Precisely diagnosing and effectively treating cryopyrin-associated periodic syndrome (CAPS), an inflammatory condition linked to gain-of-function NLRP3 inflammasome mutations, poses challenges. A novel classification approach may help inform therapeutic decisions and offer valuable insights into broader inflammatory conditions (Cosson et al. J. Exp. Med. 2024. https://doi.org/10.1084/jem.20231200).
Inflammation is estimated to kill millions of individuals per year worldwide, strongly contributing to the top four causes of death, namely heart ischemia, stroke, pulmonary disease, and infections. Understanding its mechanistic basis is vital for the development of new anti-inflammatory therapies.
The NLRP3 inflammasome is a key molecular driver of inflammation in settings ranging from infections to neurodegeneration but has mainly been studied in mouse models (reviewed in Weber et al., 2020). Therefore, human auto-inflammatory disease (AID) conditions with a clear NLRP3 association have the potential to provide novel pathophysiological insights. CAPS—short for cryopyrin-associated periodic syndromes, and increasingly called NLRP3-AID—is an AID genetically linked to NLRP3 gain-of-function mutations but with a wide range of severity and often non-specific symptoms. Hence, researchers and clinicians struggle to categorize the 271 currently reported NLRP3 CAPS mutations reported in the InFevers database (Milhavet et al., 2008) in a way that would allow reliable predictions of therapy outcomes. Which treatment (mostly but not always anti-IL-1 inhibitor biologicals or—in atypical or milder phenotypes—the microtubule inhibitor, colchicine) for which patient is oftentimes determined by trial and error (Welzel et al., 2021a, 2021b). As a delay to effective treatment may have severe and debilitating long-term consequences, especially for children diagnosed with CAPS, there is a desperate need to link a given NLRP3 mutation with a clear pathophysiological inflammasome profile, enabling confident and rapid therapy choices. Although other studies of certain NLRP3-AID variants in immortalized mouse macrophages (Molina-Lopez et al., 2024), primary mouse macrophages or mouse models in vivo (Vande Walle et al., 2019), or patient-derived primary blood cells ex vivo (Corcoran et al., 2020; Weber et al., 2022) have been insightful, a systematic analysis or classification based on standardized functional study has been missing.
Insights from Alexander N.R. Weber, Maria Mateo Tortola, and Jasmin B. Kuemmerle-Deschner.
This is where the recent work by Cosson et al. (2024) in this issue of JEM comes in and provides an insightful framework towards genotype–phenotype-informed disease classifications and possibly therapeutic decisions. By systematically and functionally connecting an unprecedented number of CAPS NLRP3 mutations (altogether 34) with key inflammasome outcomes (i.e., cytokine release and cell death) in a tour de force, the French team generated an impressive panel of immortalized monocytic cell lines, each expressing a patient-derived mutant NLRP3 variant alone or in combination with WT NLRP3 (which is typically the case in patients, as mutations tend to be heterozygous germline or mosaic somatic). Through systematic in vitro stimulation, analysis, and unbiased clustering of results, the authors for the first time identified five functional classes of CAPS mutations, each with different requirements for triggering the inflammatory outcomes relevant to the disease (Fig. 1). Whereas group 5 variants were so active that conventional NLRP3 triggers seemed obsolete for driving cell death and IL-1β release, group 2 and 3 variants retained at least one of the two typical requirements (checkpoints) for NLRP3 functionality, namely priming, and activation: WT NLRP3 requires both transcriptional and post-translational “priming” (sufficient to trigger group 3 NLRP3 variants) as well as actual activation (sufficient for group 2) mediated by conformational change, trafficking, and nucleotide exchange (reviewed in Weber et al., 2020). Thus, in group 2 and 3 variants, either priming or activation checkpoints, respectively, were selectively compromised. Group 4 showed an intermediate phenotype between group 5 and 2/3 in that priming and activation were both individually sufficient to trigger cell death. Notably, the researchers found that group 1 variants showed no major differences compared to WT NLRP3 and hence appear to not qualify as bona fide CAPS disease–causing variants. This is a very relevant insight that can help avoid misclassification of carriers of group 1 variants as CAPS patients.
