Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (145K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (10)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Willis, S.
Right arrow Articles by Seyfred, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Willis, S.
Right arrow Articles by Seyfred, M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 1995 Oxford University Press 1065-1072

Footnote

Pituitary-specific chromatin structure of the rat prolactin distal enhancer element

Pituitary-specific chromatin structure of the rat prolactin distal enhancer element Sherry D. Willis and Mark A. Seyfred*

Department of Molecular Biology, Vanderbilt University, Nashville , TN 37235, USA

Received November 29, 1995; Revised and Accepted January 24, 1996

ABSTRACT

The location of target DNA sequences within chromatin may affect the ability of trans -acting factors to bind cis -elements and regulate gene transcription. To examine the effect of chromatin structure on the ability of the estrogen-estrogen receptor complex (E2R) to bind its respective DNA binding element within the rat prolactin (rPRL) gene and modulate rPRL gene expression, we have developed cell lines derived from the rPRL-expressing (rPRL + ) rat pituitary cell line GH3 and the rPRL-non-expressing (rPRL - ) rat embryo fibroblast cell line Rat1. These cell lines contain minichromosomes composed of the 5 ' upstream regulatory region of the rPRL gene driving expression of a reporter gene, Tn 5 , within a bovine papillomavirus (BPV) vector. The rPRL -Tn 5 gene retains the characteristics of cell-specific expression and estrogen inducibility of transcription displayed by the endogenous rPRL gene. The distal enhancer region, which contains an estrogen response element, was found to exist in a nucleosome-free region in pituitary-derived cells even in the absence of estrogen. In contrast, the rPRL distal enhancer in fibroblast cells was found to be randomly packaged into nucleosomes. These results indicate that DNA sequence is not sufficient to position nucleosomes in the rPRL gene. Rather, it suggests that cell-specific factors are present in pituitary cells that modify the chromatin structure of the distal enhancer which allow E2R to bind to its response element.

INTRODUCTION

DNA does not exist as a protein-free entity within the eukaryotic nucleus. Instead, eukaryotic DNA is associated with many proteins in large hierarchical complexes called chromatin. The most basic packaging unit of chromatin is the nucleosome, which contains 146 bp of double-stranded DNA wrapped around an octamer core of histone proteins. This packaging naturally limits the accessibility of trans -acting factors to specific DNA sequences. To overcome this impediment, genes that are constitutively expressed must maintain nucleosome-free regions which ensure that transcription factors have access to important DNA regulatory elements. In contrast, the degree of chromatin packaging and, consequently, the degree of transcription factor accessibility to regulatory regions of inducible genes may vary in response to specific stimuli that may be a function of a stage of development or a specific cell type (reviewed in 1 ).

Transcription factors differ in their ability to bind nucleosomal templates. Both GAL4 ( 2 ) and the glucocorticoid-glucocorticoid receptor complex ( 3 - 5 ) can bind their respective cis -acting elements when the DNA sequence is wrapped around a nucleosome, although the affinity of binding is strongly dependent upon both the rotational and translational position of the recognition element with respect to the surface of the nucleosome ( 6 ). In contrast, other transcription factors, including heat shock factor (HSF) ( 2 ), nuclear factor 1 ( 3 , 4 ) and TFIID ( 7 ), are unable to bind their response elements within a nucleosome. To accommodate these varying binding affinities the chromatin structure of genes is organized differently. Under basal conditions a nucleosomal array exists over the glucocorticoid regulatory elements (GRE) in both the mouse mammary tumor virus long terminal repeat in mouse C127 cells ( 8 ) and the tyrosine aminotransferase gene in liver cells ( 9 ). Upon glucocorticoid treatment the nucleosome encompassing the GRE is disrupted, providing access for additional transcription factors to the DNA sequences within this region. In contrast, other genes have transcriptional regulatory elements located in nucleosome-free regions even in the uninduced state. The heat shock protein (HSP) genes are examples of these preset or primed genes. Under non-stress conditions both RNA polymerase II ( 10 ) and the TFIID complex ( 11 ) are bound to HSP promoters, while two HSF binding sites are free of nucleosomes ( 11 , 12 ). Upon heat shock, HSF can bind its response elements located in these nucleosome-free regions leading to activation of transcription.

In vivo, the rat prolactin gene ( rPRL ) is expressed primarily in lactotrophic and somatotrophic cells of the anterior pituitary gland. Expression of the gene is governed by a number of tissue-specific factors and hormones that act through two regulatory domains, the distal enhancer (-1500 to -1800) and the proximal promoter (+1 to -200), located 5' upstream of the transcriptional start site of the gene (reviewed in 13 ). The tissue-specific expression of rPRL appears to be determined to a large extent by the pituitary-specific factor Pit-1 ( 14 - 16 ). In vitro Pit-1 binds to DNA sequences located in both the distal enhancer and the proximal promoter regions of the rPRL gene ( 17 ). The steroid hormone 17[beta]-estradiol (E2) has been shown to induce rPRL transcription ( 18 ). Deletion studies have identified a DNA sequence motif or response element located in the distal enhancer that is required for E2-mediated transcriptional activation ( 19 , 20 ). For E2-mediated induction of rPRL transcription to occur, it is imperative that the E2-estrogen receptor complex (E2R) locate and bind to its response element. In vitro studies by Schlid et al . ( 21 ) demonstrated that the estrogen receptor was incapable of binding to an estrogen response element (ERE) which was occupied by a nucleosome. Therefore, in vivo the accessibility of the ERE to E2R and subsequent induction of transcription may be strongly influenced by the chromatin structure of the distal enhancer.

