Dysregulation of interleukin 5 expression in familial eosinophilia

Familial eosinophilia (FE) is a rare autosomal dominant inherited disorder characterized by the presence of lifelong peripheral eosinophilia (>1500/μL). Mapped to chromosome 5q31‐q33, the genetic cause of FE is unknown, and prior studies have failed to demonstrate a primary abnormality in the eosinophil lineage.

eosinophilia in affected family members, clinical manifestations related to the eosinophilia are uncommon. 2,3 This "benign" phenotype is likely related to a relative lack of eosinophil activation as evidenced by cellular morphology, surface activation markers, and release of eosinophil granule proteins. 3 Although the gene responsible for FE has been mapped by genomewide linkage analysis in one large kindred to a region on chromosome 5q containing the cytokine gene cluster (multipoint logarithm of odds score of 6.49), targeted sequencing of candidate genes, including IL-5, IL-3, and GMCSF, and their promoters has failed to identify the causative mutation. 4 Attempts to identify the causative mutation by whole-genome sequencing are ongoing, but have been unsuccessful to date.
The increased levels of morphologically normal peripheral blood eosinophils observed in subjects with FE could be due to a primary abnormality in the eosinophil lineage or to mediators produced by cell types other than the eosinophil. Prior studies, including morphologic examination, ex vivo differentiation of eosinophils from CD34+ cells, and eosinophil survival assays, have provided little evidence for a primary eosinophil defect. 3 Although serum cytokine levels measured previously were similar between affected and unaffected family members, the assay used to measure IL-5 was relatively insensitive, detecting IL-5 in sera from only three of 14 affected and three of 22 unaffected family members. 1 The aim of this study was to identify the cell population(s) responsible for driving the eosinophilia in FE.

| Study subjects
Affected and unaffected family members from two kindreds with autosomal dominant hypereosinophilia (HE; absolute eosinophil count >1500/lL) documented over at least three generations were evaluated on institutional review board-approved clinical protocols to study FE (NCT00091871 and a prior protocol that preceded clinicaltrials.gov) (see pedigrees in Fig. S1). Healthy controls (HC) without eosinophilia were recruited on a separate institutional review board-approved clinical protocol to obtain normal blood samples for in vitro research (NCT00001846). All participants gave written informed consent. As described previously, 3 subjects underwent an extensive clinical evaluation to exclude secondary causes of eosinophilia and assess end organ manifestations of eosinophilia. Routine laboratory testing, including complete blood counts, serum immunoglobulin levels, and assessment of T, B, and NK cell subsets in whole blood by flow cytometry, was performed in the Department of Laboratory Medicine, NIH Clinical Center. Mast cells were quantified in bone marrow core biopsies by a single hematopathologist following tryptase immunostaining. Results are reported as the average count over 10 high-powered fields.

| Peripheral blood mononuclear cell purification
Peripheral blood mononuclear cell (PBMC) were isolated from peripheral blood by density gradient centrifugation (Ficoll-Paque PLUS; GE Healthcare, Uppsala, Sweden). Red cells were lysed for 10 minutes at room temperature with ammonium-chloride-potassium (ACK) lysing buffer (Invitrogen, Carlsbad, CA, USA). Cell number and viability were determined by staining with trypan blue. Aliquots of PBMC were cryopreserved in liquid nitrogen in freezing medium with C-RPMI containing 10% fetal bovine serum (Invitrogen) and 7.5 % dimethyl sulfoxide (DMSO; Thermo Fisher Scientific, Waltham, MA, USA) or stored in TRIzol (Invitrogen) at À80°C for future use.

| Flow sorting of PBMC subsets
Peripheral blood mononuclear cell from affected family members and HC were stained with antibodies to CD3, CD4, CD8, CD19, and CD14 for isolation of CD3+CD4+ and CD3+CD8+ T cells, CD19+ B cells, and CD14+ monocytes. Fresh PBMC were stained with a cocktail of lineage markers to deplete lineage positive cells and with subset markers to sort innate lymphoid cells (ILC) subsets. Innate lymphoid cells were defined as LIN-CD45+CD127+ and ILC subsets as ILC1 (CD117À CRTH2À), ILC2 (CRTH2+), and ILC3 (CRTH2À CD117+) as previously described. 5 A complete listing of antibodies used can be found in Table S1.

