help button home button Endocrine Society Molecular Endocrinology ENDO 08 Sessions Library
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Molecular Endocrinology, doi:10.1210/me.2006-0147
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
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 Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Trujillo, M. A.
Right arrow Articles by Eberhardt, N. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Trujillo, M. A.
Right arrow Articles by Eberhardt, N. L.
Molecular Endocrinology 20 (10): 2559-2575
Copyright © 2006 by The Endocrine Society

The Human Growth Hormone Gene Contains a Silencer Embedded within an Alu Repeat in the 3'-Flanking Region

Miguel A. Trujillo, Michiko Sakagashira and Norman L. Eberhardt

Departments of Medicine (M.A.T., N.L.E.) and Biochemistry and Molecular Biology (N.L.E.), Mayo Clinic/Mayo Foundation Rochester, Minnesota 55905; and Third Department of Internal Medicine (M.S.), Wakayma Medical University, Wakayama 641-8510, Japan

Address all correspondence and requests for reprints to: Norman L. Eberhardt, Ph.D., Mayo Clinic, Endocrine Research Unit, 200 First Street Southwest, Rochester, Minnesota 55905. E-Mail: eberhardt{at}mayo.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Alu family sequences are middle repetitive short interspersed elements (SINEs) dispersed throughout vertebrate genomes that can modulate gene transcription. The human (h) GH locus contains 44 complete and four partial Alu elements. An Sx Alu repeat lies in close proximity to the hGH-1 and hGH-2 genes in the 3'-flanking region. Deletion of the Sx Alu repeat in reporter constructs containing hGH-1 3'-flanking sequences increased reporter activity in transfected pituitary GC cells, suggesting this region contained a repressor element. Analysis of multiple deletion fragments from the 3'-flanking region of the hGH-1 gene revealed a strong orientation- and position-independent silencing activity mapping between nucleotides 2158 and 2572 encompassing the Sx Alu repeat. Refined mapping revealed that the silencer was a complex element comprising four discrete entities, including a core repressor domain (CRD), an antisilencer domain (ASE) that contains elements mediating the orientation-independent silencer activity, and two domains flanking the CRD/ASE that modulate silencer activity in a CRD-dependent manner. The upstream modulator domain is also required for orientation-independent silencer function. EMSA with DNA fragments representing all of the silencer domains yielded a complex pattern of DNA-protein interactions indicating that numerous GC cell nuclear proteins bind specifically to the CRD, ASE, and modulator domains. The silencer is GH promoter dependent and, in turn, its presence decreases the rate of promoter-associated histone acetylation resulting in a significant decrease of RNA polymerase II recruitment to the promoter. The silencer may provide for complex regulatory control of hGH gene expression in pituitary cells.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
ALU FAMILY SEQUENCES are middle repetitive short interspersed elements (SINEs) dispersed throughout vertebrate genomes. Although most Alu repeats are transcriptionally silent, these elements are potentially functional class III genes, derived from a 7SL gene, and can be transcribed by RNA polymerase III. Once transcribed into RNA, Alu sequences can be reverse transcribed and then reintegrated into the genome. Alu sequences may account for about 10% of the whole human genome. Alu repeats provide cis-acting sequences that affect chromatin structure, initiation site of DNA replication, sites of recombination, and RNA processing and stability (1, 2).

In silico analysis of the human Alu family of retroposons exposed an increased concentration of consensus transcription factor binding sites, suggesting that the majority of Alu repeats have the potential for regulating gene expression via modulation of RNA polymerase II-dependent transcription (3). Indeed, it has been observed that, when inserted in the near vicinity of genes, Alu elements may control gene transcription both positively and negatively. For example, Alu sequences have been identified in several positive cis-acting regulatory elements such as the T cell-specific enhancer of the human CD8a gene (4) and estrogen receptor-dependent enhancers (5). Alu sequences enhance the transcription of the liver-specific Haptoglobin-related promoter in hepatoma cell lines, but not in HeLa cells (6). Additionally, the Fc{epsilon} RI-{gamma} chain gene possesses two adjacent cis-acting regulatory elements that are part of an Alu repeat. The first confers positive transcriptional regulation on basophils and T cells, whereas the second inhibits transcription in basophils, but acts as a positive element in T cells (7). Thus, Alu sequences can influence transcription control positively as well as negatively. Indeed, Alu sequences have been shown to function as silencers (8). Likewise, silencers have been found embedded within Alu repeats in, among others, the WT1 (9), the human poly(ADP-ribosyl) transferase (10), and the human {epsilon}-globin genes (11).

GH belongs to a family of hormones that include chorionic somatomammotropin (CS), and prolactin that have evolved from a common precursor. The human (h) GH locus spans about 66.5 kb and is located on chromosome 17 (12). It comprises two GH genes, two CS genes, and a CS pseudogene (13). The hGH locus contains 44 complete Alu elements and four partial Alu elements that account for approximately 20% of the sequence of this locus (14). The arrangement pattern of these Alu repeats indicates that these elements may be involved in the duplications that gave rise to the hGH locus (12, 14). However, it is not known whether any of these Alu repeats are also involved in the regulation of the expression of the genes of the GH locus.

The hGH-1 gene is specifically expressed in pituitary somatotrophs. A locus control region (15) and a promoter, both of which are dependent on the binding of the pou homeoprotein Pit-1 (16, 17, 18, 19), dictate tissue specificity. In addition, a negative thyroid responsive element has been identified in the 3'-untranslated region (UTR) region of the hGH-1 gene (20). However, it is not known whether sequences in the 3'-flanking region (3'-FR) of the gene have any influence on the transcriptional regulation of the hGH-1 gene. This region of the hGH-1 gene contains an Sx Alu repeat between nucleotides (nts) 2228 and 2501 (+559 = starting codon) (14). The close proximity of this repeat to the hGH coding region makes it an ideal candidate to study whether cis-regulatory sequences are present within this particular Alu retrotransposon. In this study, we undertook a comprehensive functional analysis of the hGH-1 3'-FR sequences and found that a silencer/antisilencer cassette resides in this region between nts 2158 and 2572.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
hGH Gene Expression in GC Cells Is Regulated by Sequences Embedded within an Alu Middle-Dispersed Repeat Located in the 3'-FR
To determine whether the Alu repeat located in the 3'-FR of the hGH gene was capable of regulating gene expression, we constructed an expression plasmid in which sequences 3' from nt 2158 were deleted. This deletion eliminates the 3'-Alu element closest to the hGH-1 gene (nts 2228–2501). We consistently observed that hGH expression in transfected GC cells was 4-fold higher in plasmid p{Delta}2158 than pGL3.GH (Fig. 1AGo). This result indicates that a repressor element is located within the hGH 3'-FR. To determine the position of the repressor element, additional reporter genes were constructed that contained progressive 3'-deletions from nts 2572–2158. As shown in Fig. 1AGo, the lowest activity was observed with constructs containing the full-length hGH 3'-UTR/3'-FR. There was an increase in reporter activity with progressive 3'-deletions reaching a maximum at nt 2306. Subsequent deletion to nts 2236 and 2158 resulted in a 20% decrease in activity; however, this decrease was not statistically different from p{Delta}2306. These data indicate that a major repressor element is located between nts 2306 and 2465. Because the deletion series between nts 2306 and 2572 exhibits a gradual increase in activity, the data suggest that multiple repressor elements may be distributed over this 266-bp sequence. This region includes the majority of the sequences that constitute the Alu middle-dispersed repetitive element, indicating that a repressor is localized with the Alu element.


Figure 1
View larger version (12K):
[in this window]
[in a new window]
  		Fig. 1. A Silencer Element Maps to the 3'-FR of the hGH Gene

A, Analysis of basal expression of the 3'-FR deletion series. Plasmid pGL3.GH includes the GH promoter, the GH cDNA as a reporter, and the wild-type hGH 3'-UTR/3'-FR (nts 2039–2572). Transfection and analysis were performed as described in Materials and Methods. Plasmid pGL3.hActp.luc+.SV3' was used to monitor transfection efficiency. The data represent basal GH expression normalized to luciferase activity from five independent experiments. Error bars represent SD. B, Position and orientation effects of the hGH 3'-FR. The hGH sequences from nts 2158–2572 were inserted in both orientations 3' to the SV40 3'-UTR or 5' to the hGH promoter in plasmid pGL3.GHp (pGL3.GHp.LUC.SV40 3'-UTR). Luciferase activity in transfected GC cells was measured and normalized to protein concentration. The fold inhibition was calculated by dividing the normalized expression values of the control plasmid by the normalized expression values obtained with plasmids carrying the insert. Multivariate ANOVA were performed for all data and were significant at the P < 0.0001 level. The Bonferroni/Dunn classes (letter designation above the error bars) indicate significant differences at the P < 0.05 level for the post hoc tests of individual constructs. Note in this analysis that all data groups are being compared with each other and that only data groups with unique class designations are significantly different from one another.