To investigate the role of chromatin structure in the regulation of rPRL gene expression we have created a series of cell lines derived from rPRL-expressing (rPRL + ) rat pituitary GH3 cells and rPRL-non-expressing (rPRL - ) rat embryo Rat1 fibroblasts that contain bovine papillomavirus vectors as multicopy nuclear episomes ( 22 ). These episomes are packaged in nucleosome-like structures and contain a hybrid gene composed of the initial 1941 bp of the 5' flanking sequences of the rPRL gene driving expression of the Tn 5 gene (rPRL minichromosomes) ( 23 ). In this study we have used these cell lines to determine the accessibility of the distal enhancer region of the rPRL -Tn 5 gene to trans -acting factors. The results from DNase I hypersensitivity studies and nucleosome mapping using micrococcal nuclease (MNase) indicate that the chromatin structure of rPRL is primed for E2-mediated transcriptional activation in pituitary cells, but not in fibroblasts. This priming event occurs independent of E2 and results in the creation of a nucleosome-free region in the rPRL distal enhancer in rat pituitary cells, whereas in rat fibroblasts the distal enhancer is packaged randomly into nucleosomes. Pituitary-specific factors, such as Pit-1, perhaps in combination with unidentified ubiquitous factors, are most likely responsible for priming the rPRL chromatin.

MATERIALS AND METHODS

Cell culture conditions and isolation of nuclei

The creation and characterization of GH3 and Rat1 cell lines that contain the rPRL -Tn 5 -BPV minichromosome have been described ( 22 ). To examine the effects of hormones on chromatin structure and transcription, G1I cells were plated in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum at 3-4 * 10 6 cells/75 cm 2 flask. After 18-24 h the cells were rinsed and incubated in DMEM without phenol red containing 10% newborn calf serum (low steroid medium). Cells were grown for 2-3 days in low steroid medium, refeeding every 24 h, and then challenged with hormone. After 18 h hormone treatment the cells were harvested and resuspended in cell lysis buffer (20 mM triethanolamine-HCl, 60 mM KCl, 15 mM NaCl, 2 mM EDTA, 0.5 mM EGTA, 0.5 mM spermidine, 0.2 mM spermine, 5% sucrose, 7 mM [beta]-mercaptoethanol, 0.25 mM phenylmethylsulfonyl flouride) at a concentration of 1-2 * 10 7 cells/ml. All subsequent procedures were performed at 0-4oC. The cells were lysed by addition of Nonidet P-40 to 0.25% and incubated for 5 min. The nuclei were recovered by centrifugation at 750 g for 5 min. The supernatant was removed and the pellet resuspended in cell lysis buffer at a concentration of 2-4 * 10 7 nuclei/ml. The nuclei were layered over a 3 ml cushion of cell lysis buffer containing 20% sucrose and centrifuged for 15 min at 1000 g . The resulting nuclear pellet was used for chromatin structure analysis.

Transcription rates

G1I and R1L cells were cultured as described above. Nuclei were isolated and rates of Tn 5 gene transcription were determined by run-on transcription assays, as previously described ( 22 ).

Chromatin structure analysis

Deoxyribonuclease I analyses. Deoxyribonuclease I (DNase I) hypersensitivity studies were performed as previously described ( 23 ).

