help button home button Endocrine Society Molecular Endocrinology ENDO 08 Sessions Library
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fenstermacher, D. A.
Right arrow Articles by Joseph, D. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fenstermacher, D. A.
Right arrow Articles by Joseph, D. R.
Molecular Endocrinology 11 (9): 1387-1400
Copyright © 1997 by The Endocrine Society

DNA Sequences and Their Binding Proteins Required for Sertoli Cell-Specific Transcription of the Rat Androgen-Binding Protein Gene

David A. Fenstermacher and David R. Joseph

The Curriculum in Genetics and Molecular Biology (D.A.F.) Department of Pediatrics (D.R.J.) The Laboratories for Reproductive Biology The University of North Carolina Chapel Hill, North Carolina 27599


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The rat androgen-bindng protein (ABP) gene is transcriptionally regulated from two promoters: the P1 promoter regulates expression of transcripts starting at exon 1, whereas PA regulates transcripts containing exon A. The P1 promoter directs cell-specific gene regulation of ABP secreted by Sertoli cells. In this study, the Sertoli cell-regulatory sequences of P1 were further examined using a luciferase reporter system with three cell lines, including a Sertoli cell line (MSC-1) that expresses the ABP gene. Deletion mapping experiments determined that the sequences required for full activity in MSC-1 cells were included within 619 bp of the start site and identified several regions that demonstrated increased luciferase activity: the -583 bp to -564 bp, -503 bp to -484 bp, and -114 bp to -65 regions. The activities contributed by each region were much higher (up to 120-fold) in MSC-1 cells than in MA10 Leydig or NIH3T3 fibroblast cells. Nuclear-binding proteins and their binding sequences were identified using several molecular biology techniques. Complexes formed by nuclear proteins of MSC-1, MA10, and NIH3T3 cells, which bind specifically to the -114 to -65-bp region, were identified using gel retardation assays. Furthermore, the inverted repeat sequence in this region, 5'-AGGGTCAGTGTCCCT-3' was identified by deoxyribonuclease (DNase) I footprinting. The regulatory element contained within the -503 to -484-bp region was identified by scanning mutagenesis, but no protein was found that bound to this sequence by gel retardation or DNase I protection assays. This element is characterized by the core sequence, 5'-GGAGGC- 3'. The third regulatory region (residues -583 to -564) bound a protein complex that retarded mobility of the free DNA probe in a gel shift assay. Using several techniques, the binding sequence was identified as 5'-TTCATAGTATCCATTAAAC-3'. In summary, these data have identified several transcriptional regulatory sequences and their binding proteins, which appear to play a role in the Sertoli cell-specific expression of the ABP gene.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The rat androgen-binding protein (ABP) gene encodes a 90-kDa homodimer protein that is secreted by testicular Sertoli cells (1, 2, 3, 4). The extracellular protein is characterized by its high-affinity binding to testosterone, dihydrotestosterone, and estradiol (5, 6). The ABP gene is expressed in the adult rat testis and is also responsible for encoding the hepatocyte-secreted protein, sex hormone-binding globulin (SHBG) (1, 7). ABP and SHBG are thought to regulate the bioavailability of sex steroids in extracellular spaces and likely have a much broader function, possibly as a hormone (4, 8, 9). In addition, ABP has been used extensively as a marker of Sertoli cell function (10).

Previous regulation studies using hypophysectomized rats have implicated hormones in the regulation of testicular ABP. These studies suggested that testosterone and FSH regulated ABP and ABP mRNA production but did not determine whether the effect was indirect or direct on the gene (11, 12, 13, 14, 15). Hansson et al. (16) described that testicular ABP levels increased in Tfm rats, which lack the androgen receptor, thereby providing evidence that androgens do not directly regulate ABP. Experiments with primary Sertoli cell cultures demonstrated that androgens and FSH did modestly increase secreted ABP levels in the medium, but the data could not determine whether the regulation affected transcription/translation or stability of the ABP protein (17). Later, the use of DNA hybridization analysis techniques demonstrated that the addition of androgens to Sertoli cell cultures did not affect ABP mRNA levels (18). Furthermore, dihydrotestosterone did not alter gene transcription in vitro with a promoter-luciferase reporter construct (our unpublished results). In addition, FSH did not increase ABP mRNA levels in primary Sertoli cell cultures but caused dramatic increases in the mRNA levels of c-fos, c-jun, inhibin, and tissue plasminogen activator (20, 21).

The rat ABP gene has been sequenced and its activity partially characterized (our unpublished results and Refs. 22, 23). The gene consists of promoters P1 and PA; P1 regulates the synthesis of the secreted testicular ABP. Several characteristics of a GC-rich "housekeeping" gene are present within the ABP P1 promoter region, including localized high GC content and several putative Sp1 sites (our unpublished results and Ref. 22). These characteristics are even more prevalent in the alternate ABP promoter PA, located 15 kb upstream of the P1 region (23), but this promoter does not show Sertoli cell-specific expression (23, 24). Based on sequence analysis and mutagenesis experiments, no TATA and/or CCAAT box sequences were identified within the ABP P1 promoter (our unpublished results and Ref. 22). The ABP P1 promoter also possesses several characteristics not consistent with a GC-rich type of promoter, such as site-specific initiation of transcription (22, 23, 25, 26). The major transcriptional start site has been previously mapped by primer extension, RNase protection, and primer walking; it is located 36 bp upstream of the translational start site (our unpublished results and Refs. 22, 23). The nucleotide sequences flanking the major start site are consistent with the presence of an initiator element (our unpublished results and Refs. 22, 23, 27). A minor start site is apparently located upstream of the major site (our unpublished results).

Several other genes, which have been characterized as markers of Sertoli cell function, have been studied to ascertain the mechanism of transcriptional regulation. The FSH receptor (FSHR), tissue plasminogen activator, mullerian inhibitory substance (MIS), and the inhibin Bß-subunit genes have specific start sites, but no TATA sequences have been identified (28, 29, 30, 31). The transferrin gene, which is also expressed by Sertoli cells, contains the classic TATA and CCAAT sequences that direct the expression of the gene (32). No Sertoli cell-specific cis-regulatory sequences have been identified in any of these Sertoli cell-expressed genes.

Probably the most interesting characteristics of genes that are specifically expressed by Sertoli cells are the DNA sequences that dictate cell type specificity. In this manuscript, the Sertoli cell-specific transcriptional regulation of the ABP gene P1 promoter was characterized using a luciferase reporter system. Three cis-acting sequences were identified, which appear to play a key role in Sertoli cell-specific expression of the ABP gene. In addition, the results identify several protein complexes and their DNA-binding sequences that appear to be involved in the Sertoli cell-specific transcription of the ABP gene.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Deletion Mapping of ABP Promoter-Enhancer Region
Transgenic mouse experiments demonstrated that 1.5 kb of upstream sequence with intron-exon sequences was sufficient to specifically direct rat ABP gene expression in the testis (33). To identify important regulatory sequences within this 1.5-kb region, deletion mapping experiments were performed. Various length fragments of the 1.5-kb 5'-regulatory sequence were amplified by PCR and cloned into the reporter vector pXP1 DNA, which contains a promotorless luciferase cDNA (34). The DNA:pXP1 constructs were used to transfect MSC-1 Sertoli cells, and promoter activities were measured with luciferase activity. Initial experiments revealed that sequences within the ABP promoter region expressed one tenth the luciferase activity compared with a construct containing a viral thymidine kinase (TK) promoter. In addition, full luciferase activity was obtained using 619 bp upstream of the major transcriptional start site (data not shown). The entire 1.5-kb construct had only 41% of the activity associated with the DNA-619 construct.

