HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 34, 26144-26149, August 25, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/34/26144    most recent M003405200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lyngsø, C. Articles by Jørgensen, P. Search for Related Content PubMed PubMed Citation Articles by Lyngsø, C. Articles by Jørgensen, P. Social Bookmarking                      What's this?

Interaction between the Transcription Factor SPBP and the Positive Cofactor RNF4

AN INTERPLAY BETWEEN PROTEIN BINDING ZINC FINGERS^*

Christina Lyngsø, Guillaine Bouteiller, Christian K. Damgaard, Dorthe Ryom, Sylvia Sanchez-Muñoz, Peder L. Nørby, Bjarne J. Bonven^§, and Poul Jørgensen

From the Department of Molecular and Structural Biology, University of Aarhus, C. F. Møllers Alle 130, DK-8000 Aarhus C., Denmark

Received for publication, April 19, 2000, and in revised form, June 5, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The activator of stromelysin 1 gene transcription, SPBP, interacts with the RING finger protein RNF4. Both proteins are ubiquitously expressed and localized in the nucleus. RNF4 facilitates accumulation of specific SPBP-DNA complexes in vitro and acts as a positive cofactor in SPBP-mediated transactivation. SPBP harbors an internal zinc finger of the PHD/LAP type. This domain can form intra-chain protein-protein contacts in SPBP resulting in negative modulation of SPBP-RNF4 interaction.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The murine transcription factor SPBP plays an important role in regulation of stromelysin 1 gene expression. The palindromic SPBP binding site, SPRE, is situated in the stromelysin 1 promoter/enhancer region, where it mediates mitogen activation of the gene (1, 2). SPBP does not belong to any of the classical transcription factor families, but the protein features several sequence motifs that may be of functional significance. Two runs of polyglutamine near the amino terminus may harbor a transactivation function and a leucine zipper in the central part may be involved in heteromerization with other transcription factors. A basic region similar to the DNA binding domain in leucine zipper transcription factors is believed to mediate binding of SPBP to DNA (1, 2). At the extreme carboxyl terminus, the protein contains an atypical PHD/LAP domain similar to PHD/LAP domain 4 in the trx trithorax proteins.

PHD/LAP domains conform to the general consensus Cys-X₂-Cys-X_8-21-Cys-X_2-4-Cys-X_4-5-His-X₂-Cys-X_12-46-Cys-X₂-Cys (3-5). The PHD/LAP domain in SPBP deviates moderately from the consensus around the latter two cysteines. PHD/LAP domains occur in a number of transcriptional regulators including trithorax (4, 5) and the TIF factors (6-10), but their functional roles are poorly understood. In this report, we provide evidence that the PHD/LAP domain in SPBP can act as an intra-chain protein-protein interaction module.

The RING finger is a protein-protein interaction module occurring in a large, functionally diverse group of proteins. The RING finger consensus sequence is Cys-X₂-Cys-X_9-39-Cys-X_1-3-His-X_2-3-Cys-X₂-Cys-X_4-48-Cys-X₂-Cys (11). The RING finger is structurally unique, in that the two first cysteines coordinate a zinc atom together with the fourth and fifth cysteines in the motif. A second zinc atom is held by the third cysteine, the histidine and the sixth and seventh cysteines. The resulting "cross-brace" motif constitutes an independent structural unit that is clearly distinct from the classical tandem array of fingers (8, 12). RING finger proteins with a role in transcriptional control include the polycomb complex-associated negative regulators Bmi-1; Mel18; and RING1 (13-15); the TIF family of cofactors (6, 7, 9, 10, 16), MDM2, which can act both positively and negatively in a transcription factor-dependent manner (17, 18); BRCA1, which can be found in association with the RNA polymerase II holoenzyme (19) or act as a bona fide transactivator (20); the repressor PML (21, 22); and viral transactivators (11, 23-25). Recently, it was reported that the small nuclear RING finger protein RNF4 (SNURF) interacts with steroid hormone receptors, notably the androgen receptor, and Sp1. Transcriptional coactivation was observed with both the androgen receptor and Sp1, and in addition, RNF4 moderately increased basal transcription from some promoters (26).

