Two Promoters Mediate Transcription from the Human LHX3 Gene: Involvement of Nuclear Factor I and Specificity Protein 1
http://www.100md.com
《内分泌学杂志》
Department of Cellular and Integrative Physiology (S.J.R.), Indiana University School of Medicine, Indianapolis, Indiana 46202
Department of Biology (B.C.Y., M.G.), Indiana University-Purdue University Indianapolis, Indianapolis, Indiana 46202
U.S. Department of Agriculture/Agricultural Research Service (T.P.L.S.), U.S. Meat Animal Research Center, Clay Center, Nebraska 68933
Abstract
The LHX3 transcription factor is required for pituitary and nervous system development in mammals. Mutations in the human gene are associated with hormone-deficiency diseases. The gene generates two mRNAs, hLHX3a and hLHX3b, which encode three proteins with different properties. Here, the cis elements and trans-acting factors that regulate the basal transcription of the two mRNAs are characterized. A comparative approach was taken featuring analysis of seven mammalian Lhx3 genes, with a focus on the human gene. Two conserved, TATA-less, GC-rich promoters that are used to transcribe the mRNAs precede exons 1a and 1b of hLHX3. Transcription start sites were mapped for both promoters. Deletion experiments showed most activity for reporter genes containing the basal promoters in the context of –2.0 kb of hLHX3a and 1.8 kb of intron 1a (hLHX3b). Transfection, site-directed mutation, electrophoretic mobility shift, Southwestern blot, and chromatin immunoprecipitation approaches were used to characterize the interaction of transcription factors with conserved elements in the promoters. Specificity protein 1 is a regulator of both promoters through interaction with GC boxes. In addition, a distal element within intron 1a that is recognized by nuclear factor I is critical for hLHX3b promoter function. We conclude that dual promoters allow regulated production of two hLHX3 mRNAs.
Introduction
THE Lhx3/P-LIM/LIM3 GENE encodes LIM-homeodomain transcription factors with essential roles in neuroendocrine development (reviewed in Ref.1). During anterior pituitary development, Lhx3 participates with other regulatory genes, such as Sf1, Prop1, Pit1, Egr1, Pitx1, Pitx2, and Tpit, to guide the specification of differentiated hormone-secreting cells (reviewed in Refs.2, 3, 4). In mammals, the Lhx3 gene is expressed in the embryonic nervous system and in the primordial and mature pituitary gland (5, 6, 7, 8). Mice homozygous for a null Lhx3 allele die after birth and feature incomplete structural development of the pituitary and defective motor neuron specification (9, 10, 11). In these animals, some anterior pituitary corticotrope cells are functional, but expression of the characteristic hormones of the gonadotrope, thyrotrope, somatotrope, and lactotrope cell types is not detectable, indicating that Lhx3 is required for the eventual differentiation of these specialized lineages (9). These observations are consistent with molecular studies demonstrating that LHX3 and related LIM-homeodomain factors can activate anterior pituitary expressed genes, including those encoding GSU, PRL, FSH, TSH, the GnRH receptor, and the Pit-1 transcription factor (e.g. Refs.6 and 12, 13, 14, 15, 16). Furthermore, mutations in the human LHX3 (hLHX3) gene that are predicted to lead to the production of disabled proteins cause recessive genetic diseases featuring combined pituitary hormone deficiency and other symptoms (17, 18, 19).
The hLHX3 gene contains seven coding exons and six introns that span approximately 8.7 kb located within the subtelomeric region of chromosome 9 (Fig. 1A) (20). The mouse Lhx3 gene has a similar organization (21). The single hLHX3 gene produces two major mRNAs known as hLHX3a and hLHX3b (13). Translation from the first methionine codons of the hLHX3a and hLHX3b mRNAs generates the LHX3a and LHX3b protein isoforms (13). These proteins have identical LIM domains, a central DNA-binding homeodomain and a carboxyl terminus that contains the major activation domain but have distinct amino termini resulting from alternate use of 5' exons in the gene (20, 22). A third protein isoform, M2-LHX3, is generated by preferential translation of the second in-frame methionine codon of the hLHX3a mRNA (22). The three LHX3 protein isoforms display different biochemical and functional properties (13, 14, 22, 23).
In this study, we investigated the transcriptional mechanisms by which mammalian Lhx3 genes generate multiple mRNAs that encode proteins with diverse regulatory properties. Phylogenetic comparisons and functional tests were used to map two conserved, TATA-less, GC-rich promoters that guide transcription of the two mRNAs. Two kilobases of the hLHX3a promoter and 1.8 kb of intron 1a (a region that contains the hLHX3b promoter) mediate basal activity. The specificity protein 1 (Sp1) transcription factor binds to proximal GC boxes and is a strong regulator of both promoters. Furthermore, a critical upstream element within intron 1a allows regulation by nuclear factor I (NFI) family transcription factors.
Materials and Methods
Cloning of mammalian Lhx3 gene and cDNA sequences
Fragments of the hLHX3 gene (20) were amplified by touchdown PCR using bacterial artificial chromosome clone RPC11–83N9 (Sanger Centre, Cambridge, UK) or normal human genomic DNA as substrates. Primers were designed based on human genome sequence data accessed through GenBank at the National Center for Biological Information (NCBI). The upstream (hLHX3a promoter) region of the gene was amplified in approximately 500-bp increments using an antisense primer (5'-cctcctaggtcagcgtcccctg-3') and one of the following sense primers: 5'-gtcttgagtcctcagcagtgct-3' (580 bp upstream of exon 1a), 5'-gagactcacaggacaagacccttga-3' (1.096 kb), 5'-agagctggatgccacctttagg-3' (1.581 kb), 5'-tgcttcgtgtctcactcgagag-3' (2.080 kb), 5'-tgcacacacaagcatctcactc-3' (2.701 kb), 5'-cctgctccaggctgccaagtgt-3' (3.240 kb), and 5'-gtagtccggaaagggccagtgt-3' (4.8 kb). For the hLHX3b promoter upstream in intron 1a, three regions were amplified using an antisense primer (5'-cgccactctccagtcccgaacttt-3') and one of the following sense primers: 5'-gcgtgtgctccagctcaggcctct-3' (1804 bp upstream), 5'-ggtaacaagtgctgtgcaaagtga-3' (1267 bp), and 5'-agtgcccgtcagctcttgcacaca-3' (418 kb). PCR was performed with Pfu Ultra polymerase (Stratagene) and MasterAmp PCR optimization buffers (Epicentre, Madison, WI) (if required due to high GC content of the target sequences). To create luciferase reporter genes, fragments of the hLHX3a promoter upstream region or of the hLHX3b promoter/intron 1a region were cloned into the pGL2-basic plasmid (Promega, Madison, WI). All plasmids were confirmed by DNA sequencing (Biochemistry Biotechnology Facility, Indiana University School of Medicine).
To obtain the sequence of the bovine Lhx3 gene, the Trace Archive of the bovine genome project at the NCBI containing raw reads from the first 3-fold genome coverage (12 million reads at the time of screening) was searched via basic local alignment tool nucleotide (BLASTN) using the full-length cDNA sequence of the hLHX3a cDNA. Trace files whose sequence showed highly significant (scores of >300) match to the cDNA, as well as the mate-pair end sequences from the respective clones, were collected in a directory and used to construct initial genomic contigs via phred (24) and phrap (25) algorithms. Contig sequences were masked for repetitive elements using RepeatMasker (Smit, A. F. A., and P. Green, unpublished results; http://ftp.genome.washington.edu/RM/RepeatMasker.html) and used to search for overlapping trace files in the archive, which were added to the directory for reconstruction of contigs. The process was repeated until none of the contigs in the phrap output identified trace files not already in the directory. This resulted in construction of four contigs containing portions with high similarity to exons of the hLHX3 cDNA, leaving three gaps in the gene sequence. Primers then were designed to span the gaps by PCR, and sequence was obtained by amplification of bovine genomic DNA from the same animal used in the whole genome shotgun sequencing. The PCR products were sequenced with the amplification primers, nested primers, or both. The resulting 12,883-bp contig was edited by manual inspection using the Consed viewing program (25), and areas of low sequence quality or areas where read overlap was exclusively from low-complexity sequence were targeted for finishing using additional PCR-based amplification and sequencing. To obtain confirming bovine Lhx3 cDNA sequence, primers were designed based on the cDNA sequence predicted from the first set of genomic trace files obtained. Primers 5'-gagatcccgctgtgtgcc-3' and 5'-ccttgcagtacacgctctcc-3' were designed from the putative exon 2 sequence and used to obtain a full coding sequence bovine Lhx3b cDNA clone via iterative screen (26) of a pooled-tissue cDNA library that included pituitary gland [library 1BOV (27)]. The clone obtained had an insert of 2,390 bp, and the complete insert was sequenced. The edited bovine Lhx3 gene and cDNA sequences have been submitted to GenBank with accession nos. AY923832 and AY923833, respectively. Intron 1a sequences of the rhesus monkey and porcine Lhx3 genes were amplified by the PCR. Templates were 150 ng of genomic DNA from adult rhesus monkey (kindly provided by Dr. T. Golus, Wisconsin Primate Research Center, Madison, WI) or adult pig (female Yorkshire cross). Reactions included MasterAmp PCR optimization buffer G (Epicentre), and 5'-atgctgctggaaacggggctcga-3' (monkey exon 1a), 5'-gaatctctcggcgcaggtccgcc-3' (monkey exon 1b) or 5'-atgctgctggaaacggagctggc-3' (pig exon 1a), and 5'-ggatctctcggcgcaggtcctcc-3' (pig exon 1b) primers were used. PCR products from multiple independent reactions were sequenced on both strands. Sequences were submitted to GenBank (accession nos. AY879262 for pig intron 1a and AY879263 for rhesus monkey intron 1a).
RNA ligase-mediated rapid amplification of cDNA ends
The transcription start sites of human and mouse Lhx3 genes were deduced by 5' RNA ligase-mediated rapid amplification of cDNA ends (RLM-RACE) performed using the GeneRacer protocol (Invitrogen, Carlsbad, CA) according to the instructions of the manufacturer, as we have described (28). First-strand cDNA was generated from adult human pituitary gland RNA (13) or from pituitary T3–1 pregonadotrope cell (29) total RNA and primed with 5'-ggcagtcgctgcacttgagacactt-3'. Second-strand cDNA was generated using primers from the manufacturer and the following gene-specific primers: 5'-gagacgcgctcctccgaggtca-3' (5' RACE Lhx3a) and 5'-tctccagtcccgaacttt-3' (5' RACE Lhx3b).
S1 nuclease assays
S1 nuclease assays were performed as described (30). Briefly, 32P end-radiolabeled single-stranded DNA probes for the hLHX3a and hLHX3b promoters were generated. S1 digestion reactions contained 15 μg of T3–1 RNA hybridized with labeled probes. Radiolabeled DNA products were analyzed by electrophoresis through 12% polyacrylamide 8 M urea gels. A 32P end-labeled 1-kb extension ladder (Invitrogen) was used as a molecular marker.
Cell culture and transfection
Human embryonic kidney (HEK) 293T and rodent pituitary cell lines were cultured and transfected as described (14). Typical transfections contained 2 μg of a luciferase reporter gene and 500 ng of an expression vector (if any). Control parallel samples received empty vector DNA. All assay groups were performed in triplicate. Forty-eight hr following transfection, cells were lysed in 25 mM Tris-Cl (pH 7.8), 2 mM dithiothreitol, 1% Triton X-100, 2 mM EDTA (pH 8.0), 10% glycerol. The lysate supernatant was assayed for luciferase activity using a luciferin substrate (Promega) and a Beckman Coulter luminometer (Fullerton, CA). Total cell protein was determined by the Bradford method (Bio-Rad, Hercules, CA), and luciferase activity was normalized to the amount of protein present. Expression vectors included human LHX3a, human LHX3b (13), human PROP1 (31), mouse SF1 (a gift from Dr. Holly Ingraham, University of California, San Francisco, CA), mouse EGR1 (a gift from Dr. Eileen Adamson, Burnham Institute, La Jolla, CA), rat Sp1 (a gift from Michael Wegner, University of Erlangen, Erlangen, Germany).
Cell protein extraction
Nuclear or whole cell protein extracts from HEK 293T and rodent pituitary cells were prepared as we have described (14, 32). Rat pituitary GH3 somatolactotrope nuclear extracts were purchased from Active Motif, Inc. (Carlsbad, CA).
EMSAs
EMSAs were performed as we have described (14) using radiolabeled double-stranded oligonucleotide probes with the results visualized by autoradiography or using a Storm phosphorimager (Amersham Biosciences, Piscataway, NJ). Cell extracts were prepared as described above. Human recombinant Sp1 protein was purchased from Promega. In some experiments, 2 μg of anti-Sp1, anti-NFI, or anti-SOX5 antibodies (Santa Cruz Biotechnology Santa Cruz, CA) were added to the binding reactions and incubated for an additional 30 min.
Chromatin immunoprecipitation
The chromatin immunoprecipitation (ChIP) method was adapted from that of Petz et al. (33) and was performed as we have described (14) using reagents from the ChIP Assay kit (Upstate Biotechnology, Lake Placid, NY). Approximately 1 x 106 LT2 cells were cross-linked with formaldehyde and then lysed. Cellular DNA was sonicated to fragments of 200-1000 bp. The supernatant from the sonicated lysate was then precleared with salmon sperm DNA/protein A agarose. Next, either Sp1- or NFI-containing complexes were immunoprecipitated using specific antibodies (Santa Cruz Biotechnology). Complexes were collected using protein A agarose. After washing and elution, cross-linking was reversed and DNA was extracted. The purified DNA was analyzed by PCR using the following primers: 5'-agtcagacccagccctagagtga-3' and 5'-actaatccagtggttcgtgcggg-3' (mouse Lhx3a promoter region); 5'-aaaggcctggggctggtcctag-3' and 5'-ggtcagggaacactagcttggag-3' (NFI site region in mouse intron 1a/Lhx3b promoter); and 5'-ctcctgctcgaagtctagagc-3' and 5'-gctggtgataagtggccttgg-3' (mouse actin gene). PCR products were analyzed by agarose gel electrophoresis, and the identity of observed DNA fragments was confirmed by cloning into pTOPO vectors (Invitrogen) and DNA sequence analysis.