Deletion mapping experiments were continued by analyzing 12 fragments within the 619-bp DNA region; each DNA contained the gene promoter and varied in length at the 5'-end by approximately 50 bp (Fig. 1Go). The fragments were unidirectionally cloned into the luciferase vector pXP1, and the constructs were assayed on the three cell lines: MSC-1, MA10 Leydig, and NIH3T3 fibroblast cells. Data are expressed as the relative increase over the smallest construct (DNA-14:pXP1), which contains the major transcriptional start site and 14 bp of upstream sequence. The use of DNA-14 as the reference point provides an internal control that reduces biases caused by transfection efficiency variations in the three cell lines. Therefore, absolute luciferase activities between cell lines are not directly compared. The low activities of pXP1:DNA-14 in the three cell lines were approximately the same. Figure 2AGo shows that transfections of the DNA -619:pXP1 construct yielded a 362-fold increase in luciferase activity on MSC-1 cells as compared with the DNA-14:pXP1. Thus, the insertion of the 619-bp upstream sequence with the minimal ABP gene promoter increased the activity 362-fold in MSC-1 cells. On the contrary, the activity was only modestly increased in MA10 and NIH3T3 cells, 5-fold and 3-fold, respectively. Figure 2AGo summarizes the results with the other deletion mutants. With each construct, increased activity was higher in the MSC-1 cells than in the other cell lines. As the fragments were extended by 50 bp, the activity changed in increments to the maximum activity with DNA-619 construct. As the length of the fragment increased, three fragments demonstrated dramatic increases in activity as compared with the adjacent smaller fragment: DNA-114, DNA-543, and DNA-619 (Fig. 2AGo). Interestingly, only one fragment demonstrated a dramatic increase with MA10 cells; DNA-114 yielded a 13-fold increase over DNA-63. There was little activity with any fragment on NIH3T3 cells. These data suggest the presence of important regulatory elements between residues -65 to -114, -484 to -543, and -544 to -619. Moreover, the two upstream elements act specifically in MSC-1 Sertoli cells and not MA10 or NIH3T3 cells. Furthermore, an inhibitory activity is associated with residues -164 to -114.



View larger version (48K):
[in this window]
[in a new window]
  		Figure 1. The 5'-Regulatory Sequence of the Rat ABP Gene

The residue numbering on each flank is based on the transcriptional start site (*) 36 bp upstream of the initiating Met codon (ATG). An arrowhead marks the putative minor start site (22 ). The underlined sequences indicate some of the oligonucleotides used as primers for PCR. Other primers are shown in Materials and Methods. In addition to the sequence indicated above, each primer contained a sequence creating a unique restriction site (forward primers — SstI, reverse primer — HindIII) and 6 bp of nonsense sequence at the 5'-end.

 


View larger version (19K):
[in this window]
[in a new window]
  		Figure 2. Deletion Mapping Analysis of the ABP Gene

Fragments were amplified by PCR and unidirectionally cloned into pXP1 DNA, and the constructs were used to transfect MSC-1, MA10, and NIH3T3 cell lines as described in Materials and Methods. Fragments are labeled at the most 5'-nucleotide as numbered negatively from the transcriptional start site. Each DNA fragment contained the identified transcription start site but not the initiating Met residue. Luciferase activities were calculated by subtracting the relative light units (RLU) obtained with the pXP1 plasmid with no insert and dividing this activity by the DNA-14 construct activity. This DNA-14 construct contained the transcriptional start site and 14 bp of upstream sequence. A, Mapping experiments with fragments differing by 50 bp within the 619-bp region. B, Deletion mapping of the -484-bp to -619-bp region using amplified fragments differing by 20 bp. Values represent mean fold increase (±SEM; n = 7). The RLU activities represent promoter activity.

 
The -619 to -484-bp region of the ABP gene appeared to contain sequences that act as Sertoli cell-specific transcriptional regulatory elements. To determine the sequences within this region that may contribute to increased luciferase activities, further mapping was performed. Fragments differing by 20 bp in length were cloned into the pXP1 plasmid DNA and assayed as described above (Fig. 2BGo). Constructs of ABP gene fragments containing 503, 523, 543, and 563 bp upstream of the start site resulted in less activity than the DNA-583 construct. These data further pinpointed the location of upstream regulatory elements between residues -583 to -564 and -503 to -484. These changes in luciferase activities were only observed in the MSC-1 cell line and not in the MA10 or NIH3T3 cell line (Fig. 2BGo). The deletion mapping experiments presented here have identified three regions of the ABP gene that appear to play a critical role in Sertoli cell regulation. The regions are -114 to -65 bp, -503 to -484 bp, and -583 to -564 bp.

Analysis of the -114 to -65-bp Region
Mutagenesis of Putative Regulatory Sequences
To further investigate the sequences important for ABP gene expression, several mutants were constructed within the -65 to -114-bp region. The sequences were selected by identifying regions of high homology between the rat and human ABP genes. Two regions were identified: one contains an inverted repeat sequence from -87 to -101 bp whereas the second contains a long stretch of pyrimidines from -61 to -78 bp. Both sequences were independently mutated using oligonucleotides 19 and 20 in DNA-114:pXP1. Each mutant plasmid was used to transfect MSC-1, MA10, and NIH3T3 cell lines to assess luciferase expression. Figure 3Go demonstrates that mutation of either region in the DNA-114 construct resulted in a 66% or 79% increase in activity with MSC-1 cells. Transformation of MA10 cells with the -87 to -101-bp mutant in the DNA-114: pXP1 construct resulted in a decrease of 59%, whereas the -61 to -74-bp mutant resulted in 10-fold activity increase (Fig. 3Go). In contrast, luciferase activity change after transformation of NIH3T3 cells was not significant for the -87 to -101-bp mutant. Similar to the large increase observed in MA10 cells, the -61 to -74-bp mutant of DNA-114 using NIH3T3 cells yielded a significant increase (2.5-fold) in activity (Fig. 3Go).



View larger version (24K):
[in this window]
[in a new window]
  		Figure 3. Mutagenesis of Putative Regulatory Sequences within the -114 to -65-bp Region

Mutants were generated using the DNA-114:pXP1 construct and assayed for luciferase activity in MSC-1, MA10, AND NIH3T3 cells. Mutations altered the -101 to -87-bp region (inverted repeat, 5'-AGGGTCAGTGTCCCT-3' to 5'-CTCGAGCTGACTATG) and residues -74 to -61 (pyrimidine stretch, 5'-CCTTCTTCCCCCGG-3' to 5'-GTAATCATGAG-CTC-3'). To assess the changes due to the mutation the activities were presented as a percent of the parent construct activity (100%). Values represent mean percent difference (±SEM, n = 4–8). All calculations were performed after subtracting background activity of the pXP1 plasmid.

 
Gel Retardation
To identify transcription factors that bind to these sequences, gel retardation assays were conducted using nuclear proteins from MSC-1, MA10, and NIH3T3 cell lines. A 32P-labeled DNA probe based on the -133 to -46-bp sequence was amplified by PCR. Several parameters were tested to optimize protein binding to double-stranded DNA and gel resolution of the DNA-protein complexes according to previously published reports (35, 36, 37). The optimal conditions for the binding of MSC-1 nuclear proteins to DNA-133 to -46 were found to contain 50 or 100 mM KCl in the binding buffer (see Materials and Methods). Although the presence of magnesium increased binding in the presence of 25 mM KCl, the presence of 5 or 10 mM MgCl2 with higher KCl concentrations had no apparent effect (data not shown). All subsequent gel retardation experiments using the DNA-133 to -46 probe were performed in a reaction containing 50 mM KCl without MgCl2. Four individual bands (A-D) are visible in Fig. 4AGo, lanes 1 and 3 that contain 20 or 40 µg of MSC-1 nuclear protein. Similar complexes were observed with MA10 and NIH3T3 nuclear proteins. In addition, band E was present with MA10 and NIH3T3 proteins (Fig. 4AGo, lanes 5, 7, 9, and 11). With a short exposure time discrete E bands are clearly visible (data not shown). The specificity of the reaction was tested by including an excess of unlabeled probe in the reaction. The addition of excess unlabeled DNA-133 to -46 eliminated or reduced the amount of bands A-D using MSC-1 nuclear proteins (Fig. 4AGo, lanes 2 and 4). MA10 and NIH3T3 complex E was also competed by an excess of unlabeled probe (Fig. 4AGo, lanes 6, 8, 10, and 12). Whereas the homologous DNA efficiently competed for complex formation, separate experiments demonstrated that the presence of Sp1, glucocorticoid regulatory element (GRE), and AP-1 consensus sequences in the binding reaction did not affect formation of the DNA-protein complexes in any cell line (data not shown). These data provide evidence that DNA-protein complex formation depends on a specific DNA sequence(s). Thus, it appears that several nuclear proteins bind to DNA-133 to -46, generating multiple bands. Interestingly, the formation of some complexes is cell specific.



View larger version (76K):
[in this window]
[in a new window]
  		Figure 4. Gel Retardation Assays of the -114 to -65-bp Sequence

The 32P-labeled DNA probe containing the -114 to -65-bp sequence was incubated with nuclear proteins from MSC-1, MA10, or NIH3T3 cells, and the products were analyzed by PAGE. A, Cell specificity of mobility shift assays. Binding conditions were the optimal conditions determined as described in Materials and Methods. Lanes 1–12, MSC-1, MA10, and NIH3T3 nuclear extracts in the presence or absence of competitor DNA (unlabeled DNA-114 to -65). Specific complexes are labeled A-F. B, Gel retardation assay using the DNA-114 to -65 probe containing the -101 to -87-bp mutation with 20 or 40 µg of MSC-1, MA10, or NIH3T3 nuclear protein (lanes 2–7). Lane 1, Native DNA-114 to -65 without mutation.

