 μg/mL in PBS for 1 hour at room temperature and assayed as previously described (30). NaSCN displacement was performed at a serum dilution of 1:100. After washing of serum samples, NaSCN was added at various (0, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0 M) concentrations in PBS for 15 min followed by 5 washes in EWB. The assay was then completed as above.

ELISPOT Assay

Splenocyte T and B cell ELISPOT reagents were obtained from Mabtech (Mariemont, OH). H1HA specific T cells were quantified per manufactures instructions. Positive controls were stimulated with 20 ng/ml PMA (Sigma-Aldrich, St. Louis, MO) and 500 ng/ml ionomycin (Sigma-Aldrich). The H1HA relevant peptide used was a CD8⁺ cell-restricted HA peptide (IYSTVASSL). H1HA-specific antibody secreting cells were detected by coating of MAIPSWU (Millipore, Billerica, MA, USA) plates with the transiently expressed H1HA antigen utilized for ELISA (~1.0 μg/well). Positive spots were visualized on a CTL Imager and counting was performed with Immunospot software (Cellular Technology Ltd., Shaker Heights, OH)

In Vivo Caspase-1 activation

Aim2^+/+ and Aim2^−/− mice were shaved and immunized with 100 μg H1-TX04-09.tPA DNA vaccine plasmid intramuscularly injected into the hind quad muscle. Punch biopsies were harvested from the site of immunization at 6 hours, 12 hours, and 24 hours post immunization and snap frozen. Cryopreserved tissue sections were generated and adhered to glass slides. Samples were then stained with a caspase-1 FAM/FLICA kit according to manufacturer’s instructions (ImmunoChemistry Technologies, Bloomington, MN). Stained slides were visualized on a confocal microscope. Sixteen independent fields were analyzed for fluorescence.

Quantitative Real Time-Polymerase Chain Reaction

Aim2^+/+ and Aim2^−/− mice were shaved and immunized with 100 μg H1-TX04-09.tPA DNA vaccine plasmid intramuscularly injected into the hind quad muscle. Punch biopsies were harvested from the site of immunization at 6 hours, 12 hours, and 24 hours post immunization and snap frozen. RNA was isolated from tissues biopsies using TRIzol Reagent (Life Technologies #15596-026), and cDNA was generated using the Bio Rad iScript cDNA synthesis kit (Bio-Rad, Hercules, CA). cDNA was then used for rt-PCR reactions on a Bio-Rad CFX-96 cycler. Primers sequences are available upon request.

Statistical analysis

All data is presented as the mean of individual mice +/− standard deviation (SEM). Statistical analysis was performed using a Student’s t test, a one-way ANOVA followed by a Tukey post-test, or a two-way ANOVA followed by a Bonferonni post-test.

DNA vaccine plasmid induces expression of Aim2, caspase-1 and the inflammasome

While previous studies have mainly used non-coding DNA plasmid or DNA vaccines coding for marker proteins to study DNA-elicited innate immune responses, the current study tested a DNA vaccine (pH1HA) expressing the HA antigen of the type A influenza virus subtype H1N1 virus which was responsible for a pandemic influenza in 2009. HA is the major protective antigen in clinically licensed inactivated and live-attenuated influenza vaccines. DNA vaccines expressing HA have been shown to be immunogenic in eliciting HA-specific antibodies in both animal and human studies (2, 30–34). The expression of HA antigen by pH1HA used in the current study was confirmed by Western blot and its immunogenicity to elicit HA-specific antibody response was verified in a pilot mouse study (data not shown).

