Vitamin D enhances production of soluble ST2, inhibiting the action of IL-33.

To the Editor:

Vitamin D insufficiency is an environmental factor that has been strongly associated with asthma and its severity. The genes IL33 and IL1RL1 have been repeatedly identified as predisposing to asthma risk in genome-wide association studies. IL-33 is an alarmin cytokine that acts on multiple pulmonary cell types, including TH2 lymphocytes, mast cells, and innate lymphoid cells, to promote TH2-type cytokine secretion and airway inflammatory responses of the kind observed in asthmatic patients. The receptor for IL-33 is encoded by IL1RL1; differential splicing of the gene can produce a functional membrane-bound receptor (ST2L) or a soluble decoy receptor (sST2). Therefore we investigated whether IL1RL1 is regulated by vitamin D in cells relevant to asthma. We did this first by measuring gene expression by means of quantitative real-time PCR with 2 different TaqMan probe sets: Hs01073300, which detects mRNA splice variants encoding both the membrane-bound and soluble receptors (the total mRNA for IL1RL1), and Hs00249389, which detects mRNA only for the splice variant encoding the membrane-bound receptor ST2L (Fig 1, A). Methods are provided in this article's Online Repository at www.jacionline.org.

Fig 1

Vitamin D induces expression of sST2. A, Schematic for IL1RL1 mRNA: the 2 splice variants are differentially detected in PCR by using the probe sets Hs01073300 and Hs00249389. B-D,IL1RL1 transcripts detected by each probe set in cultures of HBECs at 24 hours (Fig 1, B; n = 7) and CD4 (Fig 1, C) and CD8 (Fig 1, D) lymphocytes at 7 days (n = 6). E-H, sST2 protein in culture supernatants of HBECs at 24 hours (Fig 1, E; n = 6), CD4 lymphocytes at 7 days (Fig 1, F; n = 6), and HBECs (Fig 1, G; n = 6) and HNECs at 24 hours (Fig 1, H; n = 5). ns, Not significant. *P ≤ .05, **P ≤ .01, and ***P ≤ .001.

Both probe sets display similar efficiency of amplification (data not shown). Human primary bronchial epithelial cells (HBECs), CD4 lymphocytes, CD8 lymphocytes, eosinophils, and LUVA mast cells were cultured in the presence or absence of 1α,25-dihydroxyvitamin D3 (1,25[OH]D3), the active form of vitamin D (Fig 1, B-D, and see Fig E1, A and B, in this article's Online Repository at www.jacionline.org). Addition of 1,25(OH)D3 significantly increased the total number of IL1RL1 mRNA transcripts expressed by HBECs and CD4 and CD8 lymphocytes, as measured by using the Hs01073300 probe set. However, 1,25(OH)D3 did not significantly increase expression of IL1RL1 mRNA transcripts by primary eosinophils and LUVA mast cells, despite it significantly increasing expression of the gene cathelicidin antimicrobial peptide (CAMP), which is known to be induced by vitamin D (see Fig E1, A and B).

Fig E1

Further data on the effect of vitamin D on IL1RL1 expression. A and B, 1,25(OH)D3 (100 nmol/L) increases the expression of mRNA for CAMP in primary human eosinophils (Fig E1, A) and LUVA mast cells (Fig E1, B) but not mRNA for IL1RL1, as measured by using the Hs01073300 probe set in either cell type (n = 5 for both cell types). C and D, 25(OH)D3 (100 nmol/L) increases the expression of mRNA for both IL1RL1 and CAMP in HBECs (Fig E1, C) but not in CD4+ lymphocytes (Fig E1, D; Fig E1, C: n = 5; Fig E1, D: n = 3). *P ≤ .05 and **P ≤ .01.

Higher IL1RL1 expression levels were detected with the Hs01073300 probe set than with the Hs00249389 probe set in HBEC and lymphocyte cultures. Although 1,25(OH)D3 increased the total copies of IL1RL1 mRNA measured by using the Hs01073300 probe set in these cell types, it did not increase the number of transcripts encoding ST2L detected with Hs00249389, indicating that vitamin D selectively upregulates the expression of mRNA for the soluble decoy receptor sST2. sST2 concentrations in culture supernatants were measured by means of ELISA to confirm the findings of gene expression studies at the protein level. sST2 concentrations were significantly increased by 1,25(OH)D3 in both HBEC and CD4 lymphocyte cultures (Fig 1, E and F). In contrast, CD8 lymphocytes produced less than 200 pg/mL sST2 in all conditions, and there was no evidence of upregulation by 1,25(OH)D3 (data not shown).

Vitamin D circulates primarily as the inactive precursor 25-hydroxyvitamin D3 (25[OH]D3); however, epithelial cells constitutively express CYP27B1, the enzyme that converts 25(OH)D3 to 1,25(OH)D3. HBECs cultured with 100 nmol/L 25(OH)D3, a physiologic concentration, increased expression of IL1RL1 (see Fig E1, C) and produced significantly greater amounts of sST2 (Fig 1, G). Primary human nasal epithelial cells (HNECs) cultured with a similar concentration range of 25(OH)D3 also showed a clear concentration-dependent increase in sST2 production (Fig 1, H). CD4 lymphocytes did not respond to 25(OH)D3, which is consistent with their lower expression of CYP27B1 (see Fig E1, D).

