Nuclear factor I/thyroid transcription factor 1 interactions modulate surfactant protein C transcription

Abstract

Surfactant protein C (SP-C; Sftpc) gene expression is restricted to pulmonary type II epithelial cells. The proximal SP-C promoter region contains critical binding sites for nuclear factor I (NFI) and thyroid transcription factor 1 (TTF-1; also called Nkx2.1). To test the hypothesis that NFI isoforms interact with TTF-1 to differentially regulate SP-C transcription, we performed transient transfection assays in JEG-3 cells, a choriocarcinoma cell line with negligible endogenous NFI or TTF-1 activity. Cotransfection of NFI family members with TTF-1 induced synergistic activation of the SP-C promoter that was further enhanced by p300. TTF-1 directly interacts with the conserved DNA binding and dimerization domain of all NFI family members in coimmunoprecipitation and mammalian two-hybrid experiments. To determine whether SP-C expression is regulated by NFI in vivo, a chimeric fusion protein containing the DNA binding and dimerization domain of NFI-A and the Drosophila engrailed transcriptional repression domain (NFIen) was conditionally expressed in mice under control of a doxycycline-inducible transgene. Induction of NFIen in a subset of type II cells inhibited SP-C gene expression without affecting expression of TTF-1 in doxycycline-treated double-transgenic mice. Taken together, these findings support the hypothesis that NFI family members interact with TTF-1 to regulate type II cell function.

FIG. 1.

Coexpression of NFI isoforms in JEG cells modulates SP-C promoter activity. JEG-3 cells were transiently transfected in 6-well plates with 2 μg of p0.32SPCluc, the indicated amounts of expression plasmids for NFI family members, and 0.5 μg of pCMV-βgal/well. Relative luciferase activity was determined by correcting for β-Gal activity and then setting the activity of p0.32SPCluc cotransfected with empty vectors to 1. Data are shown as means ± SE (n = 9; three independent experiments). All cotransfections with other NFI family members were significantly different from that with NFI-C alone (P < 0.01).

FIG. 2.

Homodimers of all NFI isoforms bind similarly to a palindromic NFI site probe (NFI) but with varying weaker affinities to C1 and C3 NFI half-site probes from the SP-C promoter. (A) Western blot analysis of nuclear extracts from JEG cells transiently transfected with HA-tagged NFI family members detected by anti-HA monoclonal antibody. (B) EMSA analysis of selected nuclear extracts from panel A. Extracts were incubated with the indicated ³²P-labeled double-stranded oligonucleotides and electrophoresed on nondenaturing gels. The sequence of each EMSA probe is shown, and the NFI consensus binding sites are underlined.

FIG. 3.

Synergistic SP-C promoter activation by TTF-1 and NFI family members. JEG cells were cotransfected with 1.5 μg of p0.32SPCluc per well, 1 μg of the indicated HA-tagged NFI isoform per well, and 0.5 μg of either pCMVTTF-1 (+) or pCDNA3 empty vector (−) per well. Relative SP-C promoter activity was calculated as described for Fig. 1. Data are plotted as means ± SE (n = 4 to 6).

FIG. 4.

Coimmunoprecipitation of NFI-TTF-1 complexes. (A) Nuclear extracts of SL2 cells transfected with HA-tagged NFI-A1 or NFI-C2 were incubated as indicated with extracts of SL2 cells transfected with TTF-1, and complexes were immunoprecipitated (IP) with polyclonal anti-HA antiserum and protein G-agarose. Immunoprecipitated proteins or equivalent amounts of input protein were resolved by SDS-polyacrylamide gel electrophoresis (PAGE), transferred to Hybond membranes, and sequentially probed with monoclonal antibodies against HA and TTF-1. (B) Nuclear extracts of SL2 cells transfected with the indicated NFI isoforms or a truncation mutant of NFI-A containing the DNA binding and dimerization domain (dbd) was incubated with TTF-1, and complexes were collected and visualized as for panel A.

FIG. 5.