Figure 1. Novel screening approach by Cosson et al. (2024), phenotyping CAPS mutations extensively in vitro for disease classification. Combined with in vitro drug sensitivity studies, a new classification of CAPS subgroups based on standardized functional analysis of variants may help to tailor therapy decision to each individual group and eventually patient.
What makes the in vitro analysis clinically relevant is the fact that the authors sampled primary blood cells from patients covering the five groups and used them to validate the results from the cell lines in an ex vivo setting—a formidable accomplishment, given that CAPS is a very rare disease with an estimated prevalence of 2.7–5.5 per 1 million in Europe (Lainka et al., 2010).
To complement the diagnostic usefulness and foreshadow an approach for approximating therapy responses, the authors also profiled their cell lines for sensitivity to known inhibitors—either targeting NLRP3 directly (using the small molecules MCC950 and CY-09), or indirectly via regulatory proteins involved in the NLRP3 activation process at the priming or trafficking level, such as the deubiquitinase BRCC3 inhibitor, G5 (Cheng et al., 2020), or the protein kinase D (PKD) inhibitor, CRT0066101 (Zhang et al., 2017). Encouragingly, most NLRP3 mutants were sensitive to direct NLRP3 inhibition by MCC950, which locks NLRP3 in an inactive, ADP-bound oligomer (Hochheiser et al., 2022), in good agreement with our previous ex vivo study of CAPS peripheral blood mononuclear cells (Weber et al., 2022) and another recent study (Molina-Lopez et al., 2024). However, exceptions were also observed depending on whether the inhibitor was administered before priming or activation or both: for example, some group 5 (D303H and E525K) and group 3 (G301D, L353P, and G569R) variants showed resistance to MCC950. Interestingly, these mutations are located in areas critical for ATP binding and hydrolysis, hinting that once active and ATP-bound, NLRP3 might become insensitive to MCC950. Conversely, a covalent inhibitor, CY-09 (Wang et al., 2021), like G5, exhibited a broad effectiveness across all tested mutants.
Regarding CAPS, this highlights the need for exploring different inhibitor approaches via NLRP3 or its regulators. Whereas none of the inhibitors tested by Cosson et al. (2024) are currently clinically relevant, profiling compounds like ibrutinib, a Food and Drug Administration–approved inhibitor targeting the NLRP3-modifier Bruton’s tyrosine kinase (BTK) (Bittner et al., 2021), might prove insightful in the future (Fig. 2). Another open question to address is the dependence of different CAPS variants on a well-established oligomeric NLRP3/Golgi/MTOC-dependent NLRP3 pathway (Andreeva et al., 2021; Magupalli et al., 2020) versus a novel oligomer- and MTOC-independent pathway (Mateo-Tórtola et al., 2023, Preprint). Given the observed resistance of CAPS mutants to the inhibitor of PKD—a kinase supposedly involved in the Golgi/MTOC pathway (Zhang et al., 2017) like BTK (Bittner et al., 2021)—we speculate that these mutants do not require a pre-formed oligomeric cage (Andreeva et al., 2021) and circumvent the need for MTOC association. Also, testing the sensitivity to the aforementioned microtubule inhibitor, colchicine, might help rationalize the different suggested phenotypic groups by the pathway they employ and may simultaneously shed light on the observed colchicine treatment response of CAPS patients with mild or atypical symptoms who do not qualify for IL-1 therapy.