| Microarray analysis
Total RNA was isolated from freshly purified PBMC stored in TRIzol according to the manufacturer's instructions (Invitrogen) prior to reverse transcription using a T7 oligo dT 6   levels were quantified in serum using an in-house chemiluminescence capture ELISA as previously described. 8

| Statistical analysis
Geometric means (GM) were used for measurements of central tendency. Group means were compared using the Mann-Whitney test, and paired samples were analyzed using the Wilcoxon signed rank test. A P-value of <.05 was considered statistically significant for all tests. For the microarray data, the table of gene probe set expression estimates (using the MAS5 algorithm) for 17 arrays (nine affected subjects, eight unaffected) was transformed by quantile normalization before calculating expression differences between the two groups using the SAS mixed-effects ANOVA procedure in JMP/Genomics    (Fig. S4). As ILCs are a known source of IL-5 and could account for the increased IL-5 seen in the CD3À PBMC fraction, ILC subsets were assessed in affected family members (n=3) and compared to those from normal controls (n=6). Whereas the ILC2 subset accounted for a similar proportion of the total ILC population in the two groups (GM 7% (range 2-10%) in HC compared to 4% (range 2-10%) in the affected family members; Figure 6A), flow-sorted ILC2s from the affected family members produced increased amounts of IL-5 following stimulation with PMA/ionomycin (GM net IL-5 of 185.9 vs 6.194 in HC; Figure 6B).

| DISCUSSION
Familial eosinophilia is a rare genetic disorder in which seemingly normal eosinophils are increased in the peripheral blood without Whereas absolute eosinophil counts have been linked to chromosome 5q31-q33 in other settings, including asthma in children, 10,11 schistosomiasis 12 and, more recently, in adults with a high risk of cardiovascular disease, 13 little is known about the specific mechanisms underlying these associations. Furthermore, the complex genetics of these conditions and the presence of elevated IgE and other findings suggestive of a more generalized type 2 response complicate the interpretation of functional studies. In this context, FE provides a unique opportunity to study the effect of a single genetic abnormality in chromosome 5q31-q33 on the regulation of eosinophilia. The major limitation of the present study is the fact that the studies to date have been restricted to one large kindred.
Although a second family with asymptomatic eosinophilia over three generations and increased production of IL-5 has recently been identified, the small number of affected family members available for study and logistical constraints have precluded additional analyses.
Given the lack of evidence pointing to an abnormality in the eosinophil lineage in FE, the current study focused on the role of circulating mononuclear cells in driving the eosinophilia. Although microarray analysis revealed few differences in PBMC gene expression between affected and unaffected family members, a relative increase in IL-5 expression was observed and confirmed by qRT-PCR. Of note, mild but significant increases in basophils and mast cells, the only other lineages that express IL-5Ra in humans, were also seen in affected family members. Other lineages were similar between the two groups (Fig. S5). Increased IL-5 expression was also significant at the protein level in serum and in supernatants from stimulated PBMCs. In contrast to the generalized Th2 responses seen in allergen or helminth-driven eosinophilia, however, there was no concomitant increase in PBMC expression of IL-4 or IL-13 mRNA, and serum IL-13 and IgE levels were significantly decreased in affected family members.
The genes for the three major Th2 cytokines, IL-5, IL-4, and IL-13, are clustered in a 160-kb region of chromosome 5 and are most often expressed together. 14,15 Nevertheless, selective regulators of expression of IL-4 and IL-13 have been described. 16  In contrast, prior to the present study, selective production of IL-5 had only been convincingly demonstrated in murine T-cell clones stimulated with IL-2 20 and in clonal T-cell populations from some patients with lymphocytic variant HES. 21,22 Whereas coordinated expression of IL-4, IL-5, and IL-13 has been shown to be regulated by a locus control region (LCR) containing the 3 0 portion of the RAD50 gene, 15,23 the genes encoding IL-4 and IL-13 are adjacent to each other and to the LCR, but separated from the IL-5 gene by the single large gene encoding RAD50. Furthermore, IL-5 transcription occurs in a direction opposite from that of IL-4 and IL-13. Consequently, it is not difficult to imagine a scenario in which specific regulatory elements might lead to isolated expression of IL-5. 24 In fact, isolated upregulated IL-5 transcription has been shown to occur in response to mIR-1248. 6 In the present study, the lack of concomitant upregulation of RAD50 is consistent with the hypothesis that IL-5 is not being expressed under control of the previously described LCR and suggests a novel regulatory mechanism for IL-5 mRNA expression. The fact that all cell types examined expressed increased levels of IL-5 mRNA implies that this dysregulation of expression is lineage independent.
In summary, FE is a rare genetic disorder of unknown etiology characterized by persistent eosinophilia and increased IL-5 production by PBMCs. The central role of IL-5 in the production, activation, and survival of eosinophils has made it a prime target for therapeutic intervention in asthma and other eosinophilic disorders. Familial eosinophilia provides a unique opportunity to explore a novel mechanism of IL-5 regulation in humans. Ultimately, this could lead to the development of new therapeutic strategies for the treatment of eosinophilic disorders.

ACKNOWLEDG MENTS
The authors would like to thank the many members of the Clinical Parasitology Section of the Laboratory of Parasitic Diseases who have participated in the protocol management and patient care that made this study possible.

CONFLI CTS OF INTEREST
The authors declare that they have no conflicts of interest.