 
Characterization of the Repressor Sequences of the hGH Gene
We next sought to establish the orientation and position dependence of the repressor sequences in the hGH 3'-FR. We amplified the sequences spanning nts 2158–2572 of the 3'-FR of the hGH gene and inserted them into the luciferase reporter vector pGL3.GHp, which harbors the hGH promoter and the simian virus 40 (SV40) 3'-UTR. This 414-bp fragment was inserted into the upstream and downstream regions in both forward and reverse orientations. The resulting constructs were transfected into GC cells, and their activity relative to pGL3.GHp was established. In all constructs harboring the 3'-FR of the hGH gene, luciferase expression was inhibited regardless of the orientation and position of the 2158–2572 fragment (Fig. 1BGo). However, there were significant quantitative differences linked to the position of this fragment. Thus, when cloned in the 3'-position, the inhibitory effect was almost 2-fold larger than the constructs in which it was inserted at the 5'-position. Additionally, when cloned in the 3'-position, there was a statistically significant difference between the forward-oriented and the reverse-oriented element with the activity of the latter being about 20% higher. Conversely, there was no significant difference linked to the cloning orientation when the fragment was cloned in the 5'-position. In general, silencers operate independently of position and orientation (21) although silencers that are strictly orientation dependent (21) or strictly position dependent (22) have been described. Here, although the level of repression differs significantly according to the orientation and position, repression is always observed in all four constructs. Thus our data indicate that a position- and orientation-independent silencer is encoded in the 3'-flanking regions of the hGH gene within the Alu element.

Functional Analysis of the hGH Gene Silencer
The results presented in Fig. 1AGo indicate that the silencer element maps between nts 2306 and 2465. Indeed, a 159-bp fragment spanning these sequences retained full repressor activity (pGLS1, Fig. 2AGo), whereas their removal from the 414-bp fragment spanning nts 2158–2572 resulted in complete loss of silencer activity (pGLS8, Fig. 2AGo). Nevertheless, when the 159-bp fragment was cloned in the opposite orientation, repressor activity was lost, indicating that sequences flanking this minimal internal repressor region were required for orientation-independent silencer activity (pGLS1rev, Fig. 2AGo). Accordingly, this 159-bp fragment appears to represent a core repressor domain (CRD) because the repressor activity is lost in a further deleted fragment encompassing nts 2350–2404 (pGLS2, Fig. 2AGo). Moreover, simultaneous deletion of nts 2280–2350 and nts 2401–2464 virtually abrogated repression (pGLS5, Fig. 2AGo). Despite the fact that, by themselves, sequences between nts 2350 and 2401 lacked repressor activity, their removal in plasmid pGLS19 resulted in a 50% drop in activity, indicating that they are also implicated in the CRD function or that the sequences on either side require strict phasing for repressor function. Thus, sequences spanning nts 2306–2465, designated as the core repressor domain or CRD, are essential for silencer function.


Figure 2
View larger version (23K):
[in this window]
[in a new window]
  		Fig. 2. Functional Analysis and Mapping of the Silencer Elements

In all experiments below, deletion constructs are identified by their nucleotide number with respect to the start site of transcription of the hGH gene, and orientation is designated by the direction of the arrow. GC cells were transfected as described in Materials and Methods, and the fold inhibition was calculated as described in the legend to Fig. 1Go. A, Deletion analysis of the region between nts 2300 and 2464. B, Effect of neighboring 5'- and 3'-sequences to modulate silencer activity. C, Identification of sequences imparting orientation independence. The cloning orientation of the silencer mutants pGLS12, pGLS13, and pGLS14 was inverted as indicated by the "rev" designation. All data were analyzed by a single, combined multivariate ANOVA, and the data were significant at the P < 0.0001 level with respect to the control plasmid pGL3.GHp (pGL3.GHp.LUC.SV40 3'-UTR). The Bonferroni/Dunn classes (designated by uppercase letters above the error bars) indicate significant differences at the P < 0.05 level for the post hoc tests of individual constructs. Note in this analysis that all data groups are being compared with each other and that only data groups with unique class designations are significantly different from one another.

 
Sequences Adjacent to the Core Repressor Domain Contribute to the Overall hGH Silencer Activity
The finding that there was no significant difference in repressor activity between pGL2.3 (nts 2158–2572) and pGLS1 (nts 2306–2465), whereas plasmid pGLS1rev lacked all silencer activity (Fig. 2AGo), indicated that sequences adjacent to the CRD were involved in silencer function, including sequences that confer orientation-independent activity. Consequently, we studied additional constructs to establish the location of sequences adjacent to the core repressor domain that contribute to silencer function. Interestingly, a construct that contained the intact CRD and included the sequences immediately downstream (nts 2464–2501) exhibited a complete loss of repressor activity (pGLS10, Fig. 2BGo), indicating that the immediate downstream region contains an antisilencer element (ASE). Antisilencers have been found only in a handful of eukaryotic genes and are defined as cis-acting sequences that repress silencers but are otherwise inactive by themselves (23, 24, 25). This result also implies that the regions upstream and/or downstream from the CRD may contain sequences that counteract the antisilencer. Indeed, a plasmid containing the CRD and antisilencer as well as the upstream region, nts 2158–2300, exhibited strong repressor activity (pGLS13, Fig. 2BGo) that was significantly greater than that of the parent plasmid (pGLS2.3, Fig. 2AGo). This result indicates that either the upstream region contains independently acting repressor elements or contains sequences that are capable of modulating the activity of the CRD, at least in part, by inactivation of the ASE. We next examined a plasmid containing a truncated CRD that contained the intact upstream region (pGLS3, Fig. 2BGo) and found that it lacked repressor activity, indicating that the region between nts 2158 and 2300 contained sequences that acted in a CRD-dependent manner, but do not contain independently acting repressor elements. These sequences were subsequently localized to the more upstream portion between nts 2158 and 2228, because plamid pGLS9, which lacked these sequences, failed to repress reporter gene activity (Fig. 2BGo). Further support that the region between nts 2158 and 2228 contains a CRD-dependent modulator element was provided by plasmid pGLS12 (Fig. 2BGo), which lacked the antisilencer domain, and exhibited repressor activity that was almost 3-fold higher than either plasmid pGLS2.3 or pGLS1 (Fig. 2AGo). This result suggests that the region upstream from the CRD/ASE interacts with the CRD, resulting in increased silencer activity. This would provide a potential mechanism to account for antisilencing modulator activity. Finally, we examined plasmids containing additional downstream sequences to determine whether modulator elements were contained in these regions. Plasmid pGLS14 exhibited very strong repressor activity and only differed from pGLS9 in that it contained sequences immediately downstream (nts 2501–2572) of the ASE (Fig. 2BGo), indicating that this region also contains sequences capable of independent or CRD-dependent repression. A plasmid containing a truncated CRD and the downstream sequences (nts 2464–2572) lacked repressor activity (pGLS4, Fig. 2BGo). Accordingly, the data support the concept that the sequences immediately downstream of the core repressor element contain an antisilencer modulator domain and function in a CRD-dependent manner. Thus our results strongly support the concept that the hGH 3'-FR silencer region is organized in at least four different elements: a central core repressor, CRD, followed by an antisilencer, ASE, flanked by 5'- and 3'-modulator regions, both of which appear to be CRD/ASE dependent.

We next sought to localize the sequences that impart orientation independence to the silencer. As shown in Fig. 2CGo, reversing the orientation of the hGH 3'-FR in plasmids pGLS12rev (3'-deletion) and pGLS14rev (5'-deletion) resulted in a drastic reduction of silencer activity. This is especially significant for plasmid pGLS12rev because the ASE is deleted in this construct, and plasmid pGLS12 exhibited the largest repressor activity (Fig. 2BGo). In contrast, plasmid pGLS13rev resulted in a 20% increase of silencer activity (Fig. 2CGo), which is comparable to the difference between pGL2.3 and pGL2.3rev (Fig. 1BGo). These data demonstrate that both the upstream modulator element (nts 2158–2228) and the ASE modules are essential for the functional independence of silencer orientation. Thus the interactions between the 5'-module, the ASE module, and the CRD include not only the regulation of repression, but also the regulation of the orientation-independent silencer properties.

