This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Steele-Perkins, G. Articles by Gronostajski, R. M. Search for Related Content PubMed PubMed Citation Articles by Steele-Perkins, G. Articles by Gronostajski, R. M.

 Previous Article  |  Next Article 

Molecular and Cellular Biology, January 2005, p. 685-698, Vol. 25, No. 2
0270-7306/05/$08.00+0     doi:10.1128/MCB.25.2.685-698.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

The Transcription Factor Gene Nfib Is Essential for both Lung Maturation and Brain Development

George Steele-Perkins,^1, Céline Plachez,² Kenneth G. Butz,¹ Guanhu Yang,³ Cindy J. Bachurski,³ Stephen L. Kinsman,⁴ E. David Litwack,² Linda J. Richards,² and Richard M. Gronostajski^1*

Department of Biochemistry and the Program in Neuroscience, State University of New York at Buffalo, Buffalo, New York,¹ Division of Pulmonary Biology, Cincinnati Children's Research Foundation, Cincinnati, Ohio,³ Department of Anatomy and Neurobiology and Program in Neuroscience,² Department of Pediatrics, University of Maryland, Baltimore, School of Medicine, Baltimore, Maryland⁴

Received 25 September 2004/ Accepted 10 October 2004

	ABSTRACT

Top Abstract Introduction Materials and Methods Results Discussion References

 
The phylogenetically conserved nuclear factor I (NFI) gene family encodes site-specific transcription factors essential for the development of a number of organ systems. We showed previously that Nfia-deficient mice exhibit agenesis of the corpus callosum and other forebrain defects, whereas Nfic-deficient mice have agenesis of molar tooth roots and severe incisor defects. Here we show that Nfib-deficient mice possess unique defects in lung maturation and exhibit callosal agenesis and forebrain defects that are similar to, but more severe than, those seen in Nfia-deficient animals. In addition, loss of Nfib results in defects in basilar pons formation and hippocampus development that are not seen in Nfia-deficient mice. Heterozygous Nfib-deficient animals also exhibit callosal agenesis and delayed lung maturation, indicating haploinsufficiency at the Nfib locus. The similarity in brain defects in Nfia- and Nfib-deficient animals suggests that these two genes may cooperate in late fetal forebrain development, while Nfib is essential for late fetal lung maturation and development of the pons.

	INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

 
Nuclear factor I (NFI) transcription and replication proteins function both in adenoviral DNA replication (12, 43, 44) and in the regulation of transcription throughout development (21). There are four NFI genes in mammals (Nfia, Nfib, Nfic, and Nfix) and single NFI genes in Drosophila melanogaster, Caenorhabditis elegans, Anopheles spp., and other simple animals (21, 30, 50). No NFI genes have been found in plants, bacteria, or single-cell eukaryotes. In mammals, NFI proteins function as homo- or heterodimers and are expressed in complex, overlapping patterns during embryogenesis (6, 31). NFI proteins bind to a dyad-symmetric binding site (TTGGCN₅GCCAA) with high affinity (20, 40), and NFI proteins have been shown to either activate or repress gene expression depending on the promoter and cellular context (21, 42). The presence of four NFI genes in mammals with possibly overlapping functions makes it a challenge to identify in vivo targets of individual NFI proteins and the roles of NFI genes in development.

We showed previously that disruption of Nfia causes late gestation neuroanatomical defects, including agenesis of the corpus callosum, size reductions in other forebrain commissures, and loss of specific midline glial populations (11, 56). In contrast, disruption of Nfic results in early postnatal defects in tooth formation, including the loss of molar roots and aberrant incisor development (61). In a previous study, insertion of a lacZ reporter gene into the Nfib locus resulted in defects in lung maturation but no apparent defects in brain development (22). Here we report the replacement of the essential exon 2 of the Nfib gene with a lacZ reporter gene and show that mice homozygous for our replacement mutation have major neuroanatomical defects similar to, but more severe than, those of Nfia^–/– mice. These defects include callosal agenesis, aberrant hippocampus and pons formation, and loss of specific midline glial populations. In addition, as was seen in the previously described Nfib insertion mutant, these Nfib-deficient animals show defects in lung development and die at birth with immature lungs. The loss of one copy of Nfib results in callosal agenesis in some animals and reduced lung maturation in all animals, demonstrating haploinsufficiency at the Nfib locus. Together, these defects indicate that Nfia and Nfib may function cooperatively during mouse forebrain development and that Nfib alone is essential for late gestational lung maturation and hippocampus and pons development.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

 
Gene targeting and mouse strains. Genomic clones containing Nfib exons 1 and 2 were isolated from a mouse 129/Sv phage library as described previously (11, 61). We modified lacZ to encode the simian virus 40 T antigen nuclear localization signal (NLS) at the ß-galactosidase (ß-Gal) N terminus and to include a phosphoglycerate kinase (PGK) gene poly(A) signal at its 3' end (64), generating NLS-ß-Gal-pA. Using a convenient EcoRI site at the 5' end of exon 2 we cloned NLS-ß-Gal-pA in frame with a 5.65-kb Nfib fragment ending at base 9 of exon 2 to form a translational fusion protein with the product of exon 2 of Nfib. A PGK-Neo-bGHpA cassette (60) was cloned downstream of the lacZ fusion cassette, followed by 3 kb of intron 2. The NLS-ß-Gal-pA and Neo cassettes replace 700 bp of Nfib, including all but 9 bp of exon 2. A diphtheria toxin A chain-expressing negative selection cassette (PGK-DTA-bGHpA) was assembled and cloned downstream of the 3' homologous region in the opposite transcriptional orientation (67). E14-1 mouse embryonic stem (ES) cells (5 x 10⁶) were electroporated with 5 µg of gel-purified 15.2-kb XhoI targeting vector fragment and then selected in 0.2-mg/ml G418 (GIBCO/BRL) for 6 days. Colonies were picked and expanded twice in 96-well plates. Two plates were frozen, and two were used for DNA isolation and Southern blot analyses.