 
As described above, mutants of the -114 to -65-bp region (inverted repeat) were used to determine the importance of specific sequences on ABP promoter activity. One mutation altered the -101 to -87-bp sequence (an inverted repeat) while another mutated the -74 to -61-bp sequence (a pyrimidine stretch). To further investigate the cell-specific regulation of this region, probes containing the mutated sequences were prepared to test nuclear protein binding by gel retardation. Figure 4BGo shows the effects of nuclear proteins from MSC-1, MA10, and NIH3T3 on mutant DNA migration. The reaction using the altered -101 to -87-bp DNA and MSC-1 nuclear proteins yielded only wild type complex D and a small amount of complex B. Complexes A, C, and E were absent, but an additional complex, F, which was not a product of wild type DNA, appeared and migrated slower than complex D. Similarly, the MA10 and NIH3T3 nuclear proteins yielded complex patterns with the mutant DNA similar to the pattern observed with MSC-1 cell proteins. (A small amount of complex B and E was obtained with MSC-1 and NIH3T3 nuclear proteins, respectively.) The most notable difference was the disappearance of complex E with MA10 proteins. The formation of a small amount of complex E with NIH3T3 nuclear proteins using the mutant -101 to -87-bp probe suggests that it may be different from the MA10 complex, even though they both migrate at approximately the same rate. These data further demonstrate that complex A, B, C, and E formation is specific, relying on specific DNA sequences. The formation of complex F, which was not present in the MSC-1 wild type reaction products, appears only when the mutated probe is used. Thus, the mutation of region -101 to -87 sequence reduces transcriptional activity in MA10 cells and alters the binding of nuclear proteins. The -74 to -61-bp mutant, which removed a pyrimidine tract, did not affect the formation of DNA-protein complexes in any nuclear extract (data not shown).

DNase I Footprinting
The results in the previous section indicated that several nuclear proteins bind to sequences within the -114 to -65-bp region. To further characterize this DNA, DNase I footprinting was used to identify the binding sequence(s). Comparison of DNA probes digested in the presence or absence of nuclear proteins reveals regions protected by DNA-binding proteins (38, 39). This protection identifies the nucleotide residues that are directly involved with DNA-protein interactions. The identical probe is also cleaved with base-specific (G/A) reagents to generate a ladder of molecular weight markers that are used to locate the protected region (40). DNase I digestion of a 224-bp fragment containing the -114 to -65-bp region was performed after a binding reaction using the optimized conditions determined for the gel retardation assays. Figure 5AGo (lanes 2 and 3) shows that the presence of MSC-1 nuclear proteins yielded an electrophoretic pattern identical to the pattern without nuclear proteins. On the contrary, MA10 cell nuclear proteins appeared to protect a specific region of DNA (residues -101 to -87 bp, 5'-AGGGTCAGTGTCCCT-3') (Fig. 5AGo, lane 4). The addition of a 200-fold excess of the unlabeled DNA probe restored the DNase I digestion pattern to a pattern similar to the control reaction (Fig. 5AGo, lane 5), indicating that the protection depends on specific sequences. Because the presence of MA10 nuclear proteins appeared to also affect the DNase I digestion of the DNA probe (i.e. the level of small molecular weight bands is decreased in the presence of nuclear proteins compared with the control), a DNase I titration was performed to equalize the level of digestion between the control and the MA10 DNA-binding reactions. Figure 5BGo (lanes 2–9) demonstrates that to achieve equal DNase I digestion in the presence of MA10 proteins, at least 3 times the amount of DNase I was required. Nevertheless, at DNase levels of 1.0–4.0 U, residues -101 to -87 were protected from digestion. Similarly, DNase I experiments using the complementary strand as labeled probe showed partial protection of this region (data not shown). These data demonstrate that MA10 cell nuclear proteins bind to the -101 to -87-bp region. As shown above, mutation of this region causes a 60% decrease in the transcription rate as measured by luciferase activity in MA10 cells, but not MSC-1 cells. The specific formation of complex E, with MA10 nuclear proteins, provides evidence that DNase protection has mapped a binding site for this unique complex.



View larger version (86K):
[in this window]
[in a new window]
  		Figure 5. Identification of the Binding Site within the -114 to -65-bp Region

The DNase I footprint was obtained using a nuclear extract purified from the MA10 Leydig cell line. Protected sequences were revealed by comparing the digestion patterns of reactions containing nuclear proteins to a control reaction that contained 100 µg BSA. A, Lanes 2–5: the -163 to +36 strand-specific end-labeled probe digested with 0.5 U of DNase I. The reaction in lane 5 contained competitor DNA, unlabeled probe in a 200-fold molar excess over the labeled probe. Lane 1 contained the DNA probe cleaved at G and A (G/A) residues with Maxam-Gilbert (40 ) reagents. This digest was used as a molecular weight marker to identify the residues that were protected. B, DNase I titration of the binding reactions. Lane 2 contained 100 µg BSA and 0.5 U of DNase I. Lanes 3–9 contained 100 µg MA10 nuclear extract and the end-labeled -163 to +36 bp probe. The protected sequences are indicated.

 
Analysis of the -503 to -484-bp Region
Gel retardation assays were performed using a DNA probe amplified by PCR containing residues -523 to -464. Although several binding parameters were tested to identify proteins that bind to this important regulatory sequence, no DNA-binding proteins that decreased the mobility of the free probe and exhibited sequence specificity were identified. The possible reasons for the lack of binding activity are presented in Discussion. Alternatively, scanning mutagenesis was used to determine the sequences that account for increased gene expression in the -503 to -464-bp region (41). Four mutants were created over 24 bp spanning residues -505 to -482 (Fig. 6AGo). The mutant primers contained a restriction endonuclease site to make the mutant clones identifiable by restriction fragment analysis. Each mutation was created using the DNA-619:pXP1 construct and the oligonucleotides described in Materials and Methods. The mutated constructs were used to transfect MSC-1 cells, and the cell extracts were assayed for luciferase activity, which measures promoter activity. The significance of each sequence was evaluated by comparing the results with the activity of the wild type DNA-619:pXP1. The activity of the DNA-482 :pXP1, which has the complete region deleted, served as a baseline for activity without the -503 to -484 region. The results of the luciferase assays (Fig. 6BGo) demonstrated that mutation of the sequence 5'-GGAGGC-3' (-498 to -493 bp) decreases expression to near the level of the activity obtained with DNA-482:pXP1. These data extend the previous mutagenesis experiments and provide evidence that the sequence 5'-GGAGGC-3' is important for transcription of the ABP gene.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 6. Scanning Mutagenesis of the -502 to -483-bp Region

The -503 to -464-bp sequence was analyzed by scanning mutagenesis to determine the sequences that are needed for activity. A, A diagrammatic representation of the scanning mutants created in the DNA-619:pXP1 construct. The -503 to -484-bp region is indicated in bold type. B, The effects of mutagenesis on transcriptional activity. Mutant and wild type DNA constructs were used to transfect the MSC-1 cell line. Luciferase activities were measured 48 h after transfections and are represented as the percent activity of the wild type DNA-619:pXP1 construct (±SEM; n = 4). The activities of the mutant DNAs were compared with the luciferase activities of DNA-619:pXP1 (wild type) and DNA-482:pXP1, which contains a complete deletion of residues upstream of residue -483.

 
Analysis of the -583 to -564-bp Region
Gel Retardation
The deletion mapping experiments revealed that the -583 to -564-bp sequence was important for maximum ABP gene expression in Sertoli cells. To identify transcriptional proteins that bind to this sequence, a probe including residues -619 to -544 was amplified by PCR and labeled with 32P. The initial gel retardation experiments, using MSC-1 nuclear proteins, identified a major DNA:protein complex that migrated at 10% the distance compared with the free DNA probe. As described above, various salt and Mg concentrations were tested to determine the optimal binding conditions for the complex. The optimal binding conditions for the major complex G include 100 mM KCl + 15 mM MgCl2 or 300 mM KCl with little obvious differences in binding with the addition of MgCl2 (data not shown). Gel retardation assays also were used to assay for DNA-protein complexes in MA10 and NIH3T3 nuclear proteins. Although no DNA-protein complexes were detected using NIH3T3 nuclear proteins, the MA10 nuclear proteins formed a complex (complex G1) with a mobility similar to MSC-1 complex G. However, the optimal binding conditions for formation of the MA10 complex required salt concentrations differing dramatically from MSC-1 complex G (data not shown). Optimal complex G1 formation occurred in the presence of 25 mM KCl, and formation was inhibited by MgCl2, whereas only minimal binding occurred at 300 mM KCl. Although both MSC-1 and MA10 nuclear proteins also yielded six to eight minor products that migrate between the probe and the major complex, the pattern of these bands was very different. The differences in optimal binding conditions suggest that complexes G and G1 are not the same complex of DNA-binding proteins, but the complexes may have common components. Likewise, complexes G and G1 may share components with the minor faster migrating species.

