In the last few years substantial progress has been made in unravelling the genetic basis of susceptibility to Crohn’s disease. Three CARD 15 (previously called NOB₂) genetic variants, resulting in proteins with modified carboxy terminal regions, have been implicated.^1,2 43% of patients with Crohn’s disease carry at least one of these CARD 15 mutations (compared with 15% of healthy controls).¹ Mutations in CARD 15 have also been identified in affected members of families with Blau syndrome.^3,4 This is a rare autosomal dominant disorder, sometimes referred to as familial sarcoidosis, characterised by granuloma formation in joints, skin, and uvea.

Card 15 is a microbial sensing protein involved in innate immunity. It recognises conserved structural components of microorganisms (bacterial muramyl dipeptide, MDP and peptidoglycan, PGN) and is part of the danger signal pattern recognition network which forms the front line of protective immunity. Mutations associated with Crohn’s disease render the molecule insensitive to MDP and interfere with the downstream activation of NF-κB. One potential result of this may be the persistence of an antigen resulting in engagement of other arms of the adaptive immune system and formation of granulomas. Expression of CARD 15 in monocytes (precursors of macrophages and granulomas) further supports its potential role in granuloma formation. We hypothesised that mutations in CARD 15 may be a unifying defect in the multisystemic granulomatous diseases of Crohn’s disease, sarcoidosis, and Blau syndrome.

To investigate this we recruited a cohort of 29 patients with sarcoidosis from the Oxford Centre for Respiratory Medicine. All had a typical clinical picture of sarcoidosis and either histologically proven disease or characteristic Lofgren’s syndrome (defined as an acute onset of disease with erythema nodosum, joint pains, and bilateral hilar lymphadenopathy). The diagnosis of sarcoidosis was also supported by the characteristic appearance of the lungs on a high resolution computed tomographic (CT) scan in all patients. The definition and diagnosis of sarcoidosis adheres to the statement on sarcoidosis adopted by the joint committees of WASOG/ATS/ERS.⁵ The patients were first diagnosed between the ages of 25 and 40 years and had been followed up for at least 1 year before recruitment. All were white and one third presented with Lofgren’s syndrome. Written informed consent was obtained for genetic analysis and the study was approved by the local ethics committee.

The entire coding region of CARD 15 (11 exons and flanking intronic sequences) was screened for the presence of mutations. In brief, the CARD 15 gene was amplified from the genomic DNA samples by polymerase chain reaction (PCR) using primers as previously described⁶ and sequenced on an ABI 377 automated sequencer using a Dye Terminator Cycle Sequencing Ready reaction kit (Perkin Elmer Applied Biosystems, CA, USA). Sequence data were then aligned using the Sequence Navigator analysis software Version 1.0.1 (Perkin Elmer Applied Biosystems) and compared with the known CARD 15 sequence (EMBL accession number AJ303140).

435 sequence analyses were performed in the 29 patients. We were therefore able to detect new alleles putatively associated with the sarcoidosis phenotype with frequencies as low as 2%. The three mutations associated with Crohn’s disease were specifically examined. The R702W mutation was observed in one patient with sarcoidosis while the G908R and the 1007fs mutations were not found. These results were not different from those reported previously in control populations where the mutations R702W, G908R and 1007fs were present in 4%, 1%, and 2% of 103 European healthy controls.⁶ Furthermore, they do not differ from data derived from healthy controls recruited in the UK.^7,8 The codons 334 and 469 reported to be involved in Blau syndrome were also carefully scrutinised but we did not detect any genetic variation in our sarcoidosis patients. No additional mutations were seen within the rest of the coding region of the gene, suggesting that there were no specific alternative CARD 15 mutations associated with sarcoidosis. Schurmann and colleagues⁹ recently found no correlation between specific CARD 15 polymorphic alleles and patients with sarcoidosis from families with more than one member afflicted by the disease. This study provides evidence that CARD 15 is not associated with non-familial sarcoidosis (in patients with a white ethnic background) and that there are no mutations in any part of the coding region of the CARD 15 gene in these patients.

There is little doubt that there is a genetic predisposition in sarcoidosis, as indicated by the presence of familial clustering, ethnic susceptibility, and recent evidence of an association with HLA-DRB1.¹⁰ The search for this susceptibility gene(s) has recurrently pointed to a locus near the HLA DR region on chromosome 6. This and the exclusion of the NOD2 locus has focused attention on abnormalities in antigen presentation and cytokine/chemokine receptors as a potential basis for the aetiology of sarcoidosis.