Figure 2. Open questions. Molecularly, it will be interesting whether the classification of variants will match a preference for MTOC-dependent (Magupalli et al., 2020; Andreeva et al., 2021) or MTOC-independent (Mateo-Tórtola et al., 2023, Preprint) signaling and possibly a sensitivity to microtubule or NLRP3 modifier (e.g., BTK) inhibitors like colchicine or ibrutinib, respectively. Although in principle each new mutation can be tested in vitro using the methodology proposed by Cosson et al. (2024), it will be interesting to determine whether (or at which point of refinement of the model) current cell line data, possibly in combination with in silico structural analysis, may allow for predicting the classification of new or so far functionally untested mutations. Ultimately, it remains to be investigated by clinicians if in vitro sensitivities and classifications faithfully predict in vivo therapy success.
Clinicians may also welcome testing Cosson et al.’s grid for patient sub-stratification—e.g., aligning observed symptoms, clinical parameters, and treatment responses with the five groups—which may ultimately help to confirm or refine the functionality-based classification system introduced here. This will be an important next step. Moreover, whether new variants or the >200 known variants not assessed in the cell line screen so far could be categorized de novo without specific experimental exploration waits to be established; the existing data for 34 variants combined with the authors’ exploration of recent cryo-EM data (Hochheiser et al., 2022) could, however, be envisaged to reach refinement at a level at which function- and structure-guided modeling may allow for reliable predictions (Fig. 2). It may take a while to get there, but not having to test every new NLRP3 variant functionally first to draw conclusions would be ideal for clinicians and patients alike.
Although CAPS is rare, and hence drug companies are unlikely to develop new inhibitors specifically for CAPS, approaches like Cosson et al.’s could also be highly useful for developing new treatments for non-genetic NLRP3-mediated inflammation in two ways: firstly, via new insights gained on the WT NLRP3 activation process through the study of CAPS; and secondly, because CAPS is a vital proof-of-concept population for new NLRP3-centered therapies: thus, the insights provided by Cosson et al. (2024) may be important to interpret the results from the NCT04086602, NCT04868968 (both phase 1), and NCT05186051 (phase 2a) clinical trials of new NLRP3 compounds in CAPS patients, and safely extrapolate from them to WT NLRP3. In a similar way, blocking strategies targeting IL-1 were first trialed in CAPS, and this laid the foundation for assessing the effect of blocking IL-1–mediated inflammation to prevent cardiovascular events in the CANTOS trial (Ridker et al., 2018). Investing in better understanding NLRP3 function in CAPS may thus contribute to better understanding (and eventually targeting) non-genetic NLRP3-linked inflammatory processes. We believe Cosson et al.’s formidable work represents an important milestone in this process and may, in the meantime, provide a more accurate and precise classification of NLRP3-AID.
Acknowledgments
We thank Libera Lo Presti for helpful manuscript comments and apologize to those authors whose work we could not cite directly due to length restrictions. Figures were created with Biorender.com.
The study was supported by the Deutsche Forschungsgemeinschaft (German Research Foundation, DFG) grant We-4195/15-1 (to A.N.R. Weber), and the Medical Faculty of the Eberhard Karls University Tübingen and the European Reference Network for rare or low prevalence complex diseases “Immunodeficiency, Autoinflammatory and Autoimmune Diseases (ERN RITA)” (to A.N.R. Weber and J.B. Kuemmerle-Deschner). Infrastructural funding was provided by the University of Tübingen, the University Hospital Tübingen and the DFG Clusters of Excellence “iFIT—Image-Guided and Functionally Instructed Tumor Therapies” (EXC 2180, to A.N.R. Weber) and “CMFI—Controlling Microbes to Fight Infection” (EXC 2124, to A.N.R. Weber). Gefördert durch die DFG im Rahmen der Exzellenzstrategie des Bundes und der Länder—EXC 2180 and EXC 2124.
Author contributions: The work was conceptualized and visualized by A.N.R. Weber, but all authors contributed equally to background research and writing.
Disclosures: A.N.R. Weber and J. Kuemmerle-Deschner have received research grants from IFM Therapeutics and Novartis funding aspects of studies related to CAPS. None of these or other funders had a role in data collection and analysis, decision to publish, or preparation of the manuscript.