Other experiments were performed to test the formation specificity of complexes G and G1. Figure 7AGo demonstrates that the binding of both complexes was reduced by the addition of the unlabeled DNA-619 to -544. The competition with complex G increased as the amount of unlabeled probe increased (lanes 2–6). Also, the presence of an oligonucleotide containing the consensus Sp1 sequence did not affect the complex formation (Fig. 7AGo, lanes 7–10). The binding of two minor DNA-protein complexes H and I, formed with MSC-1 nuclear extracts, were also reduced by the addition of unlabeled probe. The band observed under complex I was not affected by the presence of competitor DNA and was therefore not formed by specific binding. Similarly, MA10 nuclear proteins form a complex migrating at the same apparent rate as complex G; however, the amount of the MA10 complex was much less than the amount formed with MSC-1 proteins. This complex formation with MA10 nuclear proteins was reduced with an excess of the DNA probe (Fig. 7AGo, lanes 13–20). The two bands migrating in the same region as MSC-1 complex H (Fig. 7AGo, lanes 13–20) were not affected by excess probe DNA and do not represent specific DNA-protein complexes.



View larger version (54K):
[in this window]
[in a new window]
  		Figure 7. Characterization of MSC-1 and MA10 Nuclear Proteins That Bind to the -583 to -564-bp Region

A, Gel retardation assays to determine the specificity of complex G and G1 formation. Specificity was tested by competition with unlabeled DNA probe, -619 to -544-bp DNA (lanes 1–10, 13–20). Several other sequence-specific complexes were revealed with MSC-1 nuclear proteins and are designated as complexes H and I (lanes 2–6). Lanes 11 and 12 contain 0.2 µg of recombinant Sp1 protein and exhibited no binding to the probe. B, UV cross-linking of protein complexes using MSC-1 nuclear proteins and DNA-619 to -544. Binding reactions were performed as described in Materials and Methods. The products were exposed to a 254 nm UV light source for 15, 30, and 45 min and fractionated by SDS-PAGE. Prestained protein molecular weight markers were fractionated in an adjacent lane. Two distinct products were identified and labeled as "1" and "2" (lanes 3–5). The probe migrated with the mobility of a 20-kDa protein.

 
To further characterize MSC-1 Sertoli cell complex G, UV-cross-linking experiments were performed. The preformed complex was exposed to UV light for various times to covalently link the DNA-binding protein(s) to the 32P-labeled DNA probe. Electrophoresis of the products yielded two labeled bands, migrating as 80-kDa and 90-kDa proteins (Fig. 7BGo, lanes 3–5). UV treatment of the DNA without nuclear proteins yielded no visible product (Fig. 7BGo, lane 1), indicating the high molecular weight bands were not formed by cross-linking of the DNA alone. The probe alone migrated at a position equivalent to a 20-kDa protein (Fig. 7BGo, lanes 1–5). Because the DNA probe migrates as a 20-kDa protein, the sizes of the DNA binding proteins are estimated to be 60 kDa and 70 kDa.

DNase 1 Footprinting
The identification of the DNA-binding sequence of complex G was examined using DNase I footprinting as described above. A 208-bp probe was amplified from residues -619 to -435 using 32P-labeled primers to specifically label one strand. The optimal binding conditions for complex G formation contained 300 mM KCl, which inhibited DNA digestion by DNase I. To obtain suitable digestion of the DNA probe, the binding reaction conditions were changed to 300 mM KCl and 10 mM MgCl2, which had no apparent effect on formation of complex G. Even with the addition of MgCl2 to the binding reaction, DNase I digestion of the free probe required 9 U of the enzyme. Under these conditions there was no obvious protection with 25 µg or 50 µg of MSC-1 cell nuclear proteins; however, in the presence of 100 µg of protein, there were two areas of limited protection separated by several nucleotides (Fig. 8Go). The two protected sequences included the sequence 5'-TTCTAGTATCCATTAAACACAGAAAGA-3' (residues -573 to -547) with the unaffected site from -556 to -554 bp (underlined). Several attempts were made to optimize the DNase I footprinting of complex G, including increasing the specific activity of the probe and the concentration of nuclear proteins, but no other DNA-protein interactions that protected against DNase I digestion were identified.



View larger version (44K):
[in this window]
[in a new window]
  		Figure 8. Determination of the Binding Sequence within the -583 to -564-bp Region

DNase I footprinting used a strand-specific DNA probe containing the -583 to -564-bp sequence. Lane 1 contained a Maxam-Gilbert G and A (G/A) sequencing reaction of the probe, which was used as a molecular weight marker. Protected sequences were revealed by comparing lanes 3–5 to the BSA control, lane 2. Two regions of limited protection (A and B) were identified, and the sequences are presented on the right.

 
To test the ability of the sequence identified by footprinting to form a complex, the sequence was tested for binding. Complementary oligonucleotides 34 and 35 containing the sequence protected by DNase I footprinting (residues -573 to -554) was 32P-labeled and used as a probe in the gel retardation assay with MSC-1 nuclear proteins. Figure 9AGo demonstrates that a complex forms in the presence of MSC-1 nuclear proteins with migration properties similar to complex G. Further characterization of this sequence was performed by creating a mutation in DNA-619:pXP1, a construct used for mutagenesis experiments. The mutation replaced the central portion of the binding sequence 5'-TATCCA-3' with 5'-GTCGAC-3'. Both the wild type and mutated constructs were used to transfect MSC-1, MA10, and NIH3T3 cells, and the luciferase activities were used to compare expression. The mutant construct reduced the luciferase activity expressed from the ABP promoter by 51% in MSC-1 cells and 33% in MA10 cells but did not significantly affect activity in NIH3T3 cells (Fig. 9BGo). These data support the DNase I footprinting results, which identified residues -573 to -554 as the DNA-binding site in Sertoli cell complex G formation.



View larger version (42K):
[in this window]
[in a new window]
  		Figure 9. Verification of the Binding Sequence within the -583 to -564-bp Region

A, Gel retardation assay of DNA -573 to -554 using MSC-1 nuclear proteins. The binding conditions and gel conditions were as described in Fig. 7Go. The identified complex is indicated with an arrow (lane 2). B, Transcription assays of the residues -565 to -560 mutant in DNA-619:pXP1. Mutant and wild type constructs were used to transfect MSC-1, MA10, and NIH3T3 cells, and luciferase assays were performed 48 h later. Luciferase activities are expressed as the percent activity compared with the wild type DNA-619:pXP1 construct for each cell type (±SEM; n = 4).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Previous studies using transgenic mice demonstrated that a 5.5-kb genomic DNA fragment containing 1.5 kb of upstream sequence and all eight exons was capable of directing Sertoli cell-specific transcription (33). The study presented here demonstrates that the 1.5 kb of upstream DNA directed expression in the Sertoli cell line MSC-1, whereas expression was much less in MA10 Leydig or NIH3T3 fibroblast cells. It would appear that regulatory DNA elements in the introns or exons are not needed for cell-specific expression. Moreover, as little as 619 bp of sequence upstream of the transcription start site was sufficient for full activity in each cell line. The addition of upstream sequences -619 to 1500 bp actually inhibited activity. This negative regulatory effect by repetitive DNA sequences has been previously described for several genes, including the rat insulin 1 gene (42). These findings suggest that the ABP promoter may be capable of directing Sertoli cell-specific expression of heterologous genes in transgenic mice. Sertoli cell expression of the ABP promoter constructs should be later during development than MIS promoter constructs (43), and the ABP gene should direct expression in the adult Sertoli cell.

DNA-619 increased luciferase activity more than 300-fold in MSC-1 cells. This high activity with only 619 bp of upstream sequence was not surprising since the sequence -620 to -1500 in the 1.5 kb DNA consists of species-specific repetitive elements in the rat and human gene (7, 22). The high sequence homology of rat and human gene sequence from -1 to -600 (rat gene residue numbers) has been previously presented (4). Interestingly, there was not a single Sertoli cell cis-regulatory region, but three regions appeared to contribute to the Sertoli cell specificity. The deletion mapping experiments identified regions -583 to -564, -503 to -484, and -114 to -65 as containing important regulatory elements. Regions -583 to -564 and -503 to -484 appear to stimulate activity only in the Sertoli cell line, but not the other lines, whereas region -114 to -65 acts in the Sertoli cell line MSC-1 and the Leydig cell line MA10, but not the fibroblast line NIH3T3. In addition, the ABP promoter region contains both positive and negative regualtory sequences that direct Sertoli cell-specific expression of the ABP gene. This is demonstrated by the presence of a polypyrimidine stretch in the -114 to -65-bp region, which decreased expression of the ABP gene, in MA10 cells. The sequences in none of these three regions had high homology with any known regulatory elements in the GCG Findpatterns database.