We first wanted to profile key immune response genes following DNA vaccine pH1HA using the Nanostring nCounter gene expression system, which includes a custom array encoding 50 innate immunity targets. Gene induction was quantified from wild-type C57BL/6 mice immunized with the pH1HA DNA vaccine. Messenger RNA was isolated and the expression of innate immune genes profiled using the Nanostring nCounter, and changes in gene induction quantified. Notably, Aim2 was induced ~6 fold within 12 hours of immunization when compared to naïve samples. Aim2 is a type I IFN inducible gene suggesting a potent ability of cells at the site of vaccination to recognize cytosolic plasmid vaccines (Figure 1A). In accordance with the induction of Aim2, caspase-1 was also highly upregulated. (Figure 1B). Most striking were the high levels of the inflammatory cytokines IL-1α and IL-1β (Figure 1C,D). These observations indicate that inflammasome components were present at the site of vaccination. To test if the inflammasome pathway was active at the site of vaccination we utilized a caspase-1 specific FAM/FLICA fluorescent stain to covalently label catalytically active caspase-1. Mature caspase-1 became apparent within 6 hours of immunization and reached a peak at 12 hours (Supplemental Figure 1). Collectively, these results led us to examine the role of the Aim2 inflammasome pathway in antigen specific immune responses elicited by pH1HA DNA vaccination.

Involvement of Aim2 in cellular responses to DNA vaccine plasmid

We first evaluated the ability of macrophages and dendritic cells to recognize pH1HA DNA vaccine by performing in vitro experiments. IL-1β production in response to pH1HA DNA vaccine was evaluated in BMDM collected from either Aim2^+/+ or Aim2^−/− mice (Figure 2A). As expected, both the synthetic B-form dsDNA poly(deoxyadenylic-deoxythymidylic) and pH1HA DNA vaccine induced a robust IL-1β response in Aim2^+/+ BMDM as measured by ELISA. However, IL-1β production was abolished in Aim2^−/− BMDM. Similar results were seen in BMDC (data not shown). Next, the pan-caspase inhibitor Z-VAD-FMK was included in BMDM cultures prior to adding pH1HA DNA vaccine (Figure 2B). This resulted in inhibited IL-1β production in Aim2^+/+ macrophages, yielding IL-1β levels comparable to Aim2^−/− wells, supporting the role of Aim2 in DNA vaccine mediated IL-1β maturation. Finally, pH1HA DNA vaccine induced an inflammatory form of cell death (pyroptosis) in Aim2^+/+ macrophages, as measured by lactate dehydrogenase release (Figure 2C). This response was attenuated in Aim2^−/− cells. Collectively, these data indicate that Aim2 acts as a sensor of DNA vaccine plasmid and regulates caspase-1 dependent IL-1β production and pyroptotic cell death in response to pH1HA DNA vaccines in vitro.

Effects of Aim2 deletion on pH1HA induced HA-specific immune responses

As it is evident that the Aim2 inflammasome recognizes and responds to pH1HA DNA vaccine in cultured cells, the role of Aim2 in pH1HA DNA vaccination was next examined in Aim2-deficient (Aim2^−/−) and wild-type Aim2^+/+ mice. The pH1HA DNA vaccine induced high-level HA-specific antibody responses in Aim2^+/+ mice, but significantly lower antibody titers in Aim2^−/− mice (Figure 3A). This reduction is isotype-dependent as Aim2^−/− mice exhibited significantly lower levels of HA-specific IgG1, IgG2b, and IgG2c responses (data not shown). Likewise, Aim2^−/− mice exhibited significantly reduced HA-specific circulating B cells as well as IFN-γ secreting CD8⁺ T cells in the spleen (Figure 3C,D). The role of Aim2 in regulating the maturation process of pH1HA-induced antibody responses was further confirmed by measuring the avidity of serum HA-specific antibodies in these mice (Figure 3B). Aim2^+/+ mice required high concentrations of the chaotropic agent NaSCN to disrupt antigen/antibody complexes, while much lower concentrations of NaSCN were required for disassociation in Aim2^−/− mice. To confirm the requirement for inflammasome signaling in DNA vaccine immunogenicity, we also quantified the adaptive response in Asc^−/− mice. Asc-deletion similarly inhibited the generation of optimal HA-specific immune responses (Supplemental Figure 2A–C).

Aim2 deficient mice fail to cleave caspase-1 into its active form

Since DNA vaccination resulted in high levels of caspase-1 activation in Aim2^+/+ mice (Supplemental Figure 1), we analyzed Aim2^−/− mice for their ability to generate catalytically active caspase-1 using the FAM/FLICA assay (Figure 4). Aim2^−/− mice demonstrated a clear reduction in caspase-1 activation at the 12-hour peak time point when compared to Aim2^+/+ controls.

