| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Testis |
Research Service (151), Overton Brooks Veterans Administration Medical Center, Shreveport, Louisiana 71101-4295
Department of Biochemistry and Molecular Biology, Louisiana State University Health Science Center, Shreveport, Louisiana 71130-3932
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
developmental biology, gene regulation, spermatogenesis, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
60%) in primary spermatocytes, with H1a and H1c contributing approximately 30% and 10%, respectively [1]. Histone H1 synthesis is undetectable following the primary spermatocyte stage, but H1 and core histones are maintained through two meiotic divisions. The histones are replaced by transition proteins during the midspermatid stage of spermatogenesis. [2]. Histone H1t expression is regulated by stringent transcriptional control. In nongerminal cells, H1t transcription is repressed with contributions from silencing elements found upstream [3] and within the H1t proximal promoter [4] as well as by methylation at specific sites within its proximal promoter [5]. Although the proximal promoter contains all of the consensus elements common to the somatic promoters, including an H1/AC box, a GC box, an H1/CCAAT box, and a TATA box, transcription of the H1t gene is not detected in cells other than testis pachytene primary spermatocytes in animal studies [6–8]. Reporter gene expression, promoted by the testis-specific histone H1t promoter, occurs at a low level in somatic cell lines following transient transfections [8–10].
H1t synthesis is dependent on transcriptional upregulation of the gene in pachytene primary spermatocytes [7, 8]. Regulatory sequence elements that are necessary and sufficient for spermatocyte-specific expression of histone H1t are present within 2.5 kilobases (kb) of upstream DNA and 3.8 kb of downstream DNA flanking the H1t transcriptional start site [11]. The promoter, truncated to within 141 base pairs (bp) of the transcriptional start site in constructs containing the native gene and 800 bp of downstream sequence, maintains testis-specific activity in transgenic mice [12]. Previous studies revealed the importance of the proximal promoter and of testis element (TE) located between the H1/AC box and the H1/CCAAT box [13, 14]. When TE is replaced by a heterologous DNA sequence, expression of the mutant rat H1t gene is disrupted in transgenic mice [11, 15].
In the present study, we examine a repressor element (RE), which is located upstream from the H1/AC box. The DNase I footprinting and electrophoretic mobility shift assays (EMSAs) show that nuclear proteins from primary spermatocytes bind to the 5'-end of the element to generate a unique protein-DNA complex. This testis complex is formed by proteins from the cells in which the H1t gene is transcribed. The level of the testis complex drops in nuclear extracts from early spermatids, which correlates with downregulation of the gene. The EMSA competitions were used to delineate regions within the 5'-end of this bifunctional RE that is responsible for specific nuclear protein binding. Ultraviolet (UV) cross-linking of nuclear proteins bound to the RE revealed three protein/DNA electrophoretic bands with relative molecular weights of 59, 52, and 50 kDa. These binding proteins are candidates for histone H1t transcriptional activators.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animal studies were conducted in an AAALAC-accredited facility in accordance with the Guiding Principles for the Care and Use of Research Animals promulgated by the Society for the Study of Reproduction.
Reagents and Supplies
Radiochemicals were purchased from New England Nuclear (Boston, MA). X-OMAT XAR-5 x-ray film (Eastman Kodak, Rochester, NY) was purchased from Sigma (St. Louis, MO). Deoxynucleotides and Ampli-Taq for polymerase chain reaction (PCR) were purchased from Perkin-Elmer (Foster City, CA). The DNA sequence used for site determination in DNase I footprinting was performed using 32P-labeled sequencing primers with the fmol DNA Cycle Sequencing System from Promega (Madison, WI). T4 polynucleotide kinase was obtained from New England Biolabs (Beverly, MA). Oligonucleotides were purchased from Genosys (Woodlands, TX).
Oligonucleotides
Primers for PCR amplification included a primer (5'-CAGAGAGCACACATGCAA-3') located from 126 to 143 bp upstream from the H1t transcriptional start site and a primer (5'-AAGTACAGGAGAGCGAAC-3') located from 28 to 8 bp downstream from the H1t transcriptional initiation point. Before amplification, the 5'-terminus of the upstream primer was labeled using [
32P]ATP with T4 polynucleotide kinase [16]. Following PCR amplification (30 cycles of 94°C for 2 min, 54°C for 1 min, and 72°C for 1 min) with plasmid pPS3 as template [8], the 174-bp product was isolated from a 0.7% low-melting agarose gel [17] before use in footprinting. Double-stranded fragments for EMSA probes and for competitions were prepared from oligonucleotides based on sequences shown in Figure 1.