The hGH 3'-FR Silencer Functions in Pituitary and Placental Cells
To study the cell type specificity of the hGH 3'-FR silencer, we compared the activity of the control plasmid pGL3.GHp, which contains the luciferase reporter gene under the control of the hGH promoter and the SV40 3'-UTR, to plasmids that additionally contains the full hGH-1 3'-silencer region (nts 2158–2572). These constructs were transiently introduced into four cell types: GC (rat anterior pituitary), BeWo (human placental scyntiotrophoblast), HeLa (human cervical cancer), and COS-1 (monkey kidney). Because the hGH promoter activity is dominantly dependent on the pituitary-specific factor Pit-1 (26, 27), we first established that sufficient promoter activity could be detected in nonpituitary cells to allow meaningful interpretation of the data. Surprisingly, luciferase activity driven by the hGH promoter was equally strong in GC and COS-1 cells (Fig. 3AGo). This may be due to the fact that either the SV40 T-antigen may somehow affect transcription from the hGH promoter or affect expression through interactions with the SV40 3'-UTR. In both BeWo and HeLa cells, hGH promoter-driven luciferase expression was 50 times lower than in GC cells. However, sufficient activity above background was readily detected in both BeWo and HeLa cells (signal-noise ratios of 14:1 and 6:1, respectively) to allow assessment of silencer activity.


Figure 3
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 3. Cell Type Specificity of the hGH Silencer

A, GC (open bar), BeWo (stippled bar), HeLa (striped bar), and COS-1 (dark-striped bar) cells were transfected with plasmid pGL3.GHp as described in Materials and Methods to ascertain basal levels of GH promoter-driven luciferase expression. Results are expressed in light units/µg protein. B, GC (open bar), BeWo (stippled bar), HeLa (striped bar), and COS-1 (dark-striped bar) cells were transfected with control plasmid pGL3.GHp, plasmids harboring the GH silencer in the forward (pGL2.3, p2.3GL) or reverse orientation (pGL2.3rev, p2.3revGL), and downstream of the SV40 polyadenylation signal (pGL2.3, pGL2.3rev) or upstream of the promoter (p2.3GL, p2.3revGL). Data reflect the averages of six independent transfection experiments. Luciferase activity normalized to total cellular protein was measured in the transfected cells, and fold inhibition was calculated by dividing the control plasmid values by the values obtained with plasmids carrying the silencer insert. The data were subjected to ANOVA followed by post hoc Bonferroni/Dunn tests (P < 0.05) to compare individual constructs. Significant differences are denoted by the letters above the error bars (SD), which indicate the Bonferroni/Dunn classes. Note in this analysis that all data groups are being compared with each other and that only data groups with unique class designations are significantly different from one another.

 
The results presented in Fig. 3BGo confirm the occurrence of a silencer in the hGH-1 3'-FR that is position (compare pGL2.3 with p2.3GL) and orientation independent (compare pGL2.3 with pGL2.3rev and p2.3GL with p2.3revGL). However, transcriptional repression of the luciferase reporter gene was completely extinguished when constructs harboring the silencer sequence were introduced into HeLa and COS-1 cells (Fig. 3BGo). The absence of repression activity in these two cell lines suggested that the hGH-1 silencer might be only active in cells that can express the GH gene family members. We tested this hypothesis using the human placental cell line BeWo, because the related GH-2 and CS genes are expressed in placental syncytiotrophoblasts and the presence of hGH-2 has been verified in maternal serum during pregnancy (28). The data in Fig. 3BGo demonstrate a strong silencer activity in BeWo cells transfected with constructs containing the hGH-1 3'-FR sequences. As in GC cells, transcriptional repression was observed regardless of the cloning orientation and position (Fig. 3BGo). Thus, our results demonstrate that the hGH-1 silencer is as active in BeWo cells as in GC cells, supporting the concept that the hGH 3'-FR silencer is a tissue-specific element that restricts gene expression only in cells capable of expressing GH-related genes.

Multiple Specific Protein-DNA Interactions Occur within the hGH Silencer Region
As discussed above we have identified four specific regions between nts 2158 and 2572 that account for and/or modulate the hGH 3'-silencer function. To verify the ability of the silencer element to interact with trans-acting factors, we performed EMSA on the entire silencer domain (nts 2158–2572). The silencer domain was divided into 15 overlapping 50-bp fragments as depicted in Fig. 4AGo. Each fragment was PCR amplified, 32P labeled, and incubated with 20 µg of protein prepared from GC cell nuclear extracts, and EMSA was performed. The specificity of the DNA-protein complexes revealed by EMSA was tested by competition with either unlabeled self-competitor or a nonspecific unlabeled probe.


Figure 4
View larger version (29K):
[in this window]
[in a new window]
  		Fig. 4. EMSA

EMSA was performed as described in Materials and Methods. A, Structure of the hGH 3'-FR from nts 2158–2572. The position of the 3'-UTR, polyadenylation site, and relative position of the 3'-Alu repeat are depicted. The position of the 15 overlapping DNA fragments tested by EMSA with respect to the various functional domains are indicated by arrows. The specificity of the DNA-protein complexes was tested by competing with either self-unlabeled probe (S) or with a nonspecific (NS) mutated thyroid hormone response element (DR4) unlabeled probe. The fold molar excess is indicated at the top of each well. B, Analysis of 5'-ASE modulator and orientation module with the 5'-probes spanning nts 2158–2300 (FP1 –FP3). C, Analysis of probes spanning nts 2275–2475 (FP4–FP7) that contains the CRD. D, Analysis of probes spanning the 3'-moiety between nts 2464 and 2572 (FP8–FP9) that harbors the ASE and 3'-antisilencer and orientation module. P, Protein-DNA complex; FP, DNA fragment.

 
The 5'-portion of the silencer containing the 5'-CRD/ASE modulator and orientation module yielded seven specific DNA-protein complexes (Fig. 4BGo, bands P1–P7) that bound to the DNA between nts 2158 and 2283. Within the region containing the core repressor domain, CRD, a total of six specific DNA protein complexes were obtained with the seven probes (Fig. 4CGo, P8–P13). In the region 3' to the CRD (nts 2464–2572) containing the ASE/orientation module and 3'-modulator domains, seven specific DNA-protein complexes were observed (Fig. 4DGo, P14–P20). Several other complexes appear to be nonspecific, because either self or nonspecific competitor DNA does not inhibit them. Additionally, some bands, including those located between P11 and P12 as well as the band below complex P11 (Fig. 4CGo), were not always reproducible when individual nuclear extract preparations were compared, suggesting that these complexes may correspond to degradation products, which are still capable of binding to DNA in a specific manner. Taken together, the results presented in this and the previous section indicate that the silencer located on the hGH 3'-FR is a composite element consisting of multiple DNA regions that are specifically contacted by multiple trans-acting nuclear proteins.

The hGH Silencer Function Is Optimal with the hGH Promoter
We substituted the GH promoter by the thymidine kinase (TK) or the SV40 promoters into the constructs containing the complete hGH silencer sequences placed downstream of the reporter expression constructs in both orientations. Surprisingly, as shown in Fig. 5AGo, when the silencer was in the forward orientation, silencer activity was only observed in the construct harboring the hGH promoter. When the silencer was present in the reverse configuration, there was weak, but significant, activity above baseline in constructs containing both the SV40 and TK promoters. However, this activity was significantly reduced relative to that obtained with the hGH promoter (Fig. 5AGo). This unexpected result suggests that the hGH silencer is promoter-dependent and suggests that the silencer may interact with specific components of the hGH promoter to repress transcription.


Figure 5
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 5. Promoter Specificity of the hGH Silencer

A, Analysis of silencer function in forward and reverse orientation placed downstream of the SV40 polyadenylation site in reporter constructs driven by the GH promoter (open bars), SV40 promoter (stippled bars), or with the TK promoter (striped bars). B, Effect of GH promoter mutations on hGH silencer activity. Individual mutations in the GH promoter for the consensus binding sites for the transcriptional factors Zn-15 (pGL3.ZN), SP-1 (pGL3.SP1), NF-1 (pGL3.NF1), proximal and distal Pit-1 (pGL3.PIT1), and proximal and distal CRE (pGL3.CRE) were compared with constructs containing the silencer sequences placed downstream of the SV40 polyadenylation site in each of the respective vectors (pGL3.ZN2.3, pGL3.SP1_2.3, pGL3.NF1_2.3, pGL3.PIT1_2.3, and pGL3.CRE2.3) after transfection into GC cells. Luciferase activity in transfected cells was measured and normalized to protein. Data represent the average of six independent transfection experiments. The data were subjected to ANOVA followed by post hoc Bonferroni/Dunn tests to compare individual constructs. Significant differences (P < 0.05) are denoted by the letters above the error bars (SD), which indicate the Bonferroni/Dunn classes. Note that only those comparisons with unique letter designations are significantly different from one another.