Correctly targeted ES cells were injected into C57BL/6 blastocysts, and stable transmission of the targeted allele was assessed by PCR genotyping of the progeny of founder male chimeras. The targeted (knockout [KO]) allele was crossed into Black Swiss (Taconic), C57BL/6 (Taconic and Jackson Laboratories), and FVB/N (Jackson Laboratories) strains for analysis. Mice derived from two independent targeted ES cell clones were analyzed and had indistinguishable phenotypes. Since some 129 strains of mice have a recessive callosal agenesis phenotype (35, 36) and since the ES cells used were derived from 129S6 mice, all brain sections were prepared from mice backcrossed into the Black Swiss (one generation) or C57BL/6 (more than five generations) strain to avoid the as yet uncharacterized 129 recessive acallosal-susceptibility locus. The lung morphology phenotypes of Nfib mutant animals of the Black Swiss, C57BL/6, and FVB/N strains were indistinguishable. Each figure legend contains the strain used in the figure.

PCR genotyping. Postnatal day 10 (P10) to P14 0.5-cm tail tip biopsy samples were shaken at 55°C overnight in 0.5 ml of tail lysis buffer (100 mM Tris-Cl [pH 8.5], 5 mM EDTA, 200 mM NaCl, 0.2% sodium dodecyl sulfate) including 0.1 mg of proteinase K/ml. One microliter of the lysate diluted 1:10 in H₂O was analyzed by multiplex PCR (0.5 U of Platinum Taq DNA polymerase [Invitrogen], 1.5 mM MgCl₂, 0.2 mM [each] deoxynucleoside triphosphate, 1x Rediload [Research Genetics], 1x PCR buffer [Invitrogen; 10 mM Tris-Cl {pH 8.3}, 50 mM KCl, 0.02 mg of bovine serum albumin/ml], five primers at 0.2 µM each). Primers were Nfib specific (a and b), lacZ specific (c), or mouse Y chromosome specific (SRY-1 and -2); sequences are as follows: a, 5'-GCTGAGTTGGGAGATTGTGTC-3'; b, 5'-TTCTGCTTGATTTCGGGCTTC-3'; c, 5'-CATCGTAACCGTGCATCTGCC-3'; SRY-1, 5'-AACAACTGGGCTTTGCACATTG-3'; SRY-2, 5'-GTTTATCAGGGTTTCTCTCTAGC-3'). PCR products were resolved on 2.0% agarose gels and visualized with ethidium bromide.

PCR and QPCR transcript analysis. Total RNA was isolated with TRIzol reagent (Invitrogen), and cDNA was generated from 2 to 5 µg of RNA with Superscript II reverse transcriptase (Invitrogen) and random primers as recommended by the manufacturer. Nfib transcripts were analyzed by PCR with an exon 1-based sense primer (5'-GATCGGCCTCACGGGCCGATGATGTATTCTCCCATCTG-3') and an exon 4-based antisense primer (5'-CTCTGATACATTGAAGACTCCG-3'), and products were resolved on a 2.5% agarose gel. Transcript levels of all genes shown were quantified by quantitative PCR (QPCR) with a Bio-Rad iCycler real-time PCR machine and gene-specific primers that span multiple exons as described previously (61). Primer sequences for the genes quantified are available upon request.

Tissue preparation and lacZ detection. Embryos were harvested at various times after detection of a coital plug. For total lung DNA analysis, either the dissected left lobe (embryonic day 18.5 [E18.5]) or the whole lung (E15.5) was harvested into 1 ml of 2 M NaCl-2 mM EDTA in phosphate-buffered saline (PBS) and sonicated on ice, and DNA content was assayed fluorimetrically as described previously (32) with a 96-well plate fluorimeter and purified calf thymus DNA as a standard. Whole thoraces were immersed in cold 4% paraformaldehyde (PFA) overnight, washed in PBS, dehydrated in ethanol, embedded in paraffin, and transversely sectioned at 5 µm for hematoxylin and eosin (H&E) staining and immunohistochemistry.

ß-Gal staining was performed on cryostat sections of frozen fixed tissues. Lungs were fixed in PBS containing 0.2% glutaraldehyde, 5 mM EGTA (pH 7.3), and 0.1 M MgCl₂ for 3 h on ice and rinsed three times for 15 min with cold PBS. Fixed tissues were cryoprotected overnight in cold 30% sucrose in 1x PBS and then embedded in OCT embedding medium and snap-frozen on dry ice. Cryostat sections of frozen tissue (10 µm) were loaded onto polylysine-coated slides and allowed to dry at room temperature for 2 h and incubated in ß-Gal solution (5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 1 mg of X-Gal [5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside]) in buffer (100 mM sodium phosphate [pH 7.3], 2 mM MgCl₂, 0.01% sodium deoxycholate, 0.02% NP-40) at room temperature for 6 h. Stained sections were washed three times in PBS and then postfixed in distilled water containing 2% glutaraldehyde, 2% PFA, and 0.1 M sodium cacodylate for 1 h at 4°C. The slides were rinsed in three changes of cold PBS for 10 min each, cleared in distilled water for 5 min, and then counterstained with 0.1% nuclear fast red in 5% aluminum sulfate for 2 min, washed in running water, dehydrated, and topped with coverslips for analysis.

For whole-mount staining fixed lungs were washed and immediately incubated in ß-Gal solution at room temperature overnight without cryoprotection. Stained lung lobes were washed three times with PBS and postfixed as described above, and washed left lobes were photographed through a dissecting microscope.