Analysis of the -114 to -65-bp region identified several nuclear proteins that specifically bind to these sequences. Other experiments found that a pyrimidine stretch at residues -74 bp to -61 caused decreased expression of transcriptional activity in MA10 and NIH3T3 cells; there was a 10-fold increase in activity in MA10 cells after mutagenesis of this region. It is somewhat of a paradox that no detectable DNA-binding proteins interacted specifically with this region (i.e. substitution mutagenesis had no effect on protein complexes in band shift assays, and no protection was observed with DNase 1 footprinting), but this sequence acted as a negative regulator of expression in cells other than Sertoli cells. Possible explanations for the apparent lack of specific nuclear proteins are described below. In addition, another cis-regulatory element between residues -101 and -87 was identified; this sequence contains an inverted repeat sequence with a stem-loop structure. Mutation of this sequence reduced gene activity dramatically in MA10 cells. Gel retardation and DNase I footprinting assays were used to identify binding proteins and their specific binding sequences. Mobility shift assays revealed several DNA-protein complexes with nuclear proteins from MSC-1, MA10, and NIH3T3 cells. DNase I footprinting with MA10 nuclear proteins determined that the binding sequence was 5'-AGGGTCAGTGTCCCT-3' (residues -101 to -87). Taken together, these data suggest that a Leydig cell transcription factor complex and its corresponding binding sequence were identified. Although the element increased gene activity in MA10 Leydig cells in vitro, the gene is not known to be active in Leydig cells. Interestingly, an analysis of the FSHR gene (which in vivo is restricted to Sertoli cell expression) found that in vitro gene activities were greater in MA10 cells than MSC-1 cells (28).

Whereas the -114 to -65-bp region stimulates gene activity in a nonspecific manner, two upstream regions increase gene activity specifically in Sertoli cells. Residues -503 to -484 generated a 5-fold increase in transcriptional activity; deletion-mapping experiments demonstrated that this increase was Sertoli cell-specific. In this study, numerous attempts were made to identify nuclear-binding proteins that bind to this sequence. Varying both DNA-binding reaction conditions and electrophoresis parameters failed to reveal evidence of protein binding. However, scanning mutagenesis identified a core sequence of 5'-GGAGGC- 3' (residues -498 to -493). Removal of the sequence reduces gene activity to near the core level obtained with DNA-482:pXP1. There are several possible explanations of why no binding proteins were observed. Although unlikely, the cis-acting element may not be acting via a trans-acting protein or the factors may be present but not detectable by the assays for several reasons. By nature, for the DNase 1 protection assay to be successful, the binding protein molarity must exceed the probe concentration. For many binding proteins this concentration is difficult to achieve without further purification. Furthermore, even though several conditions were used to test binding, the binding conditions may not have been optimal. Nevertheless, a Sertoli cell-specific cis-acting regulatory element, which included the sequence 5'-GGAGGC-3', was identified.

The final regulatory region revealed by mapping was bounded by residues -583 and -564. This region was necessary for full ABP promoter activity in MSC-1 cells, but not heterologous cell lines. In these experiments, gel retardation experiments identified a major DNA-protein complex G. The large reduction in the migration rate for this complex suggests an interaction of several proteins existing as a homo- or heterocomplex. This type of interaction has been described for many transcription factors including the homo-tetramer protein complex, Sp1 (44, 45). Evidence for a multi-polypeptide complex was revealed by UV cross-linking; two species with estimated molecular masses of 60 kDa and 70 kDa were identified. DNase I footprinting of the upstream regulatory region revealed limited protection of the sequence 5'-TTCATAGTATCCATTAAA-3' (-573 to -555 bp). Furthermore, band shift and mutagenesis experiments confirmed the identity of this sequence. Thus, this regulatory element appears to act as a Sertoli cell enhancer.

These data have defined various aspects of the Sertoli cell-specific transcriptional regulation of the rat ABP gene. Three regions upstream of the transcriptional start site that increase promoter P1 transcription were identified; two of these elements act in a Sertoli cell-specific manner. Both positive and negative regulatory sequences appear to be involved with directing expression of the ABP gene in a cell-specific manner. The identification of both cis- and trans-regulatory elements within these three sequences has identified several putative Sertoli cell-regulatory elements. Although the promoters for several Sertoli cell expressed genes, such as the FSHR, tissue plasminogen activator, MIS, transferrin, and the inhibin Bß-subunit genes (28, 29, 30, 31), have been characterized, no Sertoli cell-specific regulatory elements have been described in these genes. Further studies are needed to determine the similarities between the regulatory mechanisms of these Sertoli cell-specific genes.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Construction of Plasmids for Promoter and Enhancer Assays
DNA fragments corresponding to the rat ABP 5'-regulatory region were amplified by PCR using the thermophillic DNA polymerase, Amplitaq (Perkin-Elmer, Norwalk, CT) (46). DNA amplification was accomplished using reverse primer 13 (based on sequence immediately upstream of the initiating Met codon) and forward primers 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 14, 15, 16, and 17 with the 5.5-kb SstI genomic fragment (in pUC 19) as template. Each forward primer contained a SstI site, and the reverse primer contained a HindIII site for cloning. After amplification, the reaction products were analyzed by agarose gel electrophoresis to ensure the product was the correct size, cleaved with SstI and HindIII, and purified by Qiaex Gel Extraction (Qiagen, Chatsworth, CA). The cleaved fragments were then cloned unidirectionally into pXP1 DNA (46) cleaved with the same enzymes.

Mutagenesis of ABP promoter sequences was performed using the double-stranded mutagenesis system, Chameleon (Stratagene, La Jolla, CA) (47). Oligonucleotides used for each mutagenesis are defined below. Each plasmid DNA construct was isolated and its nature verified by restriction endonuclease fragment analysis and DNA sequence analysis. Plasmid DNA was purified by affinity chromatography using QiaFilter Maxiprep Isolation kit (Qiagen). For each construct the plasmid DNA from two unique isolates was purified for assays. At least one isolate of each construct was sequenced to confirm the identity of the DNA insert. DNA was sequenced at the University of North Carolina at Chapel Hill Automated DNA Sequencing Facility on a model 373A Applied Biosystems DNA Sequencer using the Taq DyeDeoxy Terminator Cycle Sequencing Kit (P-E Applied Biosystems, Foster City, CA).

Transfections and Luciferase Assays
The mouse Sertoli cell line, MSC-1 (43, 48), which expresses the ABP gene, was cultured in DMEM containing 10% FBS at 32 C and 5% CO2. MA10, a Leydig cell line (49), was cultured at 37 C and 5% CO2 in Waymouth’s medium containing 15% horse serum, and the mouse fibroblast cell line NIH3T3 was propagated in DMEM containing 10% FBS at 37 C and 5% CO2. Cells were plated in 100-mm dishes (in duplicate) in the appropriate medium and cultured until each culture was approximately 70% confluent. Plasmid DNA was used to transfect each cell line using LipofectAmine (Life Technologies, Bethesda MD), according to the manufacture’s protocols. Two isolates of each variant were tested for activity. Briefly, the plasmid DNA (10 µg/plate) was mixed with 800 µl of OptiMEM Reduced Serum Medium (Life Technologies) and followed by the addition of 30 µl of LipofectAmine in 800 µl of OptiMEM. Liposomal complexes were allowed to form at room temperature for 20 min, and cells were incubated at 32/37 C for 5 h. The appropriate medium containing double the serum concentration (6.4 ml) was added to each plate, and the cells were incubated for 18 h. The medium was removed from each plate and changed as described above. After the transfection process, the cells were incubated for 48 h under the appropriate conditions. For the luciferase assays, the cells were harvested, washed twice with ice-cold PBS, pH 7.5, and lysed with a solution of 1% Triton X-100, 25 mM glycyl glycine (GlyGLy), 15 mM MgSO4, 4 mM EGTA, 1 mM dithiothreitol (DTT), and 0.17 mg/ml phenylmethylsulfonyl fluoride (PMSF), pH 7.80. The cell lysates were removed from the plate and sedimented at 14,000 x g for 5 min at 4 C to remove cellular debris. The supernatant fluid was analyzed by adding 100 µl cellular lysate + 300 µl reaction buffer (25 mM GlyGly, 15 mM sodium phosphate buffer, pH 7.80, 15 mM MgSO4, 4 mM EGTA, 2 mM ATP, and 1 mM DTT, pH 7.80) and 100 µl of 200 µM D-luciferin (sodium salt, Sigma Chemical Co., St. Louis, MO). Luciferase activity (relative light units) was measured for 20 sec using a Monolight 2010 luminometer (Analytical Luminescence Inc, San Diego, CA). The luciferase data were very consistent with an SEM of less than ±15% (n = 4–10). A change in activity was considered significant if 1) it was greater than 30%, 2) the results were reproducible, and 3) the equivalent results were obtained with at least two isolates.