 
Mutations of the hGH Promoter cis-Acting Elements Reveal a Stringent Silencer-Promoter Cooperation
We next sought to test whether a particular element of the hGH promoter was implicated in promoter-silencer interaction. To this end we independently mutated all of the consensus sequences that are known to regulate GH promoter activity. Figure 5BGo shows that, as expected, mutations of the SP-1 and Pit-1 binding sites resulted in a significant decrease of promoter activity whereas mutations of the nuclear factor 1 (NF-1) consensus binding site increased promoter trans-activation. Two unexpected results, however, were observed. First, mutation of the Zn-15 consensus-binding site did not alter promoter activity. This is unusual because it was previously demonstrated that binding of the zinc finger protein Zn-15 to its cognate sequence was essential for GH promoter activity in heterologous cells (29). Second, mutation of the two cAMP response element (CRE) sites also resulted in a significant increase in GH promoter activity. Although an increase in promoter activity resulting from mutating the two CRE sites of the hGH promoter was previously observed (30), this increase was not as large as the one described here. Paradoxically, all mutations tested resulted in a significant decrease in silencer activity (Fig. 5BGo). Mutations of the Zn-15, SP-1, and Pit-1 binding sites virtually abrogated the GH 3'-silencer activity, because the differences detected did not reach statistical significance. Only the constructs containing the NF-1 and CRE mutations retained significant silencer activity, although the inhibition relative to the wild-type promoter was reduced (44% and 59% of wild-type activity, respectively for NF-1 and CRE mutants). These data indicate that the hGH silencer activity is strictly dependent on the integrity of the hGH promoter and that, as a consequence, a stringent cooperation between silencer binding proteins and GH promoter binding proteins must be established to bring about full silencer activity.

The Silencer Inhibits Transcription by a Mechanism Involving Histone Deacetylation
Down-regulation of the hGH gene by its 3'-FR silencer could be due to altered transcriptional activity resulting from changes in DNA methylation (31, 32), or by the recruitment of factors that attenuate promoter activity through histone modification (33, 34), and/or preinitiation complex formation (35). We do not favor the DNA methylation hypothesis because 1) we observe silencer-mediated down-regulation of gene expression in transiently transfected cells, and 2) computer analysis using the CpG Island Searcher program (http://cpgislands.usc.edu/) did not reveal a significant score for CpG islands within the hGH promoter. On the other hand, histone acetylation and methylation have been shown to mediate locus control region regulation of GH/CS gene expression (36, 37), which suggests that the hGH gene may be largely controlled through histone modifications. Moreover, we showed that the silencer function is strictly dependent on the integrity of the hGH promoter. In particular, mutations of the two Pit-1 sites as well as SP-1 abolish silencer activity. This is of importance because Pit-1, as well as SP-1, interact with corepressors and coactivators that tether histone deacetylase or histone acetyltransferase enzymes to their binding site (38, 39, 40, 41, 42). Thus we hypothesized that the acetylation status of promoter-associated histones is a key component for silencer activity. We tested the effect of the histone acetyl transferase inhibitor trichostatin A (TSA) on silencer activity. TSA stimulated luciferase gene expression in the control pGL3.GHp by 25-fold, but the luciferase expression in the silencer containing plasmid pGL2.3 rose by 100-fold (data not shown). Because the silencer inhibits luciferase expression by a factor of 4, when the fold inhibition by silencer sequences was calculated under each regimen, the net result was a complete silencer shut down by TSA (Fig. 6Go). This suggests that the silencer inhibits transcription by a mechanism involving histone deacetylation.


Figure 6
View larger version (13K):
[in this window]
[in a new window]
  		Fig. 6. TSA Inhibits hGH Silencer Activity

GC cells were transfected with the control plasmid pGL3.GHp or pGL2.3, harboring the hGH 3'-FR silencer. Transfected cells were treated with and without 200 ng/ml of TSA as described in Materials and Methods and TSA was added upon transfection. After 24 h, cell lysates were harvested and assayed for luciferase activity.

 
To directly demonstrate that the silencer impinges on the status of promoter-associated histone acetylation, we performed chromatin immunoprecipitation (ChIP) assays using antibodies against acetylated histone H3 (AcH3) (43, 44, 45). We also included antibodies against RNA polymerase II to detect changes in the polymerase load on the promoter. PCR was performed with primers that amplify the entire 496-bp hGH promoter. A strong signal was obtained with the AcH3 antibody in GC cells transfected with the control plasmid pGL3.GHp. Correlated with the acetylated status of histones, we were also able to amplify DNA precipitated with the RNA polymerase II antibody (Fig. 7AGo). By contrast, only a very faint signal was detected with either the AcH3 or RNA polymerase II antibody in cells transfected with pGL2.3 at comparable input levels, indicating that the levels of these products were greatly diminished in the presence of the silencer. In three independent experiments we observed that the signal obtained with the AcH3 and RNA polymerase II antibodies were barely above that of the negative control (Fig. 7BGo). These data indicate that histone acetylation and RNA polymerase II loading on the GH promoter are greatly diminished in the presence of the silencer. These results demonstrate that the silencer hampers promoter-associated histone acetylation, resulting in the inhibition of RNA polymerase II recruitment to the promoter.


Figure 7
View larger version (16K):
[in this window]
[in a new window]
  		Fig. 7. Silencer Sequences Affects the Acetylation Status of Promoter-Associated Histones

GC cells were transfected with 10 ng of either pGL3.GHp (A) or pGL2.3 (B). After 24 h, posttransfection cells were harvested and ChIPs performed using the Easy-ChIP kit assay (Upstate Technology). Immunoprecipitated chromatin with input, mouse anti-IgG (mIg), anti-AcH3, or anti-RNA Polymerase II (Pol II) was PCR amplified using a set of primers (GH5/GH3) that encompass the entire 496-bp hGH promoter.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
In the current study we describe a context-, position-, and orientation-independent GH promoter-dependent silencer/ASE cassette located in the proximal 3'-FR of the hGH gene. Interestingly, a middle-dispersed Alu repeat is embedded within the GH 3'-FR silencer and, as discussed below, likely contributes several functional domains of this regulatory element. Classically, enhancer elements that up-regulate transcription and silencers that inhibit it have been defined as cis-acting elements that share orientation- and position independence properties (46, 47). However, more recently, the designation "silencer" has also referred to cis-acting domains that are capable of extinguishing gene transcription over long chromosomal distances (48, 49). Although we prefer to retain the classical definition of silencer for the repressor element in the 3'-FR of the GH gene, we emphasize that the ability of this element to act over long chromosomal distances is not known. Whether the silencer can act over shorter distances to affect the downstream hCS genes is also not known. The silencer is approximately 6 kb downstream from the hCS-5 or hCS-L gene. However, because this gene is a pseudogene (12), any effect of silencer function on transcription of this gene would not likely be physiologically relevant.

Unlike most silencer elements that are usually short elements (50, 51, 52), the hGH silencer spans a 414-bp long fragment and is composed of at least four distinct functional modules. Because the silencer activity is not equal to the sum of the activity of each individual subfragment, it follows that the silencer is a compound element comprised of several interacting domains. A number of previously identified silencer elements restrict gene expression in nonpermissive cells (22, 53, 54, 55). However, our data indicate that the silencer functions in a tissue-specific manner that includes pituitary and placental cells. This indicates that silencer function appears to be specific to cell lineages that express either the hGH-1 or the highly related hGH-2 and hCS genes. This places the GH gene into a class of genes that harbor silencer elements that function within their expressing cells. This family of silencers includes among others the B29 (56), {lambda}5 (57), bcl2 (58), and hETS-1 genes (59). Thus, the hGH 3'-FR evokes a unique functional and structural organization for cis-repressor elements.

The concept that the silencer is composed of multiple interacting modules is further supported by the fact that EMSA revealed 19 sites of DNA-protein interaction. Obviously, binding does not equate to function; therefore, it is possible that among the DNA-protein complexes detected, some may be nonfunctional. However, the functional data indicate that all of these DNA binding sites support critical overall silencer/ASE functions.

