Characterization of Novel Pathogenic Variants Leading to Caspase-8 Cleavage-Resistant RIPK1-Induced Autoinflammatory Syndrome

Autoinflammatory diseases (AIDs) represent a special type of inborn errors of immunity (IEI) characterized by recurrent episodes of fever and sterile inflammation, which are a consequence of the overproduction of proinflammatory cytokines and increased cell death [1]. Different types of programmed cell death (PCD) have been molecularly characterized, with apoptosis, pyroptosis, and necroptosis as the most relevant ones [2, 3]. All types of PCD are physiological processes, essential for normal tissue development and homeostasis. However, an excessive cell death may amplify the inflammatory cascade and even provokes uncontrolled life-threatening events.

Necroptosis is a specific type of PCD triggered after the binding of certain cytokines (i.e., TNF) to cell surface receptors containing cytosolic death domains (DD). Unlike apoptosis and similarly to pyroptosis, necroptosis is associated with loss of membrane integrity, release of sterile intracellular content to the extracellular space, and the induction of a strong inflammatory response [4]. To prevent harmful reactions, necroptosis is tightly regulated, with the mixed-lineage kinase domain-like (MLKL), receptor-interacting serine/threonine kinase-1 (RIPK1), and RIPK3 proteins as its most important regulatory elements [2, 4]. Two different IEI have been associated with pathogenic variants in the RIPK1 gene. On one side, the deficiency of RIPK1 is a recessively inherited disease characterized by severe immunodeficiency, arthritis, and early-onset inflammatory bowel disease that is a consequence of biallelic, loss-of-function RIPK1 variants [5‐8]. On the other side, dominantly-inherited variants at the amino acid residue Asp324 have been associated with the cleavage-resistant RIPK1-induced autoinflammatory (CRIA) syndrome, a recently described monogenic AID characterized by early-onset periodic fever and lymphadenopathies [9, 10].

RIPK1 is a key regulator of different types of PCD, including necroptosis, and its activity can be down-regulated by the caspase-8. The caspase-8 binding and cleavage site occurs at the ³²¹LQLD³²⁴ sequence of RIPK1, and the cleavage process separates the kinase domain from the intermediate and DD domains [14, 16]. A novel monogenic AID named CRIA syndrome has been described in twelve patients as a direct consequence of heterozygous variants at amino acid residue Asp324, located at the ³²¹LQLD³²⁴ sequence [9, 10]. These variants render RIPK1 resistant to caspase-8-cleavage, provoke the loss of this regulatory mechanism, and enable its oligomerization, the RIPK3 recruitment, the MLKL phosphorylation, and the pore forming at cell membrane that finally led to both cell permeabilization and death. Therefore, a functional impairment in the caspase 8-mediated cleavage of RIPK1 resulting in exacerbated necroptosis represents the main mechanism of the pathogenesis of CRIA syndrome [9, 10].

In the present work, we have identified two novel missense variants in the ³²¹LQLD³²⁴ sequence of the RIPK1 in patients with a dominantly inherited, undefined AID, which displayed a perfect intrafamilial phenotype-genotype relationship. On the basis of their location in the RIPK1, we hypothesized that these novel variants may be the cause of the observed manifestations and different experiments were performed to characterize their pathogenicity. Structural analyses revealed that the Leu321 and Asp324 residues were located in a surface-accessible region of the RIPK1, and the detected variants hypothetically provoke a conformational change and structural re-modeling of the protein, probably towards an α-helix conformation, making it less accessible to and/or less cleavable by caspase-8. Additional in vitro studies have clearly demonstrated that the two novel variants were caspase-8 cleavage-resistant, as those RIPK1 variants previously reported as causing the CRIA syndrome [9, 10]. Altogether, these evidences supported the pathogenicity of the novel RIPK1 variants here described and enabled us to establish the definitive diagnosis of CRIA syndrome in all enrolled patients increasing thus the total number of patients affected by this rare monogenic AID.

From a clinical point of view, the main manifestations detected in the enrolled patients are markedly similar to those previously described, with minor differences in the incidence of splenomegaly (16.7% vs 58.3); abdominal pain (16.7% vs 41.7%); oral ulcers (66.7% vs 71.4%); arthralgias (83.3% vs 37.5%), and headache (83.3% in our series vs not specified) [9, 10]. These data clearly indicate a relatively homogenous clinical picture for the CRIA syndrome and markedly different to the manifestations described in the deficiency of RIPK1, which is characterized by profound cellular immunodeficiency, with recurrent and severe infections, progressive polyarthritis, and early-onset inflammatory bowel disease [5‐8].

The CRIA syndrome displays marked similarities with two well-known AIDs, the HIDS syndrome and the periodic fever, aphthosis, pharyngitis, and adenitis (PFAPA) syndrome, and consequently it should be considered in the differential diagnosis of these diseases. Some clinical features are frequent in all these three diseases, including early age at disease onset, the periodic nature of inflammatory episodes, and the presence of fever, lymphadenopathies, oral ulcers, and increased acute phase reactants [17, 18]. In contrast, there are some differences that may help clinicians to distinguish these entities. These differences include the inheritance pattern of the disease (dominant for the CRIA syndrome, recessive for HIDS, and polygenic inheritance for PFAPA); the near absence of inflammatory manifestations at serosa; skin and eye in CRIA syndrome; the tendency in PFAPA syndrome to spontaneously subside the inflammatory episodes during puberty and adolescence in contrast to HIDS and CRIA syndromes, where these episodes are still present during adulthood; certain differences in the hematological tests (tendency to normal values of leukocyte, neutrophil and platelet counts in CRIA syndrome); and the different outcome of frequently administered treatments. Finally, the deficiency of caspase-8 may be also considered in the differential diagnosis of CRIA syndrome, because both diseases share some pathophysiological mechanisms (partial or total impairment of normal caspase-8 function) and clinical features such as lymphadenopathies and splenomegaly [19‐21]. However, the differences with regard to their inheritance pattern, age at disease onset, frequencies of severe infections and mucosal involvement, and mortality rates may help clinicians to clearly distinguish these two conditions. Table 3 includes a comparative summary of the main similarities and differences observed among the above mentioned diseases.

The precise mechanisms by which all these cytokines are increased in the CRIA syndrome are still under investigation. Previous evidences have shown that caspase-8 deficiency and defects of RIPK1 processing are connected with activation of NLRP3-inflammasome by different mechanisms [22‐24], in particular by necroptosis-induced intracellular potassium efflux that is a well-established NLRP3 activator that triggers the activation of the inflammasome [22, 25]. Recent evidences also support the hypothesis of a direct connection between RIPK1 and the increase of inflammatory cytokines through the pyroptosis-related proteins gasdermin E (GSDME) and gasdermin D (GSDMD). Thus, it has been recently described that GSDME, a protein related with pyroptosis and IL-1β release, is activated in neutrophils in a RIPK1-dependent manner [26] and that the FADD-RIPK1-caspase 8 complex is recruited after infection to cleavage GSDMD and initiates the inflammatory cell death [27]. The loss of regulatory mechanisms of RIPK1 activity due to cleavage-resistant variants may generate an overproduction of inflammasome-related cytokines (IL-1β, IL-18), and inflammatory cell death through both necroptosis and pyroptosis, which in its turn may provoke the increase of plasma levels of IL-6 and TNF. This profile of circulating cytokines has similarities with those detected in other monogenic AIDs, specially associated with increased pyroptosis (inflammasomopathies, TNF-receptor-associated periodic syndrome), which are usually well controlled with anti-IL-1 and/or anti-TNF [28‐30]. Unlike these diseases, patients with CRIA syndrome only displayed positive responses to the anti-IL-6 drug tocilizumab [9, 10]. To our knowledge, the patient P5 here described is the first one in whom the IL-1 blockade with the long-lasting monoclonal antibody canakinumab showed clear effectiveness to control the disease. The different effectiveness of treatments targeting inflammatory cytokines in CRIA syndrome will be a matter of future investigations that probably will require the identification of novel patients and the collection of additional data about inflammatory mediators. Our data on patients’ cytokine profiles has limitations due to the small sample size. Moreover, the serum samples were collected during active disease and inactive intervals and from patients with different RIPK1 mutant genotypes, which may lead to differences in the degree of necroptosis and the overproduction of specific cytokines. As CRIA is a rare disease and patients undergo various treatment modalities, defining the cytokine profile will be challenging. Nonetheless, our results are concordant with the limited data on circulating cytokines reported in CRIA syndrome [10].

In summary, we have shown herein the effectiveness of analyses based on next-generation sequencing methods to identify the genetic defects in patients with undiagnosed AID, as has been previously reported in other groups of AIDs and in other different diseases [31‐35]. Our analyses identified two novel RIPK1 variants at the consensus sequence of caspase-8-mediated cleavage that lead to an exacerbated necroptosis. These evidences confirmed the diagnosis of CRIA syndrome in all enrolled patients and clearly differentiate them from patients with RIPK1 deficiency. Altogether, our results add novel data about the clinical and genetic diversity of the recently characterized CRIA syndrome. Before establishing the patients’ diagnosis, the treatments were empirically administered. With the novel data of patients’ cytokines profile during active disease, potential novel targeted treatments may be proposed for the treatment of future patients.