All animal procedures were carried out according to the provisions of the EU Protection of Animals Used for Scientific Purposes Regulations 2012 (SI No. 543 of 2012) as amended and Directive 2010/63/EU of the European Parliament issued from the Health Products Regulatory Authority Ireland—license number AE 19132/P030. Human blood samples were collected from anonymous healthy blood donors from the Irish Blood Transfusion Service (IBTS) under license number (BI-AG-300919) issued from the School of Biochemistry and Immunology Research Ethics Committee, Trinity College Dublin in accordance with the Declaration of Helsinki.

All animals used in this research project were healthy Holstein-Friesian calves under 1 year old (unless stated). Buffy coats were collected from anonymous healthy human blood donors.

The stimuli used to activate cells were LPS, Escherichia coli Serotype R515 (Enzo Life Sciences) and alhydrogel (Brenntag Biosector). Inhibitory molecules used in this research included: MCC950 (Cayman Chemical), caspase1-Z-YVAD-FMK (Bachem), caspase8-Z-IETD-FMK (Bachem), CA-074 (Sigma-Aldrich), and cathepsinB-CA-074-Me (Sigma-Aldrich). Antibodies used for western blotting included: polyclonal anti bovine IL-1β (Bio-Rad), polyclonal (N-15-R) antimouse ASC (Santa Cruz Biotechnology sc-22514-R), and monoclonal (AC-74) β-actin (Sigma-Aldrich). ELISA kits used to detect bovine IL-1β and human IL-1β were sourced from ThermoScientific and R&D Systems, respectively. The FLICA™ Assay Kit (FAM-YVAD-FMK) for caspase-1 detection was acquired from ImmunoChemistry Technologies.

Bovine PBMCs were isolated from whole blood samples collected in 9 ml vacutainers containing Heparin anticoagulant. Human PBMCS were extracted from buffy coats. PBMCs were isolated using leucosep tubes (Greiner Bio-One, Storehouse, UK) and a density gradient histopaque 1077 (Sigma-Aldrich). Red blood cell contamination was eliminated using sterile 0.25% sodium chloride (Baxter) as a lysis buffer. The cells were subsequently centrifuged twice in PBS at a speed of 400 g for 10 min. For innate cytokine analysis, cells were incubated at 37°C with 5% CO₂ in RMPI 1640 medium (Biosera) enriched with heat-inactivated fetal calf serum (Biosera), l-glutamine (Gibco), and penicillin (Gibco).

Supernatants from treated cells were used to measure IL-1β secretion by ELISA as per the manufacturer’s protocols. Absorbance was read on a Multiscan FC plate reader and analyzed with SkanIt for Multiscan FC software (Thermo Scientific). The limit of detection was between 31.25 and 2,000 pg/ml.

Peripheral blood mononuclear cells were lysed using 100 µl of Laemmli buffer (4% SDS, 10% 2-mercaptoethanol, 20% glycerol, 0.004% bromophenol blue, 0.125 M Tris–HCl). The lysates were transferred onto a 0.2 µm PVDF membrane (Millipore) and probed with anti-IL-1β, anti-ASC, and anti-β-actin antibodies. The blots were developed using a Bio-Rad ChemiDoc Imaging system (Bio-Rad).

In preparation for caspase-1 analysis, cells were seeded at a density of 1 × 10⁷/ml and incubated with stimuli. Following incubation, the cells were resuspended and then centrifuged at 1,200 rpm for 5 min to pellet the cells. The supernatants were discarded and the cells were resuspended in FACs buffer (1% FCS in PBS) and incubated for 30 min with caspase-1 specific probe. The cells were centrifuged and washed three times in FACs buffer and analyzed by flow cytometry.

Peripheral blood mononuclear cells (0.5 × 10⁶ cells/ml) were plated in cRPMI on 35 mm glass bottom tissue dishes. Cells were treated with alum (50 µg/ml), LPS (1 pg/ml), alum + LPS, and or RPMI only and stained with calcein as outlined by Khormaee et al. (13). Cells were viewed using a Point Scanning Confocal Microscope with a heated stage and CO₂ chamber (Olympus FV100 LSM Confocal Microscope).

Statistical analysis was performed using Graphpad Prism 5 software. The means for two groups were compared using a paired T-test. The means for three or more groups were compared using one-way ANOVA. A p-value of <0.05 was taken as statistically significant.

It has been previously established in murine and human cells that a combination of LPS and alum effectively drives NLRP3 inflammasome dependent IL-1β secretion (14), although this has not been investigated in cattle. Bovine PBMCs were isolated from Friesian calves (<6 months old) and treated with a range of LPS and alum concentrations for 24 h. Bovine PBMCs secreted small amounts of IL-1β in response to LPS (10 ng/ml) but secretion was significantly elevated when alum (100 µg/ml) was added (Figure 1A).

Incubation of PBMCs with a fixed dose of alum (50 µg/ml) and LPS concentrations from 1 pg/ml to 1 µg/ml resulted in higher IL-1β secretion than the corresponding amounts of LPS alone (Figure 1B). A surprising observation was that cells stimulated with alum in the absence of LPS secreted IL-1β (Figure 1B). This observation contrasts what is seen in murine and human macrophages and dendritic cells. A more detailed analysis found that addition of alum to cells at 25, 50, or 100 µg/ml enhanced IL-1β secretion from LPS-primed PBMCs (Figure 1C). To address whether the IL-1β promoting ability of alhydrogel also applied to aluminum phosphate and calcium phosphate, PBMCs were treated with these clinically applied adjuvants alone or following priming with LPS. As seen with alhydrogel, both calcium phosphate and aluminum phosphate were potent stimuli for promoting IL-1β secretion by bovine cells (Figures 2A,B).

To secrete IL-1β, innate immune cells generally require two signals in the form of a PAMP agonist and an inflammasome activator (15, 16). Typically, toll-like receptor (TLR) agonists prime cells to synthesize inactive pro-IL-1β while activation of the inflammasome triggers caspase-1 activation and the cleavage and secretion of processed IL-1β. Given the remarkable observation that alum could trigger the secretion of IL-1β in the absence of priming with LPS (Figure 1B), it was hypothesized that PBMCs from some calves may contain preformed pro-IL-1β and as such require no TLR priming. To address this, PBMCs were isolated from 17 calves (aged < 1 year old) and were treated with LPS, alum or a combination of LPS and alum for 24 h. The control cells were left unstimulated. Cells stimulated with LPS and alum together for 24 h secreted high concentrations of IL-1β (Figure 3A). Moreover, PBMCs from 10 of 17 animals secreted IL-1β when incubated with alum in the absence of LPS (Figure 3A).

To address whether this response was bovine specific, PBMCs from 12 humans were isolated and treated with LPS (100 pg/ml) and alum (50 µg/ml). In contrast to the findings with bovine PBMCs there was no detectable secretion of IL-1β by human PBMCs following incubation with alum when LPS was absent (Figure 3B). However, in response to LPS and alum combined, human PBMCs secreted high concentrations of IL-1β (Figure 3B).

To determine whether alum drives the expression of pro-IL-1β as well as IL-1β secretion, PBMCs were isolated from 8 calves and cultured in the presence or absence of LPS and/or alum for 6 h. The 31 kDa pro-IL-1β band was detected in cells stimulated with LPS (Figure 3C). Interestingly, in four of the eight animals tested, pro-IL-1β was also present in unstimulated cells thus suggesting that some bovine monocular cells are already primed to secrete IL-1β (Figure 3C).

In humans, it has been demonstrated that alum-mediated IL-1β secretion is caspase-1 dependent (8). Conversely, although alum has been incorporated into numerous veterinary vaccines, its signaling capacity in a bovine context remains unknown. As a result, the role of caspase-1 in alum-induced bovine IL-1β secretion was next investigated.

When bovine PBMCs were incubated with a caspase-1 inhibitor, Z-YVAD-FMK (10 µm), prior to addition of alum, and after stimulation with LPS, a substantial reduction in IL-1β secretion was observed (Figure 4A). To further prove that caspase-1 is involved in alum-induced IL-1β secretion, cells from four animals were stimulated with cRPMI or alum. Activated caspase-1 was detected using a caspase-1 detection probe termed FLICA 660-YVAD-FMK. This probe is comprised of an affinity peptide sequence specific for caspase-1 (YVAD), a far-red fluorescent 660 dye and a fluoromethyl ketone reactive moiety. This reagent enters cells and irreversibly binds to active caspase-1 thereby emitting and retaining the red signal inside cells positive for caspase-1. The results were subsequently analyzed by flow cytometry. The results demonstrate that the percentage of cells positive for caspase-1 activity is increased in response to alum (Figure 4B).

In mice, it has been demonstrated that alum-mediated IL-1β secretion is NLRP3 dependent (14, 17). The role of caspase-8 in alum-induced IL-1β responses has received less attention. Recent publications have indicated that in response to TLR4 activation, caspase-8 is required for NLRP3 inflammasome-dependent IL-1β processing in murine cells (18). The role of NLRP3 and caspase-8 in alum-induced bovine IL-1β secretion has not been previously investigated.

Having established that alum-induced IL-1β secretion was caspase-1 dependent in bovine cells, the potential role of the NLRP3 inflammasome was addressed. LPS-primed PBMCs were pretreated with MCC950, a specific NLRP3 inflammasome inhibitor (19), prior to the addition of alum. After 24 h the supernatants were analyzed by ELISA for IL-1β secretion. LPS and alum stimulation induced significantly less IL-1β secretion from cells cocultured with 500 nM MCC950 (Figure 5A). In the presence of 100, 50, 10, and 1 nM of MCC950, alum-induced IL-1β secretion was not abrogated (Figure 5A).

To determine whether alum-induced IL-1β secretion is caspase-8 dependent, LPS-primed PBMCs were incubated with Z-IETD-FMK, a caspase-8 inhibitor, after which they were cocultured with alum for 24 h. IL-1β secretion was suppressed by caspase-8 inhibition after 24 h (Figure 5B). These data suggest that caspase-8 and caspase-1 both contributed to alum-induced IL-1β secretion by bovine PBMCs.

Inflammasome activation involves NLRP3 recruiting and binding to the adaptor protein ASC through interactions involving the PYD domain on both molecules (20). When bound to NLRP3, ASC oligomerizes to form a pyroptosome which then binds to pro-caspase-1 via homotypic interactions (21). The assembled inflammasome facilitates the cleavage and activation of caspase-1, which cleaves pro-IL-1β to its bioactive form resulting in its secretion from the cell (21, 22). In murine models, it has been demonstrated that ASC oligomerization is crucial for the activation of caspase-1 (22–24).

To assess the effects of alum on ASC oligomerization in bovine cells, PBMCs were cultured at a density of 10⁷ /ml. Cells were left unstimulated, treated with LPS (100 pg/ml), alum (50 µg/ml), or a combination of both for 3 h. Afterward, the cells were lysed and probed for ASC oligomerization by western blotting. In response to alum alone or together with LPS, the oligomerization of ASC was evident by detection of an 80–90 kDa molecular weight band (Figure 6; Figure S3 in Supplementary Material). The oligomer was absent in unstimulated cells or those incubated with LPS alone.

Studies on murine macrophages and dendritic cells have demonstrated that alum induces lysosomal disruption resulting in activation of the NLRP3 inflammasome (25, 26). Additionally, lysosomal damage and release of hydrolases has been shown to up-regulate caspase-1 activity (27). To determine whether lysosomal proteases play a role in regulating bovine IL-1β secretion, PBMCs were cultured for 1 h with an inhibitor (CA-074-Me) targeting cathepsin B, a protease released during lysosomal damage. Following that, the cells were stimulated with alum (50 µg/ml) for 24 h after which the supernatants were collected and tested for IL-1β. Cells stimulated with LPS and alum produced substantially less IL-1β when incubated in the presence of the cathepsin B inhibitor than those cultured without the inhibitor (Figure 7A). Since at high concentrations this inhibitor can also inhibit other lysosomal cysteine cathepsins (28), we confirmed these results with CA-074 which although less cell permeable than CA-074-Me, has been shown to have enhanced selectivity for cathepsin B (28) (Figure 7B). CA074 inhibited alum-induced IL-1β secretion at a concentration of 50 µM. Finally to confirm that alum results in lysosomal damage, we stained cells with calcein. Calcein is endocytosed by cells in a non-specific manner, but cannot diffuse across endosomal membranes. This leads to calcein becoming trapped in endosomes resulting in a “punctate” staining pattern (13), as observed when bovine PBMCs are cotreated with RPMI or LPS in Figure 7C. However, when taken up by PBMCs that have been costimulated with alum, calcein homogenously stains the entire cell as a result of alum-driven lysosomal disruption and subsequent leakage of lysosomal contents (including calcein) into the cell cytoplasm.