Brain sectioning and staining. Brains were perfused with 4% PFA, fixed overnight in 4% PFA, and stored in PBS until use. Brains were blocked in 3% agar and cut at 40 to 50 µm on a vibratome (Leica, Deerfield, Ill.). The sections were counterstained with 2% Mayer's hematoxylin (Sigma Chemical Co.) for 6 min, or alternatively with 1% thionin, rinsed, dehydrated, and mounted for analysis.

Immunohistochemistry. All procedures were performed at room temperature. Brain sections were washed three times in PBS and blocked in a solution of 2% (vol/vol) serum and 0.2% (vol/vol) Triton X-100 (Sigma) in PBS for 2 h. Normal goat serum (S-1000; Vector Laboratories, Burlingame, Calif.) or normal donkey serum (017-000-121; Jackson ImmunoResearch Laboratories, West Grove, Pa.) was used as the blocking agent. The sections were incubated with the primary antibody, rabbit anti-glial fibrillary acidic protein (GFAP; 1:30,000; Z0334; Dako, Glostrup, Denmark) or rat anti-L1 (1:5,000; MAB5272; Chemicon, Temecula, Calif.), overnight. Sections were then washed three times in PBS and incubated with the biotinylated secondary antibody (biotinylated goat anti-rabbit secondary antibody [1:500; Vector Laboratories] or biotinylated donkey anti-rat secondary antibody [1:500; Jackson ImmunoResearch Laboratories]) for 2 h. After three washes in PBS, sections were incubated in avidin-biotin solution (1:500; Vector Laboratories) for 1 h, followed by three washes in PBS. Sections were then immersed in a nickel-3,3'-diaminobenzidine (DAB; D-5905; Sigma)-chromogen solution (2.5% nickel sulfate and 0.02% DAB in 0.175 M sodium acetate) activated with 0.01% (vol/vol) H₂O₂ until a dark purple-black precipitate formed. Sections were washed in PBS, mounted, and placed on coverslips in DPX mounting medium (Electron Microscopy Services). Labeling was analyzed with an upright light microscope (Leica). Images were made with a PowerPhase digital camera (PhaseOne, Copenhagen, Denmark) directly into Adobe Photoshop software.

Immunohistochemistry for smooth muscle actin (SMA) was performed essentially as described previously (41). The smooth muscle differentiation marker SMA was detected with mouse monoclonal anti-SMA 1A4 (1:20,000; Sigma). Staining was performed with the Mouse on Mouse kit (Vector Laboratories). Biotinylated secondary antibodies were used with a nickel-enhanced DAB protocol essentially as described for the brain except that slides were counterstained with nuclear fast red.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

 
Nfib gene targeting and confirmation of Nfib null allele. We constructed a replacement-type targeting vector by replacing the most 3' 523 bp of exon 2 and the first 177 bp of intron 2 with a translational-fusion ß-Gal reporter gene and a neomycin resistance gene (lacZ and Neo, respectively, in Fig. 1A). Since exon 2 encodes both the DNA-binding and dimerization activities of NFI, homologous recombination of the native gene and the targeting vector should generate a null allele. SacI digests of targeted ES cell genomic DNA analyzed with a DNA probe that maps external to the targeted region generated 8.2-kb targeted-allele and 22.7-kb native-allele fragments (Fig. 1A) in heterozygous targeted ES cells. A Southern blot screen of 80 G418-resistant ES clones identified 6 clones harboring a single targeted mutation, and a composite of the results of the screen is shown (Fig. 1B). Further Southern blot analyses using Neo- and lacZ-specific probes showed that Nfib-targeted clones harbored no other vector integrations. PCR analyses confirmed homologous recombination of the 3' flanking region (data not shown).

View larger version (39K): [in this window] [in a new window]   FIG. 1. Disruption of Nfib gene and generation of null allele. (A) Nfib targeting vector (top) generated by replacement of most of exon 2 with the coding sequence for an NLS-ß-Gal translational fusion protein and a PGK-neo selection cassette. Correct targeting of the Nfib locus causes loss of a 22.68-kb SacI fragment and gain of an 8-kb SacI fragment (arrows) on screening with the indicated probe. (B) Detection of targeted ES clones by Southern blotting. Genomic DNA was digested with SacI, analyzed on a 0.8% agarose gel, transferred to a nylon membrane, and hybridized with the 32P-labeled probe fragment indicated in panel A. The WT allele yields an 23-kb band, while the targeted allele shows an 8-kb band. Lanes 1, 2, and 6 contain both WT and targeted alleles. (C) Loss of full-length Nfib transcripts from brains, livers, and lungs of Nfib–/– mice. RNA from the indicated tissues of WT, Nfib–/+, and Nfib–/– mice was isolated and reverse transcribed, and the cDNA was analyzed by PCR with the exon 1 and exon 4 primers described in Materials and Methods. PCR products were analyzed on 2.0% agarose gels, and three products were detected, a major 684-bp band containing exons 1 to 4 seen in cDNA from WT and Nfib–/+ mice and a much weaker 152-bp band lacking exon 2 seen in cDNA from Nfib–/+ and Nfib–/– mice. Another low-abundance product, migrating at approximately 450 bp, was generated from WT and Nfib–/+ cDNA but not from Nfib–/– cDNA. This product may represent aberrantly spliced Nfib transcripts. Lane Mr, molecular weight markers.