Site-directed mutagenesis
Site-directed mutagenesis was performed as described (32) using the QuikChange Site Directed Mutagenesis Kit (Stratagene, La Jolla, CA). Sequences containing predicted transcription factor-binding sites were mutated using pGL2 plasmid substrates containing either intron 1a of the hLHX3 gene (i.e. the hLHX3b promoter) or –2.7 kb of hLHX3a 5' flanking sequence. Mutagenic oligonucleotides were as follows: 5'-cagaggggtgaggttgggggctgccttgag-3' (–291 Sp1 site in hLHX3a promoter); 5'-cccgggaggtggttgggcgcgcgggcgggg-3' (–181 Sp1 site in hLHX3a promoter); 5'-gcgacccccggccccttcccttcccgttgccctttcccccggccg-3' (–203/–185 Sp1 sites in intron 1a); and 5'-ggagggctggcgggatgccagcagggtgggccgcc-3' (NFI site in intron 1a).
Southwestern analysis
For Southwestern blotting experiments, pituitary protein extracts were separated by standard SDS-PAGE and transferred to nitrocellulose membranes by electrophoresis. Membranes were then incubated in TNED renaturation buffer (10 mM Tris-Cl pH 7.5, 0.1 mM EDTA pH 7.5, 50 mM NaCl, 1 mM dithiothreitol, 5% nonfat dry milk) at room temperature with slow rotation in a hybridization oven. The membranes were then incubated with approximately 8 x 107 cpm/ml of a 32P end-radiolabeled DNA probe (30–60 bp) overnight at room temperature in TNED buffer supplemented with 0.25% nonfat dried milk and 0.005 mg/ml sheared salmon sperm. After binding of the DNA probe, the protein blot membranes were washed three times for 5 min with 20 ml of TNED buffer plus 0.25% milk. Membranes were then air dried, and results were analyzed using a Storm Phosphorimager (Molecular Dynamics, Piscataway, NJ).
RT-PCR of NFI isoforms
Total RNA was isolated from mouse LT2 gonadotrope pituitary and HEK 293T human embryonic kidney cell lines. RT of 1 μg of RNA was performed using the oligo-dT primers and the SuperScript First-Strand Synthesis System (Invitrogen). NFI transcripts were detected by primers that were designed to unique regions within the coding sequences of the four mouse/human NFI isoforms accessed through GenBank. These were 5'-gaagtcttggtttcagcagccc-3' and 5'-aatgggttgtgcaccttgcctt-3' for the NFI-A isoform; 5'-gggaactggagtcaacttccca-3' and 5'-ggtggagttcgagttgagatga-3' for NFI-B; 5'-acttccaggagagctttgtcac-3' and 5'-tggggggacgggctgttgaatg-3' for NFI-C; and 5'-aagtactgatggggagcggctc-3' and 5'-tgctggtggaaggaggggaggt-3' for NFI-X.
Results
Analysis of Lhx3 gene and cDNA nucleotide sequences from humans and mice suggested that the two major mRNAs are generated from two TATA-less promoters featuring high GC contents (Fig. 1 and data not shown). To characterize the transcription start sites in the human and mouse Lhx3 genes, we performed RNA ligase-mediated RACE and S1 nuclease-mapping experiments using human pituitary gland RNA and mouse pituitary T3–1 pregonadotrope cell (29) RNAs as substrates. These experiments revealed two major transcription start sites for a hLHX3a promoter upstream of exon Ia, one for the mouse Lhx3a promoter and one for a hLHX3b promoter upstream of exon Ib in intron 1a (Fig. 1, B and C, and data not shown).
To test the functional properties of the two identified hLHX3 gene promoters, we created luciferase reporter genes containing fragments of the hLHX3a and hLHX3b promoter 5' flanking regions (Fig. 2). These reporter genes were transfected into cultured mouse pituitary LT2 gonadotrope or T3–1 pregonadotrope cells, and their activities were measured. Both of these cell types express the mouse Lhx3a and Lhx3b mRNA transcripts (14, 22). The hLHX3a promoter reporter genes were active in these pituitary cells with the –2701- and –2080-bp constructs displaying the highest activities (Fig. 2A and data not shown). The region between –4824 and –2701 in the hLHX3a upstream sequence appears to contain negatively acting elements. In addition, positive regulatory elements appear to lie between –2080 and –1581 bp of the hLHX3a promoter. The hLHX3b promoter reporter gene containing the entire intron 1a sequence (–1804 bp) was also active in the pituitary cells (Fig. 2B). Deletion of the distal region of this sequence to leave –1267 bp reduced the activity of the promoter to a level similar to that of a construct retaining only 418 bp of 5' flanking sequence (Fig. 2B). This observation suggested that intron 1a contains proximal elements that are important for basal transcription of the hLHX3b promoter and additional regulatory elements located between –1804 and –1267 that confer higher levels of expression (see below).
We next used a comparative strategy to examine conservation of Lhx3 gene promoter sequences from primate, ungulate, and rodent mammals (Figs. 3 and 4). As part of these studies, the sequence of the entire bovine Lhx3 gene was determined. First, BLASTN searches of the NCBI Trace Archives containing whole genome shotgun reads from the bovine genome sequencing project were performed using the full-length cDNA sequence of the hLHX3a cDNA as a query. Recovered sequences were collected and aligned into four contigs containing portions with high similarity to exons of hLHX3 (see Materials and Methods), leaving three gaps in the gene sequence. The PCR was then used to span these gaps and to confirm all regions of low sequence quality or complexity by amplification and sequencing of bovine genomic DNA from the same animal used in the whole genome shotgun sequencing project. The resulting edited 12,883-bp gene contig was submitted to GenBank as accession no. AY923832. To confirm the exons predicted from the gene, a full-length bovine Lhx3b cDNA sequence was cloned by iterative screening of a multitissue cDNA library that included pituitary gland. The 2,390-bp bovine Lhx3b cDNA clone was submitted to GenBank as accession no. AY923833. This clone predicts a 403 amino acid protein with 95% primary sequence identity to human LHX3b (data not shown). The genome sequence encompasses the entire observed cDNA sequence, including 5,000 bp of sequence upstream from the first exon, and displays consensus splice boundary and polyA addition signal sequences (data not shown). The predicted bovine Lhx3a promoter 5' flanking region DNA sequence was aligned with the equivalent regions from the human, chimp, mouse, and rat genomes (Fig. 3). Chimp, mouse, and rat sequences were identified by BLASTN searches of NCBI databases. The Lhx3a promoter sequences are very GC rich (e.g. the human and bovine promoters have 79% and 76% GC content in this region, respectively) and lack obvious TATA boxes (Fig. 3). Two GC boxes located at –181 bp and –165 bp of the human sequence (the LHX3a protein first codon is considered to be position +1) appear to be conserved in the examined mammalian sequences (Fig. 3). An additional element at –291 bp (in humans) is observed also in the chimp with the bovine sequence having a similar, more proximal element (Fig. 3).
To examine conservation of mammalian Lhx3b promoter sequences, the Lhx3 intron 1a sequences of the rhesus monkey and pig genomes were also cloned and sequenced (see Materials and Methods). The intron sequences were submitted to GenBank (accession nos. AY879263 for rhesus monkey and AY879262 for pig). As described above, the corresponding chimp, mouse, and rat sequences were obtained by BLASTN searches of GenBank databases. The aligned mammalian Lhx3 gene intron 1a DNA sequences display two regions of strong similarity: the proximal region around the transcription start site (Fig. 4). Similar to the Lhx3a promoter, the Lhx3b proximal regions are GC rich (e.g. the human and bovine promoters are 80% GC content in this region) and have no obvious TATA elements. Six GC boxes are found in the human sequence (Fig. 4). Of these, the distal three sequences (–345, –308, and –286) are also found in the other primate sequences (chimp and rhesus), with the –286 sequence also found in the cow. Two closely aligned central GC boxes (–203 and –185) are found in all of the mammalian sequences, and a proximal element at –133 is found in all examined mammals, except rodents (Fig. 4).
The observed clusters of GC boxes in the hLHX3 promoters led us to test the hypothesis that the Sp1 transcription factor can recognize these elements and regulate hLHX3 gene transcription. Luciferase reporter genes representing the most active 5' flanking regions of the hLHX3a and hLHX3b promoters were transiently cotransfected into human embryonic kidney (HEK) 293T cells with Sp1 transcription factor expression vectors. In these experiments, the activity of the hLHX3a promoter was increased approximately 40-fold (Fig. 5A) and that of the hLHX3b promoter was boosted approximately 7-fold (Fig. 5B) in comparison with negative controls. For comparison, we tested whether other pituitary transcription factors could activate or repress transcription from these promoter constructs. LHX3a and EGR1 increased transcription from the hLHX3a promoter (5-fold and 4-fold, respectively; Fig. 5A). LHX3b, SF1, and PROP1 had very modest effects on the hLHX3a reporter gene (Fig. 5A). The hLHX3b reporter gene was moderately induced by LHX3a and inhibited by LHX3b, EGR1, and PROP1 (Fig. 5B).
To better characterize the effects of the Sp1 transcription factor, we performed EMSA experiments to test the Sp1 interaction properties of GC box-containing sequences in the hLHX3 promoters. After extended exposures of EMSA gels to film, the –291 element from the hLHX3a promoter formed a very weak complex with proteins from LT2 pituitary cells that was disrupted by the addition of anti-Sp1 antibodies (data not shown). In addition, a faster-migrating complex was observed that was not affected by anti-Sp1 (Fig. 6A). This –291 sequence was bound by purified, recombinant Sp1 protein (Fig. 6A, right panel). The –181 hLHX3a GC box interacted with LT2 cell proteins, and the interaction was prevented by the anti-Sp1 antibodies (Fig. 6A). Consistent with this observation, this site was strongly bound by recombinant Sp1 protein. The –165 element formed a faster-migrating complex with pituitary cell proteins, but separate experiments using anti-Sp1 antibodies or pure Sp1 protein indicate that this is likely not a high-affinity Sp1-binding site. We conclude that pituitary cell proteins recognize the GC boxes in the hLHX3a promoter and that the –291 and –181 sites are likely weak and strong Sp1 interaction elements, respectively.
We similarly examined protein/DNA interaction properties with GC boxes in the hLHX3b proximal promoters. In these studies, EMSA experiments using pituitary LT2 or HEK cell extracts, anti-Sp1 antibodies, or purified Sp1 protein to probe protein/DNA interactions indicated that the –308 element can be weakly bound by Sp1 and the –203/–185 region is strongly recognized by Sp1 (Fig. 6B and data not shown). As for the hLHX3a promoter analysis, for some sites (–345 and –308), faster-migrating complexes that are not disrupted by anti-Sp1 were also observed.
To assess Sp1 association with the endogenous mouse Lhx3a gene promoter, we performed ChIP experiments. These studies demonstrate Sp1 occupation of the proximal region of the Lhx3a promoter in LT2 pituitary cells (Fig. 7A). Parallel negative controls showed no nonspecific recovery of an unrelated actin gene (Fig. 7A). To date, we have been unable to obtain similar data showing Sp1 association with the mouse Lhx3b gene promoter. Likely technical explanations for this result include the high-GC content of this genomic region, a condition that makes conventional PCR challenging.
We next examined the importance of the identified Sp1-binding sites in the transcriptional activity of the hLHX3 promoters. Mutation of either the –181 or –291 element within the hLHX3a promoter reduce the activity of the promoter in pituitary cells (Fig. 7B). Interestingly, these mutations reduce promoter activity to similar, low levels. This observation may indicate the elements in this region act together rather than in additive fashion. Similarly, mutation of the major Sp1-interacting region of the hLHX3b promoter (–203/–185) compromised activity by approximately 5-fold (Fig. 7C). We conclude that Sp1 is an important regulator of the hLHX3 gene promoters.
Experiments described above indicate that the 5' end of intron 1a (upstream of the hLHX3b promoter) contains a regulatory element that is critical for hLHX3b activity in pituitary cells (Fig. 2B). To further define this potential positive regulatory region, we first scanned the entire intron for trans-acting protein interactions using large (400 bp), overlapping probes in EMSA experiments. All of the tested sequences displayed some protein binding, but the major complex-forming regions were –1504/–1084 and the two most proximal sequences that encompassed the GC boxes, including the identified Sp1 element (asterisks, Fig. 8A). Based on this observation and the functional data shown in Fig. 2B, we therefore concentrated on the –1504/–1084 region. EMSA experiments using shorter probes eventually refined the primary binding site within this region to an element located between –1444 and –1414 that interacted strongly with proteins from both LT2 pituitary cells and HEK cells (Fig. 8, B and C). In Southwestern blot experiments, radiolabeled probes representing this intronic region interacted with proteins of approximately 60–65 kDa from cultured rodent GC, T3–1, and LT2 pituitary cells (Fig. 8D and data not shown). Sequence analyses of the –1444 to –1414 DNA sequence suggested the presence of a possible nuclear factor I (NFI) transcription factor-binding site. To test the possibility that NFI family proteins interacted with this region, we performed additional EMSA assays. Anti-NFI antibodies disrupted the protein/DNA complexes in EMSA experiments using LT2 pituitary cell extracts, and a supershifted complex was observed (Fig. 8E). Antibodies to non-NFI proteins did not affect the protein/DNA complexes in parallel negative controls. To further examine NFI interaction with this element within mammalian Lhx3 genes, we performed ChIP experiments targeting the endogenous mouse intron 1a region. Consistent with the EMSA and Southwestern data, anti-NFI antibodies precipitated chromatin complexes containing this genomic sequence from LT2 pituitary gonadotrope cells but not a region of an actin gene in parallel negative controls (Fig. 8F).