Although most silencers are not promoter specific (50, 56, 58, 59, 60, 61), the hGH-1 silencer ranks within the rising class of promoter-dependent silencers (53, 62). This dependence on promoter activity may, in turn, determine the mechanism of action of the silencer. The presence of silencer sequences results in a net decrease of histone acetylation at the promoter with a concomitantly poor RNA polymerase II recruitment. Two general models of silencer action can be envisioned (Fig. 8Go). In the first model, silencer sequences compete with promoter sequences for a limited supply of coactivators. As a result, the silencer will recruit coactivators, thus targeting corepressors to promoter sequences. In the second model, silencer sequences will loop over, and silencer-bound proteins will interact with promoter-bound proteins, thus preventing tethering of coactivators. In both cases, the multiprotein complex at the promoter will have a different composition depending whether it is linked to the silencer or not. Because we showed that silencer function is altered when promoter sequences are mutated, the looping mechanism appears most likely to account for silencer activity.


Figure 8
View larger version (13K):
[in this window]
[in a new window]
  		Fig. 8. Models of Silencer Mechanism

A, Silencer-bound proteins sequester coactivators resulting in corepressor and histone deacetylase (HDAC) binding to the promoter. B, Silencer-bound proteins interact with promoter-bound proteins, thus inhibiting the recruitment of coactivators and histone acetyltransferase (HAT) to the promoter. NCoR, Nuclear receptor corepressor; CBP-p/CAF, cAMP response element binding protein (CREB)-binding protein-p/CAF.

 
The antisilencer element (ASE) located between nts 2464 and 2501 also displays distinctive properties. It restricts silencer activity in somatotrophic GC cells but does not inhibit it completely. This markedly distinguishes the hGH ASE from other ASEs, including the pp52 (LSP1) ASE. In this case, and in the absence of its ASE, the pp52 silencer is functional in its own expressing cells; however, the pp52 silencer element is completely overridden by its ASE in B cells (23). EMSA with a probe encompassing the ASE region produced a single specific band. Thus, unlike the silencer, the ASE may represent a simple element. Computer analysis of the FP9 sequence scored a homology to the binding sites of the SP-1 family of transcriptional factors. In preliminary studies we demonstrate in vitro and in vivo binding of SP-2 to the ASE region (our unpublished data). The functional data however indicate that, in addition to the CRD, the ASE may closely interact with silencer sequences upstream of nt 2228 and downstream from nt 2501 because these sequences counteract its effect. Furthermore, our functional data indicate that, in conjunction with the 5'-domain, the ASE sequences are implicated in silencer orientation independence. At the present time the mechanisms of how ASE elements might function are poorly understood, and the required DNA sequence(s) and the proteins that bind to these cis-acting elements have not been elucidated. The only case in which an ASE sequence and its associated factor have been identified is in the vimentin gene. Sequence analysis revealed that the vimentin ASE harbors cognate sequences for signal transducer and activator of transcription (STAT) proteins. Indeed, it was later found that this ASE binds signal transducer and activator of transcription-3, which in turn interacts with the repressor ZBP-89 (63).

The Alu repeats of the GH locus may have played a role not only in the generation of the locus, through involvement in gene duplication (13, 14, 64), but also, at least in the case of the Alu repeat just downstream of the hGH-1 gene, have evolved cis-acting elements that control the transcriptional regulation of the duplicated genes. This raises the question whether other Alu repeats in the locus can affect the transcription of neighboring genes. At this point, we can only speculate based on the homologies of the Alu sequences that flank the GH and CS genes (Fig. 9AGo). The 3'-Alu repeat of the hGH-1 and hGH-2 genes belongs to the Sx subfamily whereas the Alu sequences that flank the CS genes are of the Sz class in the 5'- and Sq subfamily in the 3'-regions (14). In fact, the whole 3'-FRs of the hGH-1 and hGH-2 genes are almost 92% identical (Fig. 9BGo). Thus, it is likely that the Alu element within the 3'-FR of the hGH-2 gene also possesses a silencing element. On the other hand, the homology with the hGH Alu sequences decreases in the repeats flanking the 3'- and 5'-regions of the hCS genes (Fig. 8BGo). Remarkably, the region that is most conserved among all Alu repeats that flank the hCS and hGH genes is the segment corresponding to the hGH-1 ASE sequences, suggesting that evolutionary pressure to preserve inhibition of silencing is operating. Taking into account the pattern of expression of the hCS and hGH-2 genes during pregnancy, a mechanism that represses silencer function is required, because these genes are abundantly expressed in placenta. In this regard, the presence of the highly active hCS enhancers might serve this function (65, 66).


Figure 9
View larger version (50K):
[in this window]
[in a new window]
  		Fig. 9. Comparison of Different Alu Elements in the hGH/hCS Gene Locus

A, Sequence comparison by PILEUP analysis (GCG Wisconsin Package, San Diego, CA) of the various types of Alu elements that map nearest to the hGH and hCS genes. The locations and directions of the various elements are given by the number of designations, indicating the position within the 66,495-bp sequence (HUMGHCSA) and conforms to the designation of Toda and Tomita (14 ). B, Relative homology of each of the Alu elements was done by GAP analysis (GCG Wisconsin Package) relative to the Alu element in the hGH-1 gene 3'-FR (6 Sx 6093<7168). The genes and the relative position (5' or 3') of the nearest Alu are shown.

 
Finally, it is of interest to consider when the regulatory elements within the hGH-1 Alu sequence may have arisen and whether they may be generally distributed in human Alu sequences. Because the duplications that give rise to the modern hGH locus occurred 15–60 million yr ago and after the mammalian radiation (12), it is possible that these regulatory sequences are relatively recent evolutionary acquisitions. To address this issue, we compared the hGH-1 and hGH-2 Alu sequences with a number of rat B1 sequences, which are evolutionarily related to the human Alu sequences (67) as shown in Fig. 10Go, A and B. Interestingly, just as in the case of the comparison of the different classes of human Alu repeats discussed above (Fig. 9AGo), the highest degree of homology with the B1 rodent sequences is observed within the ASE element (P14, 82% homology) followed by the three CRD domains (P13, P11–P12, and P10 with 48%, 37%, and 32% homology, respectively). This suggests that the ASE at least was likely present at the time of the mammalian radiation (100 million yr ago) and may even be older, because, based on phylogenetic reconstruction, the precursors to the human Alu and rodent B1 families appear to have originated before the divergence between the primate and the rodent lineages (67). Although Alu sequences have been associated with silencing activity in many genes (8, 9, 10, 11), to our knowledge the data presented here represent the most detailed dissection of the specific Alu sequences that mediate silencing activity. Consequently, further studies will be required to establish whether the domains in the hGH-1 3'-Alu element are related to those that mediate repression in other gene systems or whether the hGH-associated 3'-Alu members have evolved specialized regulatory features, including those domains that lie outside the Alu repeat, which are involved in modulating silencer function.


Figure 10
View larger version (55K):
[in this window]
[in a new window]
  		Fig. 10. Comparison of Rat B1 and Human Alu Sequences Associated with the hGH-1 and hGH-2 3'-FR

A, Sequence comparison by PILEUP analysis (GCG Wisconsin Package) of the rat B1 and human hGH-1 (ghlau6) and hGH-2 (ghalu30) Alu sequences. The locations of the ASE and CRD are indicated. Bold uppercase letters denote conserved sequences within the ASE and CRD. B, Identity and locus designation of the rat B1 sequences used in the comparison in panel A. P11, P12, Primers 11 and 12 etc.

 

   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Materials
The Mayo Clinic Molecular Biology Core Facility performed oligonucleotide synthesis and sequencing. Dithiothreitol and ATP were obtained from Sigma Chemical Co. (St. Louis, MO). [{gamma}32P]ATP was purchased from Amersham Biosciences (Piscataway, NJ). TEMED was from Roche Diagnostics (Indianapolis, IN) and acrylamide/bisacrylamide from Bio-Rad Laboratories (Hercules, CA). Luciferin was from Molecular Probes, Inc. (Eugene, OR). The numbering system for the hGH gene constructs is based on the hGH-1 gene sequence (GenBank accession no. M13438).

