| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Division of Cell and Developmental Genetics, Department of Medicine, VA Medical Center, University of California, San Francisco, California 94121
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
developmental biology, early development, male sexual function, signal transducers, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The KRAB domain is a conserved domain found in approximately one-third of the Krüppel-type (C2H2-type) zinc finger proteins [12]. The zinc finger domain is a DNA-binding domain, even though its target sequences or sequence specificities have not been well characterized except for a few examples [13, 14]. A KRAB domain is subdivided into KRAB-A and KRAB-B boxes. The KRAB-A box is highly conserved, and the KRAB-B box is divergent or absent in subfamilies of KRAB-zinc finger proteins [15]. The KRAB-A box mediates a protein-protein interaction and acts as a potent transcriptional repressor when it is tethered to target genes via known DNA-binding domains [16–18]. Furthermore, the KRAB domain is postulated to mediate transcriptional regulation by binding KRAB-associating protein 1 (KAP1; also known as transcriptional intermediary factor 1ß) and KRAB-interacting protein 1 (KRIP1) [19–22]. The KAP1 is a universal corepressor for KRAB domain containing zinc finger proteins and binds histone-modifying enzymes, histone deacetylase and histone H3 lys-9 methyltransferase, and heterochromatin protein 1 (HP1). The chromatin modification by these histone-modifying enzymes and HP1 is thought to be one of epigenetic mechanisms of gene expression in various organisms, including yeast, fly, and mammals [23]. The complex formation of KRAB domain with KAP1 is critical for the function of KRAB-containing proteins as transcriptional repressors. Such transcriptional repression is postulated to be associated with the formation of heterochromatin structure [24, 25]. In the present study, we identified an SRY-interacting protein that harbors only a KRAB domain (KRAB-A and -B subdomains), which is named KRAB-O (KRAB Only). We show that this interaction leads to indirect association of SRY with KAP1 and HP1. The identification of KRAB-O as an interactive partner for SRY may provide insights regarding the molecular mechanisms by which SRY mediates the regulation of genes involved in the mammalian sex determination.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To construct the bait for the two-hybrid screen, a DNA fragment coding for the N-terminal 124 amino acids of the mouse SRY was cloned into the GAL4 DNA-binding domain vector, pGBKT7 plasmid vector (Clontech, Palo Alto, CA). Various deletion constructs were amplified by polymerase chain reaction (PCR) with specific primers flanking respective protein-coding domains and were cloned into the pGBKT7 vector. Additional yeast expression constructs containing the open reading frames (ORFs) of KRAB-O and Zfp208 were cloned into GAL4-activation domain vector, pACT2. The FLAG-SRY was constructed by ligating the B6 Sry ORF in the SmaI site of the p3XFLAG-CMV-7 vector (Sigma-Aldrich, St. Louis, MO). For FLAG-KRAB-O and FLAG-ZFP208, the respective ORF was amplified by PCR using SalI-linked primers and ligated at the SalI site of p3XFLAG-CMV-7 vector. For FLAG-KAP1, the ORF of KAP1 from I.M.A.G.E. clone (no. 5293191; Invitrogen, Carlsbad, CA) was excised with MluI and XbaI, blunt-ended by Klenow polymerase, and ligated to the SmaI site at the p3XFLAG-CMV-7 vector. For C-terminal V5 epitope-tagged proteins, ORFs of B6 Sry and KRAB-O without a stop codon were amplified by PCR and ligated into pcDNA3.1/V5-His vector (Invitrogen). For glutathione S-transferase (GST)-fusion protein expression, PCR-amplified ORFs were ligated into pET41a(+) vector (Novagen, Madison, WI). All the nucleotide sequences of constructs were verified by DNA sequencing.
Yeast Two-Hybrid Library Construction and Screening
A cDNA library was constructed using the Matchmaker library construction and screening kit (Clontech) following the manufacturer's instructions. Briefly, 1 µg of total RNA from E11.5 gonad-mesonephros complexes was used to synthesize first-strand cDNA using reverse transcriptase and oligo d(T) primer. The resulting cDNA was subjected to 20 cycles of PCR amplification with specific primers supplied by the vendor. The amplified cDNA was cotransformed into yeast strain, AH109 (MATa, trp1–901, leu2–3, 112, ura3–52, his3–200, gal4
, gal80, LYS2: GAL1UAS-GAL1TATA-HIS3, GAL2UAS-GAL2TATA-ADE2, URA3:: MEL1UAS-MEL1TATA-lacZ), together with predigested pGADT7-Rec vector containing a Leu-selectable marker. The transformants were plate amplified on selective media plate (SD/-Leu), collected, aliquoted, and stored as a pretransformed yeast library. The primary library size is estimated to contain 4 x 106 independent clones.
The bait construct, pGBKT7-SRY124 (with selectable marker, Trp), was transformed into a yeast strain, Y187 (MAT
, ura3–52, his3–200, ade2–101, trp1–901, leu2–3, 112, gal4, gal80, met-, URA3::GAL1UAS-GAL1TATA-lacZ MEL1) containing the mating-type capable of mating with the library-strain AH109. Such mating introduced the bait into host cells harboring the cDNA library. When plated on selective media (QD, synthetic drop-out media, -Leu/-Trp/-Ade/-His), specific cDNA coding for an interactive protein with the SRY bait would survive such selection. The selected colonies were further tested for -galactosidase activity on QD/ -galactosidase plates and for ß-galactosidase activity by colony-filter assay. Colonies positive for all reporter genes (Ade, His, -galactosidase, and ß-galactosidase) were further analyzed as candidates for SRY-interactive proteins.
Analysis of Gene Structure of KRAB-O
The genomic sequences coding for KRAB-O were analyzed using the BLAST-like alignment tool of the Genome Browser at the University of California, Santa Cruz (UCSC; http://genome.ucsc.edu), with the mouse genome assembled in February 2003 [26].
DNA Isolation and Reverse Transcription-PCR
Total RNA from mouse embryonic or adult tissues was isolated using Trizol reagent (Invitrogen). First-strand cDNA was synthesized with oligo d(T) primer using Superscript II cDNA synthesis kit (Invitrogen). One microliter of cDNA from the first-strand synthesis reaction was used for PCR with a common 5' primer A (GTCACCATGGAGGAAATGCTGTCATTCAG) in combination with specific KRAB-O 3' primer B (GGTGTGAAAGAAATTTATTTGTGTCTATACTGTATTATAAT) or Zfp208 3' primer C (TTATCCGGAATGAATACTTAGGTGGGTAGCAAGTTCTAAGT). A touchdown PCR amplification strategy from 68 to 60°C was used with 35 cycles at 60°C annealing temperature and elongation at 72°C for 45 sec for KRAB-O and 2 min and 30 sec for Zfp208.
Recombinant Protein Purification and GST Pull-Down Assay
The GST-fusion protein constructs were transformed into Escherichia coli strain, BL21(DE3), and the expression of recombinant proteins were induced by 1 mM isopropyl-ß-D-thiogalactopyronoside (IPTG). Cells were collected after 5 h of induction and lysed by sonication in lysis buffer (20 mM Tris-HCl [pH 8.0], 0.5 M NaCl, 0.5% NP-40, 10 mM dithiothreitol [DTT], 1 mg/ml of lysozyme, and protease-inhibitor cocktail). Lysates were cleared by centrifugation at 20 000 x g for 30 min, and the supernatant was incubated with glutathione sepharose beads (Amersham Biosciences, Piscataway, NJ). The beads were washed with buffer (20 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.1% NP-40, and 10 mM DTT) four times. The washed beads were stored in the same buffer containing 20% glycerol and protease inhibitors at 4°C.
For a GST pull-down assay, target proteins were synthesized and labeled with S35-Met using in vitro transcription and translation kit (Promega, Madison, WI). Approximately 1 µg (as judged on SDS-PAGE) of GST or GST-fusion proteins bound to glutathione beads prepared as described above was blocked with 1 mg/ml of BSA in wash buffer for at least 2 h and then incubated with labeled target proteins in wash buffer overnight at 4°C. The bound proteins were washed five times with wash buffer. Bound proteins were eluted by boiling the beads in SDS sample buffer and were subjected to SDS-PAGE. The gels were dried and exposed to x-ray film (Kodak Scientific Imaging Systems, New Haven, CT).