Mammalian NFI factors include the NFI-A, NFI-B, NFI-C, and NFI-X isoforms (reviewed in Ref.34). To investigate which NFI family isoforms are expressed in the LT2 pituitary cells used in our studies and in pituitary glands, we performed RT-PCR experiments using cDNA derived from LT2, human adult pituitary, mouse adult pituitary, or HEK cells. Whereas all four NFI isoforms were expressed in the adult pituitary tissues and in HEK cells, only the A, C, and X isoform mRNAs were detected in the mouse LT2 pituitary gonadotrope cells (Fig. 9A). During the course of our studies, another group reported a RT-PCR assay of NFI isoform expression in LT2 cells with similar results, except that these authors did not detect NFI-A in their LT2 cells (35). In Northern blot experiments using pituitary GC somatolactotrope cell RNA, a separate laboratory detected expression of the A, C, and X NFI mRNAs (36). We conclude that all members of the NFI family are expressed in the pituitary and that subsets are found in differentiated pituitary cell types.
To better understand the interaction of NFI factors with the hLHX3 intron 1a element, we performed further structure/function studies of the DNA element. Inspection of the –1454 to –1426 region of intron 1a revealed two NFI half-sites (beginning at –1441 bp) surrounding an E-box, i.e. a DNA sequence matching the CANNTG consensus that can be recognized by members of the basic helix-loop-helix transcription factor superfamily (Fig. 9B). This sequence is therefore positioned 1325 bp upstream of the major transcription start site. Upstream of these sequence features is an imperfect potential NFI half-site (Fig. 9B). Oligonucleotide probes representing the wild-type human sequence and variants with specific mutations in each of these sequence features were tested in EMSA experiments using LT2 cell extracts. These experiments demonstrated that, whereas the two downstream NFI half-sites were critical for protein/DNA interaction, the E-box and the upstream element were not required for formation of the complex (Fig. 9C). Similar results were obtained in experiments using protein extracts from rat pituitary somatolactotrope GH3 cells (Fig. 9C). These observations are consistent with our initial mapping experiments (Fig. 8). Mutation of the downstream NFI half-sites of the intron 1a luciferase reporter gene severely compromised its activity in LT2 cells (Fig. 9D). Intriguingly, mutation of the E-box similarly affected reporter gene function (Fig. 9D). It is also interesting to note that although both mutation of the NFI site and deletion of the region containing the site both significantly reduce activity of the Lhx3b promoter/intron 1a (Figs. 2 and 9), the mutation results in a more severe reduction in activity. One explanation for this is that a repressive element in the intron is also removed in the deletion experiment. The intron region containing the NFI site is conserved in primates, and a similar sequence is found in a 5'-shifted location in other examined species (Fig. 9E).
Discussion
In this study, we present the first characterization of the mechanisms that regulate transcription of the Lhx3 gene from any species. Two conserved, TATA-less, GC-rich promoters located upstream of exons 1a and 1b of mammalian Lhx3 genes function to initiate basal transcription of the Lhx3a and Lhx3b mRNAs. Interestingly, the proximal Lhx3b promoter is more strongly conserved in mammals than is the Lhx3a promoter (Garcia, M., and S. J. Rhodes, unpublished data). To date, molecular assays have demonstrated that the LHX3a protein is significantly more active than LHX3b in DNA binding and pituitary gene activation assays because of repressive properties conferred by the LHX3b-specific amino terminus (13, 22, 23). However, the observation that the hLHX3b promoter is conserved, together with the previous report that the LHX3b-specific amino terminus protein sequence is better conserved than that of the LHX3a-specific amino-terminal domain (22), suggests that LHX3b plays important and, perhaps, unique roles in neuroendocrine development.
The TATA-binding protein-associated factor components of the TFIID complex are required for basal transcription at TATA-less promoters but TATA-binding protein itself may not always be required for initiation from this type of promoter (37). Classically, TATA-less promoter organizations were associated with housekeeping genes, which lacked precise temporal and spatial expression patterns. However, it is becoming apparent that TATA-less promoters may be more common than TATA-containing promoters (38) and that TATA-less promoters are often a feature of tissue-specific and regulatory genes (e.g. Refs.39 and 40). The results described here for the hLHX3 gene are consistent with these findings.
We have shown that some of the conserved GC boxes within the hLHX3 promoters are Sp1-binding sites, that expression of Sp1 results in increased promoter activity, and that the Sp1 sites are important contributors to basal promoter function (Figs. 5–7). An organization including the presence of multiple Sp1 binding sites in a GC-rich, TATA-less promoter exhibiting several initiation sites has been observed for other tissue-specific genes that encode regulatory transcription factors. For example, the mouse and human Wilm’s tumor suppressor gene (wt1), the expression of which is regulated spatially and temporally during urogenital development, has all of these features (41).
Mice lacking the Sp1 gene die in utero by embryonic d 10 demonstrating that Sp1 is critical for development. However, in these animals expression of suggested Sp1 target genes is nevertheless detectable (42). Although our data here demonstrate that Sp1 proteins in pituitary cells do occupy functionally important, GC-rich elements within the hLHX3a and hLHX3b promoters, it is possible that, at specific times, other members of the Sp protein family might interact with these and other hLHX3 promoter elements. Mammalian genomes encode multiple Sp1-related/XKLF transcription factors (43) and some members of this family exhibit restricted expression patterns and play roles in the development of specific tissues (e.g. Ref.44).
Some of the identified GC elements within the hLHX3 promoters conform to consensus sites for the EGR1/NGF1A/KROX24 zinc finger transcription factor. EGR1 plays a direct role in the transcriptional control of the LH subunit gene (reviewed in Ref.45), and gene inactivation experiments demonstrate its importance in pituitary somatotrope and gonadotrope cell development (46, 47). In transfection experiments, EGR1 boosted transcription from the hLHX3a promoter and reduced the activity of a hLHX3b reporter gene (Fig. 5), suggesting that EGR1 may differentially regulate hLHX3 promoter activities. The activity of EGR1 is often a response to environmental signals such as growth factors, neurotransmitters, and hormones, and the hLHX3 GC boxes may allow control of hLHX3 promoter activities through competitive interplay among factors such as EGR1, Sp1, and the Sp1-related Sp3 protein, as has been described for other promoters (e.g. Refs.45 and 48, 49, 50).
We have shown in this report that a conserved, positively acting cis-element lies at approximately –1442 bp of intron 1a in the hLHX3b promoter. EMSA, Southwestern blot, and ChIP assays indicate that NFI proteins interact with this region. NFI/CTF family transcription factors (34) have been suggested to contribute to both basal and tissue-restricted gene activation and repression, including in the pituitary and nervous systems (e.g. Refs.35, 36 , and 51, 52, 53). NFI and Sp1/Sp3-binding sites have been found in other genes with expression patterns in endocrine tissues, such as the ADAMTS-1 gene (54). Intriguingly, considering the involvement of Sp1 in hLHX3 promoter regulation, NFI family members have also been demonstrated to interfere with Sp1 activities in some promoters (55). All four of the four major NFI isoforms are found in pituitary cell types (Refs.35 and 36 and this study). To test the potential roles of individual NFI proteins in hLHX3b promoter activation, we have performed transfection experiments in pituitary cells with expression vectors for all four NFI factors and a hLHX3b promoter reporter gene. In these assays, overexpression of NFI did not significantly affect the activity of the promoter (Garcia, M., and S. J. Rhodes, unpublished observations), likely due to the presence of endogenous NFI proteins in the transfected cells. However, mutation of the NFI element compromised the activity of the hLHX3b promoter in pituitary gonadotrope cells (Fig. 9D). Interestingly, a mutation of an E-box-like sequence that did not affect gonadotrope cell NFI protein binding also strongly reduced the activity of the promoter (Fig. 9A), suggesting that this sequence is important for NFI-mediated transcription but not DNA interaction or that this sequence is recognized by other proteins that migrate within the same DNA/protein complex in EMSA experiments.
The embryonic expression patterns of the four murine NFI genes in many tissues including the pituitary suggest roles for the NFI factors in developmental regulation (Ref.56 and Lyons, G., personal communication). Gene ablation experiments in mice are beginning to dissect the unique roles of individual NFI genes. Loss of the Nfia gene results in perinatal death associated with nervous system defects (57, 58). Mice lacking functional Nfib genes also are not viable and have incomplete nervous system and lung development (59). The Nfic gene appears to play functions in mouse tooth root development (60). To date, NFI gene knockout animals have not displayed overt pituitary-associated phenotypes, but the roles of NFI factors in pituitary development may have yet to be revealed due to compensation or redundancy phenomena.
In this study, we have characterized the proximal elements that coordinate the basal production of the two major Lhx3 gene mRNAs and have identified two classes of trans-acting factors that regulate this activity. Our studies indicate that both positive and negative cis-acting sequences function within the approximately –4.8 kb upstream of the hLHX3a promoter (Fig. 2A). A recent annotation of the human genome draft (accessed at the NCBI) indicates that the QSCN6L1/SOXN gene encoding the quiescin Q6-like 1 protein, a putative sulfhydryl oxidase (61), may be located as close as 1.2 kb from the hLHX3a transcription initiation sites. The close location of this gene is an important consideration in interpretation of future studies investigating transcriptional control elements that lie upstream of hLHX3. Furthermore, the basal promoters described here have some activity in nonpituitary cells: additional gene regulatory pathways in addition to those characterized herein are required to correctly guide the restricted temporal and spatial expression of the LHX3 mRNAs. Future experiments will map the genomic regions required for these repressive and/or activating pathways and to understand how the promoters are differentially or coordinately controlled.
Autoregulation appears to be an important mechanism by which pituitary transcription factor genes participate in the establishment of stable cell fates during development (e.g. Refs.62 and 63). The experiments shown in Fig. 5A suggest that this may also be true for the hLHX3a promoter. In addition, recent reports have demonstrated that paired-like homeodomain (PITX)-class transcription factor genes act upstream of Lhx3 in pituitary development (64, 65). In preliminary experiments, we have tested the responses of the hLHX3a and hLHX3b promoters described in this report in cotransfection assays using rodent PITX1 and PITX2c expression vectors. In the presence of either PITX1 or PITX2c, transcription from the –2.7 kb hLHX3a promoter is moderately increased (2- to 3-fold). The intron 1a/hLHX3b promoter is induced to a similar degree by PITX2c, but is not affected by PITX1 (Yaden, B. C., and S. J. Rhodes, unpublished observations). These data suggest that PITX proteins might exert some of their effects by actions at proximal Lhx3 promoters, but further studies will be required to examine whether additional direct and indirect mechanisms mediate the induction of Lhx3 by PITX-dependent pathways.
To date, the mutations in the hLHX3 gene that have been associated with combined pituitary hormone-deficiency diseases are located within the protein-coding regions of the gene (19). The transcriptional regulatory elements that we have characterized here provide candidate regions for diagnostic analyses looking for genetic lesions causing combined pituitary hormone-deficiency diseases of unknown etiology.
Acknowledgments
B.C.Y. and M.G. contributed equally to this study. We thank Drs. E. Adamson, L. Cushman, T. Golus, R. Gronastaski, H. Ingraham, G. Lyons, P. Mellon, A. Nardulli, A. Showalter, M. Wegner, and F. Witzmann for providing reagents and sharing results and advice. We are grateful to Linda Flathman and Kevin Tennill for expert assistance. We thank Chad Hunter, Jesse Savage, and Kyle Sloop for comments on the manuscript.
Footnotes
This work was supported by grants (to S.J.R.) from the National Institutes of Health (HD42024) and the National Science Foundation (IBN 0131702).
B.C.Y., M.G., T.P.L.S, and S.J.R. have no conflicts of interest to declare.
First Published Online September 22, 2005
Abbreviations: BLASTN, Basic local alignment tool nucleotide; ChIP, chromatin immunoprecipitation; NFI, nuclear factor I; PITX, paired-like homeodomain transcription factor; RLM-RACE, RNA ligase-mediated rapid amplification of cDNA ends; Sp1, specificity protein 1.
Accepted for publication September 15, 2005.