Lung maturation defects in Nfib-deficient mice. A previous study showed that mice with an insertion of a lacZ translational fusion gene into the same EcoRI site in exon 2 of Nfib targeted here exhibited defects in lung maturation (22). We therefore examined lung development in our Nfib-null mice. Mouse lung maturation occurs relatively late in gestation, with final prenatal maturation (sacculation) occurring from E16.5 to P3 (10). Histological examination of lungs in E17.5 Nfib-deficient mice showed that lungs from Nfib^–/– mice were severely defective in maturation, while those from Nfib^–/+ animals had a clear developmental delay in maturation. WT lungs at E17.5 were well sacculated (Fig. 2A), while lungs from Nfib^–/+ mice had smaller saccules (Fig. 2B) and lungs of Nfib^–/– mice completely lacked saccules (Fig. 2C). The morphology of the E17.5 Nfib^–/– lungs resembled that of less-mature WT E15.5 lungs (Fig. 2, compare panels C and D). To assess the approximate time in development at which maturation was arrested, we examined the morphology of E15.5 WT, Nfib^–/+, and Nfib^–/– lungs (Fig. 2D to F, respectively). Unlike the case at E17.5, at E15.5 the WT and mutant lungs had similar morphologies with dense mesenchyme and no saccules. In addition, levels of smooth muscle actin expression around the bronchioles (Fig. 2D to F) and vessels (Fig. 2D and F) in WT and mutant mice were similar. These data indicate that lung development in Nfib^–/– mice appears arrested between E15.5 and E16.5 at the late-pseudoglandular-to-early-canalicular stage of development and that Nfib^–/+ mice have a less severe phenotype of delayed lung maturation.

View larger version (93K): [in this window] [in a new window]   FIG. 2. Decreased lung maturation in Nfib–/+ and Nfib–/– mice. Transverse sections of E17.5 lungs (A to C) were stained with H&E, and E15.5 lungs (D to F) were stained with a monoclonal antibody against a smooth muscle actin (SMA) and counterstained with nuclear fast red. E17.5 Nfib+/+ lungs (A) show normal sacculation (open areas that produce a "Swiss cheese" appearance), while Nfib–/+ lungs (B) have reduced sacculation and Nfib–/– lungs (C) show essentially no sacculation. E15.5 lungs of all three genotypes (Nfib+/+ [D], Nfib–/+ [E], and Nfib–/– [F]) showed similar cellular morphologies and SMA staining in smooth muscle surrounding bronchioles (arrows) and vessels (arrowheads). Mice were in the Black Swiss (A to C) and FVB/N (D to F) backgrounds.

View larger version (75K): [in this window] [in a new window]   FIG. 3. Gross morphological changes in Nfib–/– lungs. Dissected left lobes are shown unstained at E15.5 (A), whole mount stained for ß-Gal at E18.5 (B; ßgal), and H&E stained in transverse section at E18.5 (B; H&E). Although there is no gross change in overall size and shape, minor surface indentations seen in E15.5 lungs appear more striking as deep clefts or ruffles at E18.5. lacZ expression is detected in Nfib–/– lungs but not Nfib+/+ lungs (B). Arrows, surface indentations and clefts.

Nfib::lacZ expression during lung development. Lung development requires many reciprocal epithelial-mesenchymal inductions (see references 25 and 54 for reviews). While our previous in situ data showed that Nfib is highly expressed in the developing lung (6), the cellular expression pattern was not determined. To identify the specific cell compartments where Nfib is expressed during lung development, we stained lungs of mice heterozygous for the Nfib::lacZ allele for ß-Gal activity. Nfib::lacZ was uniformly expressed at high levels throughout the mesenchyme at E14.5, whereas weaker expression was detected in a few scattered epithelial cells at this time (Fig. 4A). Nfib::lacZ was expressed at similar levels in most epithelial and mesenchymal cells at E16.5, with increased expression in smooth muscle (Fig. 4B). At E18.5, Nfib::lacZ was expressed in a subset of both proximal and distal epithelial cells, while mesenchymal expression was decreased but still present in the smooth muscle (Fig. 4C and D). LacZ staining was not detected in WT E18.5 lung (Fig. 4E). In adult lung Nfib::lacZ was expressed primarily in the bronchiolar epithelium and type II cells (Fig. 4F). Preliminary immunohistochemistry with Nfib-specific antibodies supports this pattern for endogenous Nfib in WT mice (C. Bachurski, unpublished data). These data indicate that Nfib could play essential roles in either the lung mesenchyme or epithelium during late prenatal lung maturation.

View larger version (87K): [in this window] [in a new window]   FIG. 4. Expression of ß-Gal from the Nfib-lacZ allele in Nfib–/+ lungs. Frozen sections of lungs of E14.5 (A), E16.5 (B), E18.5 (C and D), and adult Nfib–/+ (F) or E18.5 Nfib+/+ (E) mice were stained for ß-Gal activity. ß-Gal expression was predominantly mesenchymal at E14.5. Smooth muscle was strongly positive from E16.5 to E18.5, with reduced staining in adult lung (arrows). Expression decreased in the mesenchyme and increased in both proximal and distal epithelial cells from E16.5 through E18.5. ß-Gal expression was mainly in a subset of the bronchiolar epithelium and type II cells in adult lung. Br, bronchiole; V, vessel; arrows, smooth muscle; arrowheads, epithelium. The scale bar (E) applies to all panels. Mice were in the FVB/N background.

View larger version (96K): [in this window] [in a new window]   FIG. 5. Agenesis of the corpus callosum in Nfib–/– mice. Coronal 50-µm vibratome sections from E17.5 Nfib brains on either a Black Swiss (A to C) or C57BL/6 (D to F) background were mounted and stained with hematoxylin (A to F) or stained free-floating with an antibody against L1 CAM (G to I). Arrows in each panel indicate the presence or absence of the corpus callosum; arrowheads (G to I) indicate the presence of the fornix. Probst bundles were evident in Nfib–/+ mice on a mixed Black Swiss background (B). Nfib–/– mice displayed enlarged ventricles (C and F; asterisks). Scale bars: 2 mm (F; applies to panels A to F) and 1 mm (I; applies to panels G to I).