|
In Vitro DNase I Footprinting
The DNase I footprinting reactions were performed using nuclear extracts from unfractionated testis cells and enriched populations of pachytene primary spermatocytes. The DNase I digestions were conducted in a total volume of 40 µl as previously described [3]. Following DNase I digestion, the samples were extracted twice with phenol:chloroform (1:1) and DNA precipitated from the aqueous phase with 2.5 volumes of ethanol at -20°C for 1 h. The DNA was recovered by centrifugation for 20 min at 16 000 x g at 4°C. Ethanol was removed and the DNA pellet was dried partially and dissolved in 3 µl of loading solution before electrophoresis [3]. Samples were heated at 90°C for 3 min before loading onto a denaturing polyacrylamide sequencing gel (6% polyacrylamide [19:1, acrylamide: bisacrylamide] containing 7 M urea and 1x TBE [0.09 M TrisBorate, 0.002 M EDTA, pH 8.3]). After electrophoresis at 1200 mV for 120–165 min, the gels were soaked for 20 min in a solution containing 10% methanol and 10% acetic acid before transfer to Whatman 3MM paper (Whatman Laboratory Products, Clifton, NJ) and drying. Dried gels were exposed to Kodak X-OMATIC XAR 5 film using a Sigma Lite Plus intensifying screen (Sigma Chemical Company, St. Louis, MO).
Preparation of Nuclear Extracts and EMSAs
Nuclear protein extracts were prepared from crude nuclei derived from several rat organs (testis, liver, brain, spleen, and kidney) as previously described [13, 14, 18]. Nuclear extracts were also prepared from enriched populations of rat pachytene spermatocytes and early spermatids obtained by centrifugal elutriation as previously described [8]. Nuclear extracts enriched for the testis-binding proteins were obtained by ion-exchange chromatography. Briefly, 1 ml of testis nuclear extract was fractionated by binding to 0.5 ml of phosphocellulose P11 in a batch procedure in a 1.5-ml microfuge tube at 4°C. After addition of testis extract, the phosphocellulose was washed five times (0.5-ml washes) with Dignam Buffer D containing 0 M KCl [18]. The phosphocellulose was then washed with Buffer D containing 0.3 M KCl to elute proteins. The 0.3 M KCl fraction was dialyzed against Buffer D containing 0 M KCl to obtain an extract that is enriched for proteins that binds to the RE. Binding of nuclear proteins to labeled DNA probes for EMSA was performed on ice as described previously [13] in a total volume of 20 µl with 500 ng of the nonspecific competitor poly dG-dC added to each assay. When unlabeled specific competitor DNA was used, it was added 15 min before binding of nuclear proteins to labeled probe. Following binding to radiolabeled probe, samples were electrophoresed through 4% polyacrylamide gels (60:1, acrylamide:bisacrylamide) using low-ionic-strength buffer containing 6 mM Tris-HCl (pH 7.9), 3.3 mM sodium acetate, and 1 mM EDTA with buffer recirculation. After electrophoresis, gels were dried, exposed to a SR phosphor screen, and processed using a Cyclone PhosphorImager (both from Packard Instrument Company, Meriden, CT).
Cross-Linking of Protein and DNA
Proteins bound to radiolabeled DNA in a 40-µl EMSA reaction (twice the standard EMSA reaction) were cross-linked to the DNA by three pulses of 254-nm UV light (120 mJ/cm2) in a Stratalinker 1800 (Stratagene, La Jolla, CA). The SDS-PAGE sample loading mix was added to give a final concentration of 0.06 M Tris (pH 6.8), 2% (w/v) SDS, 0.72 M 2-mercaptoethanol, and 0.0005% bromophenol blue [19]. Samples were boiled for 5 min and loaded on an SDS polyacrylamide gel (width, 8 cm; length, 7 cm; 12.0% separating gel with a 5% stacking gel) and electrophoresed at 200 V for 1.25 h in buffer (pH 8.3) containing 25 mM Tris, 192 mM glycine, and 0.1% SDS. The gel was dried at 60°C, exposed to a SR phosphor screen, and processed using a Cyclone PhosphorImager. The relative molecular masses of the protein/DNA adducts were determined using Quantity One software (Bio-Rad, Hercules, CA) and Precision Plus Protein standards (Bio-Rad).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The testis-specific H1t promoter sequence has been described previously [7, 8, 20], and a schematic diagram of the promoter elements essential for transcription are presented in Figure 1. TE, located between the CCAAT box and the AC box, is essential for spermatocyte-specific transcription of the H1t gene in transgenic mice. A new element, discovered in part by DNase I footprinting, is located adjacent to and upstream from the H1/AC box between nucleotides -129 and -107. It was designated the RE, because the 3'-end serves as a binding site for a transcriptional repressor in some nongerminal cell lines [20, 39].