Plasmid Construction
The expression cassette containing the hGH promoter (GHp), hGH cDNA, and GH 3'-UTR/3'-FR was excised from plasmid pWZ_GH (20) with EcoRI, blunt ended, and inserted into the pGL3-Basic vector (Promega Corp., Madison, WI) restricted by SmaI/BamHI and blunt ended to create pGL3.GH. In plasmids p{Delta}2158 the BamHI/HindIII region encompassing the hGH cDNA and 3'-UTR is replaced with the corresponding fragments from plasmids pWZ.GHp.GHc.GH3'[{Delta}2158] (20). Deletion mutants, p{Delta}2236, p{Delta}2306, p{Delta}2353, p{Delta}2465, were generated by standard PCRs using a forward primer situated at nt 901 (Table 1Go, primer GHc_13) and a reverse primer that encompasses the corresponding hGH 3'-deletion (Table 1Go, primers GH3'{Delta} _14, GH3'{Delta} _15, GH3'{Delta} _16, and GH3'{Delta} _17, respectively). The PCR-generated products were then digested with BglII/SalI and inserted into the BglII/SalI fragment derived from pGL3.GH.


View this table:
[in this window]
[in a new window]
  		Table 1. Primers and Synthetic Oligonucleotides for GH3'-UTR/3'-FR Mutagenesis and Deletion Constructs.

 
The hGH promoter was inserted in the pGL3-Basic expression vector at the SmaI site to construct pGL3.GHp. Plasmid pGL3.GH was cut with NheI/BglII, blunt ended, and self-ligated to generate pT-3' that served as a template for standard PCR. The primer sets NTRE2-HIND-L/NTRE2-HIND-R, NTRE3-ECO-L/NTRE3-ECO-R, and R2401/ L2350 (Table 1Go) were used to generate plasmid pHdII, pRI3, pHdII_RI3, and p{Delta}50 by inverse PCR according to Hemsley et al. (68). These inverse PCRs replace nts 2280–2350 by a HindIII site in pHdII, nt 2401–2464, by an EcoRI site in pRI3, combines the two deletions of pHdII and pRI3 in pHdII_RI3, and deletes nts 2350–2401 in p{Delta}50. Plasmid pHdII_RI3 was restricted with HindIII/EcoRI, blunt ended, and self-ligated to generate pT{Delta}2280–2464. Plasmid pT-3' served as template to generate inserts for plasmids: pGL2.3, pGL2.3rev, p2.3GL, p2.3revGL, pGLS1, pGLS1rev, pGLS2, pGLS3, pGLS4, pGLS9, pGLS10, pGLS12, pGLS13, pGLS14, pGLS12rev, pGLS13rev, and pGLS14rev by standard PCR with the primers described in Table 1Go. Plasmid pT{Delta}2280–2464 was the template to generate the insert of plasmid pGLS8, pHdII_RI3 for pGLS5, and p{Delta}50 for pGLS19. Primers for these standard PCRs are shown in Table 1Go. All hGH 3'-UTR mutations were cloned into pGL3.GHp at the 3'-BamHI site, except for plasmids p2.3GL and p2.3revGL in which the insert was cloned at a 5'-MluI site. The identity of each mutant was verified by sequencing.

Promoter Swap and GH Promoter Mutants Construction
To construct plamid pTK, containing the luciferase reporter, the TK promoter was excised from plasmid pRMTKCAT (provided by Dr. K. L. Parker, Duke University Medical Center, Durham, NC) by SalI/BglII digestion followed by blunt ending with Klenow and cloned into the vector pGL3Basic (Promega) that had been restricted by SmaI/BglII and blunt ended. Plasmid pGL3.GH was cut with NheI/BglII, blunt ended, and self-ligated to generate pT-3' that served as a template to generate inserts for plasmids pTK2.3 and pTK2.3rev by standard PCR with the primers described in Table 1Go. The 414-bp fragment from nts 2158–2572 corresponding to the hGH silencer was then cloned into pTK at the 3'-SalI site. The expression vector pGL2 (Promega), harboring the SV40 promoter, was used to generate pGL2_2.3 and pGL2_2.3rev, in which GH 3'-silencer sequences were inserted at the 3'-BamHI site after PCR amplification using the primers described in Table 1Go. Plasmid pGL3.GHp was digested with HindIII/BamHI, blunt ended with T4 DNA polymerase, and self-ligated to generate plasmid pTGHp. Using pTGHp and the primers described in Table 2Go, inverse PCRs were performed according to the method of Hemsley et al. (68) to generate a series of mutated GH promoter constructs. The resulting plasmids were excised with NheI/XhoI, and the GH promoter mutant was inserted into pGL3.GHp and pGL2.3, restricted with NheI/XhoI, and dephosphorylated with calf intestinal alkaline phosphatase to generate the following: pGL3.ZN, pGL3.ZN2.3, pGL3.SP1, pGL3.SP1_2.3, pGL3.NF1, and pGL3.NF1_2.3. Plasmids containing the GH promoter with Pit-1 and CRE proximal/distal mutation were described previously (30). These mutant promoters were PCR amplified using the primers described in Table 2Go. The amplified mutants were then inserted into pGL3.GHp and pGL2.3 restricted with NheI/XhoI and dephosphorylated with calf intestinal alkaline phosphatase to generate pGL3.PIT, pGL3.P IT2.3, pGL3.CRE, and pGL3.CRE2.3.


View this table:
[in this window]
[in a new window]
  		Table 2. Promoter Primers

 
Cell Culture and DNA Transfections
Rat anterior pituitary tumor cell GC cells were grown in DMEM (Life Technologies, Inc., Gaithersburg, MD) supplemented with 10% fetal bovine serum (BioWhittaker, Walkersville, MD), 100 U/ml penicillin, 100 mg/ml streptomycin sulfate, 0.3 mg/ml glutamine (Life Technologies, Inc.). Cells were maintained at 37 C in an atmosphere containing 5% CO2 and 100% humidity. For each transfection 1 x 107 GC cells were suspended in 250 µl PBS, 0.1% D-glucose and electroporated (Gene Pulser, Bio-Rad Laboratories) at 250 V x 500 µF with 20 µg plasmid DNA in 50 µl Tris-EDTA buffer, pH 8. When the cDNA was used as reporter gene, cells were cotransfected with 2.5–5.0 µg of pGL3.hActp.luc+.SV3', containing the human ß-actin promoter, luciferase gene, and SV40 late polyA region, to serve as a transfection efficiency control. However, we found that, under our conditions, the transfection efficiency was very reproducible, rendering the use of a control plasmid unnecessary. Transfections were done in 10 replicates unless otherwise indicated.

Reporter Assays
At 24–48 h after transfection, cell culture media were collected to measure secreted hGH. Intracellular hGH was determined from cell extracts. Cells were harvested in 200 µl lysis buffer (0.1 M K2HPO4, pH 7.8; 1 mM dithiothreitol; 0.5% Nonidet P-40), and cell extracts were collected after centrifugation (13,000 x g for 10 min). Protein concentration was measured by the Coomassie dye-binding assay (Pierce Chemical Co., Rockford, IL). Secreted and intracellular hGH concentrations were measured by a two-site immunoenzymatic immunoassay (Access Ultrasensitive hGH Reagent Pack; Beckman Coulter, Inc., Brea, CA) using a mouse-anti-hGH coupled to magnetic particles and alkaline phosphatase conjugated to goat antimouse antibody performed on an automated Beckman Dxl system. The Mayo Clinic Immunochemistry Core Facility performed the GH assays. Luciferase was assayed with 10 µl of cell extract in 25 mM glycylglycine (pH 7.8), 15 mM MgSO4, 5 mM ATP (65).

Nuclear Extract and DNA Binding Assay
GC crude nuclear extracts were prepared using a Nuclear/Cytosol fractionation kit from MBL International Corp. (Woburn, MA) according to the manufacturer. All probes were generated by PCR using the primers described in Table 3Go. Probe labeling and EMSA were performed using the Gel Shift Assay Systems kit from Promega according to the manufacturer’s specifications in the absence or presence of unlabeled double-stranded competitor. Nonspecific competitor is the following unlabeled DR4-mutant double-stranded probe: 5'-ATCCTCCAATCACAGGCAATCAGAG-3'.


View this table:
[in this window]
[in a new window]
  		Table 3. Primers for EMSA Probes

 
ChIP Assay
ChIP assays were performed with the EZ-CHIP assay kit following the recommendations of the manufacturer (Millipore, Billerica, MA).

Statistics
All results were analyzed by ANOVA. Individual differences were assessed by post hoc Bonferroni/Dunn tests and were deemed significant at the P < 0.05 level. The controls for the comparisons are designated in the figure legends.


   FOOTNOTES
 
This work was supported by National Institutes of Health Grants R01 DK41206 (to N.L.E.) and T32 DK07352 (to M.A.T.).