Midline glial populations are important for the formation of the corpus callosum, as they secrete guidance factors, such as Slit2, that guide callosal axons both before and after they cross the midline (58, 59). Figure 6 shows an immunohistochemical analysis of GFAP expression. Three midline glial populations are evident in WT animals: the glial wedge (GW), glia within the indusium griseum (IGG), and the midline zipper glia (MZG; Fig. 6D) (57). Nfib^–/– mice showed a reduction in GFAP expression at the midline, which likely reflects a loss or reduction in these glial populations. This phenotype is similar to that observed in Nfia^–/– mice (56). Nfib^–/– mice display a complete absence of GFAP labeling in the regions of both the IGG and the MZG (Fig. 6F and L). Furthermore, although a few glial cells were present, the GW was greatly reduced in the Nfib^–/– animals (Fig. 6F and L). Since there are NFI binding sites in the GFAP gene promoter, these results, together with our results for Nfia^–/– mice (56), suggest that NFI genes may regulate both GFAP expression and glial development. Further evidence for such regulation is that Nfib^–/+ mice on a C57BL/6 background display lower levels of expression of GFAP than their Nfib^+/+ littermates (compare Fig. 6K and J). Although all three midline glial populations were present in Nfib^–/+animals on both the Black Swiss and C57BL/6 background, the IGG appeared disorganized on the Black Swiss background, probably due to the formation of Probst bundles at the midline (Fig. 6E).

View larger version (137K): [in this window] [in a new window]   FIG. 6. Midline glial defects in Nfib–/– mice. Coronal sections were labeled with an antibody against GFAP to label the midline glial populations. IGG (D, E, J, and K; arrowheads) and MZG (D, E, J, and K; thick arrows) were absent in Nfib–/– mice on both the Black Swiss (F) and C57BL/6 (L) backgrounds. The GW was severely reduced in the Nfib–/– mice (F and L; thin arrows). Panels D to F and J to L are higher-power views of panels A to C and G to I, respectively. Scale bars, 2 mm (I; applies to panels A to C and G to I) and 200 µm (L; applies to panels D to F and J to L).

View larger version (129K): [in this window] [in a new window]   FIG. 7. Development of the hippocampus is severely disrupted in Nfib–/– mice. Coronal 50-µm vibratome sections from E17.5 (A to F) and E15.5 (G to I) embryos were mounted and stained with hematoxylin to reveal the structural morphology of the hippocampus. Gross regionalization of the hippocampus is evident in panel A. Note that the CA2 subfield has not yet developed by E17.5. Nfib–/– mice have a complete absence of the dentate gyrus (DG; asterisks [C and F] denote the region where the DG should have developed) and a reduction in the size of the fimbria (Fi; A and C). (A to C) Sections of brains from mice on a mixed Black Swiss background; (D to I) brains from mice on a C57BL/6 background. Scale bars, 200 µm (C; applies to panels A to F) and 50 µm (I; applies to panels G to I).

View larger version (50K): [in this window] [in a new window]   FIG. 8. Absence of the basilar pons in Nfib–/– mice. Sagittal sections of E18.5 Nfib+/+ (A), Nfib–/+ (B), and Nfib–/– (C) hindbrains stained with thionin. Brains were from litters backcrossed for five generations to C57BL/6. The basilar pons (BP) is present in panels A and B but not panel C (arrowhead). The rest of the hindbrain, including the inferior olive (IO), appears normal in the mutant, and the cerebellum (Cb) is present but appears unfoliated in panel C. Scale bar (A), 0.5 mm (applies to all panels).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Lung maturation defects in Nfib^–/– animals. Lung maturation defects that are morphologically similar to those seen in our Nfib^–/– animals were described previously for an insertion mutation in the Nfib locus (22). While that study described the defect as lung hypoplasia, our analysis shows that there is an apparent increase in the amount of DNA in Nfib^–/– lungs (Table 3), suggesting aberrant cell proliferation and/or apoptosis during lung maturation. Increased PCNA expression at E18.5 in Nfib^–/– versus WT lungs is also consistent with increased cell proliferation (Table 4). Prenatal lung maturation proceeds through a series of morphologically and biochemically defined stages (10) from budding (approximately E9.5), initial branching morphogenesis and expansion (pseudoglandular, approximately E9.5 to E16.5), continued branching of the distal epithelium and mesenchyme (canalicular; approximately E16.5 to E17.5), and formation of terminal saccules with terminal differentiation of type I and type II epithelial cells (sacculation; approximately E17.5 to P5). Final maturation of the terminal sacs into alveolar ducts and sacs (alveolization) occurs postnatally from approximately P5 to P30 (47, 65). The reduction of specific epithelial markers seen here (Table 4), together with the lack of saccule formation, indicates that the lungs in Nfib^–/– mice are arrested at the late-pseudoglandular or early-canalicular stage of maturation. Since the morbidity and mortality of premature infants are strongly associated with a failure of lung maturation (45), these mice may provide a useful model to investigate the later stages of prenatal lung maturation.

Failure of late fetal lung maturation has also been seen in mice containing targeted disruptions of the genes encoding glucocorticoid receptor (9), transforming growth factor ß3 (55), endothelial nitric oxide synthase (24), and Sp3 (5) and other genes (66, 68, 69). The pathways linking this diverse set of molecules to the developmental events essential for lung maturation have not yet been determined. Given the possible increase in cell number seen in the Nfib^–/– lungs (Table 3), it is of interest that the combined loss of p21CIP and p57KIP2, two Cdk inhibitors that mediate cell cycle arrest, results in a maturation arrest that appears morphologically similar to that seen with loss of Nfib (69). Thus it is possible that Nfib is involved in either mesenchymal or epithelial cell exit from the cell cycle and that this cell cycle arrest is essential for normal maturation of the lung. Consistent with this model are the observations that overexpression of NFIB and other NFI proteins can suppress the oncogenic transformation of chicken embryo fibroblasts mediated by a number of nuclear oncogenes (52) and that rearrangement or altered expression or both of NFIB have been associated with the overgrowth syndromes of pleomorphic adenoma (17) and polycythemia vera (29). To test this model, it will be useful in future studies to determine whether the apparent increase in cell number in the Nfib^–/– lungs is due to changes in cell proliferation, apoptosis, or both.

