| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Chemistry and Biochemistry and the School of Medicine, University of South Carolina, Columbia, South Carolina 29208
ABSTRACT
H1t is a testis-specific variant histone 1 gene transcribed in pachytene spermatocytes. As part of a program to understand its transcriptional control, we have investigated the effect of the cap-proximal, GC-rich silencer element in the context of various lengths of upstream sequence. By transient transfection of NIH 3T3 cells, we showed that a targeted mutation in the silencer has a large (>10-fold) effect on reporter gene expression, regardless of the length of upstream sequence present. No other discrete silencing activity was observed in the upstream region extending to nucleotide -1842. Similarly, when the silencer mutation was introduced into the natural gene, H1t expression was readily detected in permanently transfected cells by both RNase protection and Western blot analysis, regardless of the extent of 5' or 3' flanking genomic DNA. In constructs with the mutated silencer, we showed interdependence of the characteristic H1 AC and TG box regulatory elements. Promoter up-regulation occurred only when both were intact, and possibly identical binding factors were demonstrated for each by electrophoretic mobility shift assays. In view of its precisely regulated but limited expression, it is interesting that H1t retains all the promoter elements known to activate standard H1 genes, including the TG/AC unit, SP1 site, and CCAAT element. Their presence emphasizes the apparent dominance of the silencer element in most cells.
gene regulation, meiosis, spermatogenesis, testes
INTRODUCTION
Histone H1 is associated with the linker region of nucleosomal DNA and comprises seven subtypes in mammals [1–3]. Five of these are present to varying degrees in most somatic cells and are referred to as replicative variants, because they are synthesized primarily or exclusively during the S phase of the cell cycle [4, 5]. The H1° is a replacement variant enriched in terminally differentiated cells [6]. The focus of this report, H1t, is found only in the male germ line [7–11], appearing midway through meiosis and remaining until the nucleus undergoes condensation during the development of haploid spermatids [12–14]. It is synthesized in mid- to late pachytene spermatocytes, well after the last round of replicative DNA synthesis in preleptotene cells [14–17]. The expression of H1t is an interesting puzzle because of this gene's remarkable similarity to those of ubiquitously expressed, replicative H1s. It seems clear that H1t evolved by duplication of a typical replicative H1 gene and subsequent regulatory changes that activated expression in male meiosis but suppressed it elsewhere.
Among the conserved features of the H1t gene are the absence of introns and the presence of a downstream stem-loop that generates a nonpolyadenylated 3' terminus [18]. In the upstream transcriptional control region, no less than five conserved sequence elements are present [19–21]. These include a TATA box, CCAAT (H1TF2) element, GC box (SP1 family), AAACACA (AC box, H1TF1) element, and the more recently recognized upstream TGTGTT (TG box) element [20] (Fig. 1). In the face of all these common regulatory features, additional sequences must be sufficiently powerful to restrict expression to spermatocytes. A testis-specific binding factor for palindromic sites to either side of the SP1 element is a good candidate to activate transcription specifically in spermatocytes [22–24] (Fig. 1), but to our knowledge, it has not yet been shown to have a distinct functional effect on the gene. In vitro transcription studies in testicular extracts and transfection studies in germ cell-enriched cultures have documented the importance of the conserved SP1 and CCAAT regions of the promoter [24, 25], but so far, no stimulatory activity has been shown for the AC box, the upstream TG box element, or the spermatocyte-specific binding factor. Transgenic analysis demonstrated that the rat H1t gene will express correctly in mice [26, 27], and that as little as 141 base pairs (bp) of the rat promoter is sufficient to direct correct expression of the H1t coding region [27]. More recently, it was shown that a 1-kilobase upstream sequence directs expression of the lethal diphtheria toxin A chain gene exclusively to the male germ line [28].
|
The H1t promoter showed cell-cycle regulation in a transfected somatic cell line [29], but expression was very weak compared to that originating from the H1d control region [29]. Further comparison of these promoters led to identification of a strong silencer element located between the TATA box and the cap site [30] (Fig. 1). This silencer consists of two long runs of C. Recently, comparison of the H1t and H1d control regions in a different cell line identified a second silencer region in the -900 region [31].