Constructs with DNA inserts of various lengths were used to compare the activities of gene fragments. Because the length of the DNA could conceivably affect transformation efficiency, the efficiency of constructs of various lengths were compared. Plasmid DNAs were isolated from MSC-1 cells transfected with several fragment:pXP1 constructs 48 h after transfection as previously described (50). The isolated plasmid DNA was digested with HindIII, fractionated by electrophoresis on a 0.7% agarose gel, and transferred to a nylon membrane as described above. The membrane was hybridized with 32P-labeled pXP1 DNA in 50% formamide, 6x NaCl-sodium citrate at 42 C for 18 h. The membrane was washed at a final stringency of 0.1x NaCl-sodium citrate at 65 C and exposed to Kodak XAR x-ray film (Kodak, Rochester, NY) with intensifier screens. Densitometry (using a PhosphorImager; Molecular Dynamics, Sunnyvale, CA) of the signals identified on the Southern blot revealed that the transfection efficiency for each construct did not change due to length of the insert (data not shown).

Nuclear Extract Preparation
Cells were plated in T150 flasks and grown (see above) until the cells were approximately 90% confluent (1 x 108 cells per flask). Cells were washed twice with ice-cold PBS, pH 7.5, and scraped from the flasks in PBS, after which they were pooled and collected by centrifugation. Nuclear proteins were prepared from each cell line according to methods detailed by Kupfer et al. (51). Briefly, the cells were gently resuspended in a lysis buffer containing the protease inhibitors PMSF, pepstatin, leupeptin, and aprotinin (Sigma). Nuclei were isolated by the addition of the detergent Nonidet P-40 (NP-40) and separated by centrifugation at 1,400 x g for 5 min at 4 C. The nuclei were suspended in an extract buffer (20 mM Tris-HCl, pH 7.5, 1 mM EDTA, 0.5 M NaCl, 10% glycerol, 1 mM DTT, 0.5 mM PMSF, and 2 µg/ml of each of the other protease inhibitors), shaken vigorously for 15 min at 4 C, and centrifuged at 14,000 x g for 5 min at 4 C. The supernatant fluid was removed and dialyzed against DNA binding buffer (10 mM Tris-HCl, pH 7.5, 0.1 mM EDTA, 25 mM KCl, 10% glycerol, 1 mM DTT, 0.5 mM PMSF, and 2 µg/ml of each other protease inhibitors) for 1 h at 4 C. Nuclear protein solutions were aliquoted and stored at -80 C. Protein concentrations were determined using the Micro BCA protein assay (Pierce, Rockford, IL).

Gel Retardation Assays
Gel retardation assays were performed on PCR-amplified fragments that contained putative regulatory sequences. DNA fragments were amplified using the following primer sets: -114 to -65-bp region, primers 21 and 22; -503 to -484-bp region, primers 23 and 24; and -583 to -564-bp region, primers 25 and 26 (46). Each amplified fragment included flanking sequence on both ends of the identified regulatory sequence to ensure the entire DNA-binding site was present. Amplified fragments were labeled using T4 polynucleotide kinase in a reaction containing [32P]ATP (NEN/DuPont, Boston, MA) and used as probes to detect DNA-binding proteins from MSC-1, MA10, and NIH3T3 nuclear proteins (51, 52). Various binding and gel electrophoresis conditions were tested to optimize complex formation (35, 36, 37). The binding assay for the -583-bp to -564-bp region was performed in a 30-µl reaction containing 10 mM Tris-HCl, pH 7.5, 1 mM EDTA, 10% glycerol, 300 mM KCl, 0.5 mM DTT, 67 µg/ml poly(deoxyinosinic-deoxycytidylic)acid (Pharmacia, Uppsala, Sweden), 10,000 cpm of 32P-labeled DNA, and 20–50 µg of nuclear proteins. DNA-binding products were loaded on a 1.5-mm 5% nondenaturing polyacrylamide gel containing 10% glycerol and 0.5x Tris-borate-EDTA (TBE) buffer and developed in 0.5x TBE at 175 V for 3–4 h at 4 C (51). Gel retardation assays were performed with the -114 bp to -65-bp DNA using a 30-µl binding reaction containing 20 mM HEPES buffer, pH 7.9, 0.1 mM EDTA, 0.025% NP-40, 10% glycerol, 3.3 µg/ml BSA fraction V, 50 mM KCl, 1 mM DTT, 67 µg/ml Poly (dI/dC), 10,000 cpm labeled DNA, and 10–50 µg of nuclear proteins. Binding reactions were incubated at ambient temperature for 40 min, and the products were fractionated by gel electrophoresis. Electrophoresis was performed, at room temperature, as previously described except that EDTA and NP-40 were added to the gel and running buffer to a final concentration of 1 mM and 0.05%, respectively (32). Competition experiments using unlabeled specific and nonspecific probes were performed using amplified fragments or a series of synthetic oligonucleotides containing consensus DNA-binding sequences (Promega, Madison, WI). The gels were fixed in a solution of 10% glacial acetic acid and 15% ethanol for 15 min, transferred to filter paper, and vacuum dried at 80 C. The dried gels were exposed to XAR film (Kodak) in the presence of intensifying screens for 18–72 h.

DNase I Footprinting and UV Cross-Linking
A 224-bp fragment (-164 to +36 bp) or a 208-bp fragment (-619 to -435 bp) were amplified by PCR using primers 27 and 13, or 25 and 28, respectively. A plasmid containing the entire ABP promoter was used as template. The PCR reactions contained one 32P-labeled primer to produce a strand-specific, end-labeled probe for DNase I footprinting. DNA-binding reactions were performed before DNase I digestion (32, 51). DNase I footprinting experiments used the SureTrack DNase I footprinting system (Pharmacia, Piscataway, NJ). DNA-binding reactions were subjected to DNase digestion with 0.5 U/reaction or 9 U/reaction of DNase I, as determined by titrations for 1 min at room temperature. Digestions were terminated by the addition of a 4x solution, containing 768 mM sodium acetate, 128 mM EDTA, 0.56% SDS, and 256 µg/ml yeast RNA, followed by phenol-chloroform extraction and ethanol precipitation (46). In addition, Maxam-Gilbert G and A (G/A) sequence reactions were fractionated as size markers to identify the nucleotide sequence of the protected region (40). The reaction products were resuspended in loading buffer (deionized formamide containing 10 mM EDTA, 0.3% bromophenol blue, and 0.3% xylene cyanol), heated at 90 C for 5 min, and loaded on a 0.2-mm 6% denaturing polyacrylamide gel. The gel was developed at 45 watts in 1x TBE for 1–2 h. After electrophoresis, the gel was fixed with 10% glacial acetic acid-15% ethanol, dehydrated using a vacuum drier, and exposed to XAR film with intensifying screens at -80 C for 24–96 h

UV cross-linking (53, 54) was performed as previously described using the PCR-generated fragment containing residues -619 to -544 of the ABP promoter. The binding assays were performed as described above before UV exposure. The reaction liquid was placed on a small piece of Parafilm at 0 C and exposed 6 cm from a 15-watt UV light (254 nm) for 15–45 min. After cross-linking, 2x Laemmli buffer (55) was added to each reaction and heated at 90 C for 2 min. The cross-linked DNA-protein complexes were separated from free DNA by SDS-PAGE. Gels were processed as previously described and exposed to XAR film with intensifying screens at -80 C.

PCR Primers, Mutagenesis, and Probe Oligonucleotides
Oligodeoxynucleotides (primers) were synthesized and purified by HPLC by the Oligonucleotide Synthesis Facility, Department of Pathology, University of North Carolina at Chapel Hill. Restriction endonuclease sites were included in the oligonucleotides to aid in cloning or mutant identification. Endonuclease sites are underlined.

