In this issue of Blood, Pagel et al1 carefully deliniate some fascinating phenotype/genotype correlations in a larger cohort than in their earlier reports of familial hemophagocytic lymphohistiocytosis (HLH).2,3
The protein coded for by STXBP2, MUNC18-2 impacts the formation and dissolution of SNARE (soluble N-ethylmaleimide sensitive factor attachment protein receptor) complex and the control of membrane fusion. Of 37 FHL5 patients evaluated, 24 had no exon 15 splice site mutations and 13 had them. The phenotype differences were striking (see Table 4 on page 6021 of the article by Pagel et al). Those with the exon 15 splice site mutations were older and had a different profile of unique clincal characteristics from those without the splice site mutations. A proline-leucine change at P477L leads to no STXBP2 expression and early onset of HLH.
MUNC18-2 protein controls T-cell toxicity but also impacts immunoglobulin synthesis, platelet function, and intestinal motility or absorption as found in B cells, epithelial tissues, kidney, and intestine. Some of the concepts used to understand MUNC18-2 come from studies of the STXBP I/Munc18-1 functions in neurons where absence of these proteins prevents neurotransmitter secretion in mice.4 Chediak Higashi and Hermansky Pudelak syndromes, also associated with the hemophagocytic syndrome, are well known to have platelet granule defects associated with possible defects in MUNC 18-2. Sandrock et al reported defects in the platelet α and δ granule secretion after thrombin stimulation in platelets from FHL5 patients.5 In an earlier report on FHL5 patients, Meeths et al evaluated 11 patients with STXBP2 mutations and showed that stimulation of the lymphocytes with IL-2 rescued the natural killer cell functional defects.6 Their patients presented from 2 months to 17 years. Four of their patients had a splice site mutation in exon 15.
Genotype/phenotype associations have been previously reported in FHL2 patients. Perforin gene mutations with the Trp374Stop/A1a91Val were found in adults and Trp374Stop mainly in younger individuals.7 Molleran Lee et al showed that patients with PRF1 mutations and some perforin expression presented at 54 months versus 3 months with no expression.8 Zhang et al reported more missense mutations in adult HLH patients with 48% having a A91V-ORF1 genotype.9 Horne et al found a higher prevalence of perforin mutations in patients from the Middle East and syntaxin mutations in Turkish patients compared with Nordic patients.10 They found no differences in presenting signs and symptoms based on genotype, but reported that patients with perforin mutations presented at median age of 2.3 months, MUNC mutations 6.2 at months, and syntaxin mutations at 14.4 months. Central nervous system disease was more prevalent in patients with perforin mutations versus syntaxin mutations. Ueda et al found that of 40 FHL patients, those with non-sense perforin mutations (FHL2) had higher sIL-2 receptor and ferritin levels and presented after 7 years of age.11
The work by Pagel et al is the result of well-organized collaborations over many years and shows the power of translational research to better inform us about a rare disease and basic biology. As more mutations in the FHL-associated genes are cataloged we may well learn of other genotype/phenotype associations. One of the important off-shoots of this work is that clinicians may now start thinking about HLH in patients who have chronic diarrhea, hypogammaglobulinemia, platelet function defects, and sensorineural hearing defects and develop cytopenias, fever, elevated ferritin, or other signs of this syndrome.
Conflict-of-interest disclosure: The author owns common stock in Johnson & Johnson and receives research support from the National Institutes of Health and clinical trial support from Glaxo Smith Kline. ■