Here, we present evidence that RNF4 can interact directly with SPBP and stimulate its transactivation potential. The interaction between RNF4 and SPBP depends critically on the RING finger in RNF4 and is negatively affected by the PHD/LAP-domain of SPBP.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- o-Nitrophenyl -D-galactopyranoside, D(+)-raffinose, heart muscle kinase, and phenylmethylsulfonyl fluoride were from Sigma; nickel-NTA¹-resin was from Qiagen; glutathione-Sepharose 4B was from Amersham Pharmacia Biotech. [-³²P]dATP was from Amersham Pharmacia Biotech, and [-³²P]ATP was from ICN. Mouse embryo and mouse multiple tissue Northern blots were from CLONTECH.

Strains and Plasmids-- The plasmid 2 µ97trp was constructed by inserting the ApaI-BamHI fragment containing the Gal4 DB expression cassette from pPC97 (27) between the ApaI and BamHI sites of pRS424 (28) and used as bait-vector in two-hybrid experiments. The library used in two-hybrid library screens was a mouse embryo cDNA library in pVP16 (a gift from S. Hollenberg (29)). The vector pET-HTG (30) was used for expression of recombinant protein in Escherichia coli, and the vector YepWOB6 (31) was used for expression of recombinant protein in Saccharomyces cerevisiae. The E. coli strain used for expression of recombinant protein was BL21 DE3 carrying the plasmid pRI952 (32). Expression of recombinant protein in S. cerevisiae was performed in the strain JEL1 (33) S. cerevisiae two-hybrid reporter strains were CBY 14a and CBY14 (34) and pPJ-4A (35). Green fluorescent protein fusion constructs were made in pEGFPc1 (CLONTECH). Full-length RNF4 and SPBP cDNAs were maintained in the vector pBNSEN (36) under the names pBN-RNF4 and pBN-SPBP, respectively. Full-length RNF4 cDNA was transferred from pBN-RNF4 as a HindIII-NotI cassette and inserted between the HindIII and Bsp1201 sites in pEGFP c1. An XmaI-NotI cassette from pBN-SPBP containing full-length SPBP was inserted between the XmaI and Bsp1201 sites of pEGFP c1 to yield pEGFP-SPBP. In both cases, in-frame fusions to the GFP-moiety are obtained. Subcloning of fragments into two-hybrid vectors or vectors for expression of recombinant protein was accomplished by inserting polymerase chain reaction (PCR) generated fragments using either pBN-RNF4 or pBN-SPBP as the template. The amino acid coordinates of the inserts will be indicated in the results section as needed. Site-specific mutagenesis was performed by the overlapping PCR mutagenesis strategy. Specific information on the primers used will be available on request.

Southwestern, Two-hybrid, and cDNA Library Screening; Isolation of cDNA for DNA-binding Proteins-- To isolate -phages with inserted cDNA fragments coding for Akv-MuLV U₃-region-binding proteins, we used the methodology developed by Vinson et al. (37). Details relevant for our procedure have been described previously (38, 39). Briefly, a random primed gt11 Sfi-Not expression library was constructed, using poly(A)⁺ RNA from the murine fibroblast cell line NIH3T3 infected with Akv retrovirus. Radioactively labeled double-stranded oligonucleotides representing mainly the enhancer region of Akv-MuLV were used in the screening of the gt11 Sfi-Not expression library.  clones encoding the DNA binding segment of SPBP were identified as binding to two independent sequences, one in the CAAT box region and one distal to the enhancer repeat region. The two-hybrid screen reported in Fig. 1 was performed using the mating-based system described by Bendixen et al. (34) and covered approximately 2 × 10⁷ clones. One among several positive isolates carried an insert spanning codons 27-189 of the murine RNF4 cDNA open reading frame. Full-length cDNAs were obtained by screening of a mouse embryo -ZAP express cDNA library (Stratagene) using the original isolates as hybridization probes.

Cell Growth and Transfection-- NIH3T3 and HT1080 fibroblasts were grown in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) supplemented with 10% newborn calf serum and 1% penicillin/streptomycin. Transient transfection experiments were performed using Superfect Transfection Reagent (Qiagen) according to the supplier's manual. In general, transfections were done with 0.1 µg of reporter plasmid, 1 µg of expression plasmid, and 0.4 µg of -galactosidase-expressing internal control plasmid. CAT assays and -galactoside assays were performed as described previously (38).

Northern Blotting-- Multiple tissue Northern blots containing poly(A)⁺ RNA from various murine tissues (CLONTECH) were hybridized with radioactive DNA probes in the supplied hybridization buffer as recommended by the manufacturer. Between hybridizations, filters were stripped by boiling in 0.5% sodium dodecyl sulfate for 5 min. Gel-purified DNA fragments were labeled using random primers, [-³²P]dATP, and DNA polymerase I (Klenow fragment). A fragment of SPBP encoding amino acids 1108-1812 was used to probe for SPBP-expression, and a fragment of RNF4 encoding amino acids 17-193 was used to probe for RNF4 expression.

Protein-Protein Interaction in Vitro-- Glutathione S-transferase (GST)-tagged full-length RNF4 and derivatives thereof were expressed in E. coli BL21(DE3) containing pRI952 (32). Cells were grown to an A₆₀₀ = 1.0 and induced in the presence of 1 mM isopropyl thiogalactoside for 3 h at 29 °C. Cells were lysed by sonication in 20 mM Tris-HCl, pH 7.9, 100 mM NaCl, 0.5% Nonidet P-40, 0.5 mM EDTA, 5 mM DTT, and 0.2 mM phenylmethylsulfonyl fluoride. The sonicate was cleared by centrifugation and filtration through a 45-µm cellulose acetate filter (Sartorius) and incubated with glutathione-Sepharose 4B beads (Amersham Pharmacia Biotech) at 4 °C for 1 h. The beads were washed five times in sonication buffer followed by elution in 50 mM Tris-HCl, pH 7.9, 100 mM NaCl, 20 mM glutathione. Labeling of GST-RNF4 with [-³²P]ATP and bovine heart muscle kinase was done as detailed previously (30). His-tagged SPBP_1336-1812 was expressed in S. cerevisiae: SPBP_1336-1218 was equipped with a heart muscle kinase phosphorylation site at the amino-terminal end and six histidine residues at the carboxyl-terminal in a polymerase chain reaction using the primers GAACCGGTAAGAAGAGCTTCAGTACCATCAAAAGAAGGTGGCCG and CAGCGGCCGCTCAATGGTGATGGTGATGGTGAGGATGAGCAGCCAGGCTCC and pBN-SPBP as the template. The amplification product was digested with AgeI and NotI and between the AgeI and NotI sites in YEpWOB6. Yeast JEL1 cells containing the resulting construct were grown in YEPR medium (20 g/liter Difco peptone, 10 g/liter yeast extract, 20 g/liter raffinose) (containing 2% raffinose) to an A₆₀₀ = 6, supplemented with 2% galactose, and grown for an additional 12 h. Pelleted cells were frozen in liquid nitrogen and ruptured in a bead beater (20 × 15 s) and extracted in 50 mM Tris-HCl, pH 7.9, 500 mM NaCl, 0.5 mM EDTA, 10% glycerol, 1 mM phenylmethylsulfonyl fluoride, 1 mM -mercaptoethanol, 10 mM imidazole. Following centrifugation and filtration through a 45-µm cellulose acetate filter, the extract was incubated with nickel-NTA resin for 1 h at 4 °C. The resin was washed three times in extraction buffer containing 20 mM imidazole and eluted by raising the imidazole concentration to 250 mM. The eluate was dialyzed against 50 mM Tris-HCl, pH 7.9, 300 mM NaCl, 0.5 mM EDTA, 10% glycerol, 1 mM -mercaptoethanol before use. In pull-down assays, 300 ng of ³²P-labeled GST-RNF4 (or GST as a control) was mixed with 400 ng of His-tagged SPBP and incubated for 30 min at room temperature in 50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 2 mM MgCl₂, 0.1 mM ZnCl₂, 0.2% Tween 20, 10% glycerol, 0.2 mg/ml bovine serum albumin. Nickel-NTA-resin was added to the reactions, and incubations were continued for 1 h at 4 °C. After washing the resins in binding buffer containing 20 mM imidazole, protein was stripped off with 1% sodium dodecyl sulfate and analyzed polyacrylamide gel electrophoresis and autoradiography.

Electrophoretic Mobility Shift Assays-- Electrophoretic mobility shift assays were performed essentially as described previously (2), except that nuclear extract was replaced with recombinant proteins in the amounts indicated in Fig. 6.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Identification of a Protein Interacting with the Central Part of SPBP-- The SPBP protein is 1965 amino acids in length. Fig. 1 gives a schematic representation indicating the positions of the polyglutamine runs (Q), the leucine zipper (Lz), the basic region (BR) and the PHD/LAP domain (PHD). The fragments shown in Fig. 1 were fused to the DNA binding region of GAL4 and used as baits in two-hybrid screens against a mouse embryo cDNA library fused to the activation domain of VP16 (29). In the screen with the shaded bait fragment (SPBP_1108-1812), the RING finger protein RNF4 (26, 40) appeared among the isolates. When retransformed into the yeast reporter strain, the isolated RNF4-VP16AD fusion interacted with the shaded bait fragment, but none of the other fragments in Fig. 1.

View larger version (8K): [in this window] [in a new window]   Fig. 1.   Overall structure of SPBP and fragments used in two-hybrid screening. An outline of SPBP is given in the top part of the figure; the major functional domains are indicated as shaded boxes. Q, polyglutamine stretch; Lz, leucine zipper; BR, basic -helical region; PHD, PHD/LAP domain. SPBP fragments used as baits in two-hybrid screening are shown below; these fragments span (listed from top to bottom) amino acids 293-668, 485-1107, 1108-1812, and 1351-1965, respectively.

RNF4 Is Coexpressed with SPBP in Several Tissues, Both Proteins Are Nuclear, and RNF4 Potentiates SPBP-mediated Transcriptional Activation-- The expression patterns of RNF4 and SPBP mRNA were analyzed by probing tissue-specific Northern blots with the corresponding cDNAs (Fig. 2A). Both mRNAs are present in total embryonic poly(A)⁺ RNA from all developmental stages tested. In adult tissues, RNA species hybridizing to both cDNAs were evident in brain, lung, liver, kidney, and testes. RNF4 mRNA was more abundant than SPBP mRNA; exposure times for the SPBP blots were four times longer than for the RNF4 blots. The expression pattern found here with murine RNF4 generally agrees with those observed with the human (40) and rat (26) orthologs.

View larger version (41K): [in this window] [in a new window]   Fig. 2.   Expression patterns, subcellular localization, and transcription activation properties of SPBP and RNF4. A, tissue-specific Northern blots (CLONTECH) were hybridized with an SPBP cDNA fragment spanning codons 1108-1812 (top) or an RNF4 cDNA fragment spanning codons 17-193 (bottom). The cDNAs were 32P-labeled to the same specific activity; the blots probed with RNF4 cDNA were exposed for 16 h, and the blots probed with SPBP cDNA were exposed for 64 h. B, the micrographs show NIH3T3 fibroblasts transfected with pEGFP (left), pEGFP-SPBP (middle), or pEGFP-RNF4 (right). C, HT1080 fibroblasts were transfected with the reporter construct PALCAT (1, 2) containing a single copy of the SPRE element. SPBP, RNF4, and a truncated derivative of RNF4 containing only the RING finger (RING) were provided in trans from the expression vector pBNSEN. pBNSEN without insert was used as a control. A -galactosidase expression vector was included in the transfections; the relative CAT activities plotted were normalized using -galactosidase activity as an internal standard.

The intracellular localization of SPBP and RNF4 was examined using green fluorescent protein (GFP) fusions. Transient expression of either RNF4-GFP or SPBP-GFP in NIH3T3 fibroblasts resulted in nuclear accumulation of fluorescence in both cases, while transfection with a control construct expressing GFP alone led to fluorescence throughout the cell (Fig. 2B). Thus, both SPBP and RNF4 are able to target GFP to the nucleus (26).

The effect of RNF4 on SPBP-mediated transactivation was assessed in transient expression experiments as shown in Fig. 2C. The reporter construct PALCAT (1, 2) contains a single SPRE (SPBP binding site) upstream to a TK-CAT expression unit. Cotransfection with a construct expressing full-length SPBP stimulated CAT expression approximately 10-fold. RNF4 alone did not affect CAT-expression. However, simultaneous expression of RNF4 (full-length or a truncation expressing only the RING finger moiety) and SPBP resulted in CAT activities 30-45-fold over the basal level. Thus, RNF4 acts to enhance SPBP-mediated transactivation, and only the RING finger moiety of the protein is needed to bring about the effect.