Micrococcal nuclease analyses. Gradient-purified nuclei were washed once with nuclease digestion buffer (10% glycerol, 60 mM KCl, 15 mM NaCl, 20 mM Tris-HCl, pH 7.5, 2.5 mM CaCl 2 , 0.5 mM dithiothreitol), then resuspended in nuclease buffer to an OD 260 of 15. The G1I and R1L chromatin were digested with a range of Mnase (Worthington Biochemical Corporation) concentrations for 5 min at ambient temperature. Purified protein-free (PF) DNA was treated with 10-fold less MNase for 2 min at ambient temperature to serve as a control for MNase digestion. The reactions were stopped with SDS/proteinase K/EDTA and the DNA purified by phenol/chloroform extractions and ethanol precipitation. Twenty micrograms of G1I genomic DNA and 40 [mu]g R1L genomic DNA were used directly to detect nucleosomal arrays flanking the enhancer region or digested with Bgl II and Ase I to determine nucleosome positions in the rPRL distal enhancer. DNA fragments were separated on 2% SeaKem GTG agarose (FMC Bioproducts) gels in 1* Tris/acetate/EDTA buffer and transferred to nylon membranes. Blots containing non-digested DNA were hybridized with a 32 P-labeled 1042 bp Ava I- Bam HI BPV probe and washed as previously described ( 23 ). To perform indirect end-labeling on Ase I and Bgl II-restricted DNA a 204 bp 3' -> 5' probe corresponding to an Ase I- Cfo I rPRL fragment was generated by polymerase chain reaction (PCR) using a 1941 bp Pst I- Pst I rPRL fragment as template and oligonucleotides IX (5'-GCGCTCCTTCCACAAATGAACTG-TA-3') and X (3'-ATTAATAAACAGAGACTAGGAGAAC-5'). Each PCR reaction consisted of 100 pg template, 1 [mu]M oligonucleotide primers, 3 mM MgCl 2 , 200 [mu]M each dATP, dGTP and dTTP, 0.7 [mu]M [[alpha]- 32 P]dCTP (3000 Ci/mmol), 1.3 [mu]M dCTP, 1* Promega Taq buffer (50 mM KCl, 10 mM Tris-HCl, pH 9 at 25oC, 0.1% Triton X-100) and 2.5 U Taq DNA polymerase (Promega) in a total volume of 50 [mu]l. Samples were preheated for 10 min at 95oC, placed on ice for 5 min and then subjected to cycling conditions of 94oC for 60 s, 55oC for 30 s and 72oC for 30 s for three cycles, followed by 27 cycles of 94oC for 15 s, 55oC for 15 s and 72oC for 15 s (BIOS Corporation Biosyclertm oven). Unincorporated nucleotides were separated from the full-length PCR-generated probes using a QIAquick-spin PCR purification kit (Qiagen) according to the manufacturer's directions. Specific activities of 1.0-3.0 * 10 9 c.p.m./[mu]g DNA were obtained. Blots of Bgl II and Ase I-restricted DNA were prehybridized for 1 h at 55oC in 5* SSPE (1* SSPE = 180 mM M NaCl, 5 mM monobasic sodium phosphate, 5 mM dibasic sodium phosphate, 1 mM EDTA, pH 7.0), 5* Denhardt's, 0.5% SDS and 100 [mu]g/ml sheared denatured herring sperm DNA. Hybridization using the 32 P-labeled Ase I- Cfo I probe was carried out for 18 h at 55oC in 5* SSPE, 0.5% SDS, 100 [mu]g/ml sheared denatured herring sperm DNA. These blots were washed sequentially with 2* SSC (1* SSC = 150 mM NaCl, 15 mM sodium citrate), 0.5% SDS for 5 min and 2* SSC, 0.1% SDS for 15 min at room temperature, 0.2* SSC, 0.5% SDS for 45 min at 37oC and 0.1* SSC, 0.1% SDS for 15 min at 65oC. All blots were exposed to Kodak XAR film for 1-4 days with one or two intensifying screens. Autoradiograms were scanned using a laser densitometer (Hoeffer) and analyzed using Macintosh-based Paedia software. Densitometry was performed on autoradiograms exposed within the linear range of the film (normally a 1 day exposure), while the figures shown are from autoradiograms exposed for 3-4 days. Areas under the repetitive signal peaks were determined using the Gaussian method of integration.

RESULTS

Expression of the rPRL -Tn 5 hybrid gene in G1I rat pituitary cells and R1L rat fibroblast cells

rPRL-expressing GH 3 pituitary cells and rPRL-non-expressing Rat1 fibroblasts were transfected with the vector rPRL -Tn 5 -BPV, selected for resistance to G418, and the clonal lines were expanded. The episomal and unrearranged state of the transfected vector within two G418-resistant clones, the GH 3 clone G1I and the Rat1 clone R1L, were verified by restriction enzyme digestion and Southern analysis of total genomic DNA. The rPRL minichromosome exists at a amplified level of ~50 copies/cell in G1I cells, while the copy number in R1L cells is ~25 ( 22 ).

The transcriptional activity of the rPRL -Tn 5 gene in G1I and R1L cells was examined by nuclear transcription run-on assays (Fig. 1 ). The maximum rate of rPRL -Tn 5 transcription in G1I cells was obtained by culturing the cells in media containing 10% fetal bovine serum plus 10 nM E2. This rate was ~20-fold greater than the rate observed in R1L cells cultured under similar conditions. This difference in transcriptional activity of rPRL -Tn 5 cannot fully be explained by the 2-fold lower copy number of rPRL minichromosomes in R1L cells compared with G1I cells, but rather reflects the requirement for pituitary-specific factors for rPRL transcription. Treatment of G1I cells in low steroid calf serum medium with E2 (CS + E2) elevated the rate of rPRL -Tn 5 transcription by ~2-fold compared with the control (CS). In contrast, treatment with E2 has no effect on rPRL -Tn 5 transcription in R1L cells, due to the absence of estrogen receptor in R1L cells ( 24 , 25 ). Treatment of cells with the estrogen antagonist monohydroxytamoxifen (CS + MHT) did not affect the rate of rPRL -Tn 5 transcription in either G1I or R1L cells compared with the control. It is important to note that although there is measurable rPRL- Tn 5 transcriptional activity in R1L cells, there was no measurable transcriptional activity of the endogenous rPRL gene in these cells (Willis and Seyfred, unpublished results; 25 ). The rPRL- Tn 5 activity in R1L cells is probably a result of the selection pressure placed on the cells by the antibiotic G418.


Figure 1 . Expression of the rPRL -Tn 5 hybrid gene in rPRL minichromosome-containing GH3 and Rat1 cells mimics cell-specific endogenous rPRL gene expression. Run-on transcription assays were performed on nuclei isolated from G1I pituitary cells or R1L fibroblast cells cultured in medium containing either 10% fetal bovine serum (FBS) plus 10 nM E 2 (FBS + E2) or 10% calf serum plus either 10 nM E 2 (CS + E2), 10 nM MHT (CS + MHT) or ethanol vehicle (CS) for 18 h. Rates of RNA synthesis (p.p.m./30 min reaction) are given for the Tn 5 gene. The results are the mean +- SE of two separately treated samples from four independent experiments.