References
Hunter CS, Rhodes SJ 2005 LIM-homeodomain genes in mammalian development and human disease. Mol Biol Rep 32:67–77
Savage JJ, Yaden BC, Kiratipranon P, Rhodes SJ 2003 Transcriptional control during mammalian anterior pituitary development. Gene 319:1–19
Zhu X, Rosenfeld MG 2004 Transcriptional control of precursor proliferation in the early phases of pituitary development. Curr Opin Genet Dev 14:567–574
Keegan CE, Camper SA 2003 Mouse knockout solves endocrine puzzle and promotes new pituitary lineage model. Genes Dev 17:677–682
Sobrier ML, Attie-Bitach T, Netchine I, Encha-Razavi F, Vekemans M, Amselem S 2004 Pathophysiology of syndromic combined pituitary hormone deficiency due to a LHX3 defect in light of LHX3 and LHX4 expression during early human development. Gene Expr Patterns 5:279–284
Bach I, Rhodes SJ, Pearse 2nd RV, Heinzel T, Gloss B, Scully KM, Sawchenko PE, Rosenfeld MG 1995 P-Lim, a LIM homeodomain factor, is expressed during pituitary organ and cell commitment and synergizes with Pit-1. Proc Natl Acad Sci USA 92:2720–2724
Zhadanov AB, Bertuzzi S, Taira M, Dawid IB, Westphal H 1995 Expression pattern of the murine LIM class homeobox gene Lhx3 in subsets of neural and neuroendocrine tissues. Dev Dyn 202:354–364
Seidah NG, Barale JC, Marcinkiewicz M, Mattei MG, Day R, Chretien M 1994 The mouse homeoprotein mLIM-3 is expressed early in cells derived from the neuroepithelium and persists in adult pituitary. DNA Cell Biol 13:1163–1180
Sheng HZ, Zhadanov AB, Mosinger Jr B, Fujii T, Bertuzzi S, Grinberg A, Lee EJ, Huang SP, Mahon KA, Westphal H 1996 Specification of pituitary cell lineages by the LIM homeobox gene Lhx3. Science 272:1004–1007
Sheng HZ, Moriyama K, Yamashita T, Li H, Potter SS, Mahon KA, Westphal H 1997 Multistep control of pituitary organogenesis. Science 278:1809–1812
Sharma K, Sheng HZ, Lettieri K, Li H, Karavanov A, Potter S, Westphal H, Pfaff SL 1998 LIM homeodomain factors Lhx3 and Lhx4 assign subtype identities for motor neurons. Cell 95:817–828
Roberson MS, Schoderbek WE, Tremml G, Maurer RA 1994 Activation of the glycoprotein hormone alpha-subunit promoter by a LIM-homeodomain transcription factor. Mol Cell Biol 14:2985–2993
Sloop KW, Meier BC, Bridwell JL, Parker GE, Schiller AM, Rhodes SJ 1999 Differential activation of pituitary hormone genes by human Lhx3 isoforms with distinct DNA binding properties. Mol Endocrinol 13:2212–2225
West BE, Parker GE, Savage JJ, Kiratipranon P, Toomey KS, Beach LR, Colvin SC, Sloop KW, Rhodes SJ 2004 Regulation of the follicle-stimulating hormone beta gene by the LHX3 LIM-homeodomain transcription factor. Endocrinology 145:4866–4879
McGillivray SM, Bailey JS, Ramezani R, Kirkwood BJ, Mellon PL 2005 Mouse GnRH receptor gene expression is mediated by the LHX3 homeodomain protein. Endocrinology 146:2180–2185
Pincas H, Amoyel K, Counis R, Laverriere JN 2001 Proximal cis-acting elements, including steroidogenic factor 1, mediate the efficiency of a distal enhancer in the promoter of the rat gonadotropin-releasing hormone receptor gene. Mol Endocrinol 15:319–337
Sloop KW, Parker GE, Hanna KR, Wright HA, Rhodes SJ 2001 LHX3 transcription factor mutations associated with combined pituitary hormone deficiency impair the activation of pituitary target genes. Gene 265:61–69
Howard PW, Maurer RA 2001 A point mutation in the LIM domain of Lhx3 reduces activation of the glycoprotein hormone alpha-subunit promoter. J Biol Chem 276:19020–19026
Netchine I, Sobrier ML, Krude H, Schnabel D, Maghnie M, Marcos E, Duriez B, Cacheux V, Moers A, Goossens M, Gruters A, Amselem S 2000 Mutations in LHX3 result in a new syndrome revealed by combined pituitary hormone deficiency. Nat Genet 25:182–186
Sloop KW, Showalter AD, Von Kap-Herr C, Pettenati MJ, Rhodes SJ 2000 Analysis of the human LHX3 neuroendocrine transcription factor gene and mapping to the subtelomeric region of chromosome 9. Gene 245:237–243
Zhadanov AB, Copeland NG, Gilbert DJ, Jenkins NA, Westphal H 1995 Genomic structure and chromosomal localization of the mouse LIM/homeobox gene Lhx3. Genomics 27:27–32
Sloop KW, Dwyer CJ, Rhodes SJ 2001 An isoform-specific inhibitory domain regulates the LHX3 LIM homeodomain factor holoprotein and the production of a functional alternate translation form. J Biol Chem 276:36311–36319
Yaden BC, Savage JJ, Hunter CS, Rhodes SJ 2005 DNA recognition properties of the LHX3b LIM homeodomain transcription factor. Mol Biol Rep 32:1–6
Ewing B, Hillier L, Wendl MC, Green P 1998 Base-calling of automated sequencer traces using phred. I. Accuracy assessment. Genome Res 8:175–185
Gordon D, Abajian C, Green P 1998 Consed: a graphical tool for sequence finishing. Genome Res 8:195–202
Smith TP, Rohrer GA, Alexander LJ, Troyer DL, Kirby-Dobbels KR, Janzen MA, Cornwell DL, Louis CF, Schook LB, Beattie CW 1995 Directed integration of the physical and genetic linkage maps of swine chromosome 7 reveals that the SLA spans the centromere. Genome Res 5:259–271
Smith TP, Showalter AD, Sloop KW, Rohrer GA, Fahrenkrug SC, Meier BC, Rhodes SJ 2001 Identification of porcine Lhx3 and SF1 as candidate genes for QTL affecting growth and reproduction traits in swine. Anim Genet 32:344–350
Showalter AD, Smith TP, Bennett GL, Sloop KW, Whitsett JA, Rhodes SJ 2002 Differential conservation of transcriptional domains of mammalian Prophet of Pit-1 proteins revealed by structural studies of the bovine gene and comparative functional analysis of the protein. Gene 291:211–221
Alarid ET, Windle JJ, Whyte DB, Mellon PL 1996 Immortalization of pituitary cells at discrete stages of development by directed oncogenesis in transgenic mice. Development 122:3319–3329
Sambrook J, Russell DW 2001 Molecular cloning: a laboratory manual. 3rd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press
Sloop KW, McCutchan Schiller A, Smith TP, Blanton Jr JR, Rohrer GA, Meier BC, Rhodes SJ 2000 Biochemical and genetic characterization of the porcine Prophet of Pit-1 pituitary transcription factor. Mol Cell Endocrinol 168:77–87
Parker GE, Sandoval RM, Feister HA, Bidwell JP, Rhodes SJ 2000 The homeodomain coordinates nuclear entry of the Lhx3 neuroendocrine transcription factor and association with the nuclear matrix. J Biol Chem 275:23891–23898
Petz LN, Ziegler YS, Schultz JR, Kim H, Kemper JK, Nardulli AM 2004 Differential regulation of the human progesterone receptor gene through an estrogen response element half site and Sp1 sites. J Steroid Biochem Mol Biol 88:113–122
Gronostajski RM 2000 Roles of the NFI/CTF gene family in transcription and development. Gene 249:31–45
Givens ML, Kurotani R, Rave-Harel N, Miller NL, Mellon PL 2004 Phylogenetic footprinting reveals evolutionarily conserved regions of the gonadotropin-releasing hormone gene that enhance cell-specific expression. Mol Endocrinol 18:2950–2966
Norquay LD, Jin Y, Surabhi RM, Gietz RD, Tanese N, Cattini PA 2001 A member of the nuclear factor-1 family is involved in the pituitary repression of the human placental growth hormone genes. Biochem J 354:387–395
Martinez E, Zhou Q, L’Etoile ND, Oelgeschlager T, Berk AJ, Roeder RG 1995 Core promoter-specific function of a mutant transcription factor TFIID defective in TATA-box binding. Proc Natl Acad Sci USA 92:11864–11868
Suzuki Y, Tsunoda T, Sese J, Taira H, Mizushima-Sugano J, Hata H, Ota T, Isogai T, Tanaka T, Nakamura Y, Suyama A, Sakaki Y, Morishita S, Okubo K, Sugano S 2001 Identification and characterization of the potential promoter regions of 1031 kinds of human genes. Genome Res 11:677–684
Alvarez M, Shah R, Rhodes SJ, Bidwell JP 2005 Two promoters control the mouse Nmp4/CIZ transcription factor gene. Gene 347:43–54
Wegner M 2000 Transcriptional control in myelinating glia: flavors and spices. Glia 31:1–14
Discenza MT, Dehbi M, Pelletier J 1997 Overlapping DNA recognition motifs between Sp1 and a novel trans-acting factor within the wt1 tumour suppressor gene promoter. Nucleic Acids Res 25:4314–4322
Marin M, Karis A, Visser P, Grosveld F, Philipsen S 1997 Transcription factor Sp1 is essential for early embryonic development but dispensable for cell growth and differentiation. Cell 89:619–628
Bouwman P, Philipsen S 2002 Regulation of the activity of Sp1-related transcription factors. Mol Cell Endocrinol 195:27–38
Bell SM, Schreiner CM, Waclaw RR, Campbell K, Potter SS, Scott WJ 2003 Sp8 is crucial for limb outgrowth and neuropore closure. Proc Natl Acad Sci USA 100:12195–12200
Jorgensen JS, Quirk CC, Nilson JH 2004 Multiple and overlapping combinatorial codes orchestrate hormonal responsiveness and dictate cell-specific expression of the genes encoding luteinizing hormone. Endocr Rev 25:521–542
Lee SL, Sadovsky Y, Swirnoff AH, Polish JA, Goda P, Gavrilina G, Milbrandt J 1996 Luteinizing hormone deficiency and female infertility in mice lacking the transcription factor NGFI-A (Egr-1). Science 273:1219–1221
Topilko P, Schneider-Maunoury S, Levi G, Trembleau A, Gourdji D, Driancourt MA, Rao CV, Charnay P 1998 Multiple pituitary and ovarian defects in Krox-24 (NGFI-A, Egr-1)-targeted mice. Mol Endocrinol 12:107–122
Huang RP, Fan Y, Ni Z, Mercola D, Adamson ED 1997 Reciprocal modulation between Sp1 and Egr-1. J Cell Biochem 66:489–499
Al-Sarraj A, Day RM, Thiel G 2005 Specificity of transcriptional regulation by the zinc finger transcription factors Sp1, Sp3, and Egr-1. J Cell Biochem 94:153–167
Pang RT, Lee LT, Ng SS, Yung WH, Chow BK 2004 CpG methylation and transcription factors Sp1 and Sp3 regulate the expression of the human secretin receptor gene. Mol Endocrinol 18:471–483
Ling G, Hauer CR, Gronostajski RM, Pentecost BT, Ding X 2004 Transcriptional regulation of rat CYP2A3 by nuclear factor 1: identification of a novel NFI-A isoform, and evidence for tissue-selective interaction of NFI with the CYP2A3 promoter in vivo. J Biol Chem 279:27888–27895
Nakazato M, Chung HK, Ulianich L, Grassadonia A, Suzuki K, Kohn LD 2000 Thyroglobulin repression of thyroid transcription factor 1 (TTF-1) gene expression is mediated by decreased DNA binding of nuclear factor I proteins which control constitutive TTF-1 expression. Mol Cell Biol 20:8499–8512
Bedford FK, Julius D, Ingraham HA 1998 Neuronal expression of the 5HT3 serotonin receptor gene requires nuclear factor 1 complexes. J Neurosci 18:6186–6194
Doyle KM, Russell DL, Sriraman V, Richards JS 2004 Coordinate transcription of the ADAMTS-1 gene by luteinizing hormone and progesterone receptor. Mol Endocrinol 18:2463–2478
Laniel MA, Poirier GG, Guerin SL 2001 Nuclear factor 1 interferes with Sp1 binding through a composite element on the rat poly(ADP-ribose) polymerase promoter to modulate its activity in vitro. J Biol Chem 276:20766–20773
Chaudhry AZ, Lyons GE, Gronostajski RM 1997 Expression patterns of the four nuclear factor I genes during mouse embryogenesis indicate a potential role in development. Dev Dyn 208:313–325
das Neves L, Duchala CS, Tolentino-Silva F, Haxhiu MA, Colmenares C, Macklin WB, Campbell CE, Butz KG, Gronostajski RM 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
Shu T, Butz KG, Plachez C, Gronostajski RM, Richards LJ 2003 Abnormal development of forebrain midline glia and commissural projections in Nfia knock-out mice. J Neurosci 23:203–212
Steele-Perkins G, Plachez C, Butz KG, Yang G, Bachurski CJ, Kinsman SL, Litwack ED, Richards LJ, Gronostajski RM 2005 The transcription factor gene Nfib is essential for both lung maturation and brain development. Mol Cell Biol 25:685–698
Steele-Perkins G, Butz KG, Lyons GE, Zeichner-David M, Kim HJ, Cho MI, Gronostajski RM 2003 Essential role for NFI-C/CTF transcription-replication factor in tooth root development. Mol Cell Biol 23:1075–1084
Wittke I, Wiedemeyer R, Pillmann A, Savelyeva L, Westermann F, Schwab M 2003 Neuroblastoma-derived sulfhydryl oxidase, a new member of the sulfhydryl oxidase/Quiescin6 family, regulates sensitization to interferon gamma-induced cell death in human neuroblastoma cells. Cancer Res 63:7742–7752
Goodyer CG, Tremblay JJ, Paradis FW, Marcil A, Lanctot C, Gauthier Y, Drouin J 2003 Pitx1 in vivo promoter activity and mechanisms of positive autoregulation. Neuroendocrinology 78:129–137
Rhodes SJ, Chen R, DiMattia GE, Scully KM, Kalla KA, Lin SC, Yu VC, Rosenfeld MG 1993 A tissue-specific enhancer confers Pit-1-dependent morphogen inducibility and autoregulation on the pit-1 gene. Genes Dev 7:913–932
Charles MA, Suh H, Hjalt TA, Drouin J, Camper SA, Gage PJ 2005 PITX genes are required for cell survival and Lhx3 activation. Mol Endocrinol 19:1893–1903
Tremblay JJ, Lanctot C, Drouin J 1998 The pan-pituitary activator of transcription, Ptx1 (pituitary homeobox 1), acts in synergy with SF-1 and Pit1 and is an upstream regulator of the Lim-homeodomain gene Lim3/Lhx3. Mol Endocrinol 12:428–441(Benjamin C. Yaden, Marin Garcia, III, Ti)
Department of Biology (B.C.Y., M.G.), Indiana University-Purdue University Indianapolis, Indianapolis, Indiana 46202
U.S. Department of Agriculture/Agricultural Research Service (T.P.L.S.), U.S. Meat Animal Research Center, Clay Center, Nebraska 68933
Abstract
The LHX3 transcription factor is required for pituitary and nervous system development in mammals. Mutations in the human gene are associated with hormone-deficiency diseases. The gene generates two mRNAs, hLHX3a and hLHX3b, which encode three proteins with different properties. Here, the cis elements and trans-acting factors that regulate the basal transcription of the two mRNAs are characterized. A comparative approach was taken featuring analysis of seven mammalian Lhx3 genes, with a focus on the human gene. Two conserved, TATA-less, GC-rich promoters that are used to transcribe the mRNAs precede exons 1a and 1b of hLHX3. Transcription start sites were mapped for both promoters. Deletion experiments showed most activity for reporter genes containing the basal promoters in the context of –2.0 kb of hLHX3a and 1.8 kb of intron 1a (hLHX3b). Transfection, site-directed mutation, electrophoretic mobility shift, Southwestern blot, and chromatin immunoprecipitation approaches were used to characterize the interaction of transcription factors with conserved elements in the promoters. Specificity protein 1 is a regulator of both promoters through interaction with GC boxes. In addition, a distal element within intron 1a that is recognized by nuclear factor I is critical for hLHX3b promoter function. We conclude that dual promoters allow regulated production of two hLHX3 mRNAs.
Introduction
THE Lhx3/P-LIM/LIM3 GENE encodes LIM-homeodomain transcription factors with essential roles in neuroendocrine development (reviewed in Ref.1). During anterior pituitary development, Lhx3 participates with other regulatory genes, such as Sf1, Prop1, Pit1, Egr1, Pitx1, Pitx2, and Tpit, to guide the specification of differentiated hormone-secreting cells (reviewed in Refs.2, 3, 4). In mammals, the Lhx3 gene is expressed in the embryonic nervous system and in the primordial and mature pituitary gland (5, 6, 7, 8). Mice homozygous for a null Lhx3 allele die after birth and feature incomplete structural development of the pituitary and defective motor neuron specification (9, 10, 11). In these animals, some anterior pituitary corticotrope cells are functional, but expression of the characteristic hormones of the gonadotrope, thyrotrope, somatotrope, and lactotrope cell types is not detectable, indicating that Lhx3 is required for the eventual differentiation of these specialized lineages (9). These observations are consistent with molecular studies demonstrating that LHX3 and related LIM-homeodomain factors can activate anterior pituitary expressed genes, including those encoding GSU, PRL, FSH, TSH, the GnRH receptor, and the Pit-1 transcription factor (e.g. Refs.6 and 12, 13, 14, 15, 16). Furthermore, mutations in the human LHX3 (hLHX3) gene that are predicted to lead to the production of disabled proteins cause recessive genetic diseases featuring combined pituitary hormone deficiency and other symptoms (17, 18, 19).
The hLHX3 gene contains seven coding exons and six introns that span approximately 8.7 kb located within the subtelomeric region of chromosome 9 (Fig. 1A) (20). The mouse Lhx3 gene has a similar organization (21). The single hLHX3 gene produces two major mRNAs known as hLHX3a and hLHX3b (13). Translation from the first methionine codons of the hLHX3a and hLHX3b mRNAs generates the LHX3a and LHX3b protein isoforms (13). These proteins have identical LIM domains, a central DNA-binding homeodomain and a carboxyl terminus that contains the major activation domain but have distinct amino termini resulting from alternate use of 5' exons in the gene (20, 22). A third protein isoform, M2-LHX3, is generated by preferential translation of the second in-frame methionine codon of the hLHX3a mRNA (22). The three LHX3 protein isoforms display different biochemical and functional properties (13, 14, 22, 23).
In this study, we investigated the transcriptional mechanisms by which mammalian Lhx3 genes generate multiple mRNAs that encode proteins with diverse regulatory properties. Phylogenetic comparisons and functional tests were used to map two conserved, TATA-less, GC-rich promoters that guide transcription of the two mRNAs. Two kilobases of the hLHX3a promoter and 1.8 kb of intron 1a (a region that contains the hLHX3b promoter) mediate basal activity. The specificity protein 1 (Sp1) transcription factor binds to proximal GC boxes and is a strong regulator of both promoters. Furthermore, a critical upstream element within intron 1a allows regulation by nuclear factor I (NFI) family transcription factors.
Materials and Methods
Cloning of mammalian Lhx3 gene and cDNA sequences
Fragments of the hLHX3 gene (20) were amplified by touchdown PCR using bacterial artificial chromosome clone RPC11–83N9 (Sanger Centre, Cambridge, UK) or normal human genomic DNA as substrates. Primers were designed based on human genome sequence data accessed through GenBank at the National Center for Biological Information (NCBI). The upstream (hLHX3a promoter) region of the gene was amplified in approximately 500-bp increments using an antisense primer (5'-cctcctaggtcagcgtcccctg-3') and one of the following sense primers: 5'-gtcttgagtcctcagcagtgct-3' (580 bp upstream of exon 1a), 5'-gagactcacaggacaagacccttga-3' (1.096 kb), 5'-agagctggatgccacctttagg-3' (1.581 kb), 5'-tgcttcgtgtctcactcgagag-3' (2.080 kb), 5'-tgcacacacaagcatctcactc-3' (2.701 kb), 5'-cctgctccaggctgccaagtgt-3' (3.240 kb), and 5'-gtagtccggaaagggccagtgt-3' (4.8 kb). For the hLHX3b promoter upstream in intron 1a, three regions were amplified using an antisense primer (5'-cgccactctccagtcccgaacttt-3') and one of the following sense primers: 5'-gcgtgtgctccagctcaggcctct-3' (1804 bp upstream), 5'-ggtaacaagtgctgtgcaaagtga-3' (1267 bp), and 5'-agtgcccgtcagctcttgcacaca-3' (418 kb). PCR was performed with Pfu Ultra polymerase (Stratagene) and MasterAmp PCR optimization buffers (Epicentre, Madison, WI) (if required due to high GC content of the target sequences). To create luciferase reporter genes, fragments of the hLHX3a promoter upstream region or of the hLHX3b promoter/intron 1a region were cloned into the pGL2-basic plasmid (Promega, Madison, WI). All plasmids were confirmed by DNA sequencing (Biochemistry Biotechnology Facility, Indiana University School of Medicine).
To obtain the sequence of the bovine Lhx3 gene, the Trace Archive of the bovine genome project at the NCBI containing raw reads from the first 3-fold genome coverage (12 million reads at the time of screening) was searched via basic local alignment tool nucleotide (BLASTN) using the full-length cDNA sequence of the hLHX3a cDNA. Trace files whose sequence showed highly significant (scores of >300) match to the cDNA, as well as the mate-pair end sequences from the respective clones, were collected in a directory and used to construct initial genomic contigs via phred (24) and phrap (25) algorithms. Contig sequences were masked for repetitive elements using RepeatMasker (Smit, A. F. A., and P. Green, unpublished results; http://ftp.genome.washington.edu/RM/RepeatMasker.html) and used to search for overlapping trace files in the archive, which were added to the directory for reconstruction of contigs. The process was repeated until none of the contigs in the phrap output identified trace files not already in the directory. This resulted in construction of four contigs containing portions with high similarity to exons of the hLHX3 cDNA, leaving three gaps in the gene sequence. Primers then were designed to span the gaps by PCR, and sequence was obtained by amplification of bovine genomic DNA from the same animal used in the whole genome shotgun sequencing. The PCR products were sequenced with the amplification primers, nested primers, or both. The resulting 12,883-bp contig was edited by manual inspection using the Consed viewing program (25), and areas of low sequence quality or areas where read overlap was exclusively from low-complexity sequence were targeted for finishing using additional PCR-based amplification and sequencing. To obtain confirming bovine Lhx3 cDNA sequence, primers were designed based on the cDNA sequence predicted from the first set of genomic trace files obtained. Primers 5'-gagatcccgctgtgtgcc-3' and 5'-ccttgcagtacacgctctcc-3' were designed from the putative exon 2 sequence and used to obtain a full coding sequence bovine Lhx3b cDNA clone via iterative screen (26) of a pooled-tissue cDNA library that included pituitary gland [library 1BOV (27)]. The clone obtained had an insert of 2,390 bp, and the complete insert was sequenced. The edited bovine Lhx3 gene and cDNA sequences have been submitted to GenBank with accession nos. AY923832 and AY923833, respectively. Intron 1a sequences of the rhesus monkey and porcine Lhx3 genes were amplified by the PCR. Templates were 150 ng of genomic DNA from adult rhesus monkey (kindly provided by Dr. T. Golus, Wisconsin Primate Research Center, Madison, WI) or adult pig (female Yorkshire cross). Reactions included MasterAmp PCR optimization buffer G (Epicentre), and 5'-atgctgctggaaacggggctcga-3' (monkey exon 1a), 5'-gaatctctcggcgcaggtccgcc-3' (monkey exon 1b) or 5'-atgctgctggaaacggagctggc-3' (pig exon 1a), and 5'-ggatctctcggcgcaggtcctcc-3' (pig exon 1b) primers were used. PCR products from multiple independent reactions were sequenced on both strands. Sequences were submitted to GenBank (accession nos. AY879262 for pig intron 1a and AY879263 for rhesus monkey intron 1a).
RNA ligase-mediated rapid amplification of cDNA ends
The transcription start sites of human and mouse Lhx3 genes were deduced by 5' RNA ligase-mediated rapid amplification of cDNA ends (RLM-RACE) performed using the GeneRacer protocol (Invitrogen, Carlsbad, CA) according to the instructions of the manufacturer, as we have described (28). First-strand cDNA was generated from adult human pituitary gland RNA (13) or from pituitary T3–1 pregonadotrope cell (29) total RNA and primed with 5'-ggcagtcgctgcacttgagacactt-3'. Second-strand cDNA was generated using primers from the manufacturer and the following gene-specific primers: 5'-gagacgcgctcctccgaggtca-3' (5' RACE Lhx3a) and 5'-tctccagtcccgaacttt-3' (5' RACE Lhx3b).
S1 nuclease assays
S1 nuclease assays were performed as described (30). Briefly, 32P end-radiolabeled single-stranded DNA probes for the hLHX3a and hLHX3b promoters were generated. S1 digestion reactions contained 15 μg of T3–1 RNA hybridized with labeled probes. Radiolabeled DNA products were analyzed by electrophoresis through 12% polyacrylamide 8 M urea gels. A 32P end-labeled 1-kb extension ladder (Invitrogen) was used as a molecular marker.
Cell culture and transfection
Human embryonic kidney (HEK) 293T and rodent pituitary cell lines were cultured and transfected as described (14). Typical transfections contained 2 μg of a luciferase reporter gene and 500 ng of an expression vector (if any). Control parallel samples received empty vector DNA. All assay groups were performed in triplicate. Forty-eight hr following transfection, cells were lysed in 25 mM Tris-Cl (pH 7.8), 2 mM dithiothreitol, 1% Triton X-100, 2 mM EDTA (pH 8.0), 10% glycerol. The lysate supernatant was assayed for luciferase activity using a luciferin substrate (Promega) and a Beckman Coulter luminometer (Fullerton, CA). Total cell protein was determined by the Bradford method (Bio-Rad, Hercules, CA), and luciferase activity was normalized to the amount of protein present. Expression vectors included human LHX3a, human LHX3b (13), human PROP1 (31), mouse SF1 (a gift from Dr. Holly Ingraham, University of California, San Francisco, CA), mouse EGR1 (a gift from Dr. Eileen Adamson, Burnham Institute, La Jolla, CA), rat Sp1 (a gift from Michael Wegner, University of Erlangen, Erlangen, Germany).
Cell protein extraction
Nuclear or whole cell protein extracts from HEK 293T and rodent pituitary cells were prepared as we have described (14, 32). Rat pituitary GH3 somatolactotrope nuclear extracts were purchased from Active Motif, Inc. (Carlsbad, CA).
EMSAs
EMSAs were performed as we have described (14) using radiolabeled double-stranded oligonucleotide probes with the results visualized by autoradiography or using a Storm phosphorimager (Amersham Biosciences, Piscataway, NJ). Cell extracts were prepared as described above. Human recombinant Sp1 protein was purchased from Promega. In some experiments, 2 μg of anti-Sp1, anti-NFI, or anti-SOX5 antibodies (Santa Cruz Biotechnology Santa Cruz, CA) were added to the binding reactions and incubated for an additional 30 min.
Chromatin immunoprecipitation
The chromatin immunoprecipitation (ChIP) method was adapted from that of Petz et al. (33) and was performed as we have described (14) using reagents from the ChIP Assay kit (Upstate Biotechnology, Lake Placid, NY). Approximately 1 x 106 LT2 cells were cross-linked with formaldehyde and then lysed. Cellular DNA was sonicated to fragments of 200-1000 bp. The supernatant from the sonicated lysate was then precleared with salmon sperm DNA/protein A agarose. Next, either Sp1- or NFI-containing complexes were immunoprecipitated using specific antibodies (Santa Cruz Biotechnology). Complexes were collected using protein A agarose. After washing and elution, cross-linking was reversed and DNA was extracted. The purified DNA was analyzed by PCR using the following primers: 5'-agtcagacccagccctagagtga-3' and 5'-actaatccagtggttcgtgcggg-3' (mouse Lhx3a promoter region); 5'-aaaggcctggggctggtcctag-3' and 5'-ggtcagggaacactagcttggag-3' (NFI site region in mouse intron 1a/Lhx3b promoter); and 5'-ctcctgctcgaagtctagagc-3' and 5'-gctggtgataagtggccttgg-3' (mouse actin gene). PCR products were analyzed by agarose gel electrophoresis, and the identity of observed DNA fragments was confirmed by cloning into pTOPO vectors (Invitrogen) and DNA sequence analysis.