Antibodies
Anti-FLAG antibodies and M2 mouse monoclonal or rabbit polyclonal antibody were obtained from Sigma (St. Louis, MO). Mouse monoclonal anti-V5 antibody, goat polyclonal anti-V5 antibody, HP1 antibody, and Alexa-conjugated secondary antibodies were obtained from Invitrogen, Bethyl (Montgomery, TX), Chemicon (Temecula, CA), and Molecular Probes (Eugene, OR), respectively. Rabbit anti-KRAB-O antiserum was produced with purified recombinant GST-KRAB-O protein as an antigen by custom antisera service at Sigma-Genosys (The Woodlands, TX).
Transient Transfection, Immunoflourescence Staining, and Confocal Microscopy
Cells were seeded 24 h before transfection on chamber slides, and a total of 0.25–0.5 µg of DNA was transfected using Fugene 6 reagent (Roche Biochemicals, Indianapolis, IN). Transfected cells were fixed with methanol at –20°C for 3 min. Slides were blocked with 5% BSA and 10% normal goat serum in PBS. Antibodies were diluted in 1% BSA in PBS. The blocked slides were incubated with the diluted primary antibodies for 1 h at room temperature, washed for 1 h in PBS, and incubated with the diluted secondary antibodies for 45 min at room temperature. After extensive washing in PBS, the slides were mounted with Vectashield mounting medium (Vector Laboratories, Burlingame, CA).
For confocal microscopy, a Leica TCS SP confocal microscope (Leica Microsystems Inc., Bannockburn, IL) was used. The images were recorded under a 100x objective with 2x zoom and then analyzed with Adobe Photoshop 7 (Adobe Systems, Inc., San Jose, CA).
Culture and Transfection of Primary Cells from Fetal Mouse Gonads
Time-pregnant CD-1 mice were purchased from Charles River Laboratory (Wilmington, MA). The Institutional Animal Care and Use Committee of the Veterans Affairs Medical Center approved all the experimental procedures in accordance with the Guide for Care and Use of Laboratory Animals. Fetal gonads were dissected from embryos at the E11.5 stage and incubated in 0.05% trypsin-EDTA at 37°C for 5 min. Gonadal cells were dispersed by repeated pipetting and seeded in chamber slides. They were cultured in Dulbecco Modified Eagle medium with 10% fetal bovine sera for 3 h before DNA transfection using Fugene 6 reagent, as described above. The cells were fixed at 16 h after DNA transfection and processed for immunofluorescence staining and confocal microscopy.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A GAL4-based yeast two-hybrid interactive cloning strategy was used to identify proteins interacting with the mouse SRY. Because the mouse Sry gene is most active in fetal gonads at E10.5–12.5 [27, 28], a yeast expression cDNA library with approximately 4 million independent clones was constructed with RNA isolated from gonad-mesonephros complexes at E11.5. The pGBKT7-SRY124 fusion protein consisting of the N-terminal 124 amino acids of the mouse SRY including the HMG and bridging peptide was used as a bait for the initial screening of this fetal gonad library. Here, we report one of the candidates that we consistently identified from this cDNA library. The nucleotide sequence of this clone was 99.9% identical to a cDNA clone from RIKEN full-length enriched cDNA library of mouse E10 whole embryo (AK011397). The cDNA insert was 2.4 kilobases in size with a poly(A) tail, as revealed by DNA-sequencing analysis. A stop codon was identified early in the 5' end, resulting in an ORF coding for a peptide of only 80 amino acids. The cDNA library from RIKEN was constructed with RNA containing a 5'-cap structure. Thus, the transcript for this cDNA most likely has a 5'-cap structure and a 3' poly(A) tail, which are characteristics of a translated transcript.
The ORF encodes a protein with a KRAB domain. The isolated clone included a 5' untranslated region (UTR) fused in-frame with the GAL4 activation domain. To confirm that the protein encoded by the ORF of the isolated cDNA clone was responsible for binding to the mouse SRY, the ORF of this cDNA without the 5' UTR was cloned in-frame with a GAL4 activation domain vector and used in the yeast two-hybrid assays with the mouse SRY bait. The result indeed confirmed its interaction with the mouse SRY (data not shown).
A BLAST search of the mouse genome database at the Genome Center of UCSC using this cDNA sequence revealed that this transcript is encoded by the Zfp208 locus on mouse chromosome 13, a homologue to the human ZNF208 locus on chromosome 19 [29]. Analysis of the genomic sequence indicates that this gene consists of five exons and produces at least two transcripts (Fig. 1). The large transcript encodes a protein harboring a KRAB domain and a DNA-binding zinc finger domain. The KRAB domain consists of two subdomains, KRAB-A and KRAB-B. These subdomains are encoded separately by exon 2 and exon 3a, respectively, similarly to other KRAB domain protein genes. The zinc finger domain is encoded by exons 4 and 5. The encoded protein has been designated as ZFP208. A smaller transcript skips the splice donor site at exon 3a and extends its sequence to exon 3b. Its ORF terminates in four amino acids after splicing into exon 3b, resulting in a small protein of only 80 amino acids. This variant protein is herein designated as KRAB-O protein. The cDNA isolated by our yeast two-hybrid screening is derived from the small KRAB-O transcript. The human ZNF208 locus also produces two transcripts similar to the mouse Zfp208 locus, which were confirmed by reverse transcription (RT)-PCR analysis of RNA from various human tissues. Both human and mouse proteins share significant homology at the KRAB domain (Fig. 1B). Most KRAB-containing proteins are associated with a Krüppel-type zinc finger domain at their carboxyl termini. Both the human ZNF208 and mouse Zfp208 genes represent the first examples of genes coding for both a KRAB-O protein and a KRAB-zinc finger protein through alternative RNA-processing mechanism.
|
To examine whether the two transcripts of Zfp208 locus are expressed in embryonic gonads, primers were designed to amplify specifically either Zfp208 or KRAB-O transcripts, a primer pair (A and C) for Zfp208 and a primer pair (A and B) for KRAB-O (Fig. 1A). The RT-PCR with RNA isolated from mouse gonads excluding the mesonephros shows that both transcripts are expressed in mouse gonads from both sexes at E11.5 and E12 (Fig. 2A). Similar analysis with other mouse embryonic tissues and adult tissues shows that both transcripts are widely expressed (Fig. 2, B and C). Because ZFP208 was not isolated in our initial screening even though ZFP208 and KRAB-O share the same KRAB domain, we addressed the question of whether ZFP208 could also interact with the mouse SRY. Under similar conditions, ZFP208 did not interact significantly with mouse SRY in the yeast two-hybrid assay and did not colocalize with mouse SRY in transfected COS7 cells (data not shown). Hence, it was not analyzed further in the present study. Using a recombinant protein, we had generated a specific antiserum against KRAB-O and used it to confirm the expression of endogenous KRAB-O protein. This antiserum was able to detect FLAG-KRAB-O from total cell extract of transfected COS7 cells but not from that of control COS7 cells, confirming the specificity of this antiserum. To detect the endogenous KRAB-O protein, whole-embryo extract at E11.5 was used for a Western blot analysis using this antiserum. The endogenous KRAB-O protein was readily detected by the anti-KRAB-O antiserum but not by the preimmune control serum (Fig. 2D).
|
Bridge Region of SRY is Necessary for Binding of SRY to KRAB-O
Because KRAB-O initially was isolated with the N-terminal fragment of the mouse SRY as a bait, we further analyzed its interaction with the full-length mouse SRY by the yeast two-hybrid and in vitro GST pull-down assays (Fig. 3, C and D). The interaction of full-length mouse SRY with KRAB-O was further confirmed by a coimmunoprecipitation assay (Fig. 3B). The V5 epitope-tagged mouse SRY and the FLAG epitope-tagged KRAB-O were expressed in COS7 cells by cotransfection. The KRAB-O was immunoprecipitated by anti-FLAG antibody. The precipitated proteins were analyzed by Western blot analysis using anti-V5 or anti-FLAG antibodies. The mouse SRY was coprecipitated with KRAB-O only by anti-FLAG antibody, not by a control antibody. The specificity was further confirmed by a coprecipitation experiment with cell extract containing only the mouse SRY-V5; in this immunoprecipitation, the mouse SRY-V5 was not precipitated by anti-FLAG antibody.
|
To map the domains within the mouse SRY that are responsible for interacting with KRAB-O protein, several deletion constructs coding for different domains of the mouse SRY (Fig. 3A) were made and assayed similarly for the ability of their encoded products in terms of binding to KRAB-O protein. The results from both assays corroborate each other, showing that the peptide between amino acids 92 and 124 within the bridge (BRG) region (residues 83– 144, between the HMG box and Q-rich domain) is necessary for binding to KRAB-O protein (Fig. 3, C and D). Although the HMG box itself is not sufficient for the SRY binding to KRAB-O, it may be necessary for optimal KRAB-O binding to the mouse SRY, because the binding of HMG-BRG was much more avid than that of BRG alone (Fig. 3D).
The BRG region has not been previously recognized as a conserved domain among the mammalian SRY proteins. Our results, so far, suggest that it might serve interactive functions with protein partners important for SRY action. To explore the possibility that such an interactive domain also exists in the human SRY, we examined whether human SRY could, indeed, interact with the mouse KRAB-O. The results show that the full-length human SRY is able to interact with the mouse KRAB-O in both yeast two-hybrid (Fig. 4B) and GST pull-down assays (Fig. 4C). To identify further the interactive domain(s) within the human SRY essential for KRAB-O binding, we generated several deletion constructs of human SRY protein and analyzed similarly as described above. First, deletion between amino acids 156 to the C-terminus did not abolish the binding to KRAB-O (Fig. 4, A and B). However, additional deletion from amino acids 138 to the carboxyl terminus abolished its interaction with KRAB-O. When the HMG domain alone was tested, it did not interact with KRAB-O. The results are consistent with those from the mouse SRY, indicating the region between amino acids 138 and 155 of the human SRY to be critical for binding to KRAB-O. Alignment of the SRY proteins from mouse, human, and other mammalian species revealed that this region constitutes a relatively conserved domain. This observation suggests a possible existence of a conserved region, beyond the HMG box in human and other mammalian SRY proteins, that is analogous to the BRG domain of the mouse SRY that is responsible for interacting with KRAB-O protein (Fig. 4D).
|
Cellular Localization of the Mouse SRY and KRAB-O
The SRY is a nuclear protein, and KRAB domains are involved in transcriptional regulation in the nucleus. Accordingly, SRY is expected to interact with KRAB-O in the nucleus. Presently, no cell lines exhibit characteristics of mouse embryonic gonads (i.e., expression of the mouse SRY and induction of male-specific genes on expression of exogenously acquired Sry) [30]. Therefore, we initially examined their cellular localizations by transient transfection of epitope-tagged transgenes, immunofluorescence, and confocal microscopy in COS7 cells. The KRAB-O was expressed as an N-terminal FLAG peptide epitope-tagged protein (FLAG-KRAB-O). The mouse SRY (B6 strain) was tagged at the C-terminus with a V5 peptide (SRY-V5). Figure 5 shows three representative images of FLAG-KRAB-O and SRY-V5 localization. The KRAB-O shows a punctuated staining in the nucleus and diffuse staining in the cytoplasm. The number and size of the punctuated dots varies among cells (Fig. 5, A–C). The merged images of Figure 5 show the colocalization of the mouse SRY and KRAB-O in these nuclear dots. To confirm further the colocalization of the mouse SRY with KRAB-O in cells physiologically relevant to SRY action, we performed the transfection, immunofluorescence staining, and confocal microscopy on primary cells from mouse embryonic gonads at E11.5 (Fig. 5D). The results show that both proteins were colocalized in the nucleus of embryonic gonadal cells, supporting our observations in COS7 cells. To confirm further the coexpression of the endogenous mouse SRY with KRAB-O in the mouse genital ridge, we performed the immunofluorescent staining using mouse monoclonal anti-mouse SRY antibody and rabbit anti-KRAB-O antibody. The mouse SRY is expressed in a subpopulation of gonadal cells, which presumably are pre-Sertoli cells [31], and KRAB-O is expressed ubiquitously (Fig. 5E).
|
These results, hence, support the notion that SRY interacts with KRAB-O in vivo in mammalian cells and primary fetal mouse gonadal cells.
SRY Associates with KAP1 and HP1 Through KRAB-O
The KRAB domains participate in transcriptional regulation and chromatin remodeling through interaction with KAP1 that, in turn, interacts with all three isoforms (, ß, and 
) of HP1 [19, 25]. To explore further the possibility that SRY may mediate such a transcriptional regulation through its interaction with KRAB-O, we systemically examined the sequence of protein-protein interactions using transfection of epitope-tagged genes, immunofluorescence microscopy, and coimmunoprecipitation techniques. To determine if KRAB-O interacts with KAP1, FLAG-KAP1, and KRAB-O-V5 constructs were transfected either independently or in combination to COS7 cells. The KRAB-O protein was immunoprecipitated from transfected cells using an anti-V5 antibody. The precipitated protein complexes were analyzed by Western blot analysis with an anti-FLAG or anti-V5 antibody. Results from this experiment show that KAP1 is coprecipitated by the anti-V5 antibody only when KRAB-O-V5 is coexpressed in the same cells (Fig. 6A), demonstrating that KRAB-O can, indeed, interact with KAP1, as observed with other KRAB domain proteins. The interaction of KRAB-O with KAP1 also were examined by indirect immunofluorescence and confocal microscopy. In transiently transfected COS7 cells, FLAG-KAP1 shows a diffuse localization along with dotted staining in the nucleus. The KRAB-O colocalizes with KAP1 in the nuclear dots (Fig. 6B). Furthermore, we noticed that more cells were expressing KRAB-O when cotransfected with KAP1 and that the number of cells showing cytoplasmic KRAB-O was very low compared with cells transfected with KRAB-O alone. Thus, it seems that exogenous KAP1 may stabilize KRAB-O protein and facilitate the transport of KRAB-O into the nucleus. Others also have observed a similar stabilizing effect of KAP1 on a KRAB domain protein [32], thereby further supporting our observations here.
|
The interaction of KRAB-O with KAP1 suggests that the mouse SRY could associate with KAP1 and, possibly, with HP1 through its binding to KRAB-O, because the mouse SRY did not bind to KAP1 directly in vitro when examined by a GST pull-down assay (data not shown). To explore this possibility, similar gene transfection and immunofluorescence analysis was performed with COS7 cells. When the mouse SRY was cotransfected with KAP1, it colocalized with KAP1 only in a few dots in the nucleus (arrows in Fig. 7, A-a). When KRAB-O was cotransfected with the mouse SRY and KAP1, the mouse SRY colocalized with KAP1 in significantly more nuclear dots (Fig. 7, A-b). Previously, we demonstrated that the mouse SRY and KRAB-O colocalized in the same nuclear dots under similar experimental conditions (Fig. 5). To examine the association of the mouse SRY with HP1, both the mouse SRY and KRAB-O were expressed in COS7 cells, stained with the antibodies (anti-FLAG [red for KRAB-O], anti-V5 [green for SRY], and anti-HP1 [blue for endogenous HP1]), and examined by confocal microscopy and sequential scanning strategy. The results show that the mouse SRY, KRAB-O, and HP1 colocalize in discrete dots (white) in the nuclei of transfected cells (Fig. 7B, merged images). The colocalization of the mouse SRY with endogenous HP1 was confirmed further in transfected primary gonadal cells (Fig. 7C). These results suggest that the mouse SRY associates with HP1 through direct and indirect interactions with KRAB-O and KAP1, respectively, in mammalian cells and embryonic gonadal cells.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The KRAB-O is a protein of 80 amino acids with only a KRAB domain, and it is encoded by an alternatively spliced transcript from the Zfp208 locus on mouse chromosome 13. The same Zfp208 gene also encodes a larger KRAB-zinc finger protein. To our knowledge, most KRAB domain proteins also harbor zinc finger domains with, so far, two identified exceptions. One is KRAB-O (present study), and the other is VHLaK, which is an alternatively spliced transcript from the ZnF197 gene [34]. Similarly, ZnF197 also encodes a larger protein with both KRAB and zinc finger domains. It is possible that other yet-to-be-identified ZNF/Zfp loci also could produce multiple transcripts coding for KRAB-containing proteins with or without a zinc finger domain. The functional relationship between KRAB domain proteins with or without zinc finger domains is currently unknown. It has been postulated that KRAB domain protein without zinc fingers may act as a dominant negative regulator for the KRAB domain protein with zinc fingers, presumably DNA-binding domains [29]. Alternatively, these two proteins from the same locus might function independently of each other. The physical separation of the KRAB domain from the DNA-binding domain may enable the KRAB domain to participate in regulation of a diverse array of genes by interacting with a variety of other DNA-binding domains without requiring a physical link to a zinc finger domain. These possibilities could be sorted out if the direct target genes regulated by KRAB domain proteins are identified.
The KRAB-zinc finger proteins are postulated to function as transcriptional regulators for their target genes, both because the KRAB domain is capable of functioning as a transcriptional repressor and because the zinc finger domain may provide target DNA-binding activity (Fig. 8A). The transcriptional repressor function of KRAB domains is mediated by interaction at the ring and coiled coil region at the amino terminus of its obligatory corepressor, KAP1. The KAP1 functions as a molecular scaffold to recruit histone deacetylase, histone H3 Lys-9 methyltransferase, and HP1, leading to gene silencing by chromatin remodeling [24, 25, 35]. When a KRAB domain protein is targeted to a euchromatic locus through a heterologous DNA-binding domain, the targeted locus is silenced and assumes a compact chromatin structure enriched with HP1 [24]. Recently, several KRAB-containing proteins have been demonstrated to be coregulators for other transcription factors [32, 34, 36]. Presumably, KRAB domain proteins may form bridging molecules between their interactive transcription factors and KAP1. Hence, such a coregulator role may be a common function for a large number of KRAB-containing proteins. The fact that KRAB-O, but not the full-length ZFP208 (i.e., KRAB-zinc finger) from the same locus, binds to SRY raises the possibility that KRAB-O proteins can act in such a capacity independently of the zinc finger domains. Based on the data from the present study, we hypothesize that the DNA-binding domain of SRY (i.e., the HMG box) provides target gene specificity, whereas the BRG domain provides the binding site for KRAB-O, which recruits transcriptional regulatory complexes containing KAP1 and HP1 (Fig. 8B). Also in the present study, we have demonstrated the indirect association of the mouse SRY with only the HP1 isoform. However, we surmise that the mouse SRY possibly could associate with other HP1 isoforms (i.e., HP1 and HP1ß), either through interaction between these isoforms and KAP1 or through heterodimerization with HP1. Furthermore, we postulate that other molecules, such as histone deacetylase and H3 K-9 methyltransferase, of the transcriptional complexes would interact with KAP1 and HP1, thereby resulting in chromatin remodeling and regulation of SRY target genes.
|
Currently, we are uncertain if other KRAB-containing proteins are expressed in fetal gonads at E10.5–E12.5 or if they, including ZFP208, are capable of interacting with the mouse SRY in vivo at the time of sex determination. Under our experimental conditions, ZFP208 barely interacted with SRY. However, we cannot rule out the possibility that they could interact in vivo in fetal gonads and/or under special conditions. Although transcriptional repressor roles have been attributed to many KRAB-associated complexes in numerous in vitro studies, the possibility of such complexes being involved in gene-activation mechanisms cannot be ruled out completely. A recent study concerning Drosophila HP1 showed that HP1 also is associated with activation of the hsp70 gene, and the presence of hsp70 transcript was necessary for deposition of HP1 on the hsp70 locus [37]. It remains to be elucidated if the association between SRY and HP1 could lead to activation of certain genes (Fig. 8C).