Disclosure: M.A.T., M.S., and N.L.E. have nothing to declare.

First Published Online June 8, 2006

Abbreviations: AcH3, Acetylated histone H3; ASE, anti-silencer domain; ChIP, chromatin immunoprecipitation; CRD, core repressor domain; CRE, cAMP response element; CS, chorionic somatomammotropin; FR, flanking region; NF-1, nuclear factor 1; nts, nucleotides; SINEs, short interspersed elements; SV40, simian virus 40; TK, thymidine kinase; TSA, trichostatin A; UTR, untranslated region.

Received for publication March 31, 2006. Accepted for publication June 1, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Tomilin NV 1999 Control of genes by mammalian retroposons. Int Rev Cytol 186:1–48[Medline]
  2. Szmulewicz MN, Novick GE, Herrera RJ 1998 Effects of Alu insertions on gene function. Electrophoresis 19:1260–1264[CrossRef][Medline]
  3. Kazakov VI, Tomilin NV 1996 Increased concentration of some transcription factor binding sites in human retroposons of the Alu family. Genetica 97:15–22[CrossRef][Medline]
  4. Hambor JE, Mennone J, Coon ME, Hanke JH, Kavathas P 1993 Identification and characterization of an Alu-containing, T-cell-specific enhancer located in the last intron of the human CD8 {alpha} gene. Mol Cell Biol 13:7056–7070[Abstract/Free Full Text]
  5. Norris J, Fan D, Aleman C, Marks JR, Futreal PA, Wiseman RW, Iglehart JD, Deininger PL, McDonnell DP 1995 Identification of a new subclass of Alu DNA repeats which can function as estrogen receptor-dependent transcriptional enhancers. J Biol Chem 270:22777–22782[Abstract/Free Full Text]
  6. Oliviero S, Monaci P 1988 RNA polymerase III promoter elements enhance transcription of RNA polymerase II genes. Nucleic Acids Res 16:1285–1293[Abstract/Free Full Text]
  7. Brini AT, Lee GM, Kinet JP 1993 Involvement of Alu sequences in the cell-specific regulation of transcription of the {gamma} chain of Fc and T cell receptors. J Biol Chem 268:1355–1361[Abstract/Free Full Text]
  8. Saffer JD, Thurston SJ 1989 A negative regulatory element with properties similar to those of enhancers is contained within an Alu sequence. Mol Cell Biol 9:355–364[Abstract/Free Full Text]
  9. Hewitt SM, Fraizer GC, Saunders GF 1995 Transcriptional silencer of the Wilms’ tumor gene WT1 contains an Alu repeat. J Biol Chem 270:17908–17912[Abstract/Free Full Text]
  10. Schweiger M, Oei SL, Herzog H, Menardi C, Schneider R, Auer B, Hirsch-Kauffmann M 1995 Regulation of the human poly(ADP-ribosyl) transferase promoter via alternative DNA racket structures. Biochimie 77:480–485[Medline]
  11. Wu J, Grindlay GJ, Bushel P, Mendelsohn L, Allan M 1990 Negative regulation of the human {epsilon}-globin gene by transcriptional interference: role of an Alu repetitive element. Mol Cell Biol 10:1209–1216[Abstract/Free Full Text]
  12. Miller WL, Eberhardt NL 1983 Structure and evolution of the growth hormone gene family. Endocr Rev 4:97–130[Abstract/Free Full Text]
  13. Hirt H, Kimelman J, Birnbaum MJ, Chen EY, Seeburg PH, Eberhardt NL, Barta A 1987 The human growth hormone gene locus: structure, evolution, and allelic variations. DNA 6:59–70[Medline]
  14. Toda Y, Tomita M 1997 Alu elements as an aid in deciphering genome rearrangements. Gene 205:173–176[CrossRef][Medline]
  15. Jones BK, Monks BR, Liebhaber SA, Cooke NE 1995 The human growth hormone gene is regulated by a multicomponent locus control region. Mol Cell Biol 15:7010–7021[Abstract]
  16. Castrillo JL, Theill LE, Karin M 1991 Function of the homeodomain protein GHF1 in pituitary cell proliferation. Science 253:197–199[Abstract/Free Full Text]
  17. Mangalam HJ, Albert VR, Ingraham HA, Kapiloff M, Wilson L, Nelson C, Elsholtz H, Rosenfeld MG 1989 A pituitary POU domain protein, Pit-1, activates both growth hormone and prolactin promoters transcriptionally. Genes Dev 3:946–958[Abstract/Free Full Text]
  18. Shewchuk BM, Liebhaber SA, Cooke NE 2002 Specification of unique Pit-1 activity in the hGH locus control region. Proc Natl Acad Sci USA 99:11784–11789[Abstract/Free Full Text]
  19. Jin Y, Surabhi RM, Fresnoza A, Lytras A, Cattini PA 1999 A role for A/T-rich sequences and Pit-1/GHF-1 in a distal enhancer located in the human growth hormone locus control region with preferential pituitary activity in culture and transgenic mice. Mol Endocrinol 13:1249–1266[Abstract/Free Full Text]
  20. Zhang W, Brooks RL, Silversides DW, West BL, Leidig F, Baxter JD, Eberhardt NL 1992 Negative thyroid hormone control of human growth hormone gene expression is mediated by 3'-untranslated/3'-flanking DNA. J Biol Chem 267:15056–15063[Abstract/Free Full Text]
  21. Le Cam A, Legraverend C 1995 Transcriptional repression, a novel function for 3' untranslated regions. Eur J Biochem 231:620–627[Medline]
  22. Avisar N, Shiftan L, Ben-Dror I, Havazelet N, Vardimon L 1999 A silencer element in the regulatory region of glutamine synthetase controls cell type-specific repression of gene induction by glucocorticoids. J Biol Chem 274:11399–11407[Abstract/Free Full Text]
  23. Malone CS, Omori SA, Gangadharan D, Wall R 2001 Leukocyte-specific expression of the pp52 (LSP1) promoter is controlled by the cis-acting pp52 silencer and anti-silencer elements. Gene 268:9–16[CrossRef][Medline]
  24. Izmailova ES, Zehner ZE 1999 An antisilencer element is involved in the transcriptional regulation of the human vimentin gene. Gene 230:111–120[CrossRef][Medline]
  25. Yannoutsos N, Barreto V, Misulovin Z, Gazumyan A, Yu W, Rajewsky N, Peixoto BR, Eisenreich T, Nussenzweig MC 2004 A cis element in the recombination activating gene locus regulates gene expression by counteracting a distant silencer. Nat Immunol 5:443–450[CrossRef][Medline]
  26. Cohen LE, Wondisford FE, Radovick S 1996 Role of Pit-1 in the gene expression of growth hormone, prolactin, and thyrotropin. Endocrinol Metab Clin N Am 25:523–540[CrossRef][Medline]
  27. Ingraham HA, Flynn SE, Voss JW, Albert VR, Kapiloff MS, Wilson L 1990 The POU-specific domain of Pit-1 is essential for sequence-specific, high affinity DNA binding and DNA-dependent Pit-1-Pit-1 interactions. Cell 61:1021–1033[CrossRef][Medline]
  28. Frankenne F, Scippo ML, Van Beeumen J, Igout A, Hennen G 1990 Identification of placental human growth hormone as the growth hormone-V gene expression product. J Clin Endocrinol Metab 71:15–18[Abstract/Free Full Text]
  29. Lipkin SM, Naar AM, Kalla KA, Sack RA, Rosenfeld MG 1993 Identification of a novel zinc finger protein binding a conserved element critical for Pit-1-dependent growth hormone gene expression. Genes Dev 7:1674–1687[Abstract/Free Full Text]
  30. Shepard AR, Zhang W, Eberhardt NL 1994 Two CGTCA motifs and a GHF1/Pit1 binding site mediate cAMP-dependent protein kinase A regulation of human growth hormone gene expression in rat anterior pituitary GC cells. J Biol Chem 269:1804–1814[Abstract/Free Full Text]
  31. Newell-Price J, Clark AJ, King P 2000 DNA methylation and silencing of gene expression. Trends Endocrinol Metab 11:142–148[CrossRef][Medline]
  32. Attwood JT, Yung RL, Richardson BC 2002 DNA methylation and the regulation of gene transcription. Cell Mol Life Sci 59:241–257[CrossRef][Medline]
  33. Peterson CL, Laniel MA 2004 Histones and histone modifications. Curr Biol 14:R546–R551
  34. Breiling A, Orlando V 2002 SET domain proteins reSET gene expression. Nat Struct Biol 9:894–896[CrossRef][Medline]