Nucleosomal arrays exist on rPRL minichromosomes of G1I rat pituitary cells and R1L rat fibroblast cells

To verify that the rPRL minichromosomes were packaged into chromatin, the rPRL minichromosomes were examined for the presence of nucleosomal arrays using the enzyme MNase, which cuts in the linker regions between nucleosomes. Southern blot analysis of DNA obtained from nuclei digested with MNase revealed a regularly spaced pattern of MNase digestion sites in both the G1I and R1L samples (Fig. 2 ). The distance between the MNase cleavages ranged from 122 to 185 bp, consistent with the presence of nucleosomal arrays. We had previously shown that rPRL minichromosomes in G1I cells were packaged into nucleosomes ( 23 ). Arrays of three to five nucleosomes extending ~1000 bp can be observed using this range of MNase concentrations and agarose gel separation conditions.


Figure 2 . Nucleosomal arrays are present over regions flanking the rPRL enhancer element of rPRL minichromosomes in both GH3 rat pituitary cells and Rat1 fibroblast cells. G1I pituitary cells and R1L fibroblasts were grown in low estrogen medium containing 10% newborn calf serum for 72 h. Isolated nuclei were treated with MNase for 5 min at ambient temperature. Southern blots containing 20 [mu]g purified G1I DNA and 40 [mu]g purified R1L DNA were probed using a 104 bp 32 P-labeled Ava I- Bam HI BPV DNA fragment. Distances between regions of enhanced MNase cleavage representing nucleosomes bound to G1I and R1L DNA are shown on either side of the autoradiogram.

DNase I sensitivity of the rPRL -Tn 5 gene in G1I rat pituitary cells and R1L rat fibroblast cells

The chromatin structure of tissue-specific regulatory elements, such as the enhancer of the serum albumin gene, is different in cells where the enhancer is active than in cells where it is inactive ( 26 ). Similarly, the rPRL 5' regulatory region may have a different chromatin structure in rat pituitary cells where rPRL is expressed than in rat fibroblasts where rPRL is not expressed. We compared the chromatin structure of rPRL minichromosomes from G1I and R1L cells for any gross differences that might support this hypothesis. When the 5'-flanking region of the rPRL -Tn 5 gene in R1L cells was examined for DNase I hypersensitivity in response to E2 or control treatments no regions in the rPRL chromatin were sensitive to DNase I digestion (Fig. 3 ). Three additional independent R1L cell preparations also failed to display any sensitivity to DNase I, even at higher DNase I concentrations (8 U/ml) and temperature (37oC). The 4000 and 3700 bp fragments that were generated were not dependent upon exogenous DNase I and probably represent non-specific binding of the hybridization probe to repetitive DNA sequences. In contrast, two regions in the 5'-flanking region of the rPRL -Tn 5 gene in G1I cells whose centers map to positions -50 and -1650 were hypersensitive to DNase I. In addition, nuclease hypersensitivity of these regions was induced by treatment of the cells with E2, as previously reported ( 23 ). These hypersensitive sites map to the two key transcription regulatory regions of the rPRL gene, the 5' distal enhancer (-1800 to -1500) and the proximal promoter (-200 to +1) ( 13 ).


Figure 3 . The 5'-flanking region of the rPRL gene in the rPRL minichromosomes of Rat1 fibroblasts is refractive to DNase I digestion. Nuclei were isolated from G1I pituitary cells and R1L fibroblast cells grown in medium containing 10% calf serum and treated as described in Figure 1 with ethanol vehicle (C) or E2 (E). The nuclei were digested with increasing concentrations of DNase I for 10 min at 30oC. Ten micrograms of purified R1L DNA and 5 [mu]g purified G1I DNA were digested to completion with Ava I and then subjected to indirect end-labeling analysis using a 1042 bp 32 P-labeled Ava I- Bam HI BPV DNA fragment. DNA fragment sizes listed on the left were obtained by hybridization to a partial Pst I digest of a pBR322-BPV 69 plasmid. The results shown are representative of four independent experiments.

Position of nucleosomes on the rPRL regulatory region in G1I rat pituitary cells and R1L rat fibroblast cells