A minimal, 141-bp H1t promoter sequence directed correct expression in transgenic mice [27], but this construct lacked the conserved H1 TG promoter (at -447 in rat H1t), which has been shown to function cooperatively with the AC box element in a human H1 gene [32]. The present study was undertaken to examine the importance of the cap-proximal silencer element in the context of much larger sequences flanking the H1t gene and also to investigate the roles of the TG and AC elements on its expression. Results to be presented demonstrate that the cap-proximal silencer has a major (>10-fold) effect on expression in somatic cells regardless of the extent of the surrounding genomic sequence, both for the natural gene and the reporter constructs. In addition, we confirm that the AC and TG boxes function cooperatively in the context of the H1t promoter.
MATERIALS AND METHODS
Plasmid Constructions
Standard methods were used to construct recombinant DNA molecules [33, 34]. To make luciferase reporter constructs, upstream sequences from H1t [7, 30] (GenBank accession no. M28409) or H1d [30, 35] (GenBank accession no. M31229) were cloned between the HindIII and NcoI site of pGL3 (Promega, Madison, WI). The Renilla luciferase expression plasmid pRL-SV40 (Promega) was used as a transfection control for the luciferase assay. Mutations in the H1t AC and TG boxes were generated by introducing BamHI sites(AAACACAA-95 > AAGGATCC; TGTGTTAC-446 > TGGGATCC) by the method of Deng and Nickoloff [36], and mutations were verified by sequencing. Plasmids to test expression of H1t in permanently transfected cells were made by ligating H1t genomic regions (with or without a mutation in the silencer element from pSC-1C [30]) into a PGK1-neo expression vector.
Cell Transfection
The NIH 3T3 cells were maintained in Dulbecco modified Eagle medium (DMEM; Life Sciences Technologies, Grand Island, NY) with 10% (v/v) calf serum in a humidified atmosphere of 5% CO2. Cells were seeded into 12-well plates (100 000 cells/well) and transfected 24 h later with 10 µg of TransFast reagent (Promega) using 2 µg of plasmid DNA per well. The Renilla control plasmid pRL-SV40 (20 ng) was cotransfected with the luciferase reporter constructs. All plasmid constructs were transfected in triplicate. The DNA-TransFast solution was added to each well in 400 µl of DMEM (minus serum). The cells were incubated for 60 min at 37°C. Then, 2 ml of complete medium were added per well, and the cells were allowed to grow for 48 h before harvest. For permanent transfections, 2 µg of PGKneo/H1t plasmid were linearized and transfected as described. Selection was made in medium containing 400 µg/ml of G418 (Life Sciences Technologies). Resistant cells were evident within 7–10 days and maintained as pooled clones. Two independent, permanent transfections were made for each construct.
Luciferase Assay
Cells were harvested with 250 µl of passive lysis buffer (Promega). The luciferase assay was performed using 20 µl of lysate according to the manufacturer's protocol for the Dual-Luciferase Reporter Assay System (Promega) and quantitated using a Turner Designs TD-20/20 luminometer (Sunnyvale, CA). Results are reported as the ratio between firefly and Renilla luciferase activities. Statistical analysis of the differences in means between sample groups was performed by Student t-test.
Electrophoretic Mobility Shift Assays
Whole-cell extracts from NIH 3T3 cells were prepared according to the method described by Jiang and Eberhardt [37], and gel mobility shift assay conditions were as described previously [30, 38]. Oligonucleotides were end-labeled using Klenow DNA polymerase [33] and 
-32P-dATP (Amersham, Arlington Heights, IL). The synthetic oligonucleotide probes are shown below, with motifs underlined:
AATGCCCG ACTCCG 
ACTTCCAT RNase Protection Assay
Riboprobes were synthesized from linearized plasmids using the MAXIscript kit (Ambion, Austin, TX) following the manufacturer's instructions. To prepare riboprobes, a +1/+180 fragment of H1t and a 260-bp PstI-RsaI fragment from the neo coding region were cloned into the Bluescript SK vector (Stratagene, La Jolla, CA). The cells were grown in 60-mm dishes until approximately 70% (w/v) confluent. The RNA was prepared from the H1t-transfected permanent cell lines using TRI Reagent (Molecular Research Center, Cincinnati, OH), yielding approximately 50 µg of total RNA per dish. The hybridization reactions were performed using the RPA III kit (Ambion). A typical reaction consisted of 10 µg of total cellular RNA, 80 000 cpm of the H1t riboprobe, and 40 000 cpm of the neo riboprobe. Hybridizations (10 µl) were incubated at 42°C overnight. After digestion with RNase A and T1, the protected products were separated by electrophoresis on a 5% sequencing gel, which was dried and exposed overnight to XAR5 film (Eastman Kodak, Rochester, NY). The gel was then exposed to a phosphor-imaging screen to permit quantitation using a STORM 860 PhosphorImager (Molecular Dynamics, Sunnyvale, CA).