Site-directed mutagenesis
Site-directed mutagenesis was performed as described (32) using the QuikChange Site Directed Mutagenesis Kit (Stratagene, La Jolla, CA). Sequences containing predicted transcription factor-binding sites were mutated using pGL2 plasmid substrates containing either intron 1a of the hLHX3 gene (i.e. the hLHX3b promoter) or –2.7 kb of hLHX3a 5' flanking sequence. Mutagenic oligonucleotides were as follows: 5'-cagaggggtgaggttgggggctgccttgag-3' (–291 Sp1 site in hLHX3a promoter); 5'-cccgggaggtggttgggcgcgcgggcgggg-3' (–181 Sp1 site in hLHX3a promoter); 5'-gcgacccccggccccttcccttcccgttgccctttcccccggccg-3' (–203/–185 Sp1 sites in intron 1a); and 5'-ggagggctggcgggatgccagcagggtgggccgcc-3' (NFI site in intron 1a).
Southwestern analysis
For Southwestern blotting experiments, pituitary protein extracts were separated by standard SDS-PAGE and transferred to nitrocellulose membranes by electrophoresis. Membranes were then incubated in TNED renaturation buffer (10 mM Tris-Cl pH 7.5, 0.1 mM EDTA pH 7.5, 50 mM NaCl, 1 mM dithiothreitol, 5% nonfat dry milk) at room temperature with slow rotation in a hybridization oven. The membranes were then incubated with approximately 8 x 107 cpm/ml of a 32P end-radiolabeled DNA probe (30–60 bp) overnight at room temperature in TNED buffer supplemented with 0.25% nonfat dried milk and 0.005 mg/ml sheared salmon sperm. After binding of the DNA probe, the protein blot membranes were washed three times for 5 min with 20 ml of TNED buffer plus 0.25% milk. Membranes were then air dried, and results were analyzed using a Storm Phosphorimager (Molecular Dynamics, Piscataway, NJ).
RT-PCR of NFI isoforms
Total RNA was isolated from mouse LT2 gonadotrope pituitary and HEK 293T human embryonic kidney cell lines. RT of 1 μg of RNA was performed using the oligo-dT primers and the SuperScript First-Strand Synthesis System (Invitrogen). NFI transcripts were detected by primers that were designed to unique regions within the coding sequences of the four mouse/human NFI isoforms accessed through GenBank. These were 5'-gaagtcttggtttcagcagccc-3' and 5'-aatgggttgtgcaccttgcctt-3' for the NFI-A isoform; 5'-gggaactggagtcaacttccca-3' and 5'-ggtggagttcgagttgagatga-3' for NFI-B; 5'-acttccaggagagctttgtcac-3' and 5'-tggggggacgggctgttgaatg-3' for NFI-C; and 5'-aagtactgatggggagcggctc-3' and 5'-tgctggtggaaggaggggaggt-3' for NFI-X.
Results
Analysis of Lhx3 gene and cDNA nucleotide sequences from humans and mice suggested that the two major mRNAs are generated from two TATA-less promoters featuring high GC contents (Fig. 1 and data not shown). To characterize the transcription start sites in the human and mouse Lhx3 genes, we performed RNA ligase-mediated RACE and S1 nuclease-mapping experiments using human pituitary gland RNA and mouse pituitary T3–1 pregonadotrope cell (29) RNAs as substrates. These experiments revealed two major transcription start sites for a hLHX3a promoter upstream of exon Ia, one for the mouse Lhx3a promoter and one for a hLHX3b promoter upstream of exon Ib in intron 1a (Fig. 1, B and C, and data not shown).
To test the functional properties of the two identified hLHX3 gene promoters, we created luciferase reporter genes containing fragments of the hLHX3a and hLHX3b promoter 5' flanking regions (Fig. 2). These reporter genes were transfected into cultured mouse pituitary LT2 gonadotrope or T3–1 pregonadotrope cells, and their activities were measured. Both of these cell types express the mouse Lhx3a and Lhx3b mRNA transcripts (14, 22). The hLHX3a promoter reporter genes were active in these pituitary cells with the –2701- and –2080-bp constructs displaying the highest activities (Fig. 2A and data not shown). The region between –4824 and –2701 in the hLHX3a upstream sequence appears to contain negatively acting elements. In addition, positive regulatory elements appear to lie between –2080 and –1581 bp of the hLHX3a promoter. The hLHX3b promoter reporter gene containing the entire intron 1a sequence (–1804 bp) was also active in the pituitary cells (Fig. 2B). Deletion of the distal region of this sequence to leave –1267 bp reduced the activity of the promoter to a level similar to that of a construct retaining only 418 bp of 5' flanking sequence (Fig. 2B). This observation suggested that intron 1a contains proximal elements that are important for basal transcription of the hLHX3b promoter and additional regulatory elements located between –1804 and –1267 that confer higher levels of expression (see below).
We next used a comparative strategy to examine conservation of Lhx3 gene promoter sequences from primate, ungulate, and rodent mammals (Figs. 3 and 4). As part of these studies, the sequence of the entire bovine Lhx3 gene was determined. First, BLASTN searches of the NCBI Trace Archives containing whole genome shotgun reads from the bovine genome sequencing project were performed using the full-length cDNA sequence of the hLHX3a cDNA as a query. Recovered sequences were collected and aligned into four contigs containing portions with high similarity to exons of hLHX3 (see Materials and Methods), leaving three gaps in the gene sequence. The PCR was then used to span these gaps and to confirm all regions of low sequence quality or complexity by amplification and sequencing of bovine genomic DNA from the same animal used in the whole genome shotgun sequencing project. The resulting edited 12,883-bp gene contig was submitted to GenBank as accession no. AY923832. To confirm the exons predicted from the gene, a full-length bovine Lhx3b cDNA sequence was cloned by iterative screening of a multitissue cDNA library that included pituitary gland. The 2,390-bp bovine Lhx3b cDNA clone was submitted to GenBank as accession no. AY923833. This clone predicts a 403 amino acid protein with 95% primary sequence identity to human LHX3b (data not shown). The genome sequence encompasses the entire observed cDNA sequence, including 5,000 bp of sequence upstream from the first exon, and displays consensus splice boundary and polyA addition signal sequences (data not shown). The predicted bovine Lhx3a promoter 5' flanking region DNA sequence was aligned with the equivalent regions from the human, chimp, mouse, and rat genomes (Fig. 3). Chimp, mouse, and rat sequences were identified by BLASTN searches of NCBI databases. The Lhx3a promoter sequences are very GC rich (e.g. the human and bovine promoters have 79% and 76% GC content in this region, respectively) and lack obvious TATA boxes (Fig. 3). Two GC boxes located at –181 bp and –165 bp of the human sequence (the LHX3a protein first codon is considered to be position +1) appear to be conserved in the examined mammalian sequences (Fig. 3). An additional element at –291 bp (in humans) is observed also in the chimp with the bovine sequence having a similar, more proximal element (Fig. 3).
To examine conservation of mammalian Lhx3b promoter sequences, the Lhx3 intron 1a sequences of the rhesus monkey and pig genomes were also cloned and sequenced (see Materials and Methods). The intron sequences were submitted to GenBank (accession nos. AY879263 for rhesus monkey and AY879262 for pig). As described above, the corresponding chimp, mouse, and rat sequences were obtained by BLASTN searches of GenBank databases. The aligned mammalian Lhx3 gene intron 1a DNA sequences display two regions of strong similarity: the proximal region around the transcription start site (Fig. 4). Similar to the Lhx3a promoter, the Lhx3b proximal regions are GC rich (e.g. the human and bovine promoters are 80% GC content in this region) and have no obvious TATA elements. Six GC boxes are found in the human sequence (Fig. 4). Of these, the distal three sequences (–345, –308, and –286) are also found in the other primate sequences (chimp and rhesus), with the –286 sequence also found in the cow. Two closely aligned central GC boxes (–203 and –185) are found in all of the mammalian sequences, and a proximal element at –133 is found in all examined mammals, except rodents (Fig. 4).
The observed clusters of GC boxes in the hLHX3 promoters led us to test the hypothesis that the Sp1 transcription factor can recognize these elements and regulate hLHX3 gene transcription. Luciferase reporter genes representing the most active 5' flanking regions of the hLHX3a and hLHX3b promoters were transiently cotransfected into human embryonic kidney (HEK) 293T cells with Sp1 transcription factor expression vectors. In these experiments, the activity of the hLHX3a promoter was increased approximately 40-fold (Fig. 5A) and that of the hLHX3b promoter was boosted approximately 7-fold (Fig. 5B) in comparison with negative controls. For comparison, we tested whether other pituitary transcription factors could activate or repress transcription from these promoter constructs. LHX3a and EGR1 increased transcription from the hLHX3a promoter (5-fold and 4-fold, respectively; Fig. 5A). LHX3b, SF1, and PROP1 had very modest effects on the hLHX3a reporter gene (Fig. 5A). The hLHX3b reporter gene was moderately induced by LHX3a and inhibited by LHX3b, EGR1, and PROP1 (Fig. 5B).
To better characterize the effects of the Sp1 transcription factor, we performed EMSA experiments to test the Sp1 interaction properties of GC box-containing sequences in the hLHX3 promoters. After extended exposures of EMSA gels to film, the –291 element from the hLHX3a promoter formed a very weak complex with proteins from LT2 pituitary cells that was disrupted by the addition of anti-Sp1 antibodies (data not shown). In addition, a faster-migrating complex was observed that was not affected by anti-Sp1 (Fig. 6A). This –291 sequence was bound by purified, recombinant Sp1 protein (Fig. 6A, right panel). The –181 hLHX3a GC box interacted with LT2 cell proteins, and the interaction was prevented by the anti-Sp1 antibodies (Fig. 6A). Consistent with this observation, this site was strongly bound by recombinant Sp1 protein. The –165 element formed a faster-migrating complex with pituitary cell proteins, but separate experiments using anti-Sp1 antibodies or pure Sp1 protein indicate that this is likely not a high-affinity Sp1-binding site. We conclude that pituitary cell proteins recognize the GC boxes in the hLHX3a promoter and that the –291 and –181 sites are likely weak and strong Sp1 interaction elements, respectively.
We similarly examined protein/DNA interaction properties with GC boxes in the hLHX3b proximal promoters. In these studies, EMSA experiments using pituitary LT2 or HEK cell extracts, anti-Sp1 antibodies, or purified Sp1 protein to probe protein/DNA interactions indicated that the –308 element can be weakly bound by Sp1 and the –203/–185 region is strongly recognized by Sp1 (Fig. 6B and data not shown). As for the hLHX3a promoter analysis, for some sites (–345 and –308), faster-migrating complexes that are not disrupted by anti-Sp1 were also observed.
To assess Sp1 association with the endogenous mouse Lhx3a gene promoter, we performed ChIP experiments. These studies demonstrate Sp1 occupation of the proximal region of the Lhx3a promoter in LT2 pituitary cells (Fig. 7A). Parallel negative controls showed no nonspecific recovery of an unrelated actin gene (Fig. 7A). To date, we have been unable to obtain similar data showing Sp1 association with the mouse Lhx3b gene promoter. Likely technical explanations for this result include the high-GC content of this genomic region, a condition that makes conventional PCR challenging.
We next examined the importance of the identified Sp1-binding sites in the transcriptional activity of the hLHX3 promoters. Mutation of either the –181 or –291 element within the hLHX3a promoter reduce the activity of the promoter in pituitary cells (Fig. 7B). Interestingly, these mutations reduce promoter activity to similar, low levels. This observation may indicate the elements in this region act together rather than in additive fashion. Similarly, mutation of the major Sp1-interacting region of the hLHX3b promoter (–203/–185) compromised activity by approximately 5-fold (Fig. 7C). We conclude that Sp1 is an important regulator of the hLHX3 gene promoters.
Experiments described above indicate that the 5' end of intron 1a (upstream of the hLHX3b promoter) contains a regulatory element that is critical for hLHX3b activity in pituitary cells (Fig. 2B). To further define this potential positive regulatory region, we first scanned the entire intron for trans-acting protein interactions using large (400 bp), overlapping probes in EMSA experiments. All of the tested sequences displayed some protein binding, but the major complex-forming regions were –1504/–1084 and the two most proximal sequences that encompassed the GC boxes, including the identified Sp1 element (asterisks, Fig. 8A). Based on this observation and the functional data shown in Fig. 2B, we therefore concentrated on the –1504/–1084 region. EMSA experiments using shorter probes eventually refined the primary binding site within this region to an element located between –1444 and –1414 that interacted strongly with proteins from both LT2 pituitary cells and HEK cells (Fig. 8, B and C). In Southwestern blot experiments, radiolabeled probes representing this intronic region interacted with proteins of approximately 60–65 kDa from cultured rodent GC, T3–1, and LT2 pituitary cells (Fig. 8D and data not shown). Sequence analyses of the –1444 to –1414 DNA sequence suggested the presence of a possible nuclear factor I (NFI) transcription factor-binding site. To test the possibility that NFI family proteins interacted with this region, we performed additional EMSA assays. Anti-NFI antibodies disrupted the protein/DNA complexes in EMSA experiments using LT2 pituitary cell extracts, and a supershifted complex was observed (Fig. 8E). Antibodies to non-NFI proteins did not affect the protein/DNA complexes in parallel negative controls. To further examine NFI interaction with this element within mammalian Lhx3 genes, we performed ChIP experiments targeting the endogenous mouse intron 1a region. Consistent with the EMSA and Southwestern data, anti-NFI antibodies precipitated chromatin complexes containing this genomic sequence from LT2 pituitary gonadotrope cells but not a region of an actin gene in parallel negative controls (Fig. 8F).