Since Nfib^–/– mice have defects in both lung maturation and commissural tract formation (Fig. 2 to 8), it is noteworthy that loss of the known axon guidance receptor gene Robo1 also causes perinatal lethality attributed to a failure of lung maturation (66). Robo1 is a mouse ortholog of Roundabout, which is essential for normal midline axonal crossing in Drosophila (26). The Robo gene family encodes receptors with homology to the neural CAM family of cell surface adhesion molecules and appears to regulate axon guidance in both Drosophila and mice (49). This gene family has been implicated in regulating axon guidance in mice and humans (27, 37, 63), but its role in lung development is unknown (1). The expression patterns for Robo and Slit family members during lung maturation are consistent with a role for Robo signaling in epithelium-mesenchyme interactions (19). Although we observed no decrease in overall Robo1 expression in Nfib^–/– brains (Table 4), analysis of the expression patterns of Robo1 and other signaling molecules by in situ hybridization in Nfib^–/– animals might reveal regional changes in expression related to the observed phenotypes in brain and lung.

It will be important to determine the specific gene expression changes in Nfib^–/– lungs that affect maturation. However, given the failure of lung maturation in the Nfib^–/– mice, it will be essential to distinguish between changes that directly affect the maturation process and those that simply reflect the immature state of the Nfib^–/– lungs. For example, while we've seen decreases in the expression of a number of type II and type I epithelial markers in E18.5 Nfib^–/– lungs (Table 4), it is possible these changes merely reflect lack of maturity of the Nfib^–/– versus Nfib^+/+ lungs. Since the expression levels of several of these genes normally increase during lung maturation, it is not surprising that their expression levels are reduced in the less mature Nfib^–/– lungs. Thus it will be important to determine the earliest time in development at which differences in gene expression between Nfib^–/– and Nfib^+/+ lungs occur and which of the genes affected at this time are essential for maturation.

One known target for NFI genes in the developing lung epithelium is SP-C. Previous studies showed that the SP-C gene promoter is activated by NFI proteins, including NFI-B2, and that a transgene-encoded dominant repressor (NFI-engrailed) inhibits the expression of SP-C in mouse lung (2, 3). Thus SP-C is a good candidate for a direct target of Nfib in the developing lung. However, SP-C is highly expressed only late in lung maturation and SP-C mutant mice do not exhibit a failure in lung maturation (18). Given the expression pattern of the Nfib::lacZ allele (Fig. 4), it is possible that Nfib regulates both mesenchymal target genes that are essential for the maturation process early in lung development and also epithelial genes that are expressed later in lung development and that are important for physiological functioning of the lung. An important future goal is to distinguish between these two potential classes of Nfib target genes.

Neurological defects in Nfib^–/– animals. While the callosal agenesis (Fig. 5 and 6) and loss of midline GFAP staining (Fig. 6) seen in Nfib^–/– animals is reminiscent of the phenotype seen previously in Nfia^–/– mice (11, 56), there are clear morphological differences between the two mutants. For example, Nfib^–/– animals usually lack large Probst bundles, regions of axonal overlap and swirling near the midline that are often present in Nfia^–/– animals (56). Such differences imply that, while both genes are essential for callosal formation, they may have both shared and distinct functions in this developmental process. Both Nfia and Nfib are expressed in the cortex (6, 56) (data not shown), but the lack of markers specific for callosally projecting neurons makes it difficult to assess the significance of this cortical expression. The similar losses of midline glial populations in Nfib^–/– mice (Fig. 6) and Nfia^–/– mice (56) suggest that both genes may cooperate in the growth and differentiation of these midline glial cells. Given the relatively restricted set of glial cells affected by loss of either Nfia or Nfib, it will be of interest to determine whether simultaneous loss of both Nfi genes produces a more severe phenotype in glial cell development.

While both Nfia- and Nfib-deficient mice exhibit agenesis of the corpus callosum, the Nfib-deficient mice have additional neurological defects not reported for Nfia-deficient animals, including severe defects in hippocampus (Fig. 7) and pons (Fig. 8) formation. Nfib had been shown previously to be expressed in both the hippocampus and the pons (6, 16), and thus it is possible that the defects in these regions are due to cell-autonomous defects in cells that normally express Nfib. Nfib is one of a small class of genes known to affect hippocampus development, a class which includes the genes Lhx-5 (70), Beta2/NeuroD (34), Lef-1 (15), CXCR4 (38), LRP6 (71), and Emx-2 (46, 51) and the Reelin-Disabled-1-beta(1)-class integrin pathway (14). The mechanisms by which these genes affect the molecular pathways and cellular processes essential for hippocampal development are beginning to be understood. LRP6-deficient mice have generalized defects in the Wnt/beta catenin signaling pathway because of the crucial function of LRP6 as a Wnt signaling coreceptor (71). These mice have severe defects in the dentate gyrus due to the decreased production of dentate granule cells. They also have abnormalities of the radial glial scaffolding in the forming dentate gyrus. By examining the hippocampal primordium at early stages, the authors showed a reduction in the number of dentate granule cell progenitors in the dentate ventricular zone prior to the emigration of the earliest granule neurons and precursors to form the dentate anlage. It is possible that loss of Nfib-regulated gene transcription may lead to a similar alteration of the hippocampal primordium, with resultant severe hypoplasia of the dentate gyrus. A more detailed analysis of early hippocampus development in the Nfib^–/– mice should allow us to test this model for Nfib function in the dentate gyrus.