Primer 14, residues -583 bp to -564 bp of ABP/SHBG gene, PCR forward primer (SstI site), deletion mapping: 5' GACTATGAGCTCGGCAGATTTCTTCATAGTAT 3'

Primer 15, residues -563 bp to -544 bp of ABP/SHBG gene, PCR forward primer (SstI site), deletion mapping: 5' GACTATGAGCTCCCATTAAACACAGAAAGACA 3'

Primer 16, residues -523 bp to -504 bp of ABP/SHBG gene, PCR forward primer (SstI site), deletion mapping: 5' GACTATGAGCTCCCACATAGGTCTGGGAAATC 3'

Primer 17, residues -503 bp to -484 bp of ABP/SHBG gene, PCR forward primer (SstI site), deletion mapping: 5' GACTATGAGCTCTAAGGGAGGCATTCATGTCG 3'

Primer 18, -5451 bp to 5482 bp of pXP1 plasmid DNA: NdeI site changed to PvuII site: 5'-GGTATTTCACAC-CGCAGCTGGTGCACTCTCAG-3'

Primer 19, -114 bp to -73 of ABP/SHBG gene: -65-bp to -114-bp region mutant: 5'-GGGCCGCATGGTCCTCGAG-CTGACTATGATCTCTTGCCCCC-3'

Primer 20, -91 bp to -47 bp of ABP/SHBG gene: -65-bp to -114-bp region mutant: 5'-CCTATCTCTTGCCCC-GTAATCATGAGCTCAGCAACCTTTAACCC-3'

Primer 21, residues -133 bp to -114 bp of ABP/SHBG gene, PCR forward primer (BamHI site), probe for gel retardation assay: 5'-GACTATGGATCCCATCTCATCTGCCTTC-AGAG-3'

Primer 22, residues -65 bp to -46 bp of the rat ABP gene, PCR reverse primer (KpnI site), probe for gel retardation assay: 5'-GACTATGGTACCAGGGTTAAAGGTTGCTCCGG-3'

Primer 23, residues -523 bp to -504 bp, PCR forward primer (XhoI site), probe for gel retardation assay: 5'-GACTATCTCGAG-CCACATAGGTCTGGGAAATC-3'

Primer 24, residues -483 bp to -464 bp, PCR reverse primer (KpnI site), probe for gel retardation assay: 5'-GACTATGGTACCCAGGCAGAATGCCCGGGATC-3'

Primer 25, residues -619 bp to -600 bp, PCR forward primer (BamHI site), probe for gel retardation assay: 5'-GACTATGGATCCGATTTTGCTGTCTCAACCTT-3'

Primer 26, residues -563 bp to -544 bp, PCR reverse primer (KpnI site), probe for gel retardation assay: 5'-GACTATGTTACCTGTCTTTCTGTGTTTAATGG-3'

Primer 27, residues -164 bp to -145 bp, PCR forward primer (SstI site), probe for DNase I footprinting: 5'-GACTATGAGCTCAAGG-GGATAGTAGTGGAAGGA-3'

Primer 28, residues -454 bp to -435 bp gene, PCR reverse primer, probe for DNase I footprinting: 5'-GCTGCTGGGAATGAGGATCG-3'

Primer 29, residues -516 bp to -484 bp, scanning mutagenesis primer (EcoRI site): 5-'GGTCTGGGAAAGAATTCG GAGGCATTCATGTCG-3'

Primer 30, residues -511 bp to -483 bp, scanning mutagenesis primer (XbaI site): 5'-GGGAAATCTAGGTCTAGA-TTCATGTCGG 3'

Primer 31, residues -505 bp to -477 bp, scanning mutagenesis primer (EcoRI site): 5'-CTAAGGGAGGCGAATTC GTCGGATCCCG-3'

Primer 32, residues -499 bp to -467 bp, scanning mutagenesis primer (EcoRI site): 5'-GGAGGCATTCATGAATTCTCCGGGCATTCTGC-3' Primer 33, residues -579 bp to -547 bp, mutagenesis primer (SalI site): 5'-GATTTC-TTCATAGGTCGACTTAAACACAGAAAG-3'

Primer 34, residues -573 bp to -554 bp, probe for gel retardation assay: 5'-TTCATAGTATCCATTAAACA-3'

Primer 35, residues -573 bp to -554 bp, probe for gel retardation assay: 5'-TGTTTAATGGATACTATGAA-3'.


   ACKNOWLEDGMENTS
 
We thank Ms. Michelle Cobb of the Laboratories for Reproductive Biology Tissue Culture Core Facilities for excellent technical assistance. These experiments would not have been possible without help from the Tissue Culture Core directed by Dr. Deborah O’Brien. We also thank Dr. James Tsuruta for helpful advice and Ronald Knight for his administrative skills.


   FOOTNOTES
 
Address requests for reprints to: Dr. David Joseph, Applied Genetics Laboratories, Biotechnology Development Institute, University of Florida, 12085 Research Drive, Alachua, Florida 32615.

This work was supported by PHS grants R01-HD21744 (PI David Joseph) and 5-P30-HD-18968 (Principal Investigator, Frank S. French, The Laboratories for Reproductive Biology). This work represents partial fulfillment of the requirements for a Ph.D. degree.