Western Blot Analysis
Permanently transfected cells were grown in 100-mm dishes under continuous selection until approximately 70% (w/v) confluent. The medium was removed, and the cells were washed with PBS (137 mM NaCl, 5 mM KCl, 1.5 mM KH2PO4, 8 mM Na2HPO4) and pelleted. The pellet was vortexed in 0.2 ml of 5% (w/v) perchloric acid, and soluble proteins were recovered following centrifugation in a microfuge. The H1 histones were precipitated with 20% (w/v) trichloroacetic acid, washed with acetone, dried, and dissolved in SDS sample buffer [39]. The acid-precipitated histones were electrophoresed in a 15% (w/v) SDS-PAGE mini gel (150 V, 90 min). The proteins were transferred to a nitrocellulose membrane in an alkaline buffer [40] (10 mM NaHCO3, 3 mM Na2CO3, 0.1% (w/v) SDS, 20% (v/v) ethanol, 100 V, 60 min). The membrane was blocked in 5% (w/v) nonfat dry milk in PBS for 3 h, incubated with H1t antiserum (1:10 000) [12], washed, incubated with peroxidase-conjugated goat anti-rabbit IgG (1:5000; Roche Molecular Biochemicals, Indianapolis, IN), and developed using enhanced chemiluminescence detection (Amersham Pharmacia Biotech, Piscataway, NJ).
RESULTS
Cap-Proximal Silencer Is the Dominant Inhibitory Element Regardless of Additional Upstream Sequence
We have been comparing the activity of the somatic H1d promoter with that of the pachytene-specific H1t control region with the goal of understanding the factors that cause the H1t promoter to be so weak in somatic cells [24, 30]. In transfected mouse fibroblasts, the reporter constructs fused to 141 bp of the H1t control region had only approximately 5% of the activity of reporters fused to a comparable region of the H1d promoter [30]. The major part of this difference was assigned to a C-rich silencer lying between the TATA box and the cap site of H1t [30]. In this earlier study, comparisons were made primarily between constructs with less than 200 bp of upstream sequence. These promoter regions therefore lacked the TG box, one of the conserved H1 elements lying in the -450 region that is common to many H1 genes, including H1t [20]. In view of this as well as of reports that significant inhibitory elements can be detected farther upstream by transfecting a mouse mammary cell line [31, 41], it was important to assess the role of the cap-proximal silencer in constructs with additional upstream sequence. Our assay system consisted of transiently transfected NIH 3T3 cells. Promoter constructs were fused to firefly luciferase as a reporter, and activities were normalized to a cotransfected plasmid containing the SV40 control region fused to Renilla luciferase.
Extending the upstream control region of somatically expressed H1d from -212 to -738, which served to include the upstream TG box at -468, led to a twofold increase in luciferase activity (Fig. 2). This increase agrees with the results of Duncliff et al. [20] for a chicken H1 and of Meergans et al. [32] for a human H1 promoter [32]. The activities of several different wild-type H1t constructs, ranging from -141 to -2149, were all at least 20-fold lower (Fig. 2). Whereas some increase in activity occurred by inclusion of the H1t TG box at -447 (H1t-515; Fig. 2), the activities of all these native gene constructs were extremely low relative to H1d-directed expression. We showed previously that an extended mutation in the cap-proximal silencer had a 10-fold stimulatory effect on promoter activity in the context of 141 bp of upstream sequence [30]. We introduced this same mutation (see Fig. 5) into constructs with varying upstream lengths and then compared their activities (Fig. 3). All constructs with the silencer mutation had markedly elevated (>10-fold) activity in transient transfections (Fig. 3). With the mutated silencer, it was possible to show a moderate, but statistically significant, increase in activity that resulted from extending the promoter from -368 to -515 (Fig. 3). This extension incorporated the TG element at -447. Further upstream extensions to -1842 led only to a gradual decrease in activity (Fig. 3).
|
|
|
Role of AC and TG Elements
The -515 H1t construct had the greatest activity, and we suspected this was due to inclusion of the -447 H1t TG element. This element was originally identified as a conserved feature of H1 upstream regions by Duncliff et al. [20] and was shown to modulate H1 promoter activity [20, 32]. It is important to establish the role of this regulatory element for H1t expression. We therefore created mutations by introducing BamHI sites in both the TG and AC elements of H1t, and we then assayed their effects both singly and in combination (Fig. 4). These mutations were assayed in the context of the silencer mutation and two different lengths of upstream sequence. In confirmation of the results reported above, the -515 construct with intact TG and AC elements was approximately twofold as active as the -141 construct. This increased activity was abolished equally by mutations in either the TG or AC elements. A double mutant was not impaired more than either single mutant. Similarly, the AC mutant had no effect in the context of just the -141 promoter region (Fig. 4). These results support the hypothesis that the TG element is the only significant feature of the upstream DNA between -141 and -515 in this assay system. Furthermore, they indicate that the TG and AC elements depend on each other, and that neither element alone conveys significant activation in transient transfection of these cells.
|
We were able to detect what appears to be a common binding factor for the AC and TG elements in cell extracts (Fig. 5). Labeled oligos for the two elements gave rise to retarded complexes of identical mobilities ("B", Fig. 5). With the labeled H1t AC probe, the retarded complex was competed with by unlabeled oligos for the H1t AC, H1t TG, and H1d AC elements, but not by an unrelated oligo (Fig. 5). Similarly, band "B," resulting from incubating the labeled H1t TG probe oligo with 3T3 cell extract, was competed for by all three unlabeled oligos as well. The simplest interpretation of these results is that the same protein factor can bind to both AC and TG elements.
Expression of H1t Protein by Permanently Transfected Cells
To our knowledge, investigations of H1t promoter activity in somatic cells have dealt primarily with fusion genes, and the role of the cap-proximal silencer has not been investigated in the context of the natural gene. In particular, it has not been determined whether mutation of this silencer might lead to significant expression of H1t protein in somatic cells. We therefore produced permanently transfected cell lines that contained the natural H1t gene surrounded by varying amounts of genomic DNA, both with and without mutations in the silencer (Fig. 6). One (-515/+916) contained just the upstream region with maximal activity in transiently transfected cells and limited downstream sequence extending just beyond the conserved histone-specific, 3'-processing signal [18]. A second (-2383/+916) extended the 5' flanking region significantly. The third (-2383/+4478) extended the downstream flanking region extensively and included the neighboring H4 gene [10]. Two independent sets of permanent transfectants were prepared for each construct by pooling G418-resistant colonies.
|
Expression was monitored at the mRNA level by RNase protection and at the protein level by Western blot analysis using a specific H1t antibody. For RNase protection, the neomycin phosphotransferase gene transcripts were used as internal controls (Fig. 7). A 340-bp probe fragment was reduced to a 226-bp protected fragment in the presence of neor mRNA. The 270-bp H1t probe fragment was converted to a set of fragments from approximately 160–180 bp. The RNA from nontransfected control cells did not give rise to any protected bands (results not shown). The RNA from neorr pools gave rise to a well-defined band for the neo probe, which showed no dependence on the state of the H1t cap-proximal silencer. The protected bands due to H1t mRNA were faint but detectable for all three genomic constructs in which the silencer was intact. Mutation of the silencer in each case led to a marked increase in the H1t signal. Using a PhosphorImager, the increase in expression of H1t mRNA resulting from mutation of the silencer was estimated to be 15- to 20-fold. No marked difference in H1t expression was found for the three constructs, although in the experiment shown, the levels are slightly higher in the -0.5/+0.9 construct.
|
The H1-enriched extracts were prepared from the permanent cell lines, resolved on SDS gels, transferred to nitrocellulose, and detected by Western blot analysis (Fig. 8). Extracts prepared from mouse testis cell suspensions served as a positive control (Fig. 8, lanes 1 and 7), and no signal was detected from lanes containing extracts from nontransfected 3T3 cells (Fig. 8, lanes 2 and 8). With cells transfected by plasmids containing purely wild-type H1t sequence, a faint H1t band was detected for the -0.5/+0.9 and -2.3/+0.9 constructs (Fig. 8, lanes 5 and 9). However, a many-fold stronger signal was obtained in lanes from each of the pooled cell lines that contained integrated copies of H1t with the mutated silencer (Fig. 8, lanes 4, 6, and 10). Thus, the relative levels of H1t mRNA and protein were in good agreement for all the permanent transfection experiments. The cap-proximal silencer is a major determinant of H1t expression in cells with integrated copies of the natural gene, just as it was in transiently transfected cells containing reporter constructs.
|
DISCUSSION
The H1t gene has unique features that invite investigation of the mechanism of its restricted expression during male germ cell development. It shares all known regulatory elements of standard H1 genes, yet instead of widespread expression during the S phase, its mRNA is present in significant amounts only during the later part of the protracted G1 phase in primary spermatocytes. Trace expression of H1t mRNA reported in early spermatocytes or spermatogonia of mice [41] agrees with results from experiments in which H1t-directed transgene expression was inferred as early as the first type A spermatogonia [28]. However, significant levels of H1t do not accumulate until late in meiosis [7, 10–16]. In the studies reported here, we have concentrated on the roles of an H1t-specific silencer element, present between the TATA box and cap site, and the H1-specific TG and AC elements, present in the upstream region of the promoter.
Cap-Proximal Silencer
The cap-proximal H1t silencer was identified by comparison of the H1t and somatic H1d promoter activities in mouse fibroblast 3T3 cells [30], and it correlates with an extended footprint found downstream of the TATA box [24]. The silencer, which consists of the C-rich pyrimidine stretches between the TATA box and cap sites, was originally studied only in the context of the -141-bp H1t promoter [30]. This limited upstream sequence is sufficient to direct correct expression in transgenic mice [27], but it does not contain the highly conserved H1 TG element located at -447, which is now known to be important for full function in a cell cycle-regulated H1 promoter [20, 32]. Furthermore, effects of mutating the silencer were studied on the expression of reporter genes, not on the expression of H1t itself. Here, we have extended the study of the effect of the silencer on much longer 5' flanking regions and, also, to the expression of the natural gene.
We used the luciferase reporter to compare activities of the widely expressed H1d promoter and the H1t promoter. The H1t-directed luciferase activity was 20-fold less than H1d-directed activity, regardless of the extent of the H1t upstream sequence over the -141 to -2149 region. The H1d-driven activity was approximately twofold enhanced by a -738 promoter fragment in contrast to the -212 promoter used in earlier experiments [30], probably due to the inclusion of the H1d TG element at -468.
An 11-bp mutation in the cap-proximal silencer was previously shown to activate reporter gene expression 10-fold in mouse 3T3 cells [30]. We introduced this same mutation into a series of nine constructs consisting of gradual deletions from -1842 to -141. Each construct with a mutated silencer had markedly elevated luciferase expression. A significant jump in activity was observed when the promoter was extended from -368 to -515. Further extension of the upstream sequence led to a gradual fall in activity (less than twofold) between -515 and -1842. As discussed further below, the extension from -368 to -515 leads to activation due to inclusion of the TG element at -447. No further silencing element was evident in this upstream region.
Recently, Wolfe et al. [31, 42] reported a silencer activity of approximately threefold in the -780 to -948 region using transient transfection in the mouse mammary C1271 cell line. We did not observe a distinct silencer activity in this region, and it may be that separate silencer elements are active in specific cell lines. We know from transgenic experiments that the -948 promoter region generates germ cell-specific expression with both lacZ and diphtheria toxin A chain reporters [27, 28]. Therefore, silencers that are sufficient to repress H1t completely in non-germ line cells must lie downstream of -948.
Some features of H1t expression are undoubtedly dependent on sequences within the gene, such as differential regulation of transcript processing or stability by the histone-specific 3' stem-loop element [18]. To examine the effects of the silencer on the natural gene, we prepared permanently transfected cell lines that contained varying amounts of upstream and downstream sequence, with or without inclusion of the 11-bp mutation in the cap-proximal silencer. The H1t mRNA was determined by RNase protection assays and was markedly elevated in all three cell lines that contained H1t with the mutated silencer. In addition, Western blot analysis was used to show that H1t protein was readily detected in the permanent cell lines that contained the mutated silencer, whereas H1t was detected in trace amounts in cell lines transfected with versions of wild type H1t.
Cell lines that express H1t may be useful to investigate functional aspects of the protein. Recently, knockout mice for H1t were reported by three laboratories [43–45], but as yet, no distinct phenotype has been linked to loss of this histone. Spermatogenesis proceeds normally, and the males are completely fertile. The possible consequences of ectopic expression of H1t in somatic cells has not been investigated, however. To understand the role of H1t as a linker histone and why its expression is kept under tight control, permanent cell lines that express the H1t gene will allow further studies on the distribution pattern of H1t protein in chromatin compared to somatic H1-type histones as well as the functional consequences of this expression.
H1 AC and GT Motifs
The conserved -70 region AC box (AACACA) of H1 promoters was identified by Coles and Wells [19], but the function of this motif was considered only in the context of the proximal promoter region for some time. A binding factor from chicken cells for the AC element was S-phase regulated [46], and mutation of the AC box led to decreased expression [47]. An AC box-binding factor (H1TF1) was also identified in HeLa cells, but its activity was not markedly cell-cycle dependent [48]. In studies from a third laboratory, two factors were identified with affinities for the general AC box region. One (HiNFA) gave a slightly retarded DNA fragment and was not observed when dIdC was used as nonspecific competitor [49]. A second (HiNFD1) generated a much larger complex. Furthermore, HiNFD1-binding activity was markedly upregulated during the S phase in nontransformed cells and was present in fetal mouse liver, though it was not detected in adult liver [50].
Later, Duncliff et al. [20] identified a fifth conserved H1 sequence element located approximately 500 bp upstream of the cap site. This element is an inverted repeat of the AC box and is characteristically located approximately 350 bp upstream of the AC box [20]. These authors showed by progressive deletions that loss of the TG box led to a two- to threefold decrease in activity of the H1 promoter in transient transfections [20]. Meergans et al. [32] were the first to demonstrate the interdependence of the AC and TG elements. By deletion analysis of the human H1.3 control region, they showed that elimination of the TG box led to a threefold reduction in promoter activity in transient transfections of HeLa cells, but that further deletions, including the AC box, had no additional effect until the SP1-binding site was eliminated. Individual mutations in the TG and AC boxes lowered promoter activity threefold, but the combination of both mutations in the same promoter had no more effect than either single mutation. This led to the suggestion that the two elements function interactively, and that a stimulatory effect depends on the combined presence of both elements. Although these authors failed to detect a binding activity for the H1.3 AC box region by electrophoretic mobility shift assay, a shifted complex formed on the TG box region was partially competed by the AC box, suggesting that a single protein has affinity for both sites [32].
Our studies confirm that the TG and AC boxes present in the H1t upstream region function co-operatively. Mutations introduced into either site separately reduced expression in transient assays by twofold, whereas the double mutation had no greater effect than either single mutation. This interdependence of the TG and AC boxes explains the results of several earlier studies in which truncated promoter regions were used and elimination or modification of the AC box had no effect on transfection assays [25, 51] or in vitro transcription [24]. Our own upstream deletions, as well as those of others [25], establish that the TG box is the only H1t-activating element known to be upstream of the AC box. Its presence explains the increase in activity observed by Kremer and Kistler [29] when the H1t promoter was extended upstream from -174 to -693 [29]. In transgenic experiments, a -141 promoter region, which lacks the TG box, led to correct testis-specific expression, with high levels in some lines [27]. It seems that expression from integrated germ line copies of the gene is only moderately dependent on these elements, just as their elimination has only a moderate effect in transient transfection assays.
In our experiments, retarded bands of identical mobilities were obtained with both TG and AC box oligos, and each complex was competed for by the other oligo. These results suggest strongly that the same factor (H1TF1) binds to each element. It is interesting that the highly conserved spacing between the TG and AC boxes of approximately 350 bp [20] is close to the length of DNA associated with one full and one core nucleosome (200 bp + 146 bp = 346 bp) [1]. This spacing suggests a model in which a dimeric factor (H1TF1) might bind to the TG and AC elements present at equivalent points on the surfaces of two adjacent nucleosomes.
Taken together, the studies reported here emphasize that the cap-proximal silencer has a major (>10-fold) effect on H1t expression in somatic cells, regardless of the length of the promoter and regardless of whether the transcript is for a reporter gene or for H1t itself. However, the putative proteins responsible for silencing remain to be identified. Previously, it was shown that members of the SP family can bind weakly to the silencer region, but they are unlikely to be responsible for silencing [30]. Additional uncharacterized binding proteins were also identified [30]. Thus, we do not yet know how this effective silencing might be overcome in pachytene cells. The simplest model is to imagine that the silencer binding protein is inactivated or repressed. An alternative is to imagine that a novel pachytene-specific activating factor could overcome the silencing effect. A testis-specific factor that binds to duplicated copies of a palindromic sequence (CCTAGG) lying on either side of the perfect GC box (Fig. 1) has been identified, and it remains the best candidate for a pachytene-specific stimulatory factor for H1t [22, 23]. Elimination of both these sites abolished expression of a rat H1t transgene in mice [52]. Interpretation of this result is not straightforward, however, because the intervening GC box was also removed. This is one of the highly conserved H1 promoter features, and several investigations have shown that elimination of the GC box drastically lowers H1 expression in general [32] and H1t expression in particular [24, 25].
The question of how a mixture of general transcription factors (SP1 family), histone-specific factors (TG/AC-binding H1TF1, CCAAT-binding H1TF2), and H1t-specific factors such as the silencer and palindrome binding factors co-operate to achieve the tightly regulated pattern of H1t expression is of intrinsic interest. The isolation and cloning of the H1t-specific factors will be a major step forward.
ACKNOWLEDGMENTS
We thank Michael Rudnicki for the PGK-neo expression plasmid and Lynette Washington and Regina York of the USC Institute for Biotechnology Research and Training for expert assistance in production of synthetic oligonucleotides and automated DNA sequencing.
FOOTNOTES
1 Supported by NIH grant HD-10793. 
2 Correspondence: W.S. Kistler, Department of Chemistry and Biochemistry, University of South Carolina, 631 Sumter St., Columbia, SC 29208. FAX: 803 777 9521; wskistler{at}sc.edu
3 Current address: North Carolina State University, College of Veterinary Medicine, Raleigh, NC 27606.
Accepted: May 21, 2001.
Received: April 9, 2001.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  W. Ma, G. C. Horvath, M. K. Kistler, and W. S. Kistler Expression Patterns of SP1 and SP3 During Mouse Spermatogenesis: SP1 Down-Regulation Correlates with Two Successive Promoter Changes and Translationally Compromised Transcripts Biol Reprod, August 1, 2008; 79(2): 289 - 300. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kim, D. Li, Y. Cui, K. Mueller, W. C. Chears, and J. DeJong Regulatory Factor Interactions and Somatic Silencing of the Germ Cell-specific ALF Gene J. Biol. Chem., November 10, 2006; 281(45): 34288 - 34298. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. C. Wilkerson, S. A. Wolfe, and S. R. Grimes H1t/GC-Box and H1t/TE1 Element Are Essential for Promoter Activity of the Testis-Specific Histone H1t Gene Biol Reprod, October 1, 2002; 67(4): 1157 - 1164. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