The molecular mechanisms leading to the loss of the basilar pons in Nfib^–/– mice are unknown. It may be that Nfib regulates one or more signaling molecules, such as DCC or Rig-1/Robo3, that control the migration of pontine neurons to the midline (39). Alternatively, Nfib may interact with other transcription factors that regulate pontine formation (4, 13, 33). Distinguishing between these and other possibilities awaits characterization of the fate of pontine neurons in Nfib^–/– mice. Since analysis of whole-brain RNA has failed to find changes in the expression of some molecules known to influence axon guidance and cell migration (Table 5), determining whether there are common Nfib target genes that mediate the defects in callosal formation, hippocampal development, and pons formation will require a detailed analysis of Nfib target genes in each of these regions of the brain.

Given the severity of the neurological phenotypes seen in the Nfib^–/– animals, it is perhaps surprising that a previous study of an insertion of a lacZ translational fusion gene into the Nfib locus found no neurological defects (22). This is despite the finding that the lung maturation defects seen in our Nfib KO animals appear morphologically similar to those seen in animals with the Nfib insertion mutation. There are several possible explanations for such differences. (i) With both our Nfia^–/– and Nfic^–/– mice we showed that deletion of exon 2 results in aberrant splicing from the first to the third exons of the genes (11, 61). We show a similar aberrant transcript from our disrupted Nfib locus (Fig. 1C). These aberrant transcripts do not encode functional proteins as they are missing DNA binding and dimerization domains and as the exons are out of frame, yielding short missense peptides. It is possible that aberrant splicing into the still-present second exon of the insertion mutant gene could generate a hypomorphic, but not null, allele in these animals. In our Nfib mutant allele the second exon is deleted, precluding this possibility. (ii) We showed previously that Nfib has an alternative first exon, expressed specifically in the brain (21, 42). Splicing from this brain-specific exon into the still present second exon of the insertion mutant gene may be why no brain phenotype for the insertion mutant was seen. Also, the report for the insertion mutant failed to show any ß-Gal activity from the targeted allele or to demonstrate loss of Nfib expression, making it difficult to compare to the present study. While further studies are needed to resolve these differences, it is clear that our Nfib^–/–mice have severe lung and brain phenotypes, making them an important tool with which to study Nfib function.

NFI gene family in mouse development. Three of the four mouse NFI genes have now been disrupted, and each mutant has unique defects in development. Disruption of Nfia results in callosal agenesis, perinatal lethality, hydrocephalus, reduction in the hippocampal commissure, and loss of specific midline glial populations (11, 56). Some of these defects are also present in the Nfib^–/– mice described here. In contrast, disruption of the Nfic gene results in postnatal defects in tooth formation, with no observable defects in neuroanatomy and no lethality if the diet is composed of soft food (61). Since the Nfib^–/– mice do not survive postnatally, it is not possible to tell if they have tooth defects similar to those seen in Nfic^–/– mice. Since the four NFI genes encode proteins with highly homologous DNA-binding domains with similar DNA-binding specificities (40), it is possible that there are common downstream targets for all four NFI genes. It is unclear whether the diverse defects seen upon loss of the individual NFI genes are due to differences in the expression patterns of the four genes, differences in downstream target specificity, differences in gene activation or repression by specific NFI isoforms, or other factors. One feature that is common in the mutants is that the defects are seen relatively late in fetal or postnatal development. In Nfia^–/– and Nfib^–/– mice the first morphological indication of differences is at E14 to E15, while for Nfic^–/– mice the first changes are seen at approximately P10. Thus at least three of the four NFI genes are essential for relatively late developmental processes in the mouse. Since the four NFI genes are expressed in an overlapping pattern during development (6) and since changes in NFI transcript levels are first detected at the four-cell stage in embryogenesis (23), it will be of interest to determine whether multiple NFI genes cooperate at earlier stages of mouse development.

It is difficult to connect the diverse defects seen in the Nfib^–/– mice, callosal agenesis, aberrant gliogenesis, hippocampus and pons development, and failure of lung maturation, into a simple model of Nfib function. It will be necessary to determine the earliest stages at which gene expression changes occur in the affected tissues in order to determine whether common or unique Nfib target genes are affected in each organ. Since different NFI proteins have been found to either activate or repress expression from the same promoter depending on the cellular context (7, 8; reviewed in references 21 and 42), it will be important to analyze mice with mutations in multiple NFI genes to determine whether they cooperate or antagonize each other at earlier developmental stages. The future identification of direct downstream targets of Nfib, together with analysis of the effects of the combined loss of Nfib together with other genes known to affect brain and lung formation, should allow us to determine the molecular mechanisms through which Nfib functions in lung and brain development.

	ACKNOWLEDGMENTS

 
We thank Valerie Stewart of the Cleveland Clinic Transgenic/Knockout facility for blastocyst injections and the generation of founder chimeras and Clemencia Colmenares (Cleveland Clinic) for advise and assistance in ES cell culture. We also thank Kimberly Valentino and Rebecca Pacifico (UMB) for excellent technical assistance and Christine E. Campbell (SUNY-Buffalo) for reading the manuscript and helpful discussions.

This work was supported in part by Public Health Service grants HD34901 from the National Institute of Child Health and Development and DK58401 and DK48796 from the National Institute of Diabetes and Digestive and Kidney Diseases to R.M.G. and HL60907 from the National Heart, Lung and Blood Institute to C.J.B.

	FOOTNOTES

 
* Corresponding author. Mailing address: State University of New York at Buffalo, Dept. of Biochemistry, 140 Farber Hall, 3435 Main St., Buffalo, NY 14214-3000. Phone: (716) 829-3471. Fax: (716) 829-2725. E-mail: rgron{at}buffalo.edu.

Present address: Buffalo, NY 14215.