Although nucleosomal arrays are present on the rPRL minichromosome in both R1L and G1I cells, it is not clear if the positions of specific nucleosomes on the rPRL minichromosome are the same in the two cell types. Since there is a marked difference in expression of the rPRL -Tn 5 gene (Fig. 1 ) and accessibility of the distal enhancer to DNase I (Fig. 3 ) in G1I versus R1L cells, it is of particular interest to determine the position of specific nucleosomes within the distal enhancer element of the rPRL -Tn 5 gene. To map the positions of these nucleosomes nuclei were isolated from G1I and R1L cells that had been grown in low steroid medium and then treated with increasing amounts of MNase. Indirect end-labeling analyses were performed to determine the position of MNase digestion sites by using a 204 bp 32 P-labeled probe complementary to a region 3' of the distal enhancer (Fig. 4 A). As a control for sequence-specific cleavage MNase analyses were also performed on PF genomic DNA isolated from G1I and R1L cells. These digestions used ~10-fold less MNase and were performed at ambient temperature for only 2 min, as compared with 5 min for the chromatin samples. Autoradiograms were analyzed by scanning densitometry to assign positions of the MNase cleavage sites within the rPRL minichromosome (Fig. 4 B and C). Comparisons between rPRL -Tn 5 chromatin and PF DNA were made on samples that were digested to a similar extent by MNase, as determined by ethidium bromide staining of the DNA fragments in agarose gels (data not shown). The densitometric scans obtained from the autoradiograms were normalized using the signal generated by hybridization of the 204 bp 3' probe to a 1675 bp repetitive fragment also observed in GH3 DNA (lane 5, Fig. 4 A). Although the general pattern of MNase cleavage sites was very similar in PF DNA compared with G1I chromatin, the amount of specific cleavage was much greater in the rPRL -Tn 5 chromatin lanes compared with the PF DNA lanes after normalization to the repetitive fragment. These results indicate that specific DNA sequences in the 5' upstream regulatory region of rPRL -Tn 5 chromatin were preferred sites of MNase digestion relative to other regions. In contrast, virtually all DNA sequences in the PF DNA sample displayed a high accessibility to MNase. The moderately intense signals observed in the PF DNA were a result of sequence-specific cleavage due to the high A/T content (~60%) of the region.


Figure 4 . Nucleosomes are specifically positioned over regions flanking the rPRL enhancer element of rPRL minichromosomes in GH3 rat pituitary cells, while they are randomly positioned in R1L fibroblast cells. G1I pituitary cells and R1L fibroblast cells were grown in low estrogen medium containing 10% newborn calf serum for 72 h. Isolated nuclei or PF genomic DNA were treated at ambient temperature with MNase for either 5 or 2 min respectively. Twenty micrograms of purified G1I DNA or 40 [mu]g purified R1L DNA were digested to completion with Ase I and Bgl II. Indirect end-labeling analysis was performed on Ase I/ Bgl II-digested DNA using a 204 bp 32 P-labeled Ase I (-1026)- Cfo I (-1230) rPRL DNA fragment (3' rPRL probe). ( A ) Autoradiogram of Southern blot containing MNase/ Ase I/ Cfo I-digested DNA hybridized with a 32 P-labeled 3' rPRL probe. DNA fragment sizes on the left were obtained by hybridization to a partial Rsa I digest of a Pst I (-1953)- Pst I (-12) rPRL DNA fragment and the corresponding rPRL map units are listed on the right side of the figure. ( B ) Densitometric scan of the autoradiogram presented in (A) corresponding to the G1I PF DNA 32 U/ml MNase lane (solid line) overlaid by a scan corresponding to the G1I chromatin 80 U/ml MNase lane. For comparison signals generated from G1I PF DNA and G1I chromatin were normalized to the peak heights of the 1675 bp repetitive fragments (rep). The areas under the peak representing the repetitive sequences were 8160 and 13 858 units for the G1I PF and G1I chromatin respectively. rPRL map units of MNase fragments formed from G1I pituitary chromatin are listed above the densitometric peaks corresponding to those fragments. ( C ) Densitometric scan of the autoradiogram presented in (A) corresponding to the R1L PF DNA 128 U/ml MNase lane (solid line) overlaid by a scan corresponding to the R1L chromatin 80 U/ml MNase lane. Signals generated from PF DNA and R1L chromatin were normalized to the 1675 bp repetitive fragment (rep) as described in (B). The areas under the peak representing the repetitive sequences were 8748 and 10 187 units for the R1L PF DNA and R1L chromatin respectively. rPRL map units of MNase fragments formed in R1L chromatin are listed above the densitometric peaks corresponding to those fragments.

Regions of enhanced cleavage in the rPRL -Tn 5 chromatin represent DNA sequences in isolated nuclei that were highly accessible to MNase and relatively devoid of proteins or protein complexes. In contrast, regions protected from MNase cleavage reflect DNA sequences that were not accessible to the enzyme as a result of protein or protein complexes binding to the DNA. Patterns of enhanced cleavage followed by protection from MNase digestion in which the protected region covers 140-200 bp are indicative of nucleosome cores binding to the protected region. Regions of enhanced cleavage by MNase represent the linker regions between nucleosomes. Thus in G1I cells nucleosomes could be assigned to positions between -1640 and -1806, -1806 and +57, +57 and +289 and +289 and +484 5' of rPRL . We were unable to map positions of nucleosomes further 5' of rPRL than +484 due to the lack of resolution of the 2% agarose gels for fragments >1300 bp. The position of these nucleosomes was verified using a 276 bp probe that is complementary to a region 5' of the distal enhancer. This probe also allowed detection of nucleosomes 3' of the distal enhancer between -1299 and -1120 and -1120 and -954 (data not shown).

In contrast, there was a striking similarity in the sites of MNase digestion and the degree of digestion of the rPRL minichromosome in R1L PF DNA (R1L PF) compared with R1L chromatin (Fig. 4 A and C). This observation was made in samples of genomic DNA that were digested by MNase to a similar extent, as judged by ethidium bromide staining of the DNA fragments in agarose gels (data not shown). Similar results were obtained when the 276 bp 5' probe was used to map MNase digestion sites (data not shown). This finding is in stark contrast to the results obtained with G1I chromatin, where the sites and extent of MNase digestion of the rPRL minichromosome were very different from G1I PF DNA. Throughout the rPRL enhancer region of rPRL minichromosomes in R1L chromatin MNase digestion sites were observed with a frequency of every 75-115 bp, much too small a fragment to accommodate a nucleosome core. However, the inaccessibility of the 5' regulatory region to DNase I (Fig. 3 ) suggests that nucleosomes are positioned throughout the 5' regulatory region of the rPRL -Tn 5 gene in R1L cells. We interpret the data to indicate that within R1L cells there are multiple populations of rPRL minichromosomes in which nucleosomes are positioned at different sites in one minichromosome molecule versus another. In other words, they are translationally out of phase. With no preferred positioning of nucleosomes all possible sites for MNase cleavage would be accessible in a heterogeneous population of packaged minichromosomes. Each of these sites, relative to the Ase I restriction sites used in the indirect end-labeling analysis, would be detected using short (~200 bp) probes. Thus if a short probe is used, a heterogeneous population of minichromosomes that contain translationally out of phase nucleosomes would give a pattern of MNase digestion sites relative to a specific DNA sequence (restriction site) that would be very similar to PF DNA.

To further illustrate the differences in the positions of nucleosomes across the rPRL enhancer region between G1I rat pituitary cells and R1L rat fibroblast cells densitometric scans obtained from autoradiograms generated by hybridization of the 3' rPRL probe to MNase-treated samples from G1I and R1L chromatin were overlaid (Fig. 5 ). Samples digested with 80 U/ml were chosen for comparison, since the degree of MNase digestion of genomic DNA was similar. The scans were normalized to the signal obtained from the 1675 bp repetitive fragment. From the data in Figure 5 it is clear that specific sites across the rPRL distal regulatory region and extending into the BPV region were much more accessible to MNase digestion in G1I nuclei compared with the same region in R1L nuclei. Particularly, the 3'-end of the enhancer element, which contains the ERE (peaks I-K) was highly sensitive to MNase in G1I cells compared with R1L cells.


Figure 5 . Nucleosomes are positioned differently on DNA sequences 5' of the rPRL distal enhancer in G1I pituitary cells compared with R1L fibroblasts. Autoradiograms of MNase/ Ase I/ Cfo I-treated chromatin samples from Figure 4A were scanned by laser densitometry. Scans corresponding to 80 U/ml MNase (G1I, solid line; R1L, dashed line) were used for both samples. Signals generated from G1I and R1L chromatin were normalized to the 1675 bp repetitive fragment (rep) as described in Figure 4B. The areas under the peak representing the repetitive sequences were 13 858 and 10 178 units for the G1I chromatin and R1L chromatin respectively. Specific regions of enhanced MNase digestion are noted by the letters A-L. The specific locations (in rPRL map units) of MNase cleavage sites within these regions are listed below.

Effect of estrogen treatment on nucleosome positions

In G1I cells sites of enhanced MNase digestion were located at -1436, -1498 and -1596 within the distal enhancer element (Fig. 4 B). However, the distance between these sites is <146 bp, the length of DNA that is wrapped around an intact nucleosome core. Similar patterns of MNase cleavage throughout the enhancer region, as well as in regions on either side of the enhancer, were obtained from nuclei isolated from E2-treated cells compared with the control (Fig. 6 ). Taken together, these results demonstrate that the distal enhancer element of rPRL -Tn 5 in rat pituitary cells is not ordered into a nucleosomal array even in the absence of E2. Furthermore, E2 treatment does not alter the positions of nucleosomes on either side of the enhancer element.


Figure 6 . Estrogen treatment of G1I cells has no affect on the position of nucleosomes on the 5' regulatory region of the rPRL -Tn 5 gene. Autoradiogram of cleavages generated by MNase treatment of G1I PF DNA and G1I chromatin from control or E2-treated G1I cells. Nuclei were isolated from G1I pituitary cells grown in low estrogen medium and treated with either ethanol vehicle (C-chromatin) or 10 nM E2 (E2-chromatin) for 16-18 h. Conditions of MNase digestion of isolated nuclei and G1I PF DNA and indirect end-labeling analysis are described in Figure 4. DNA fragment sizes on the left were obtained by hybridization to a partial Rsa I digest of a Pst I (-1953)- Pst I (-12) rPRL DNA fragment. rPRL map units assigned to MNase fragments generated in E2-chromatin are listed on the right side of the figure. The results shown are representative of three independent experiments.