The finding that both human and mouse SRY interact with KRAB-O suggests that the mechanisms by which SRY mediates its biological functions might be conserved between the human and the mouse. Previous transgenic mouse studies have demonstrated that the HMG boxes of SOX3 and SOX9 are homologous and functionally interchangeable with that of the mouse SRY [38]. This suggests that the functions of the HMG box may be common among the closely related members of the SOX family. The present study mapped the binding region for KRAB-O in both human and mouse SRY to the BRG region outside the HMG box. The amino acid sequence of this domain is conserved among SRY proteins from various mammalian species. This raises the possibility that such a conserved interactive domain outside the HMG box may be unique to SRY and that the recruitment of such a chromatin remodeling and gene-silencing machinery might be specific to the important function of SRY as a primary sex-determining factor in mammals. Indeed, an A 
T substitution at codon 143 of the SRY gene of an XY sex-reversed patient was described in a recent report [39]. This base substitution replaces a serine with a cysteine at this residue in the KRAB-O interactive domain at the BRG region (Fig. 4D). Currently, the exact nature of such a new mutation, leading to XY sex reversal, is unknown. However, this recent finding further supports the importance of this conserved domain in SRY function, either for recognition of KRAB-O or for regulation of the complex formation. Studies are currently underway to elucidate the importance of the various residues within the BRG domain that are critical for KRAB-O binding and complex formation and sex determination.
The identification of KRAB-O as an interactive protein partner for SRY has shed new and significant insights regarding the molecular mechanism of SRY function and strongly supports the hypothesis that SRY regulates the expression of genes that are critical for sex determination.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Chris Lau, Division of Cell and Developmental Genetics, Department of Medicine, VAMC-111C5, University of California-San Francisco, 4150 Clement Street, San Francisco, CA 94121. FAX: 415 750 6633; clau{at}itsa.ucsf.edu
Received: 3 August 2004.
First decision: 19 August 2004.
Accepted: 23 September 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2(XI) collagen gene. Mol Cell Biol 2002 22:4256-4267This article has been cited by other articles:
|
|
 |
|
  S. Briers, C. Crawford, W. A. Bickmore, and H. G. Sutherland KRAB zinc-finger proteins localise to novel KAP1-containing foci that are adjacent to PML nuclear bodies J. Cell Sci., April 1, 2009; 122(7): 937 - 946. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Sanchez-Moreno, R. Coral-Vazquez, J.P. Mendez, and P. Canto Full-length SRY protein is essential for DNA binding Mol. Hum. Reprod., June 1, 2008; 14(6): 325 - 330. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. DiNapoli and B. Capel SRY and the Standoff in Sex Determination Mol. Endocrinol., January 1, 2008; 22(1): 1 - 9. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. X. Jin, H. O'Geen, S. Iyengar, R. Green, and P. J. Farnham Identification of an OCT4 and SRY regulatory module using integrated computational and experimental genomics approaches Genome Res., June 1, 2007; 17(6): 807 - 817. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Wilhelm, S. Palmer, and P. Koopman Sex Determination and Gonadal Development in Mammals Physiol Rev, January 1, 2007; 87(1): 1 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Huntley, D. M. Baggott, A. T. Hamilton, M. Tran-Gyamfi, S. Yang, J. Kim, L. Gordon, E. Branscomb, and L. Stubbs A comprehensive catalog of human KRAB-associated zinc finger genes: Insights into the evolutionary history of a large family of transcriptional repressors Genome Res., May 1, 2006; 16(5): 669 - 677. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Thevenet, K. H. Albrecht, S. Malki, P. Berta, B. Boizet-Bonhoure, and F. Poulat NHERF2/SIP-1 Interacts with Mouse SRY via a Different Mechanism than Human SRY J. Biol. Chem., November 18, 2005; 280(46): 38625 - 38630. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Kanai, R. Hiramatsu, S. Matoba, and T. Kidokoro From SRY to SOX9: Mammalian Testis Differentiation J. Biochem., July 1, 2005; 138(1): 13 - 19. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