Effects of IL-1 and IL-18 deletion on vaccine induced HA-specific immune responses

As inflammasome signaling ultimately results in the downstream cleavage of pro-IL-1β and pro-IL18 into their respective active forms, the role of IL-1β and IL-18 signaling in DNA vaccine was next investigated (Supplemental Figure 3). Surprisingly, both of these cytokines were dispensable for the DNA vaccine response as mice lacking the IL-1r or the IL-18r mounted normal vaccine responses. Total serum HA-specific IgG titers were similar to wild-type C57BL/6 mice in both Il-1r^−/− and Il-18r^−/− mice. Likewise, no significant difference was seen in total HA-specific B or CD8⁺ T cell numbers as measured by ELISPOT. This data is in line with previously published reports demonstrating little impact on DNA vaccination following MyD88 deletion (20).

The immune response is lineage dependent

Previously published data has demonstrated the requirement for both hematopoietic and non-hematopoietic cell lineages in DNA vaccination (20). To further elucidate the role of Aim2, bone marrow chimeric mice were generated by transferring bone marrow from Aim2^+/+ mice into Aim2^−/− mice, or vice versa (Figure 5). Aim2^+/+ and Aim2^−/− mice reconstituted with Aim2^−/− bone marrow exhibited strong defects in both the T cell response and HA-specific IgG production. Interestingly, transfer of Aim2^+/+ bone marrow into Aim2^−/− rescued the T cell response, but only partially rescued the humoral response. While HA-specific IgG levels were impaired compared to Aim2^+/+ mice reconstituted with Aim2^+/+ bone marrow, they were significantly higher than Aim2^−/− bone marrow reconstituted mice. This would support that Aim2 is required in both the hematopoietic and non-hematopoietic lineages for optimal humoral responses, but deficiency in non-hematopoietic lineages does not affect CD8⁺ T cell responses.

Aim2^−/− mice lack IFN-αβ signaling

The failure of Aim2^−/− mice to generate optimal adaptive immune responses implies a defect in immune priming at the site of injection. While the major function of the Aim2 inflammasome is to regulate caspase-1 activation, resulting in IL-1β, IL-18 and cell death pathways, our data indicate that IL-1β and IL-18 are not responsible for the Aim2 dependent effects we observed. We therefore endeavored to quantify IFN-α/β as it has been reported to play a key role in the immune response to B-DNA (25, 35–37). In addition, it has been established that IFN-α/β signaling is required for DNA vaccination (20, 21). Aim2 does not control DNA induced IFN-α/β production directly. Rather the STING pathway mediates these effects. Since the IFN-α/β response is so critical for DNA vaccination, we performed a detailed kinetic analysis measuring IFN-α/β expression in Aim2^+/+ and Aim2^−/− mice. To ensure we only detect DNA vaccine-induced IFN-α/β, we limited our measurements to the site of immunization. Aim2^+/+ and Aim2^−/− mice were immunized with the pH1HA DNA vaccine, and punch biopsies were collected from the site of injection. Quantitative rt-PCR analysis of mRNA clearly shows that Aim2^−/− mice have reduced IFN-α and IFN-β expression compared to Aim2^+/+ controls, with expression peaking at 12 hours post immunization in wild-type mice (Figure 6). Intriguingly, IFN-αβ expression in Aim2^−/− mice peaked at 6 hours post immunization and remained static throughout the time course. Consistent with the decrease in IFN-α/β, there was a corresponding decrease in the IFN stimulated gene, IP10. We also noticed a significant decrease in TNF. Asc^−/− mice had similar levels of IFN-α/β, further confirming the requirement for inflammasome signaling (Supplemental Figure 2D). These data suggest a previously unreported role for Aim2 in regulating local IFN-α/β levels following DNA vaccination. As Aim2 controls cell death at the site of infection, it is likely that Aim2 dependent cell death liberates endogenous DAMP danger signals, which might in turn elicit IFN-α/β via the Aim2-independent STING/TBK1 pathways. This broad defect in IFN-α/β signaling likely explains the defects we observed in Aim2-deficient mice treated with DNA vaccines.