Receptor-interacting serine/threonine Kinase 1 protein, encoded by RIPK1, is a key signalling molecule in inflammation and cell death pathways, involved in pro-inflammatory and pro-survival signalling following the activation of surface receptors, such as tumour necrosis factor (TNF) receptor 1, Toll-like receptor (TLR)-3, TLR4 and interferon receptors. TNF receptor (TNFR) stimulation recruits RIPK1 and ‘TNFR1-associated death domain protein’ (TRADD) which lead to activation of nuclear factor-κB (NF-κB) pathway through the intracelluler signalling complex. Also RIPK1 activation leads to caspase-8 activation and apoptosis [11,12,13,14].

The critical role of RIPK1 in the survival of intestinal epithelial cells has been demonstrated in RIPK1-deficient mice that had developed intestinal pathology via inhibition of caspase-8-mediated apoptosis [8, 9]. Whereas intestinal cell apoptosis seems to play a crucial role in mice, the symptoms in RIPK1 deficient humans are predominantly mediated by dysregulated immune signalling [11, 12].

Biallelic loss-of-function RIPK1 variants, have been linked to severe immunodeficiency, early-onset inflammatory bowel disease and arthritis [7]. While impaired T- and B-cell differentiation, significant lymphopenia, and decreased production of proinflammatory cytokines such as IL-6, TNF, and IL-12 lead to immunodeficiency, active inflammasome formation and necroptosis might be related to the inflammatory component of the disease [9,11,12, 15]. Interestingly, recent studies show that variants which impair the caspase-8-mediated RIPK1 cleavage, confer a gain-of-function effect, leading to the autosomal dominant cleavage-resistant RIPK1-induced autoinflammatory (CRIA) syndrome [15,16,17,18].

To our knowledge, 16 patients (age at onset 1 day − 4 years old age) with autosomal recessive RIPK1 deficiency have been reported to date. They presented with colitis and recurrent infections, also some of them had polyarthritis, aphthous ulcers, and perianal disease that comparable to our patient, because of the critical role of RIPK1 in controlling human immune and intestinal homeostasis [7,8,9, 15, 19].

Patients with RIPK1 deficiency were found to have increased pro-inflammatory cytokine IL-1β and decreased IL-10 secretion, a critical cytokine in regulation of the immune response in the gut [7]. Poor treatment response was reported to immunosuppressive treatments, such as azathioprine, corticosteroids, infliximab, IL-1 receptor antagonist, and also hematopoietic stem cell transplantation (HSCT). On the other hand, several patients were reported to be alive only with intravenous immunoglobulin, antifungal and antibiotic treatment. There was no relation between treatment success and age or clinical presentation of the patients [7,8,9, 15, 19].

The ability of HSCT to treat this disease is still unknown. Given RIPK1’s functions in regulating both immunological and epithelial responses, performing HSCT to treat these sufferers should be taken with caution since it may improve immunodeficiency [7,8,9, 15, 19]. Cuchet-Lourenco et al. [7] reported three patients with RIPK1 deficiency who underwent HSCT. Of them, intestinal symptoms and arthritis of the 30-month-old age patient improved but antibiotic treatment had been continued for the chronic lung disease. The older patients aged at 12-year-old and 13-year-old patient died due to multiorgan deficiency and severe disseminated infection respectively [7]. These different clinical manifestations and response to treatment might be related to genotype-phenotype correlations, variable penetrance, or secondary factors such as microbiome [8]. With the developing genetic and functional studies, various predisposing factors that may lead to enhancement of immune dysregulation can be determined in the future.

Here, we report an infant with a severe form of infantile IBD presenting with malnutrition, recurrent severe infections, polyarthritis, and perianal fistula tract. Whole-exome sequencing followed by PCR and Sanger sequencing revealed and confirmed a homozygous deletion in RIPK1, spanning 22.7 kb and covering the last four coding exons of RIPK1 as well as the first exon of the adjacent BPHL gene (Fig. 3A). This deletion comprises more than half of the RIPK1 coding sequence and leads to the expression of two differently spliced fusion transcripts between RIPK1 and BPHL in a patient-derived B-cell line. Both detected fusion transcripts contain a frameshift followed by a premature stop codon in RIPK1 exon 6 or BPHL exon 3, respectively. Hence, the patient is predicted to express no functional RIPK1 protein (Fig. 3C, D and E). Given that the phenotype of the patient is consistent with previous reports of RIPK1 deficiency, the deletion was considered causative. Notably, both the detection of large copy-number variants, from whole-exome data as well as their validation with PCR and Sanger is challenging. Hence, such variants might evade detection during routine diagnostic testing using exome sequencing. Our approach to design suitable primers for the validation in this patient is illustrated in Fig. 3B. To our knowledge, this is first report of a pathogenic deletion affecting these two genes. It has been suggested that bilalleic loss of BPHL is tolerated [20]. Therefore, an impact of the partial BPHL deletion on the patient’s phenotype was considered unlikely. BHPL encodes Biphenyl hydrolase like, a serine hydrolase that converts valacyclovir to acyclovir and valganciclovir to ganciclovir [21]. Since the patient was not given such antiviral treatment, it is unkown whether the BPHL exon 1 deletion might have impacted the response to treatment with these drugs. Moreover, a homozygous MEFV p.Glu148Gln missense variant (E148Q) was detected in our patient. MEFV encodes pyrin, which takes part in controlling the inflammation process, regulates IL-1b and nuclear factor kappa beta (NF-kB) activation and, inhibits Caspase-1 activation and apoptosis. Defective pyrin leads to Caspase-1 activation and excessive release of IL-1β [22, 23]. It has been reported that patients with the p.Glu148Gln substitution respond well to colchicum treatment, which reduces IL-1β production [24, 25]. Aydın F et al. [24] reported E148Q alteration leads to similar clinical findings to M694V mutations, but a milder disease and good response to colchium treatment, similar to the report of Topaloglu R et al. [26]. They concluded E148Q should be considered as a disease-causing mutation [24, 26]. On the other hand, it should be noted that a pathogenic effect of MEFV p.Glu148Gln is still being debated, since the variant is relatively common in the general population (with an allele frequency of approximately 0.3 in the East and South Asian population according to the Genome Aggregation Database 2.1.1) and, hence, meets the stand-alone criterion BA1 for benignity according to the ACMG variant interpretation guidelines [27]. Urgancı N et al. [28] reported heterozygous E148Q mutation was the most common MEFV gene mutation among 597 patients (2–18 years old age) diagnosed ulcerative colitis and Crohn disease, but the relation of clinical course of the diseases and mutation is still under debate. Unfortunately, we did not observe any beneficial effect of colchicum on the clinical course of the patient. Therefore, and due to the phenotype of the patient that was highly similar to previous reports, we consider the partial RIPK1 deletion the likely cause of disease, although a modifying effect of MEFV p.Glu148Gln can not be ruled out.

In conclusion, despite the widespread utilization of genetic testing, it still has limitations in determining a specific etiology, which may prevent the decision on the appropriate treatment for the patients with VEO-IBD and associated immunodeficiencies. RIPK-1 protein deficiency and the possibility of larger copy-number variants evading detection during routine testing should be considered in the presence of VEO-IBD, polyarthritis, and recurrent infections.