Mammalian NFI factors include the NFI-A, NFI-B, NFI-C, and NFI-X isoforms (reviewed in Ref.34). To investigate which NFI family isoforms are expressed in the LT2 pituitary cells used in our studies and in pituitary glands, we performed RT-PCR experiments using cDNA derived from LT2, human adult pituitary, mouse adult pituitary, or HEK cells. Whereas all four NFI isoforms were expressed in the adult pituitary tissues and in HEK cells, only the A, C, and X isoform mRNAs were detected in the mouse LT2 pituitary gonadotrope cells (Fig. 9A). During the course of our studies, another group reported a RT-PCR assay of NFI isoform expression in LT2 cells with similar results, except that these authors did not detect NFI-A in their LT2 cells (35). In Northern blot experiments using pituitary GC somatolactotrope cell RNA, a separate laboratory detected expression of the A, C, and X NFI mRNAs (36). We conclude that all members of the NFI family are expressed in the pituitary and that subsets are found in differentiated pituitary cell types.
To better understand the interaction of NFI factors with the hLHX3 intron 1a element, we performed further structure/function studies of the DNA element. Inspection of the –1454 to –1426 region of intron 1a revealed two NFI half-sites (beginning at –1441 bp) surrounding an E-box, i.e. a DNA sequence matching the CANNTG consensus that can be recognized by members of the basic helix-loop-helix transcription factor superfamily (Fig. 9B). This sequence is therefore positioned 1325 bp upstream of the major transcription start site. Upstream of these sequence features is an imperfect potential NFI half-site (Fig. 9B). Oligonucleotide probes representing the wild-type human sequence and variants with specific mutations in each of these sequence features were tested in EMSA experiments using LT2 cell extracts. These experiments demonstrated that, whereas the two downstream NFI half-sites were critical for protein/DNA interaction, the E-box and the upstream element were not required for formation of the complex (Fig. 9C). Similar results were obtained in experiments using protein extracts from rat pituitary somatolactotrope GH3 cells (Fig. 9C). These observations are consistent with our initial mapping experiments (Fig. 8). Mutation of the downstream NFI half-sites of the intron 1a luciferase reporter gene severely compromised its activity in LT2 cells (Fig. 9D). Intriguingly, mutation of the E-box similarly affected reporter gene function (Fig. 9D). It is also interesting to note that although both mutation of the NFI site and deletion of the region containing the site both significantly reduce activity of the Lhx3b promoter/intron 1a (Figs. 2 and 9), the mutation results in a more severe reduction in activity. One explanation for this is that a repressive element in the intron is also removed in the deletion experiment. The intron region containing the NFI site is conserved in primates, and a similar sequence is found in a 5'-shifted location in other examined species (Fig. 9E).
Discussion
In this study, we present the first characterization of the mechanisms that regulate transcription of the Lhx3 gene from any species. Two conserved, TATA-less, GC-rich promoters located upstream of exons 1a and 1b of mammalian Lhx3 genes function to initiate basal transcription of the Lhx3a and Lhx3b mRNAs. Interestingly, the proximal Lhx3b promoter is more strongly conserved in mammals than is the Lhx3a promoter (Garcia, M., and S. J. Rhodes, unpublished data). To date, molecular assays have demonstrated that the LHX3a protein is significantly more active than LHX3b in DNA binding and pituitary gene activation assays because of repressive properties conferred by the LHX3b-specific amino terminus (13, 22, 23). However, the observation that the hLHX3b promoter is conserved, together with the previous report that the LHX3b-specific amino terminus protein sequence is better conserved than that of the LHX3a-specific amino-terminal domain (22), suggests that LHX3b plays important and, perhaps, unique roles in neuroendocrine development.
The TATA-binding protein-associated factor components of the TFIID complex are required for basal transcription at TATA-less promoters but TATA-binding protein itself may not always be required for initiation from this type of promoter (37). Classically, TATA-less promoter organizations were associated with housekeeping genes, which lacked precise temporal and spatial expression patterns. However, it is becoming apparent that TATA-less promoters may be more common than TATA-containing promoters (38) and that TATA-less promoters are often a feature of tissue-specific and regulatory genes (e.g. Refs.39 and 40). The results described here for the hLHX3 gene are consistent with these findings.
We have shown that some of the conserved GC boxes within the hLHX3 promoters are Sp1-binding sites, that expression of Sp1 results in increased promoter activity, and that the Sp1 sites are important contributors to basal promoter function (Figs. 5–7). An organization including the presence of multiple Sp1 binding sites in a GC-rich, TATA-less promoter exhibiting several initiation sites has been observed for other tissue-specific genes that encode regulatory transcription factors. For example, the mouse and human Wilm’s tumor suppressor gene (wt1), the expression of which is regulated spatially and temporally during urogenital development, has all of these features (41).
Mice lacking the Sp1 gene die in utero by embryonic d 10 demonstrating that Sp1 is critical for development. However, in these animals expression of suggested Sp1 target genes is nevertheless detectable (42). Although our data here demonstrate that Sp1 proteins in pituitary cells do occupy functionally important, GC-rich elements within the hLHX3a and hLHX3b promoters, it is possible that, at specific times, other members of the Sp protein family might interact with these and other hLHX3 promoter elements. Mammalian genomes encode multiple Sp1-related/XKLF transcription factors (43) and some members of this family exhibit restricted expression patterns and play roles in the development of specific tissues (e.g. Ref.44).
Some of the identified GC elements within the hLHX3 promoters conform to consensus sites for the EGR1/NGF1A/KROX24 zinc finger transcription factor. EGR1 plays a direct role in the transcriptional control of the LH subunit gene (reviewed in Ref.45), and gene inactivation experiments demonstrate its importance in pituitary somatotrope and gonadotrope cell development (46, 47). In transfection experiments, EGR1 boosted transcription from the hLHX3a promoter and reduced the activity of a hLHX3b reporter gene (Fig. 5), suggesting that EGR1 may differentially regulate hLHX3 promoter activities. The activity of EGR1 is often a response to environmental signals such as growth factors, neurotransmitters, and hormones, and the hLHX3 GC boxes may allow control of hLHX3 promoter activities through competitive interplay among factors such as EGR1, Sp1, and the Sp1-related Sp3 protein, as has been described for other promoters (e.g. Refs.45 and 48, 49, 50).
We have shown in this report that a conserved, positively acting cis-element lies at approximately –1442 bp of intron 1a in the hLHX3b promoter. EMSA, Southwestern blot, and ChIP assays indicate that NFI proteins interact with this region. NFI/CTF family transcription factors (34) have been suggested to contribute to both basal and tissue-restricted gene activation and repression, including in the pituitary and nervous systems (e.g. Refs.35, 36 , and 51, 52, 53). NFI and Sp1/Sp3-binding sites have been found in other genes with expression patterns in endocrine tissues, such as the ADAMTS-1 gene (54). Intriguingly, considering the involvement of Sp1 in hLHX3 promoter regulation, NFI family members have also been demonstrated to interfere with Sp1 activities in some promoters (55). All four of the four major NFI isoforms are found in pituitary cell types (Refs.35 and 36 and this study). To test the potential roles of individual NFI proteins in hLHX3b promoter activation, we have performed transfection experiments in pituitary cells with expression vectors for all four NFI factors and a hLHX3b promoter reporter gene. In these assays, overexpression of NFI did not significantly affect the activity of the promoter (Garcia, M., and S. J. Rhodes, unpublished observations), likely due to the presence of endogenous NFI proteins in the transfected cells. However, mutation of the NFI element compromised the activity of the hLHX3b promoter in pituitary gonadotrope cells (Fig. 9D). Interestingly, a mutation of an E-box-like sequence that did not affect gonadotrope cell NFI protein binding also strongly reduced the activity of the promoter (Fig. 9A), suggesting that this sequence is important for NFI-mediated transcription but not DNA interaction or that this sequence is recognized by other proteins that migrate within the same DNA/protein complex in EMSA experiments.
The embryonic expression patterns of the four murine NFI genes in many tissues including the pituitary suggest roles for the NFI factors in developmental regulation (Ref.56 and Lyons, G., personal communication). Gene ablation experiments in mice are beginning to dissect the unique roles of individual NFI genes. Loss of the Nfia gene results in perinatal death associated with nervous system defects (57, 58). Mice lacking functional Nfib genes also are not viable and have incomplete nervous system and lung development (59). The Nfic gene appears to play functions in mouse tooth root development (60). To date, NFI gene knockout animals have not displayed overt pituitary-associated phenotypes, but the roles of NFI factors in pituitary development may have yet to be revealed due to compensation or redundancy phenomena.
In this study, we have characterized the proximal elements that coordinate the basal production of the two major Lhx3 gene mRNAs and have identified two classes of trans-acting factors that regulate this activity. Our studies indicate that both positive and negative cis-acting sequences function within the approximately –4.8 kb upstream of the hLHX3a promoter (Fig. 2A). A recent annotation of the human genome draft (accessed at the NCBI) indicates that the QSCN6L1/SOXN gene encoding the quiescin Q6-like 1 protein, a putative sulfhydryl oxidase (61), may be located as close as 1.2 kb from the hLHX3a transcription initiation sites. The close location of this gene is an important consideration in interpretation of future studies investigating transcriptional control elements that lie upstream of hLHX3. Furthermore, the basal promoters described here have some activity in nonpituitary cells: additional gene regulatory pathways in addition to those characterized herein are required to correctly guide the restricted temporal and spatial expression of the LHX3 mRNAs. Future experiments will map the genomic regions required for these repressive and/or activating pathways and to understand how the promoters are differentially or coordinately controlled.
Autoregulation appears to be an important mechanism by which pituitary transcription factor genes participate in the establishment of stable cell fates during development (e.g. Refs.62 and 63). The experiments shown in Fig. 5A suggest that this may also be true for the hLHX3a promoter. In addition, recent reports have demonstrated that paired-like homeodomain (PITX)-class transcription factor genes act upstream of Lhx3 in pituitary development (64, 65). In preliminary experiments, we have tested the responses of the hLHX3a and hLHX3b promoters described in this report in cotransfection assays using rodent PITX1 and PITX2c expression vectors. In the presence of either PITX1 or PITX2c, transcription from the –2.7 kb hLHX3a promoter is moderately increased (2- to 3-fold). The intron 1a/hLHX3b promoter is induced to a similar degree by PITX2c, but is not affected by PITX1 (Yaden, B. C., and S. J. Rhodes, unpublished observations). These data suggest that PITX proteins might exert some of their effects by actions at proximal Lhx3 promoters, but further studies will be required to examine whether additional direct and indirect mechanisms mediate the induction of Lhx3 by PITX-dependent pathways.
To date, the mutations in the hLHX3 gene that have been associated with combined pituitary hormone-deficiency diseases are located within the protein-coding regions of the gene (19). The transcriptional regulatory elements that we have characterized here provide candidate regions for diagnostic analyses looking for genetic lesions causing combined pituitary hormone-deficiency diseases of unknown etiology.
Acknowledgments
B.C.Y. and M.G. contributed equally to this study. We thank Drs. E. Adamson, L. Cushman, T. Golus, R. Gronastaski, H. Ingraham, G. Lyons, P. Mellon, A. Nardulli, A. Showalter, M. Wegner, and F. Witzmann for providing reagents and sharing results and advice. We are grateful to Linda Flathman and Kevin Tennill for expert assistance. We thank Chad Hunter, Jesse Savage, and Kyle Sloop for comments on the manuscript.
Footnotes
This work was supported by grants (to S.J.R.) from the National Institutes of Health (HD42024) and the National Science Foundation (IBN 0131702).
B.C.Y., M.G., T.P.L.S, and S.J.R. have no conflicts of interest to declare.
First Published Online September 22, 2005
Abbreviations: BLASTN, Basic local alignment tool nucleotide; ChIP, chromatin immunoprecipitation; NFI, nuclear factor I; PITX, paired-like homeodomain transcription factor; RLM-RACE, RNA ligase-mediated rapid amplification of cDNA ends; Sp1, specificity protein 1.
Accepted for publication September 15, 2005.