A footprint covered most of the RE and part of the AC box, spanning nucleotides -125 through -100, when nuclear protein binding to the proximal promoter was examined using in vitro DNase I footprinting. Even the lowest concentration of nuclear proteins, derived from an enriched population of pachytene primary spermatocytes, generated a footprint, as shown on the right in Figure 2. The nuclease digestion pattern of free DNA is shown in the left lane of each image. The remaining lanes show the digestion patterns produced by protein binding using increasing amounts of nuclear proteins. Spermatocyte nuclear proteins also produced footprints over the TE that is located between the CCAAT box and the AC box.
|
Footprints over the RE and TE were weaker or absent when using proteins from nuclear extracts from unfractionated testis cells, as shown on the left in Figure 2. This probably results from a lower concentration of the binding proteins from primary spermatocytes in the unfractionated testis cell population. An enrichment in primary spermatocytes of at least 10-fold was found when comparing the concentration of spermatocytes in a population of unfractionated testis cell to a population of cells enriched in primary spermatocytes by centrifugal elutriation. Therefore, the binding proteins likely were derived from primary spermatocytes.
RE Is Conserved in Mouse, Rat, and Human Histone H1t Promoters
We used the rat testis-specific histone H1t promoter in the DNase I footprint experiments shown in Figure 2. Previous studies identified several sequence elements within the proximal promoter of this gene, starting upstream at the H1/AC box and ending at the GC box 2 located within the mRNA leader region downstream from the TATA box [20]. Several promoter elements important for regulating transcription of the rat H1t gene include the H1/AC box, the H1t/TE element, the H1/CCAAT box, the TATA box, and the GC box 2. All of these elements, except the GC box 2, are conserved in mice and humans.
To examine the sequence conservation of the RE in mammals, we aligned the known mouse, rat, and human H1t promoter sequences as shown in Figure 1. A significant degree of conservation of the element is found. The human and mouse sequences are 65% identical within the 23-bp boxed region; the mouse and rat sequences are 91% identical. Bases that are identical to the rat sequence are indicated by bold characters. A 6-bp core region (ACTTTT) is completely conserved among the species examined.
Primary Spermatocyte Nuclear Proteins Bind Specifically to the RE
We examined the binding affinity of nuclear proteins to the RE promoter element using EMSA. In an initial experiment, we compared binding of nuclear proteins derived from rat testis to binding of nuclear proteins derived from several other organs. The left lane of Figure 3 shows the mobility of free RE probe. The remaining lanes, from left to right, show binding of nuclear proteins from testis, liver, brain, spleen, and kidney. Testis proteins produce two low-mobility bands that are marked in the figure. The upper band, designated testis complex, is not seen in other tissues. A higher-mobility testis band, marked with an asterisk, represent nonspecific binding based on competition assays [39]. Other tissues produce only higher-mobility, nonspecific bands.
|
In another experiment, we examined binding of nuclear proteins derived from enriched populations of pachytene primary spermatocytes and early spermatids as shown on the right in Figure 3. The two major low-mobility bands produced with testis nuclear proteins were produced by these two protein samples, but the intensity of the testis complex was maximal when proteins were derived from primary spermatocytes. The intensity of the testis complex band dropped approximately 70%–80% when proteins were derived from early spermatids as determined by densitometric scanning of the digital images. Therefore, proteins that form the testis complex are germinal cell-type specific and are produced largely in pachytene spermatocytes. A strong correlation is found between the cell type where the testis-specific histone H1t gene is transcribed and where we see strong binding of nuclear proteins to the RE element of the H1t promoter to form the testis complex.
Spermatocyte Nuclear Protein Binding Occurs Within the 5'-End of the Bifunctional RE
The EMSA competition experiments were conducted to identify regions within the RE element responsible for binding nuclear proteins to produce the testis complex. In these experiments, a wild-type, double-stranded RE element and three mutant RE elements were prepared as shown on the left in Figure 4. The mutations were made by replacing regions of wild-type sequence that were 6 bp in length with heterologous sequences within the 3' end of the element (Mut1), within the central region of the element (Mut2), or within the 5' end of the element (Mut3) as shown in Figures 1 and 4.
|
The first lane (P) in Figure 4A shows free wild-type probe, and the second lane (C) shows binding of testis nuclear extract to the wild-type probe. The next four lanes show competitions with 20 µg of wild-type (Wt), Mut1 (M1), Mut2 (M2), and Mut3 (M3) unlabeled DNA competitors, respectively. An excess of wild-type or Mut1 DNA competes binding to the testis complex and the nonspecific low-mobility band. Higher-mobility bands are not competed with these probes. The Mut3 probe competes binding only to the major low-mobility, nonspecific band. On the other hand, Mut2 fails to compete the testis complex and the low-mobility, nonspecific band.
The testis complex appears to be generated by the binding of nuclear proteins to the region covering Mut2 and Mut3. Therefore, we expected to see normal binding to the Mut1 probe that contains an altered protein-binding site in the 3'-end of the RE probe. To confirm that nuclear proteins bind to the Mut1 RE probe, EMSAs were conducted comparing the binding patterns of the wild-type and Mut1 probes. In Figure 4B, the binding pattern produced by the wild-type probe is shown in lane 2 (Wt), and the binding pattern produced by the Mut1 probe is shown in lane 3. The banding patterns using these two probes are identical, suggesting that testis proteins can bind to RE that has a mutation in the 3'-end of the element, as indicated in the model shown in Figure 5B.
|
Characterization of the Testis Proteins that Bind to the RE Promoter Element
Nuclear protein extracts from unfractionated testis cells were used in the same type of DNA-binding reaction that was used for EMSA. Bound proteins were cross-linked to the radioactive DNA probe by UV irradiation and were separated by SDS-PAGE. The major band seen in a phosphorimage of the dry gel had an apparent molecular weight of 59 kDa, as shown in the left lane on the left in Figure 5A. Two lower-intensity bands with apparent molecular weights of 52 and 50 kDa were also formed. The testis nuclear protein sample used in this experiment, designated P11 +C, was partially purified by binding to phosphocellulose P11, washed, and eluted with 0.3 M KCl before the binding reaction. The +C indicates that nonspecific competitor was added to the binding reaction. The binding patterns of these partially purified proteins are indistinguishable from the original proteins, but the cleanup step lowers the background significantly in the cross-linking experiment. A faint band of approximately 250 kDa was formed when competitor was omitted, as shown in the second lane (P11 -C).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Within the H1t promoter, additional elements are important for testis-specific expression of the histone H1t gene [4, 6, 10, 11, 15, 20]. These include those that are involved in transcriptional activation of the gene in primary spermatocytes and those that are involved in transcriptional repression in other cell types. An H1t promoter element, designated TE, which is essential for activation of testis-specific transcription of the H1t gene, is located between the H1/AC box and the H1/CCAAT box [11, 13–15, 28] (Fig. 1). This element is composed of three subelements: TE1, TE2, and GC box 1.
Experiments in somatic cells indicate that both the TE1 subelement and the GC box 1 are binding sites for transcriptional activators [28, 29]. Mutation of either site reduces transcription significantly in transient expression assays [28]. However, mutation of the TE2 subelement leads to transcriptional activation in some cell lines, indicating that this site binds a transcriptional repressor [30]. A GC-rich sequence located within the leader region of the rat promoter downstream from the TATA box has been reported to function as a transcriptional repressor-binding site in some nongerminal cell lines [4]. Most of the H1t promoter regulatory elements are located within the proximal promoter, but some upstream elements also contribute to transcriptional regulation. For example, we identified a silencer region located between 875 and 948 bp upstream from the transcription initiation site [3]. In addition, an upstream TG-rich element has been identified as a binding site for a transcriptional activator [31].
To identify additional H1t promoter elements involved in either transcriptional activation or repression, we examined a small region of the H1t promoter located immediately upstream from the AC box. We identified an element, designated RE, that serves as a binding site for factors that lead to transcriptional repression in some nongerminal cell lines [39], and we found sequence conservation among the rat, mouse, and human REs. When the 3'-end of this RE is mutated, promoter activity increases in transiently transfected cell lines. Therefore, the 3'-end of the RE is conserved and appears to be involved in transcriptional repression in nongerminal cells [20, 39].
The initial evidence for binding of testis nuclear proteins to the H1t proximal promoter region adjacent to and upstream from the H1t/AC box was obtained by DNase I footprint analysis. A footprint was formed over the RE region by the binding of nuclear proteins derived from unfractionated testis cells and from an enriched population of pachytene primary spermatocytes. Although some minor differences were observed in the footprint patterns of proteins from testis and spermatocytes, these likely resulted from additional proteins present in the unfractionated testis sample derived from nongerminal cells and from germinal cells that do not express the H1t gene.
Because footprint analyses indicated binding of nuclear proteins to this region of the H1t promoter, we wanted to confirm the testis-specific binding to this site and examine in more detail the interaction of nuclear proteins with this region of the promoter. Specific binding of testis proteins to this region of the promoter was confirmed using EMSA. When assays were conducted using nuclear proteins from testis cell populations enriched in either primary spermatocytes and early spermatids, spermatocytes contained the highest level of binding activity, correlating with the maximal transcriptional activity of the H1t gene. We are not aware of another example in which a transcription factor decreases during the transition from pachytene spermatocytes to early spermatids.
The EMSA competition assays show that the testis complex is formed only by the binding of nuclear proteins to the 5'-end of the RE. Competition with Mut2 and Mut 3 (Fig. 4) fail to compete binding to the testis complex, but competition with Mut1 eliminates the testis complex, as shown in Figure 4A. It is possible that more that one protein may bind to this 5'-region. One protein could bind specifically to the sequence within the Mut3 region, and another could bind within the Mut2 region. If multiple proteins bind to these sites, the proteins may associate with each other so that a mutation in the sequence of either Mut3 or Mut2 alone will not eliminate the complex. It may be necessary to mutate both Mut3 and Mut2 to eliminate formation of the testis complex. Alternatively, the binding protein may bind to a region between the Mut3 and Mut2 sites.
Nuclear proteins bind to the Mut1 probe just as they do to the wild-type RE probe (Fig. 4B), and a model showing how these testis proteins might bind to the RE is shown in Figure 5B. In a preliminary protein-DNA cross-linking study to characterize the binding proteins, masses of the binding proteins appear to range from 50 to 60 kDa. If the binding activity contains multiple components, the entire complex may be larger, and this may explain the large complex of approximately 250 kDa that is seen when nonspecific competitor DNA is removed from the binding assay, as shown in lane 2 of Figure 5A. It is also possible that only one protein component binds to the DNA and that there may not be higher-molecular-weight bands on the gel (UV cross-linking preferentially cross-links protein to DNA rather than protein to protein). Our cross-linking conditions may make it likely that even if more than one protein is involved, only one will be cross-linked to the DNA.
Potential transcription factor consensus-binding sites are within the RE. An Oct-1 site is found within rat, mouse, and human REs, and an AP1 site is found within the rat and human REs (Fig. 5B). Both factors can form multicomponent complexes [32, 33]. Multiple testis isoforms of Oct-1 are somewhat shorter than the full-length, 766-amino acid protein [32, 34], but the exact sizes are unknown. Oct-4, which recognizes the same DNA sequence, is shorter and migrates on SDS-PAGE as a 45- to 50-kDa protein. Fos, a component of AP1, is important for spermatogenesis [35]. Mammalian c-fos is 380 amino acids in length [36], but isoforms migrate on SDS-PAGE as 40- to 60-kDa proteins. Fos B is 338 amino acids in length and migrates as a 35- to 40-kDa protein [37].
In conclusion, transcriptional regulation of the testis-specific histone H1t gene is complex. Transcription is upregulated in primary spermatocytes by binding of activators to specific promoter elements and is downregulated in other cell types by binding of repressors to the same or to other elements. The 3'-end of the bifunctional RE serves as a binding site for a transcriptional repressor in nongerminal cells, and the 5'-end serves as a binding site for a transcriptional activator in primary spermatocytes (Fig. 5B). In the present study, we show that nuclear proteins derived from primary spermatocytes bind to the 5'-end of the RE. Only nuclear proteins from pachytene primary spermatocytes bind tightly to this element to generate a low-mobility testis complex band in EMSAs. It is also possible that transacting factors from primary spermatocytes that bind to the 5'-end of the RE serve as transcriptional coactivators along with factors that bind to the AC box, the GC box 1, the TE1 subelement, and the CCAAT box.
In previous studies, we used transient transfection assays to examine activities of wild-type and mutant H1t promoters in nongerminal cells. With development of new germinal cell lines that can differentiate and progress through spermatogenesis [38], it may be possible to repeat these assays in germinal cells to examine normal and mutant H1t promoter constructs to more clearly determine the importance of the RE to transcription of the H1t gene. If this is not possible, then we can test the importance of the RE for enhanced testis-specific transcription of the H1t gene in transgenic mice [15].
|
FOOTNOTES |
|---|
2 Correspondence: Sidney R. Grimes, Medical Research Service (151), Overton Brooks Veterans Administration Medical Center, 510 E. Stoner Ave., Shreveport, Louisiana 71101-4295. FAX: 318 429 5733; srgrimes{at}prysm.net
Received: 9 December 2002.
First decision: 22 December 2002.
Accepted: 21 January 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Chowdhary, R. A. Ali, W. Albig, D. Doenecke, and V. B Bajic Promoter modeling: the case study of mammalian histone promoters Bioinformatics, June 1, 2005; 21(11): 2623 - 2628. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