Received for publication April 25, 1997. Accepted for publication May 20, 1997.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Joseph DR, Hall SH, French FS 1987 Rat androgen-binding protein: evidence for identical subunits and amino acid sequence homology with human sex hormone-binding globulin. Proc Natl Acad Sci USA 84:339–343[Abstract/Free Full Text]
  2. Feldman M, Lea OA, Petrusz P, Tres LL, Kierszenbaum AL, French FS 1981 Androgen-binding protein. Purification from rat epididymis, characterization and immunocytochemical localization. J Biol Chem 256:5170–5175[Abstract/Free Full Text]
  3. Musto NA, Gunsalus GL, Bardin CW 1980 Purification and characterization of androgen binding protein from the rat epididymis. Biochemistry 19:2853–2860[CrossRef][Medline]
  4. Joseph DR 1994 Structure, function, and regulation of androgen-binding protein/sex hormone-binding globulin. Vitam Horm 49:197–280[Medline]
  5. Westphal U 1986 Steroid-protein interactions II. Monographs on Endocrinology. Springer-Verlag, Berlin, pp 198–301
  6. French FS, Ritzen EM 1973 A high-affinity androgen-binding protein (ABP) in rat testis: evidence for secretion into efferent duct fluid and absorption by epididymis. Endocrinology 93:88–95[Abstract/Free Full Text]
  7. Hammond GL, Underhill AD, Rykse HM, Smith CL 1989 The human sex hormone-binding globulin contains exons for androgen-binding protein and two other testicular messenger RNAs. Mol Endocrinol 3:1869–1876[Abstract/Free Full Text]
  8. Rosner W 1990 The functions of corticosteroid-binding globulin and sex-hormone binding globulin: recent advances. Endocr Rev 11:80–91[Abstract/Free Full Text]
  9. Fortunati N, Fissore F, Fazzari A, Berta L, Benedusi-Pagliano E, Frairia R 1993 Biololgical relevance of the interaction between sex steroid binding protein and its specific receptor of MCF-7 cells: effect on the estradiol-induced cell proliferation. J Steroid Biochem Mol Biol 45:435–444[CrossRef][Medline]
  10. Gunsalus GL, deBesi L, Musto NA, Bardin CW 1981 Androgen binding protein as a marker for Sertoli cell function. J Steroid Biochem 15:99–106[CrossRef][Medline]
  11. Hansson V, Reusch E, Trygstad O, Ritzén EM, French FS 1973 FSH stimulation of testicular androgen binding protein. Nature New Biol 246:56–58[Medline]
  12. Elkington JSH, Sanborn BM, Martin MW, Chowdhury AK, Steinberger E 1977 Effect of testosterone propionate on ABP levels in rats hypophysectomized at different ages using individual sampling. Mol Cell Endocrinol 6:203–209[CrossRef][Medline]
  13. Means AR, Fakunding JL, Huckins C, Tindall DJ, Vitale R 1976 Follicle-stimulating hormone, the Sertoli cell and spermatogenesis. Recent Prog Horm Res 32:477–522
  14. Danzo BJ, Pavlou SN, Anthony HL 1990 Hormonal regulation of androgen-binding protein in the rat. Endocrinology 127:2829–2838[Abstract/Free Full Text]
  15. Hall SH, Joseph DR, Conti M, French FS 1986 Regulation of androgen binding protein messenger RNA. In: Stefanini M, Conti M, Geremia R, Ziparo E (eds) Molecular and Cellular Endocrinology of the Testis. Excerpta Medica, Amsterdam, pp 139–149
  16. Hansson V, Weddington SC, French FS, McLean WS, Smith A, Nayfeh SN, Ritzén EM, Hagenäs L 1976 Secretion and role of androgen-binding proteins in the testis and epididymis. J Reprod Fertil [Suppl]24:17–33
  17. Louis BG, Fritz IB 1979 Follicle-stimulating hormone and testosterone independently increase the production of androgen-binding protein by Sertoli cells in culture. Endocrinology 104:454–461[Abstract/Free Full Text]
  18. Skinner MK, Fritz IB 1985 Testicular peritubular cells secrete a protein under androgen control that modulates Sertoli cell functions. Proc Natl Acad Sci USA 82:114–118[Abstract/Free Full Text]
  19. Deleted in proof
  20. Hall SH, Conti M, French FS, Joseph DR 1990 Follicle-stimulating hormone regulation of androgen-binding protein messenger RNA in Sertoli cell cultures. Mol Endocrinol 4:349–355[Abstract/Free Full Text]
  21. Hall SH, Joseph DR, French FS, Conti M 1988 Follicle-stimulating hormone induces transient expression of the protooncogene c-fos in primary Sertoli cell cultures. Mol Endocrinol 2:55–61[Abstract/Free Full Text]
  22. Joseph DR, Hall SH, Conti M, French FS 1988 The gene structure of rat androgen-binding protein: identification of potential regulatory deoxyribonucleic acid elements of a follicle-stimulating hormone-regulated protein. Mol Endocrinol 2:3–13[Abstract/Free Full Text]
  23. Sullivan PM, Wang Y, Joseph DR 1993 Identification of an alternate promoter in the rat androgen-binding protein/sex hormone-binding globulin gene that regulates sythesis of a messenger RNA encoding a protein with altered function. Mol Endocrinol 7:702–715[Abstract/Free Full Text]
  24. Wang Y, Bayliss DA, Millhorn DE, Petrusz P, Joseph DR 1990 The androgen-binding protein gene is expressed in male and female rat brain. Endocrinology 127:3124–3130[Abstract/Free Full Text]
  25. Anderson SJ, Chou HS, Loh DY 1988 A conserved sequence in the T-cell receptor ß-chain promoter region. Proc Natl Acad Sci USA 85:3551–3554[Abstract/Free Full Text]
  26. Landau NR, St.John TP, Weissman IL, Wolf SC, Siverstone AE, Baltimore D 1984 Cloning of terminal transferase cDNA by antibody screening. Proc Natl Acad Sci USA 18:5836–5840
  27. Smale ST, Baltimore D 1989 The "initiator" as a transcriptional control element. Cell 57:103–113[CrossRef][Medline]
  28. Linder CC, Heckert LL, Goetz TL, Griswold MD 1994 Follicle-stimulating hormone receptor gene promoter activity. Endocrine 2:957–966
  29. Rickles RJ, Darrow AL, Strickland S 1989 Differentiation-responsive elements in the 5' region of the mouse tissue plasminogen activator gene confer two-stage regulation by retinoic acid and cyclic AMP in teratocarcinoma cells. Mol Cell Biol 9:1691–1704[Abstract/Free Full Text]
  30. Haqq CM, King C, Ukiyama E, Falsafi S, Haqq TN, Donahoe PK, Weiss MA 1994 Molecular basis of mammalian sexual determination: activation of Müllerian inhibiting substance gene expression by SRY. Science 266:1494–1500[Abstract/Free Full Text]
  31. Feng Z, Li Y, Chen CC 1989 Analysis of the 5'-flanking regions of rat inhibin {alpha}- and ß-B-subunit genes suggest two different regulatory mechanisms. Mol Endocrinol 3:1914–1925[Abstract/Free Full Text]
  32. Guillou F, Zakin MM 1991 Sertoli cell-specific expression of the human transferrin gene. J Biol Chem 266:9876–9884[Abstract/Free Full Text]
  33. Reventos J, Sullivan PM, Joseph DR, Gordon JW 1993 Tissue-specific expression of the rat androgen-bindinging protein/sex hormone-bonding globulin gene in transgenic mice. Mol Cell Endocrinol 96:69–73[CrossRef][Medline]
  34. Nordeen SK 1988 Luciferase reporter gene vectors for analysis of promoters and enhancers. Biotechniques 6:454–457[Medline]
  35. Carey J 1991 Gel retardation. Methods Enzymol 208:103–117[Medline]
  36. Revzin A 1989 Gel electrophoresis assays for DNA-protein interactions. Biotechniques 7:346–355[Medline]
  37. Hendrickson W 1985 Protein-DNA interactions studied by the gel electrophoresis-DNA binding assay. Biotechniques 3:198–207
  38. Buratowski S, Hahn S, Guarente L, Sharp PA 1989 Five intermediate complexes in transcription intiation by RNA polymerase II. Cell 56:549–561[CrossRef][Medline]
  39. Galas DJ, Schmitz A 1978 DNase footprinting: a simple method for the detection of protein-DNA binding specficity. Nucleic Acids Res 5:3157–3170[Abstract/Free Full Text]
  40. Maxam A, Gilbert W 1977 A new method for sequencing DNA. Proc Natl Acad Sci USA 74:560–565[Abstract/Free Full Text]
  41. Charnay P, Mellon P, Maniatis T 1985 Linker scanning mutagenesis of the 5'-flanking region of the mouse ß-major-globin gene: sequence requirements for transcription in erythroid and nonerythroid cells. Mol Cell Biol 5:1498–1511[Abstract/Free Full Text]
  42. Laimins L, Holmgren-König M, Khoury G 1986 Transcriptional "silencer" element in rat repetitive seqeunces associated with the rat insulin 1 gene locus. Proc Natl Acad Sci USA 83:3151–3155[Abstract/Free Full Text]
  43. Peschon JJ, Behringer RR, Cate RL, Harwood KA, Idzerda RL, Brinster RL, Palmiter RD 1992 Directed expression of an oncogene to Sertoli cells in transgenic mice using mullerian inhibiting substance regulatory sequences. Mol Endocrinol 6:1403–1411[Abstract/Free Full Text]
  44. Mitchell PJ, Tjian R 1989 Transcriptional regulation in mammalian cells by sequence-specific DNA binding proteins. Science 245:371–378[Abstract/Free Full Text]
  45. Pascal E, Tjian R 1991 Different activiation domains of Sp1 govern formation of multimers and mediate transcriptional synergism. Genes Dev 5:1646–1656[Abstract/Free Full Text]
  46. Sambrook J, Fritsch EF, Maniatis T 1989 Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  47. Deng WP, Nickoloff JA 1992 Site-directed mutagenesis of virtually any plasmid by eliminating a unique restriction site. Anal Biochem 200:81–88[CrossRef][Medline]
  48. McGuinness MP, Linder CC, Morales CR, Heckert LL, Pikus J, Griswold MD 1994 Relationship of a mouse Sertoli cell line (MSC-1) to normal Sertoli cells. Biol Reprod 51:116–124[Abstract]
  49. Ascoli M 1981 Characterization of several clonal lines of cultured Leydig tumor cells: gonadotropin receptors and steroidogenic responses. Endocrinology 108:88–95[Abstract/Free Full Text]
  50. Sun T, Fenstermacher DA, Vos JH 1994 Human artificial episomal chromosomes for cloning large DNA fragments in human cells. Nat Genet 8:33–41[CrossRef][Medline]
  51. Kupfer SR, Marschke KB, Wilson EM, French FS 1993 Receptor accessory factor enhances specific DNA binding of androgen and glucocorticoid receptors. J Biol Chem 268:17519–17527[Abstract/Free Full Text]
  52. Peterson GM, Tanese N, Pugh FB, Tjian R 1990 Functional domains and upstream activation properties of cloned TATA binding protein. Science 248:1625–1630[Abstract/Free Full Text]
  53. Dooley S, Welter C, Blin N 1992 DNA-protein interaction analysis using polyacrylamide gel electrophoresis and a simple and sensitive UV crosslinking procedure. Electrophoresis 13:333–334[CrossRef][Medline]
  54. Chodosh LA, Carthew RW, Sharp PA 1986 A single polypeptide possesses the binding and transcription activities of the adenovirus major late transcription factor. Mol Cell Biol 6:4723–4733[Abstract/Free Full Text]
  55. Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 277:680–685



This article has been cited by other articles:


Home page
 EndocrinologyHome page
S. Miguel-Queralt, M. Knowlton, G. V. Avvakumov, R. Al-Nouno, G. M. Kelly, and G. L. Hammond
Molecular and Functional Characterization of Sex Hormone Binding Globulin in Zebrafish
Endocrinology, November 1, 2004; 145(11): 5221 - 5230.
[Abstract] [Full Text] [PDF]

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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fenstermacher, D. A.
Right arrow Articles by Joseph, D. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fenstermacher, D. A.
Right arrow Articles by Joseph, D. R.

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