2607 - In Vivo Delivery of Alternatively Spliced Interleukin-4 Mimics Scleroderma Lung Disease

Wednesday, November 9, 2011: 10:00 AM
W192b (McCormick Place West)
Sergei P. Atamas and Irina G. Luzina, University of Maryland School of Medicine, Baltimore, MD
Presentation Number: 2607

Background/Purpose: Our previous studies and works of others have suggested that scleroderma lung disease (SLD) is driven by interleukin (IL)-4 and other type 2 cytokines. However, this notion does not explain the previously reported by us predominance of CD8+ cells in the lungs of patients with SLD. It is also contradicted by the absence of the typical type 2 phenotype in SLD, e.g., by the lack of overt eosinophilia and goblet cell hyperplasia seen in such prototypic IL-4-driven disease as asthma. Based on our previous discovery of a natural splice variant of IL-4, which omits exon 2-encoded region and is called IL-4δ2, and also based on our previous observation of elevated IL-4δ2 mRNA in patients with SLD, we hypothesized that SLD may be driven mostly by IL-4δ2 rather than by IL-4. Since the in vivo effects of IL-4δ2 are unknown, we tested whether the effects of the in vivo IL-4δ2 gene delivery will differ from those of IL-4 and will mimic SLD phenotype.

Method: Replication-deficient adenoviral constructs were created, validated, and used to deliver IL-4δ2 or IL-4 to mouse lungs in vivo. The effects of such gene delivery on the lungs were investigated, compared to each other and to a non-coding NULL adenoviral vehicle control.

Results: Delivery of IL-4δ2 (2,150 ± 470 pg/ml) and IL-4 (2,230 ± 530 pg/ml) was confirmed by ELISA assays of lung homogenates. Based on immunohistochemical analyses of lung tissue and flow cytometric analyses of BAL samples, delivery of either IL-4δ2 or IL-4 caused similar pulmonary infiltration by T and B lymphocytes; ~30 % of BAL cells were lymphocytes in either case, whereas only ~3% of lymphocytes were observed in BAL samples of NULL-challenged mice. Delivery of IL-4δ2 attracted three fold more CD8+ cells than did delivery IL-4 (15.5 ± 2.9 % vs 5.5 ± 1.7 % of BAL lymphocytes, respectively). Delivery of IL-4 induced eosinophilia (29 ± 8 % of total BAl cells) and goblet cell hyperplasia (25 ± 9 cells/field), whereas delivery of IL-4δ2 did not have such effects (~ 2 ± 2 % eosinophils in BAL and 4 ± 2 goblet cells/field histologically). Both variants stimulated proinflammatory cytokines; IL-4δ2 induced higher levels of TNF-α and IFN-γ, whereas IL-4 induced higher eotaxin, IL-17, and MCP-1 in the lungs. Differential effects of IL-4δ2 and IL-4 on global gene expression were observed by microarray analyses of lung tissue. Expression of 84 genes was similarly affected by either IL-4δ2 or IL-4, including genes for CTLA4, MHC II, C1q, and several chemokines. Expression of 38 genes was uniquely affected by IL-4δ2, including genes for Cyr61, CD48, Egr2, Lck, lysyl oxidase, and CXCL13. Total pulmonary hydroxyproline assays revealed that both IL-4δ2 and IL-4 tended to increase accumulation of collagen in the lungs.

Conclusion: Both complete (IL-4) and alternatively spliced (IL-4δ2) isoforms induce immune inflammation in the lungs. However, the differences in eosinophil accumulation, goblet cell hyperplasia, and relative accumulation of CD8+ cells caused by IL-4 versus IL-4δ2 resemble the differences between patients with asthma and patients with SLD, respectively. Combined with our previous observation of elevated IL-4δ2 mRNA, these findings suggest that new therapies for SLD need to focus on targeting IL-4δ2 rather than IL-4.

Keywords: fibroblasts, interleukins (IL), lung, lymphocytes and scleroderma

Disclosure: S. P. Atamas, None; I. G. Luzina, None.