  35. Gaston K, Jayaraman PS 2003 Transcriptional repression in eukaryotes: repressors and repression mechanisms. Cell Mol Life Sci 60:721–741[CrossRef][Medline]
  36. Kimura AP, Liebhaber SA, Cooke NE 2004 Epigenetic modifications at the human growth hormone locus predict distinct roles for histone acetylation and methylation in placental gene activation. Mol Endocrinol 18:1018–1032[Abstract/Free Full Text]
  37. Elefant F, Cooke NE, Liebhaber SA 2000 Targeted recruitment of histone acetyltransferase activity to a locus control region. J Biol Chem 275:13827–13834[Abstract/Free Full Text]
  38. Phan D, Cheng CJ, Galfione M, Vakar-Lopez F, Tunstead J, Thompson NE, Burgess RR, Najjar SM, Yu-Lee LY, Lin SH 2004 Identification of Sp2 as a transcriptional repressor of carcinoembryonic antigen-related cell adhesion molecule 1 in tumorigenesis. Cancer Res 64:3072–3078[Abstract/Free Full Text]
  39. Doetzlhofer A, Rotheneder H, Lagger G, Koranda M, Kurtev V, Brosch G, Wintersberger E, Seiser C 1999 Histone deacetylase 1 can repress transcription by binding to Sp1. Mol Cell Biol 19:5504–5511[Abstract/Free Full Text]
  40. Xu L, Lavinsky RM, Dasen JS, Flynn SE, McInerney EM, Mullen TM, Heinzel T, Szeto D, Korzus E, Kurokawa R, Aggarwal AK, Rose DW, Glass CK, Rosenfeld MG 1998 Signal-specific co-activator domain requirements for Pit-1 activation. Nature 395:301–306[CrossRef][Medline]
  41. Diamond SE, Gutierrez-Hartmann A 2000 The Pit-1ß domain dictates active repression and alteration of histone acetylation of the proximal prolactin promoter. J Biol Chem 275:30977–30986[Abstract/Free Full Text]
  42. Scully KM, Jacobson EM, Jepsen K, Lunyak V, Viadiu H, Carriere C, Rose DW, Hooshmand F, Aggarwal AK, Rosenfeld MG 2000 Allosteric effects of Pit-1 DNA sites on long-term repression in cell type specification. Science 290:1127–1131[Abstract/Free Full Text]
  43. Luo RX, Postigo AA, Dean DC 1998 Rb interacts with histone deacetylase to repress transcription. Cell 92:463–473[CrossRef][Medline]
  44. Alberts AS, Geneste O, Treisman R 1998 Activation of SRF-regulated chromosomal templates by Rho-family GTPases requires a signal that also induces H4 hyperacetylation. Cell 92:475–487[CrossRef][Medline]
  45. Braunstein M, Rose AB, Holmes SG, Allis CD, Broach JR 1993 Transcriptional silencing in yeast is associated with reduced nucleosome acetylation. Genes Dev 7:592–604[Abstract/Free Full Text]
  46. Macfarlane WM 2000 Demystified ... transcription. Mol Pathol 53:1–7[Abstract/Free Full Text]
  47. Laimins L, Holmgren-Konig M, Khoury G 1986 Transcriptional "silencer" element in rat repetitive sequences associated with the rat insulin 1 gene locus. Proc Natl Acad Sci USA 83:3151–3155[Abstract/Free Full Text]
  48. Durst KL, Hiebert SW 2004 Role of RUNX family members in transcriptional repression and gene silencing. Oncogene 23:4220–4224[CrossRef][Medline]
  49. Pirrotta V, Poux S, Melfi R, Pilyugin M 2003 Assembly of Polycomb complexes and silencing mechanisms. Genetica 117:191–197[CrossRef][Medline]
  50. Rahman M, Hirabayashi Y, Ishii T, Kodera T, Watanabe M, Takasawa N, Sasaki T 2001 A repressor element in the 5'-untranslated region of human Pax5 exon 1A. Gene 263:59–66[CrossRef][Medline]
  51. Benderdour M, Tardif G, Pelletier JP, Dupuis M, Geng C, Martel-Pelletier J 2002 A novel negative regulatory element in the human collagenase-3 proximal promoter region. Biochem Biophys Res Commun 291:1151–1159[CrossRef][Medline]
  52. Gaunitz F, Heise K, Gebhardt R 2004 A silencer element in the first intron of the glutamine synthetase gene represses induction by glucocorticoids. Mol Endocrinol 18:63–69[Abstract/Free Full Text]
  53. O’Prey J, Harrison PR 1995 Tissue-specific regulation of the rabbit 15-lipoxygenase gene in erythroid cells by a transcriptional silencer. Nucleic Acids Res 23:3664–3672[Abstract/Free Full Text]
  54. Uzan G, Prandini MH, Berthier R 1995 Regulation of gene transcription during the differentiation of megakaryocytes. Thromb Haemostasis 74:210–212[Medline]
  55. Galehshahi FS, Goke B, Lankat-Buttgereit B 2000 A novel silencer element repressing expression of the GLP-1 receptor gene in fibroblasts and pancreatic A-cells, but not in pancreatic B- and D-cells. Peptides 21:1169–1176[CrossRef][Medline]
  56. Malone CS, Omori SA, Wall R 1997 Silencer elements controlling the B29 (Igß) promoter are neither promoter- nor cell-type-specific. Proc Natl Acad Sci USA 94:12314–12319[Abstract/Free Full Text]
  57. Yang J, Glozak MA, Blomberg BB 1995 Identification and localization of a developmental stage-specific promoter activity from the murine {lambda} 5 gene. J Immunol 155:2498–2514[Abstract]
  58. Young RL, Korsmeyer SJ 1993 A negative regulatory element in the bcl-2 5'-untranslated region inhibits expression from an upstream promoter. Mol Cell Biol 13:3686–3697[Abstract/Free Full Text]
  59. Chen JH, Jeha S, Oka T 1993 Negative regulatory elements in the human ETS1 gene promoter. Oncogene 8:133–139[Medline]
  60. Nomura M, Tsujimura A, Begum NA, Matsumoto M, Wabiko H, Toyoshima K, Seya T 2000 Identification and characterization of a silencer regulatory element in the 3'-flanking region of the murine CD46 gene. Biochem J 351:353–365
  61. Baptista HA, Avellar MC, Araujo RC, Pesquero JL, Schanstra JP, Bascands JL, Esteve JP, Paiva AC, Bader M, Pesquero JB 2002 Transcriptional regulation of the rat bradykinin B2 receptor gene: identification of a silencer element. Mol Pharmacol 62:1344–1355[Abstract/Free Full Text]
  62. Maouche L, Lucien N, Cartron JP, Chretien S 1995 A CCACC motif mediates negative transcriptional regulation of the human erythropoietin receptor. Eur J Biochem 233:793–799[Medline]
  63. Wu Y, Diab I, Zhang X, Izmailova ES, Zehner ZE 2004 Stat3 enhances vimentin gene expression by binding to the antisilencer element and interacting with the repressor protein, ZBP-89. Oncogene 23:168–178[CrossRef][Medline]
  64. Chen EY, Liao YC, Smith DH, Barrera-Saldana HA, Gelinas RE, Seeburg PH 1989 The human growth hormone locus: nucleotide sequence, biology, and evolution. Genomics 4:479–497[CrossRef][Medline]
  65. Jiang SW, Eberhardt NL 1994 The human chorionic somatomammotropin gene enhancer is composed of multiple DNA elements that are homologous to several SV40 enhansons. J Biol Chem 269:10384–10392[Abstract/Free Full Text]
  66. Jacquemin P, Oury C, Peers B, Morin A, Belayew A, Martial JA 1994 Characterization of a single strong tissue-specific enhancer downstream from the three human genes encoding placental lactogen. Mol Cell Biol 14:93–103[Abstract/Free Full Text]
  67. Quentin Y 1994 A master sequence related to a free left Alu monomer (FLAM) at the origin of the B1 family in rodent genomes. Nucleic Acids Res 22:2222–2227[Abstract/Free Full Text]
  68. Hemsley A, Arnheim N, Toney MD, Cortopassi G, Galas DJ 1989 A simple method for site-directed mutagenesis using the polymerase chain reaction. Nucleic Acids Res 17:6545–6551[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
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 Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Trujillo, M. A.
Right arrow Articles by Eberhardt, N. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Trujillo, M. A.
Right arrow Articles by Eberhardt, N. L.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals