Identification of CDK4/6 Inhibitors as Small Molecule NLRP3 Inflammasome Activators that Facilitate IL-1β Secretion and T Cell Adjuvanticity

Abstract

Several FDA-approved adjuvants signal through the NLRP3 inflammasome and IL-1β release. Identifying small molecules that induce IL-1β release could allow targeted delivery and structure-function optimization, thereby improving safety and efficacy of next-generation adjuvants. In this work, we leverage our existing high throughput data set to identify small molecules that induce IL-1β release. We find that ribociclib induces IL-1β release when coadministered with a TLR4 agonist in an NLRP3- and caspase-dependent fashion. Ribociclib was formulated with a TLR4 agonist into liposomes, which were used as an adjuvant in an ovalbumin prophylactic vaccine model. The liposomes induced antigen-specific immunity in an IL-1 receptor-dependent fashion. Furthermore, the liposomes were coadministered with a tumor antigen and used in a therapeutic cancer vaccine, where they facilitated rejection of E.G7-OVA tumors. While further chemical optimization of the ribociclib scaffold is needed, this study provides proof-of-concept for its use as an IL-1 producing adjuvant in various immunotherapeutic contexts.

Figure 1.

(A) IL-1β secretion and (B) toxicity of compounds in BMDMs. LPS-primed or -unprimed BMDMs were treated overnight with the 17 analytes and Quil A as a positive control. IL-1β secretion was assayed using ELISA and quantified using recombinant protein standards. Toxicity was assayed using a secreted lactate dehydrogenase (LDH) assay and quantified as the fraction of positive (nigericin-treated) or negative (untreated) controls (1.0 = 100% LDH release relative to nigericin).

Figure 2.

Identification of ribociclib as a potent, IL-1 producing adjuvant. (A) IL-1β and (B) LDH release when LPS-primed or -unprimed BMDCs were treated with ribociclib at various concentrations. (C) LPS-primed or -unprimed BMDCs were treated with ribociclib at 125 μM, which was found to be a concentration of maximum activity. Cytokine production was assayed using Legendplex Mouse Inflammation 13-plex, and unpaired t tests were used to determine differences between groups. IL-17 and GM-CSF are not shown, these cytokines were below the limit of detection for all treatment conditions.

Figure 3.

Evaluation of ribociclib’s mechanism of IL-1 cytokine production. (A) LPS-primed BMDCs were pretreated with the indicated inhibitor compounds for 30 min, then treated with 125 μM ribociclib (or 15.6 μM, in the case of DPI). IL-1β was measured in the supernatant via ELISA. (B) LPS-primed BMDCs isolated from the indicated KO mice were treated with 125 μM ribociclib, and IL-1β was measured in the supernatant via ELISA. (C) LPS-primed or -unprimed BMDCs were treated with 125 μM ribociclib or a control, 5 μM Nigericin (N), for the indicated time points, then lysed and analyzed via Western blot in the supernatant (S) or lysate (L). Full immunoblots and conditions for antibody staining are provided in SI, Figure S4.

Figure 4.

Synthesis, formulation, and in vivo responses of ribociclib lipid derivatives. (A) Synthetic scheme for the preparation of Ribo-L in two steps from commercially available ribociclib. (B) TEM image of Ribo-L formulated at 40 μg/mL within DSPC and MPLA containing liposomes (scale bar = 250 nm). (C) Representative DLS of monodisperse liposome formulation. (D) Vaccination schedule for the characterization of ribociclib-containing formulations in vivo. (E) Splenic, antigen specific CD8+ and (F) CD4+ T cell responses 28 d after vaccination. (G) Anti-OVA IgG titer 27 d after initial vaccination.

Figure 5.

In vivo mechanism of ribociclib-mediated immunity. (A) Upregulation of the cell surface marker, CD86, and (B) enhanced presentation of MHC-I restricted antigen on dendritic cells in the draining inguinal lymph node of mice injected intramuscularly with the indicated formulations. (C) Vaccination schedule for a comparative study between wild-type (WT) and IL-1R deficient (KO) C57Bl/6J mice. (D) Systemic IL-6 in the serum of WT or KO mice vaccinated with the indicated formulations after 2 or 24 h. (E) Splenic, antigen-specific CD8+ T cell responses. (F) Total IgG titers generated in response to the indicated formulations. One-way ANOVA with Sidak’s multiple comparisons test is shown in E-F.

Figure 6.

In vivo tumor therapeutic vaccination study of Ribo-L containing liposomes and controls. (A) Vaccination schedule for the tumor challenge model (n = 10/group). Mice were implanted with 5 × 10⁴ E.G7-OVA cells subcutaneously. They were vaccinated 5 and 12 d after tumor implantation with the indicated formulations peritumorally. Tumor growth and survival were monitored 3 times/week until day 40. (B) Kaplan–Maier curve showing survival of mice treated with various formulations throughout the study. (C) Tumor growth curves of individual mice treated with each of the formulations.