An IL-33 sensitive bioassay was developed using the LUVA mast cell line to test the biological activity of sST2. Twenty-four hours of exposure of LUVA cell cultures to IL-33 resulted in marked homotypic aggregation, which was associated with increased expression of CD54 (intercellular adhesion molecule 1; see Fig E2, A and B, in this article's Online Repository at www.jacionline.org). IL-33 increased CD54 expression on LUVA cells in a concentration-dependent manner (Fig 2, A). Other stimuli can also induce mast cell CD54 expression, and a similar concentration-dependent induction of CD54 expression was achieved with a combination of IFN-γ and IL-4 (Fig 2, A). Treatment of LUVA cells with a recombinant sST2-Fc chimera significantly inhibited IL-33–induced CD54 expression (P = .0076; Fig 2, B, and see Fig E2, C) but not IFN-γ plus IL-4–induced CD54 expression (see Fig E2, D). Addition of 1,25(OH)D3 itself to LUVA cell cultures did not significantly alter constitutive or induced CD54 expression (P > .05; see Fig E2, E), as previously reported for another mast cell line.

Fig E2

Stimulation with IL-33 induces LUVA cell expression of CD54. A and B, IL-33 causes homotypic aggregation of LUVA cells (×10 magnification; Fig E2, A) associated with increased expression of CD54 (filled gray histogram, unstained; dashed line, isotype control; solid line, anti-CD54; Fig E2, B). C, Inhibition of IL-33–induced LUVA cell CD54 expression by the recombinant sST2-Fc chimera is not observed with an isotype Fc chimera (n = 3). D, IFN-γ plus IL-4–induced LUVA cell CD54 expression is not inhibited by addition of recombinant sST2-Fc chimera (n = 4). E, Effect of IL-33 on LUVA cell CD54 expression is not affected by addition of active vitamin D (100 nmol/L 1,25[OH]D3) directly to the LUVA cells (n = 4).

Fig 2

Conditioned medium from vitamin D–treated epithelial cultures inhibits IL-33–induced expression of CD54 by LUVA cells. A, Induction of CD54 on LUVA cells above that of unstimulated cells by different stimuli in the presence of conditioned medium from control epithelial cell cultures (n = 7). B, Effect of IL-33 on CD54 expression is inhibited by recombinant sST2 (n = 6). C and D, CD54 expression on LUVA cells cultured with conditioned medium with or without cell stimulation. E, Percentage reduction in stimuli-induced CD54 expression on LUVA cells cultured with conditioned medium from 25(OH)D3-treated epithelial cell cultures compared with control epithelial cell cultures (n = 7). MFI, Mean fluorescence intensity; ns, not significant. *P ≤ .05, **P ≤ .01, ***P ≤ .001, and ****P ≤ .0001.

To examine the effect of vitamin D–enhanced epithelial sST2 production on the IL-33 bioassay, HNECs were cultured in the presence or absence of 25(OH)D3 for 48 hours, and then the supernatants were added to cultures of LUVA cells stimulated with concentrations of IL-33 or IFN-γ plus IL-4 that induced comparable CD54 expression. Conditioned medium from vitamin D–treated epithelial cells significantly reduced IL-33–induced CD54 expression on LUVA cells compared with conditioned medium from matched epithelial cell cultures not treated with vitamin D (P = .015; Fig 2, C and E). In contrast, there was no significant difference in IFN-γ plus IL-4–induced CD54 expression between LUVA cells cultured with the different epithelial cell–conditioned media (P > .05; Fig 2, D and E). Although mediators in the epithelial cell–conditioned medium other than sST2 might conceivably have affected LUVA cell CD54 expression, the selective effect of vitamin D treatment on IL-33–induced CD54 expression suggests that the sST2 in the vitamin D–treated epithelial cell supernatants is biologically active.

The capacity of IL-33 to induce production of TH2-type proinflammatory cytokines by multiple cell types likely underpins the reported genetic associations of IL33 and IL1RL1 with asthma. Here we report the novel finding that 100 nmol/L vitamin D is able to augment expression by epithelial cells and lymphocytes of the soluble decoy receptor sST2, which in turn inhibits the actions of IL-33. Importantly, this effect occurs at physiologic concentrations (vitamin D sufficiency is defined as a serum 25[OH]D level of 75 to 150 nmol/L), and similar concentrations of 1,25(OH)D3 have been shown to be able to be generated from 25(OH)D3 in culture. We hypothesize that the capacity of vitamin D to augment the synthesis of an inhibitor of IL-33 in situ in the airways mucosa is of potential benefit in the limitation of asthmatic mucosal inflammation. Furthermore, this might in part account for the paradox that epidemiologic studies have repeatedly revealed associations between vitamin D insufficiency and both the risk of more severe asthma and increased serum IgE concentrations, whereas other studies have reported that in culture vitamin D can act directly on TH2 lymphocytes to promote TH2 cytokine secretion (eg, Boonstra et al). The apparent beneficial association of vitamin D in asthma in vivo, which in many patients is a TH2-type cytokine pathology, suggests that the direct action of vitamin D in promoting TH2 lymphocyte responses is less important in vivo than other vitamin D–mediated mechanisms (eg, vitamin D upregulated production of sST2 and the induction of regulatory mechanisms) in inhibiting TH2-type cytokine responses. Because the enhancement by vitamin D of sST2 production is concentration dependent, this supports therapeutic strategies to boost pulmonary vitamin D levels to reduce asthmatic inflammation.