References
Hunter CS, Rhodes SJ 2005 LIM-homeodomain genes in mammalian development and human disease. Mol Biol Rep 32:67–77
Savage JJ, Yaden BC, Kiratipranon P, Rhodes SJ 2003 Transcriptional control during mammalian anterior pituitary development. Gene 319:1–19
Zhu X, Rosenfeld MG 2004 Transcriptional control of precursor proliferation in the early phases of pituitary development. Curr Opin Genet Dev 14:567–574
Keegan CE, Camper SA 2003 Mouse knockout solves endocrine puzzle and promotes new pituitary lineage model. Genes Dev 17:677–682
Sobrier ML, Attie-Bitach T, Netchine I, Encha-Razavi F, Vekemans M, Amselem S 2004 Pathophysiology of syndromic combined pituitary hormone deficiency due to a LHX3 defect in light of LHX3 and LHX4 expression during early human development. Gene Expr Patterns 5:279–284
Bach I, Rhodes SJ, Pearse 2nd RV, Heinzel T, Gloss B, Scully KM, Sawchenko PE, Rosenfeld MG 1995 P-Lim, a LIM homeodomain factor, is expressed during pituitary organ and cell commitment and synergizes with Pit-1. Proc Natl Acad Sci USA 92:2720–2724
Zhadanov AB, Bertuzzi S, Taira M, Dawid IB, Westphal H 1995 Expression pattern of the murine LIM class homeobox gene Lhx3 in subsets of neural and neuroendocrine tissues. Dev Dyn 202:354–364
Seidah NG, Barale JC, Marcinkiewicz M, Mattei MG, Day R, Chretien M 1994 The mouse homeoprotein mLIM-3 is expressed early in cells derived from the neuroepithelium and persists in adult pituitary. DNA Cell Biol 13:1163–1180
Sheng HZ, Zhadanov AB, Mosinger Jr B, Fujii T, Bertuzzi S, Grinberg A, Lee EJ, Huang SP, Mahon KA, Westphal H 1996 Specification of pituitary cell lineages by the LIM homeobox gene Lhx3. Science 272:1004–1007
Sheng HZ, Moriyama K, Yamashita T, Li H, Potter SS, Mahon KA, Westphal H 1997 Multistep control of pituitary organogenesis. Science 278:1809–1812
Sharma K, Sheng HZ, Lettieri K, Li H, Karavanov A, Potter S, Westphal H, Pfaff SL 1998 LIM homeodomain factors Lhx3 and Lhx4 assign subtype identities for motor neurons. Cell 95:817–828
Roberson MS, Schoderbek WE, Tremml G, Maurer RA 1994 Activation of the glycoprotein hormone alpha-subunit promoter by a LIM-homeodomain transcription factor. Mol Cell Biol 14:2985–2993
Sloop KW, Meier BC, Bridwell JL, Parker GE, Schiller AM, Rhodes SJ 1999 Differential activation of pituitary hormone genes by human Lhx3 isoforms with distinct DNA binding properties. Mol Endocrinol 13:2212–2225
West BE, Parker GE, Savage JJ, Kiratipranon P, Toomey KS, Beach LR, Colvin SC, Sloop KW, Rhodes SJ 2004 Regulation of the follicle-stimulating hormone beta gene by the LHX3 LIM-homeodomain transcription factor. Endocrinology 145:4866–4879
McGillivray SM, Bailey JS, Ramezani R, Kirkwood BJ, Mellon PL 2005 Mouse GnRH receptor gene expression is mediated by the LHX3 homeodomain protein. Endocrinology 146:2180–2185
Pincas H, Amoyel K, Counis R, Laverriere JN 2001 Proximal cis-acting elements, including steroidogenic factor 1, mediate the efficiency of a distal enhancer in the promoter of the rat gonadotropin-releasing hormone receptor gene. Mol Endocrinol 15:319–337
Sloop KW, Parker GE, Hanna KR, Wright HA, Rhodes SJ 2001 LHX3 transcription factor mutations associated with combined pituitary hormone deficiency impair the activation of pituitary target genes. Gene 265:61–69
Howard PW, Maurer RA 2001 A point mutation in the LIM domain of Lhx3 reduces activation of the glycoprotein hormone alpha-subunit promoter. J Biol Chem 276:19020–19026
Netchine I, Sobrier ML, Krude H, Schnabel D, Maghnie M, Marcos E, Duriez B, Cacheux V, Moers A, Goossens M, Gruters A, Amselem S 2000 Mutations in LHX3 result in a new syndrome revealed by combined pituitary hormone deficiency. Nat Genet 25:182–186
Sloop KW, Showalter AD, Von Kap-Herr C, Pettenati MJ, Rhodes SJ 2000 Analysis of the human LHX3 neuroendocrine transcription factor gene and mapping to the subtelomeric region of chromosome 9. Gene 245:237–243
Zhadanov AB, Copeland NG, Gilbert DJ, Jenkins NA, Westphal H 1995 Genomic structure and chromosomal localization of the mouse LIM/homeobox gene Lhx3. Genomics 27:27–32
Sloop KW, Dwyer CJ, Rhodes SJ 2001 An isoform-specific inhibitory domain regulates the LHX3 LIM homeodomain factor holoprotein and the production of a functional alternate translation form. J Biol Chem 276:36311–36319
Yaden BC, Savage JJ, Hunter CS, Rhodes SJ 2005 DNA recognition properties of the LHX3b LIM homeodomain transcription factor. Mol Biol Rep 32:1–6
Ewing B, Hillier L, Wendl MC, Green P 1998 Base-calling of automated sequencer traces using phred. I. Accuracy assessment. Genome Res 8:175–185
Gordon D, Abajian C, Green P 1998 Consed: a graphical tool for sequence finishing. Genome Res 8:195–202
Smith TP, Rohrer GA, Alexander LJ, Troyer DL, Kirby-Dobbels KR, Janzen MA, Cornwell DL, Louis CF, Schook LB, Beattie CW 1995 Directed integration of the physical and genetic linkage maps of swine chromosome 7 reveals that the SLA spans the centromere. Genome Res 5:259–271
Smith TP, Showalter AD, Sloop KW, Rohrer GA, Fahrenkrug SC, Meier BC, Rhodes SJ 2001 Identification of porcine Lhx3 and SF1 as candidate genes for QTL affecting growth and reproduction traits in swine. Anim Genet 32:344–350
Showalter AD, Smith TP, Bennett GL, Sloop KW, Whitsett JA, Rhodes SJ 2002 Differential conservation of transcriptional domains of mammalian Prophet of Pit-1 proteins revealed by structural studies of the bovine gene and comparative functional analysis of the protein. Gene 291:211–221
Alarid ET, Windle JJ, Whyte DB, Mellon PL 1996 Immortalization of pituitary cells at discrete stages of development by directed oncogenesis in transgenic mice. Development 122:3319–3329
Sambrook J, Russell DW 2001 Molecular cloning: a laboratory manual. 3rd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press
Sloop KW, McCutchan Schiller A, Smith TP, Blanton Jr JR, Rohrer GA, Meier BC, Rhodes SJ 2000 Biochemical and genetic characterization of the porcine Prophet of Pit-1 pituitary transcription factor. Mol Cell Endocrinol 168:77–87
Parker GE, Sandoval RM, Feister HA, Bidwell JP, Rhodes SJ 2000 The homeodomain coordinates nuclear entry of the Lhx3 neuroendocrine transcription factor and association with the nuclear matrix. J Biol Chem 275:23891–23898
Petz LN, Ziegler YS, Schultz JR, Kim H, Kemper JK, Nardulli AM 2004 Differential regulation of the human progesterone receptor gene through an estrogen response element half site and Sp1 sites. J Steroid Biochem Mol Biol 88:113–122
Gronostajski RM 2000 Roles of the NFI/CTF gene family in transcription and development. Gene 249:31–45
Givens ML, Kurotani R, Rave-Harel N, Miller NL, Mellon PL 2004 Phylogenetic footprinting reveals evolutionarily conserved regions of the gonadotropin-releasing hormone gene that enhance cell-specific expression. Mol Endocrinol 18:2950–2966
Norquay LD, Jin Y, Surabhi RM, Gietz RD, Tanese N, Cattini PA 2001 A member of the nuclear factor-1 family is involved in the pituitary repression of the human placental growth hormone genes. Biochem J 354:387–395
Martinez E, Zhou Q, L’Etoile ND, Oelgeschlager T, Berk AJ, Roeder RG 1995 Core promoter-specific function of a mutant transcription factor TFIID defective in TATA-box binding. Proc Natl Acad Sci USA 92:11864–11868
Suzuki Y, Tsunoda T, Sese J, Taira H, Mizushima-Sugano J, Hata H, Ota T, Isogai T, Tanaka T, Nakamura Y, Suyama A, Sakaki Y, Morishita S, Okubo K, Sugano S 2001 Identification and characterization of the potential promoter regions of 1031 kinds of human genes. Genome Res 11:677–684
Alvarez M, Shah R, Rhodes SJ, Bidwell JP 2005 Two promoters control the mouse Nmp4/CIZ transcription factor gene. Gene 347:43–54
Wegner M 2000 Transcriptional control in myelinating glia: flavors and spices. Glia 31:1–14
Discenza MT, Dehbi M, Pelletier J 1997 Overlapping DNA recognition motifs between Sp1 and a novel trans-acting factor within the wt1 tumour suppressor gene promoter. Nucleic Acids Res 25:4314–4322
Marin M, Karis A, Visser P, Grosveld F, Philipsen S 1997 Transcription factor Sp1 is essential for early embryonic development but dispensable for cell growth and differentiation. Cell 89:619–628
Bouwman P, Philipsen S 2002 Regulation of the activity of Sp1-related transcription factors. Mol Cell Endocrinol 195:27–38
Bell SM, Schreiner CM, Waclaw RR, Campbell K, Potter SS, Scott WJ 2003 Sp8 is crucial for limb outgrowth and neuropore closure. Proc Natl Acad Sci USA 100:12195–12200
Jorgensen JS, Quirk CC, Nilson JH 2004 Multiple and overlapping combinatorial codes orchestrate hormonal responsiveness and dictate cell-specific expression of the genes encoding luteinizing hormone. Endocr Rev 25:521–542
Lee SL, Sadovsky Y, Swirnoff AH, Polish JA, Goda P, Gavrilina G, Milbrandt J 1996 Luteinizing hormone deficiency and female infertility in mice lacking the transcription factor NGFI-A (Egr-1). Science 273:1219–1221
Topilko P, Schneider-Maunoury S, Levi G, Trembleau A, Gourdji D, Driancourt MA, Rao CV, Charnay P 1998 Multiple pituitary and ovarian defects in Krox-24 (NGFI-A, Egr-1)-targeted mice. Mol Endocrinol 12:107–122
Huang RP, Fan Y, Ni Z, Mercola D, Adamson ED 1997 Reciprocal modulation between Sp1 and Egr-1. J Cell Biochem 66:489–499
Al-Sarraj A, Day RM, Thiel G 2005 Specificity of transcriptional regulation by the zinc finger transcription factors Sp1, Sp3, and Egr-1. J Cell Biochem 94:153–167
Pang RT, Lee LT, Ng SS, Yung WH, Chow BK 2004 CpG methylation and transcription factors Sp1 and Sp3 regulate the expression of the human secretin receptor gene. Mol Endocrinol 18:471–483
Ling G, Hauer CR, Gronostajski RM, Pentecost BT, Ding X 2004 Transcriptional regulation of rat CYP2A3 by nuclear factor 1: identification of a novel NFI-A isoform, and evidence for tissue-selective interaction of NFI with the CYP2A3 promoter in vivo. J Biol Chem 279:27888–27895
Nakazato M, Chung HK, Ulianich L, Grassadonia A, Suzuki K, Kohn LD 2000 Thyroglobulin repression of thyroid transcription factor 1 (TTF-1) gene expression is mediated by decreased DNA binding of nuclear factor I proteins which control constitutive TTF-1 expression. Mol Cell Biol 20:8499–8512
Bedford FK, Julius D, Ingraham HA 1998 Neuronal expression of the 5HT3 serotonin receptor gene requires nuclear factor 1 complexes. J Neurosci 18:6186–6194
Doyle KM, Russell DL, Sriraman V, Richards JS 2004 Coordinate transcription of the ADAMTS-1 gene by luteinizing hormone and progesterone receptor. Mol Endocrinol 18:2463–2478
Laniel MA, Poirier GG, Guerin SL 2001 Nuclear factor 1 interferes with Sp1 binding through a composite element on the rat poly(ADP-ribose) polymerase promoter to modulate its activity in vitro. J Biol Chem 276:20766–20773
Chaudhry AZ, Lyons GE, Gronostajski RM 1997 Expression patterns of the four nuclear factor I genes during mouse embryogenesis indicate a potential role in development. Dev Dyn 208:313–325
das Neves L, Duchala CS, Tolentino-Silva F, Haxhiu MA, Colmenares C, Macklin WB, Campbell CE, Butz KG, Gronostajski RM 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
Shu T, Butz KG, Plachez C, Gronostajski RM, Richards LJ 2003 Abnormal development of forebrain midline glia and commissural projections in Nfia knock-out mice. J Neurosci 23:203–212
Steele-Perkins G, Plachez C, Butz KG, Yang G, Bachurski CJ, Kinsman SL, Litwack ED, Richards LJ, Gronostajski RM 2005 The transcription factor gene Nfib is essential for both lung maturation and brain development. Mol Cell Biol 25:685–698
Steele-Perkins G, Butz KG, Lyons GE, Zeichner-David M, Kim HJ, Cho MI, Gronostajski RM 2003 Essential role for NFI-C/CTF transcription-replication factor in tooth root development. Mol Cell Biol 23:1075–1084
Wittke I, Wiedemeyer R, Pillmann A, Savelyeva L, Westermann F, Schwab M 2003 Neuroblastoma-derived sulfhydryl oxidase, a new member of the sulfhydryl oxidase/Quiescin6 family, regulates sensitization to interferon gamma-induced cell death in human neuroblastoma cells. Cancer Res 63:7742–7752
Goodyer CG, Tremblay JJ, Paradis FW, Marcil A, Lanctot C, Gauthier Y, Drouin J 2003 Pitx1 in vivo promoter activity and mechanisms of positive autoregulation. Neuroendocrinology 78:129–137
Rhodes SJ, Chen R, DiMattia GE, Scully KM, Kalla KA, Lin SC, Yu VC, Rosenfeld MG 1993 A tissue-specific enhancer confers Pit-1-dependent morphogen inducibility and autoregulation on the pit-1 gene. Genes Dev 7:913–932
Charles MA, Suh H, Hjalt TA, Drouin J, Camper SA, Gage PJ 2005 PITX genes are required for cell survival and Lhx3 activation. Mol Endocrinol 19:1893–1903
Tremblay JJ, Lanctot C, Drouin J 1998 The pan-pituitary activator of transcription, Ptx1 (pituitary homeobox 1), acts in synergy with SF-1 and Pit1 and is an upstream regulator of the Lim-homeodomain gene Lim3/Lhx3. Mol Endocrinol 12:428–441(Benjamin C. Yaden, Marin Garcia, III, Ti)