DISCUSSION

Although eukaryotic DNA is packaged into nucleosomes, DNA sequences involved in modulating transcription must be available for binding of regulatory factors. After locating its response element a transcription factor must undergo a requisite combination of productive protein-DNA interactions and protein-protein interactions to evoke transcriptional activation. The promoters of some genes have nucleosomes positioned over important regulatory elements prior to gene activation. In the yeast PHO5 gene positioned nucleosomes inhibit trans -acting factor accessibility to the TATA box and one DNA binding site for the transcription factor PHO4 ( 27 ). When phosphate starvation occurs PHO4 is activated and is able to bind its response elements in the PHO5 gene promoter, including the response element located in a nucleosome. As a consequence of PHO4 binding four nucleosomes are disrupted and this remodeling permits gene transcription ( 28 ). In contrast, other promoters are packaged in a chromatin structure such that important cis -acting elements are always accessible, even when trans -acting factors are not present. In these primed genes the positions of nucleosomes on the promoter are not altered as a result of transcription factor binding. An example of a primed gene is the HSP26 gene. Prior to activation of transcription by heat shock this gene contains two DNase I hypersensitive sites in the 5' promoter region ( 12 ). Within these sites are the DNA sequences recognized by the key transcription factors HSF and GAGA factor. Thus in the uninduced state the chromatin structure of the HSP26 gene allows for free access of HSF and GAGA to their DNA binding sites. Similarly, our data suggest that the 5' rPRL gene is a primed gene in rat pituitary cells. The region surrounding the ERE in the 5' distal enhancer appears to be free of positioned nucleosomes, allowing free access of ER and possibly other transcription factors to their DNA binding sites.

There are not only cell-specific differences in transcription of the rPRL gene, but also cell-specific differences in the chromatin structure surrounding the key regulatory elements in the 5' upstream regions of the gene. Both the distal enhancer and proximal promoter regions were hypersensitive to DNase I in rat pituitary cells, yet were refractive to DNase I in rat fibroblasts (Fig. 3 ; 23 ). Furthermore, positioned nucleosomes were mapped to sites flanking the distal enhancer in rat pituitary cells, whereas in rat fibroblasts the nucleosome array spanning this region was found to be translationally out of phase between different populations of nucleosomes (Figs 4 and 5 ). These data suggest that pituitary-specific factors act to alter the chromatin structure of the rPRL gene. The pituitary specific factor Pit-1 is required not only for basal expression ( 25 , 29 ), but also for E2-induced activation of rPRL transcription ( 24 , 30 ). Although its mechanism of action is not clear, Pit-1 may function in establishing a nucleosome-free region in the rPRL promoter and enhancer of pituitary cells, allowing binding of additional trans -acting factors. However, factors in addition to Pit-1 may also be required. Transient expression of Pit-1 in Rat1 cells failed to activate endogenous rPRL gene transcription (M. A. Seyfred, unpublished observation; 25 ). In addition, Pit-1 is found not only in lactotrophs, where the rPRL gene is expressed, but also in thyrotrophs, where rPRL is not expressed ( 29 ). These candidate factors include mammalian homologs of the yeast SWI/SNF protein complex, which have been implicated in presetting the chromatin structure of primed genes ( 31 , 32 ). The SWI/SNF proteins have also been shown to interact with the ER and affect the E2 responsiveness of reporter genes in yeast ( 33 ). Thus Pit-1 may act to localize mammalian SWI/SNF homologs through protein-protein interactions and aid in altering the chromatin structure of the rPRL gene regulatory elements to permit subsequent gene activation.

ACKNOWLEDGEMENTS

The authors would like to thank Dr Roger Chalkley for reviewing this manuscript and Dr Katherine Cullen for many helpful discussions. This work was supported by NIH grants 2T32CAO9385 and 5T32HD07043 (SDW), NIH grant DK-42371 (MAS) and NSF grant BIR-9419667.

REFERENCES

1 Wolffe,A.P. (1994) Regulation of Chromatin Structure and Function. R.G.Landes Co, Austin, TX, pp. 33-47.

2 Taylor,I.C.A., Workman,J.L., Schuetz,T.J. and Kingston,R.E. (1991) Genes Dev., 5, 1285-1298. MEDLINE Abstract

3 Pina,B., Bruggemeier,U. and Beato,M. (1990) Cell, 60, 719-731. MEDLINE Abstract

4 Archer,T.K., Cordingley,M.G., Wolford,R.G. and Hager,G.L. (1991) Mol. Cell. Biol., 11, 688-698. MEDLINE Abstract

5 Perlmann,T. and Wrange,O. (1988) EMBO J., 7, 3073-3079. MEDLINE Abstract

6 Li,Q. and Wrange,O. (1993) Genes Dev., 7, 2471-2482. MEDLINE Abstract

7 Workman,J.L. and Roeder,R.G. (1987) Cell, 51, 613-622. MEDLINE Abstract

8 Richard-Foy,H. and Hager,G.L. (1987) EMBO J., 6, 2321-2328.

9 Carr,K.D. and Richard-Foy,H. (1990) Proc. Natl. Acad. Sci. USA, 87, 9300-9304. MEDLINE Abstract

10 Rougvie,A.E. and Lis,J.T. (1990) Mol. Cell. Biol., 10, 6041-6045. MEDLINE Abstract

11 Thomas,G.H. and Elgin,S.C.R. (1988) EMBO J., 7, 2191-2201. MEDLINE Abstract

12 Cartwright,I.L. and Elgin,S.C.R. (1986) Mol. Cell. Biol., 6,779-791. MEDLINE Abstract

13 Gourdji,D. and Laverriere,J.-N. (1994) Mol. Cell. Endocrinol., 100, 133-142. MEDLINE Abstract

14 Ingraham,H.A., Chen,R., Mangalam,H.J., Elsholtz,H.P., Flynn,S.E., Lin,C.R., Simmons,D.M., Swanson,L. and Rosenfeld,M.G. (1988) Cell, 55, 519-529. MEDLINE Abstract

15 Mangalam,H.J., Albert,V.R., Ingraham,H.A., Kapiloff,M., Wilson,L., Nelson,C., Elsholtz,H. and Rosenfeld,M.G. (1989) Genes Dev., 3, 946-958. MEDLINE Abstract

16 Fox,S.R., Jong,M.T.C., Casanova,J., Ye,Z.-S., Stanley,F. and Samuels,H.H. (1990) Mol. Endocrinol., 4, 1069-1080. MEDLINE Abstract

17 Nelson,C., Albert,V.R., Elsholtz,H.P., Lu,L.I.-W. and Rosenfeld,M.G. (1988) Science, 239, 1400-1405. MEDLINE Abstract

18 Maurer,R.A. (1982) J. Biol. Chem., 257, 2133-2136. MEDLINE Abstract

19 Maurer,R.A. and Notides,A.C. (1987) Mol. Cell. Biol., 7, 4247-4254. MEDLINE Abstract

20 Waterman,M.L., Adler,S., Nelson,C., Greene,G.L., Evans,R.M. and Rosenfeld,M.G. (1988) Mol. Endocrinol., 2, 14-21 MEDLINE Abstract

21 Schlid,C., Claret,F.-X., Wahli,W. and Wolffe,A.P. (1993) EMBO J., 12, 423-433. MEDLINE Abstract

22 Seyfred,M.A., Kladde,M.P. and Gorski,J. (1989) Mol. Endocrinol., 3, 305-314. MEDLINE Abstract

23 Seyfred,M.A. and Gorski,J. (1990) Mol. Endocrinol., 4, 1226-1234. MEDLINE Abstract

24 Day,R.N., Koike,S., Sakai,M., Muramatsu,M. and Maurer,R.A. (1990) Mol. Endocrinol., 4, 1964-1971. MEDLINE Abstract

25 Kaneko,K.J., Gelinas,C. and Gorski,.J (1993) Biochemistry, 32, 8348-8359. MEDLINE Abstract

26 McPherson,C.E., Shim,E.-Y., Friedman,D.S. and Zaret,K.S. (1993) Cell, 75, 387-398. MEDLINE Abstract

27 Almer,A., Rudolph,H., Hinnen,A. and Horz,W. (1986) EMBO J., 5, 2689-2691. MEDLINE Abstract

28 Ventor,U., Svaren,J., Schmitz,J., Schmid,A. and Horz,W. (1994) EMBO J., 13, 4848-4855. MEDLINE Abstract

29 Simmons,D.M., Voss,J.W., Ingraham,H.A., Holloway,J.M., Broide,R.S., Rosenfeld,M.G. and Swanson,L.W. (1990) Genes Dev., 4, 695-711. MEDLINE Abstract

30 Nowakowski,B.E. and Maurer,R.A. (1994) Mol. Endocrinol., 8, 1742-1749. MEDLINE Abstract

31 Hirschhorn,J.N., Brown,S.A., Clark,C.D. and Winston,F. (1992) Genes Dev., 6, 2288-2298 MEDLINE Abstract

32 Kwon,H., Imbalzano,A.N., Khavarl,P.A., Kingston,R.E. and Green,M.R. (1994) Nature, 370, 477-481. MEDLINE Abstract

33 Yoshinaga,S.K., Peterson,C.L., Herskowitz,I. and Yamamoto,K.R. (1992) Science, 258, 1598-1603. MEDLINE Abstract

Return

* To whom correspondence should be addressed
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Cancer Res.Home page
C.-C. Wang, M.-F. Tsai, T.-H. Dai, T.-M. Hong, W.-K. Chan, J. J.W. Chen, and P.-C. Yang
Synergistic Activation of the Tumor Suppressor, HLJ1, by the Transcription Factors YY1 and Activator Protein 1
Cancer Res., May 15, 2007; 67(10): 4816 - 4826.
[Abstract] [Full Text] [PDF]

Home page
 J. Cell Sci.Home page
Z. D. Sharp, M. G. Mancini, C. A. Hinojos, F. Dai, V. Berno, A. T. Szafran, K. P. Smith, T. T. Lele, D. E. Ingber, and M. A. Mancini
Estrogen-receptor-{alpha} exchange and chromatin dynamics are ligand- and domain-dependent
J. Cell Sci., October 1, 2006; 119(19): 4101 - 4116.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Endocrinol.Home page
P. Kievit and R. A. Maurer
The Pituitary-Specific Transcription Factor, Pit-1, Can Direct Changes in the Chromatin Structure of the Prolactin Promoter
Mol. Endocrinol., January 1, 2005; 19(1): 138 - 147.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
X. Yang, L. Taylor, and P. Polgar
Mechanisms in the Transcriptional Regulation of Bradykinin B1 Receptor Gene Expression. IDENTIFICATION OF A MINIMUM CELL-TYPE SPECIFIC ENHANCER
J. Biol. Chem., April 24, 1998; 273(17): 10763 - 10770.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (145K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (10)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Willis, S.
Right arrow Articles by Seyfred, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Willis, S.
Right arrow Articles by Seyfred, M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?