NF45 and NF90 regulate HS4-dependent interleukin-13 transcription in T cells

Abstract

Expression of the cytokine interleukin-13 (IL13) is critical for Th2 immune responses and Th2-mediated allergic diseases. Activation of human IL13 expression involves chromatin remodeling and formation of multiple DNase I-hypersensitive sites throughout the locus. Among these, HS4 is detected in the distal IL13 promoter in both naive and polarized CD4(+) T cells. We show herein that HS4 acts as a position-independent, orientation-dependent positive regulator of IL13 proximal promoter activity in transiently transfected, activated human CD4(+) Jurkat T cells and primary murine Th2 cells. The 3'-half of HS4 (HS4-3') was responsible for IL13 up-regulation and bound nuclear factor (NF) 90 and NF45, as demonstrated by DNA affinity chromatography coupled with tandem mass spectrometry, chromatin immunoprecipitation, and gel shift analysis. Notably, the CTGTT NF45/NF90-binding motif within HS4-3' was critical for HS4-dependent up-regulation of IL13 expression. Moreover, transfection of HS4-IL13 reporter vectors into primary, in vitro differentiated Th2 cells from wild-type, NF45(+/-), or NF90(+/-) mice showed that HS4 activity was exquisitely dependent on the levels of endogenous NF45 (and to a lesser degree NF90), because HS4-dependent IL13 expression was virtually abrogated in NF45(+/-) cells and reduced in NF90(+/-) cells. Collectively, our results identify NF45 and NF90 as novel regulators of HS4-dependent human IL13 transcription in response to T cell activation.

FIGURE 1.

HS4 acts as an IL13 cis-regulatory element. A, schematic representation of DNase I HS sites (arrows) mapped throughout the IL13 locus in human primary CD4⁺ T cells (17). Exons I–IV are depicted as black boxes. B and C, Jurkat T cells were transiently co-transfected with IL13 luciferase (Luc) reporter constructs and pRLTK and harvested after 16 h of culture in the presence or absence of PMA (20 ng/ml) and ionomycin (1 μm). Results are expressed as RLA fold-induction (mean ± S.E.) measured in four (B) or seven (C) independent experiments. Statistical significance was calculated using the Wilcoxon two-sample test. n.s., not significant.

FIGURE 2.

Isolation and identification of HS4-3′-interacting proteins. A, location and sequence of biotinylated oligonucleotides used as baits in DNA affinity chromatography. The NcoI restriction site is underlined. B, biochemical approach to the isolation and identification of HS4-3′(−1527/−1496)-binding proteins. C, total numbers of protein species recovered from DNA affinity chromatography and identified by tandem mass spectrometry approaches. D, NF45 and NF90 peptides identified by tandem mass spectrometry. aa, amino acids.

FIGURE 3.

NF90 and NF45 bind HS4 in vivo. Chromatin from Jurkat T cells, resting or activated with PMA (20 ng/ml) and ionomycin (2 μm) for 4 h, was cross-linked, sonicated, and immunoprecipitated with anti-NF90 or anti-NF45 antibodies. Target enrichment was assessed by real time PCR with primers that amplify a 244-bp region spanning HS4-3′ (nucleotides −1531 to −1287) or a 249-bp negative control region within the myc1 locus. A standard curve for each experiment was generated with serial dilutions of input DNA. Results are expressed as the mean ± S.E. of the ratio between the number of HS4 and myc1 target copies immunoprecipitated by anti-NF90 or anti-NF45 antibodies in three independent experiments. P/I, PMA + ionomycin.

FIGURE 4.

DNA/protein interactions at HS4-3′-(−1527/−1496). A, sequence of oligonucleotides used as EMSA probes and competitors. Underlined is the reported (IL2pARRE) or predicted (HS4-3′-(−1527/−1496)) recognition motif for NF45/NF90. B, EMSA analysis was performed using nuclear proteins extracted from Jurkat T cells after a 4-h incubation with or without PMA (20 ng/ml) and ionomycin (2 μm). Competitors were added at 90-fold (lanes 6 and 7 and lanes 9 and 10) or 30-fold (lanes 12 and 13 and lanes 15 and 16) molar excess and are noted above the relevant lanes. The figure shows three independent gels (gel 1, lanes 1–4; gel 2,: lanes 5–10; and gel 3, lanes 11–16). Lanes that were not contiguous in the original gels and have been juxtaposed in the figure are separated by black lines. C, Jurkat T cells were incubated in the presence or absence of PMA (20 ng/ml) and ionomycin (1 μm) for 4 h. Levels of NF45 and NF45 serine phosphorylation were assessed by Western blot (WB) analysis of whole cell lysates (WCL) representing 800,000, 400,000, and 200,000 cell equivalents, using an anti-NF45 (upper panel) or an anti-phosphoserine (lower panel) antibody. P/I, PMA + ionomycin.

FIGURE 5.

CTGTT motif within HS4-3′ is critical for HS4-mediated IL13 up-regulation. CD4⁺ T cells were isolated from WT C57BL/6 mice (n = 2) and in vitro differentiated under Th2 skewing conditions for 7 days. Th2 polarization was assessed by intracellular staining for IL-4 and IL-13. After 2–3 days in the presence of plate-bound anti-CD3 and anti-CD28 mAbs, Th2 cells were nucleofected with the IL13 promoter reporter constructs HS6/Luc, HS4-HS6/Luc, HS4-3′-HS6/Luc, and HS4-3′mut-HS6/Luc. Results are expressed as the mean ± S.E. of RLA measured in seven independent experiments. Statistical significance of individual comparisons was assessed using the Wilcoxon two-sample test. n.s., not significant; Luc, luciferase.

FIGURE 6.

HS4-dependent IL13 up-regulation depends on the levels of endogenous NF45 and NF90. A, CD4⁺ T cells were isolated from WT (n = 3), NF45^+/− (n = 4), or NF90^+/− (n = 4) C57BL/6 mice and in vitro differentiated into Th2 cells for 1 week. Th2 polarization was assessed by intracellular staining for IL-4 and IL-13. PE, phycoerythrin. B, Th2 cells were re-stimulated with plate-bound anti-CD3 and anti-CD28 mAbs for 2–3 days and nucleofected with HS6/Luc or HS4-HS6/Luc. Results are expressed as the mean ± S.E. of RLA measured in nine independent experiments. Statistical significance was calculated using the Wilcoxon two-sample test (*, p ≤ 0.04). n.s., not significant.

FIGURE 7.

HS4 is a recently arisen regulatory element. The ClustalW program (40) generated a multiple sequence alignment for the HS4 region in humans, 11 primate species from distinct clades, and mice. Hominoid, Old World monkey (OWM), and New World monkey (NWM) IL13 promoter sequences were previously generated in our laboratory (19). Human, mouse, and prosimian (PS) IL13 promoter sequences were obtained from GenBank^TM accession number NC_000005, GenBank^TM accession number NC_000077, and Ensembl GeneScaffold_442, respectively. Dashes and asterisks mark conserved positions and gaps, respectively. The CTGTT motif is bracketed. Numbering is relative to the IL13 ATG.