	REFERENCES

Top Abstract Introduction Materials and Methods Results Discussion References

 
1. Anselmo, M. A., S. Dalvin, P. Prodhan, K. Komatsuzaki, J. T. Aidlen, J. J. Schnitzer, J. Y. Wu, and T. B. Kinane. 2003. Slit and robo: expression patterns in lung development. Gene Expr. Patterns 3:13-19.[CrossRef][Medline]

2. Bachurski, C. J., S. E. Kelly, S. W. Glasser, and T. A. Currier. 1997. Nuclear factor I family members regulate the transcription of surfactant protein-C. J. Biol. Chem. 272:32759-32766.[Abstract/Free Full Text]

3. Bachurski, C. J., G. H. Yang, T. A. Currier, R. M. Gronostajski, and D. Hong. 2003. Nuclear factor I/thyroid transcription factor 1 interactions modulate surfactant protein C transcription. Mol. Cell. Biol. 23:9014-9024.[Abstract/Free Full Text]

4. Ben-Arie, N., B. A. Hassan, N. A. Bermingham, D. M. Malicki, D. Armstrong, M. Matzuk, H. J. Bellen, and H. Y. Zoghbi. 2000. Functional conservation of atonal and Math1 in the CNS and PNS. Development 127:1039-1048.[Abstract]

5. Bouwman, P., H. Gollner, H. P. Elsasser, G. Eckhoff, A. Karis, F. Grosveld, S. Philipsen, and G. Suske. 2000. Transcription factor Sp3 is essential for post-natal survival and late tooth development. EMBO J. 19:655-661.[CrossRef][Medline]

6. Chaudhry, A. Z., G. E. Lyons, and R. M. Gronostajski. 1997. Expression patterns of the four nuclear factor I genes during mouse embryogenesis indicate a potential role in development. Dev. Dyn. 208:313-325.[CrossRef][Medline]

7. Chaudhry, A. Z., A. D. Vitullo, and R. M. Gronostajski. 1998. Nuclear factor I (NFI) isoforms differentially activate simple versus complex NFI-responsive promoters. J. Biol. Chem. 273:18538-18546.[Abstract/Free Full Text]

8. Chaudhry, A. Z., A. D. Vitullo, and R. M. Gronostajski. 1999. Nuclear factor I-mediated repression of the mouse mammary tumor virus promoter is abrogated by the coactivators p300/CBP and SRC-1. J. Biol. Chem. 274:7072-7081.[Abstract/Free Full Text]

9. Cole, T. J., N. M. Solomon, R. Van Driel, J. A. Monk, D. Bird, S. J. Richardson, R. J. Dilley, and S. B. Hooper. 2004. Altered epithelial cell proportions in the fetal lung of glucocorticoid receptor null mice. Am. J. Respir. Cell Mol. Biol. 30:613-619.[Abstract/Free Full Text]

10. Costa, R. H., V. V. Kalinichenko, and L. Lim. 2001. Transcription factors in mouse lung development and function. Am. J. Physiol. Lung Cell. Mol. Physiol. 280:L823-L838.[Abstract/Free Full Text]

11. das Neves, L., C. Duchala, F. Godinho, M. Haxhiu, C. Colmenares, W. Macklin, C. Campbell, K. Butz, and R. Gronostajski. 1999. Disruption of the murine nuclear factor I-A gene (Nfia) results in perinatal lethality, hydrocephalus and agenesis of the corpus callosum. Proc. Natl. Acad. Sci. USA 96:11946-11951.[Abstract/Free Full Text]

12. de Jong, R. N., and P. C. van der Vliet. 1999. Mechanism of DNA replication in eukaryotic cells: cellular host factors stimulating adenovirus DNA replication. Gene 236:1-12.[CrossRef][Medline]

13. Engelkamp, D., P. Rashbass, A. Seawright, and V. van Heyningen. 1999. Role of Pax6 in development of the cerebellar system. Development 126:3585-3596.[Abstract]

14. Forster, E., A. Tielsch, B. Saum, K. H. Weiss, C. Johanssen, D. Graus-Porta, U. Muller, and M. Frotscher. 2002. Reelin, Disabled 1, and beta 1 integrins are required for the formation of the radial glial scaffold in the hippocampus. Proc. Natl. Acad. Sci. USA 99:13178-13183.[Abstract/Free Full Text]

15. Galceran, J., E. M. Miyashita-Lin, E. Devaney, J. L. Rubenstein, and R. Grosschedl. 2000. Hippocampus development and generation of dentate gyrus granule cells is regulated by LEF1. Development 127:469-482.[Abstract]

16. Gesemann, M., E. D. Litwack, K. T. Yee, U. Christen, and D. D. O'Leary. 2001. Identification of candidate genes for controlling development of the basilar pons by differential display PCR. Mol. Cell. Neurosci. 18:1-12.[CrossRef][Medline]

17. Geurts, J. M., E. F. Schoenmakers, E. Roijer, A. K. Astrom, G. Stenman, and W. J. van de Ven. 1998. Identification of NFIB as recurrent translocation partner gene of HMGIC in pleomorphic adenomas. Oncogene 16:865-872.[CrossRef][Medline]

18. Glasser, S. W., M. S. Burhans, T. R. Korfhagen, C. L. Na, P. D. Sly, G. F. Ross, M. Ikegami, and J. A. Whitsett. 2001. Altered stability of pulmonary surfactant in SP-C-deficient mice. Proc. Natl. Acad. Sci. USA 98:6366-6371.[Abstract/Free Full Text]

19. Greenberg, J. M., F. Y. Thompson, S. K. Brooks, J. M. Shannon, and A. L. Akeson. 2004. Slit and robo expression in the developing mouse lung. Dev. Dyn. 230:350-360.[CrossRef][Medline]

20. Gronostajski, R. M. 1986. Analysis of nuclear factor I binding to DNA using degenerate oligonucleotides. Nucleic Acids Res. 14:9117-9132.[Abstract/Free Full Text]

21. Gronostajski, R. M. 2000. Roles of the NFI/CTF gene family in transcription and development. Gene 249:31-45.[CrossRef][Medli