NFI and TTF-1 fusion proteins interact via the conserved 5′ DNA binding and dimerization domain (dbd) of NFI. JEG cells were cotransfected with the indicated combinations of vectors expressing fusion proteins of the GAL dbd and the VP16 transactivation domain (tad), a luciferase reporter plasmid driven by the GAL4 upstream activator sequences, and pCMVβgal as an internal control. Relative luciferase activity was calculated as described for Fig. 1 except that luciferase activity stimulated by each GALdbd fusion protein (full-length TTF-1, NFI-A1, or NFI-A5′dbd) cotransfected with the VP16tad vector pM (control) was set to 1. Data are plotted as means ± SE (n = 6).

FIG. 6.

Cotransfection of NFI and TTF-1 with the coactivator protein p300 caused synergistic activation of the 0.32 SP-C promoter. JEG cells were cotransfected with 1.5 μg of the SP-C luciferase construct indicated at the left and combinations of expression vectors for NFI-A1 (1 μg), TTF-1 (0.5 μg), and p300 (0.5 μg), with pCMVβgal (0.5 μg) to control for transfection efficiency. The relative activity of each luciferase construct cotransfected with the relevant empty vector was set to 1 and data were plotted as means ± SE (n = 9). Schematic representations of the transcription factor interactions with each construct are shown at the left, and mutant sites are indicated by an “X.” All NFI sites are mutated in pSPC-NFI(−) and both TTF-1 sites are mutated in the C2 footprinted region of pSPC-TTF-1(−).

FIG. 7.

Dox-induced expression of the dominant-negative NFIen chimeric protein inhibited transactivation of the SP-C promoter by coexpressed NFI family members. (A) Schematic representation of the Dox-inducible chimeric dominant-negative NFI-A-engrailed (NFIen) expression vector. The HA-tagged chimeric protein consists of the DNA binding and dimerization domain of mouse NFI-A and the transcriptional repressor domain of Drosophila engrailed. Cotransfection with the rtTA and treatment with 1 μg of Dox per ml induces NFIen expression. (B) JEG cells were cotransfected with p0.32SPCluc, pUHG expressing rtTA, pCMVβgal, and expression vectors for the indicated NFI family members and were incubated for 24 h in the absence (white bars) or presence (gray bars) of Dox. *, P < 0.01 versus no Dox (means ± SE; n = 4).

FIG. 8.

Generation of double-transgenic mice. (A) Diagram of breeding scheme. Activator mice are transgenic for rtTA under control of the rat CCSP promoter, which targets expression to a subset of bronchiolar Clara cells and alveolar TII cells. Operator mice harbor the HA-Nfi-a-engrailed transgene (NFIen) under the control of the minimal CMV promoter, with seven copies of the Tet operator sequence, which is only active in the presence of rtTA bound to Dox. (B) Dox-induced expression of NFIen in double-transgenic mice inhibited SP-C mRNA but not TTF-1 mRNA. Northern blots of total lung RNA (10 μg/lane) were prepared from adult double-transgenic littermates that were untreated (lanes 1, 2, 5, and 6) or fed Dox-treated food for 3 days (lanes 3 and 4) or 8 days (lanes 7 and 8). Blots were sequentially hybridized with ³²P-labeled probes for the NFIen transgene, SP-C, TTF-1, and ribosomal protein L32 mRNA as a control for loading.

FIG. 9.

Dox-induced expression of NFIen inhibited SP-C expression but not TTF-1 expression in transgene-positive regions of lung parenchyma. Adult double-transgenic mice were fed normal (A to D) or Dox-treated (E to H) food for 8 days. Anti-HA IHC was used to detect HA-tagged NFIen transgene expression. NFIen was not detected in double-transgenic mice that were fed normal mouse chow (A) but was readily detected in a subset of TII cells in double-transgenic mice that were fed Dox for 8 days (E). TTF-1 was detected in the nuclei of TII cells in both transgene-positive and transgene-negative areas (B and F). SP-C mRNA was detected by ISH on serial sections of the same regions used for IHC, viewed by dark-field (C and G) and phase-contrast (D and H) optics. Arrows indicate representative IHC-positive TII cells. Bar = 50 μm.