The skin manifestations range from groupings of small pustules (Panel A) to a generalized pustulosis (Panel B). The bone manifestations include epiphyseal ballooning of multiple distal and proximal long bones, in the single patient from Puerto Rico (Panel C); the more typical radiographic manifestations included widening of multiple ribs (with affected ribs indicated with asterisks) and the clavicle (arrows) (Panel D), heterotopic ossification or periosteal cloaking of the proximal femoral metaphysis (arrows) and periosteal elevation of the diaphysis (arrowheads) (Panel E), and an osteolytic lesion with a sclerotic rim (Panel F, arrow).

The pedigrees of six families with deficiency of interleukin-1–receptor antagonist are presented in Panel A, according to the country or region of ancestry and the amino acid mutation. Solid symbols indicate Patients 1 through 9, open symbols unaffected relatives, squares male subjects, circles female subjects, and slashes deceased patients. For the subjects for whom the IL1RN genotype was determined, H denotes heterozygous mutant, W wild type, and M homozygous mutant. Panel B shows the structure of IL1RN (isoform 1) exons (black boxes) and the sequences of the homozygous mutations (which are named under the plots, along with the resultant amino acid mutation). A genetic isolate in northwestern Puerto Rico is a founder population for a 175-kb genomic deletion between bases 113,434,601 and 113,609,824 on chromosome 2. This deletes genes encoding six interleukin-1–related genes: IL1RN and the genes encoding interleukin-1 family, members 9 (IL1F9), 6 (IL1F6), 8 (IL1F8), 5 (IL1F5), and 10 (IL1F10) (Panel C). The centromere and the region of interest are shown in red.

Relative messenger RNA (mRNA) levels of IL1RN, encoding the interleukin-1–receptor antagonist, in whole blood were determined by means of quantitative polymerase-chain-reaction assay (Panel A). Levels are shown for seven controls, one Newfoundland subject heterozygous for the N52KfsX25 mutation, three Dutch subjects heterozygous for the E77X mutation, two Puerto Rican subjects heterozygous for the genomic deletion, and one patient homozygous for each of these three mutations. T bars indicate the standard deviations, where applicable. In Panel B, whole-blood specimens from homozygotes and heterozygotes for each of these three mutations, as well as controls (with the wild-type sequence) were stimulated ex vivo with lipopolysaccharide to induce secretion of glycosylated interleukin-1–receptor antagonist by leukocytes. Resultant protein levels, detected by means of Western blot using an antibody specific for the N-terminal of the protein, are shown. Panel C shows the results of in vitro analysis of cultured cells of the human embryonic kidney-cell line 293T, transfected with mutant or wild-type IL1RN. The 293T cells overexpressed the mRNA of both the wild-type and mutant IL1RN but secreted only the wild-type protein, as shown by analysis of the supernatant (top). The mutant proteins were detected within 293T-cell lysates at a molecular weight that suggests that the signal peptide mediating the secretion of the protein has not been cleaved (middle). Actin expression is shown as a control (bottom). The bar graph at the bottom of the panel shows the mean growth of an interleukin-1–responsive cell line in the presence of secreted interleukin-1–receptor antagonist. The wild-type protein efficiently inhibited proliferation, whereas the mutant proteins, which were not secreted, could not function in this capacity. T bars indicate the standard deviations.

Peripheral-blood granulocytes and leukocytes were obtained from seven controls with wild-type interleukin-1–receptor antagonist, six heterozygous carriers of mutant interleukin-1–receptor antagonist, and three homozygotes with mutant interleukin-1–receptor antagonist. The monocytes were stimulated with recombinant interleukin-1β for 18 hours. Panel A shows the mean production of five selected chemokines and cytokines, which were significantly up-regulated in samples from patients homozygous for mutations resulting in deficiency of interleukin-1–receptor antagonist as compared with those from heterozygotes and controls. (P values are shown for comparisons of patients with the wild-type controls for each chemokine or cytokine.) T bars indicate the standard deviations. MIP-1α denotes macrophage inflammatory protein 1α, and TNF-α tumor necrosis factor α. Panel B shows the results of cytohistochemical analysis of interleukin-17 expression in skin specimens (alkaline phosphatase stain). The interleukin was markedly up-regulated in a patient with deficiency of interleukin-1–receptor antagonist as compared with a control.

Laboratory values before and after treatment with the recombinant interleukin-1–receptor antagonist anakinra are shown in Panel A for each of the six patients who received the drug. Treatment with anakinra resulted in a rapid and sustained decline in the C-reactive protein level, erythrocyte sedimentation rate, and white-cell count; the platelet count normalized more slowly. The pink star in each plot indicates the time at which the Dutch Patient 6 had discontinuation of the therapy and a resultant flare-up of disease. Reinitiation of anakinra led to normalization of all four measures. Panel B shows an example of the clinical improvement in skin and bone manifestations after anakinra treatment was begun. Before treatment there was extensive pustulosis, diffuse erythema, and crusting on the skin of an affected child that nearly completely resolved just days after initiation of anakinra, with exfoliation of the skin occurring 7 days after (shown here). Improvement in bone findings was similarly dramatic, as seen on radiography, with osteolytic lesions (arrows) present in the proximal and distal tibial metaphysis and periosteal elevation (asterisk) before treatment and resolution 5 months after treatment.