Delimitation of the RNF4 Binding Region in SPBP-- To identify the minimal region of SPBP able to interact with RNF4, a deletion series (fragments 2-13) of the original SPBP_1108-1812 bait fragment (fragment 1) was tested against VP16-RNF4 in two-hybrid assays (Fig. 3A). The shortest interacting fragment (fragment 13) encompasses residues 1379-1743 in SPBP. This region harbors the basic segment presumed to participate in DNA binding.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Two-hybrid mapping of the interaction region in SPBP. A, a schematic representation of SPBP is shown at the top, and the original bait fragment (1108-1218; shaded in Fig. 1) is shown in enlargement. This fragment and deletions thereof were used as baits against full-length RNF4 fused to the VP16 activation domain in two-hybrid assays. -Galactosidase reporter gene activities are given in arbitrary units; duplicate determinations were made on liquid cultures stemming from three independent transformant colonies. Interacting fragments consistently activated three different reporters in the presence of VP16-RNF4: the HIS3 and -galactosidase reporters in CBY 14a(34) and the ADE2 reporter in PJ69-4A (35). The end points of the listed fragments are: 1, 1108-1812; 2, 1336-1812; 3, 1351-1812; 4, 1379-1812; 5, 1414-1812; 6, 1571-1812; 7, 1697-1812; 8, 1108-1772; 9, 1108-1743; 10, 1108-1721; 11, 1108-1691; 12, 1108-1687; 13, 1379-1743. The end point coordinates refer to amino acid residues in SPBP. B, amino acid substitutions leading to loss of interaction are highlighted.

To further characterize the interaction, we introduced point mutations in the shortest interacting fragment. This was accomplished by allowing a randomly mutagenized PCR fragment, generated by error-prone PCR, to undergo homologous recombination in yeast (41) with the bait construct containing residues 1379-1743. The recipient yeast two-hybrid reporter strain also contained VP16-RNF4, and interaction-deficient mutants could therefore be identified by their loss of reporter gene activity. The mutations were mapped by direct sequencing after isolation of the mutated plasmids. Two single substitutions (P1702S and C1736R) and one double substitution (A1629T/G1737V) that completely abolished reporter gene activity are shown in Fig. 3B. These mutations map between the basic segment and the COOH-terminal border of the shortest interacting fragment.

SPBP and RNF4 Interact in Vitro-- Affinity-purified recombinant proteins were used in pull-down experiments to assess interaction in vitro. His-tagged SPBP_1300-1812 was allowed to bind ³²P-labeled, GST-tagged full-length RNF4 prior to addition of a nickel affinity matrix. SDS-polyacrylamide gel electrophoresis analysis of the bound material is shown in Fig. 4A. Approximately 35% of the input GST-RNF4 was recovered in the bound fraction (lane 6). No retention was observed with GST alone (lane 4) or when His-tagged SPBP was omitted (lane 5). To assess the specificity, we tested the effect of mutations in either partner (Fig. 4B). The C1736R point mutation in SPBP shown above to prevent genetic interaction also blocked interaction in vitro (lane 1). Substitutions of three cysteines and a histidine in the presumptive zinc-binding center in RNF4 also led to loss of interaction (lane 2). Furthermore, SPBP did not bind to an irrelevant protein (the splicing factor PSF, lane 3). Thus, RNF4 and SPBP exhibit a robust and specific in vitro interaction that appear to be dependent on an intact RING finger in RNF4.

View larger version (34K): [in this window] [in a new window]   Fig. 4.   SPBP and RNF4 interact in vitro. Affinity-purified recombinant GST-RNF4 (32P-labeled) and His-tagged SPBP1300-1812 (His-SPBP) were preincubated prior to addition of nickel-NTA-resin (nickel beads) and centrifugation. Autoradiographs of polyacrylamide gel analyzed pellets are shown. A: lane 1, GST loaded directly (15% of the input in pull-down experiments); lane 2, GST-RNF4 loaded directly (15% of the input in pull-down experiments); lane 3, GST + nickel beads; lane 4, His-SPBP + GST + nickel beads; lane 5, GST-RNF4 + nickel beads; lane 6, His-SPBP + GST-RNF4 + nickel beads. B, mutations preventing interaction in vitro: lane 1, His-SPBP + GST-RNF4mut + nickel beads; lane 2, His-SPBP(C1736R) + GST-RNF4 + nickel beads; lane 3, GST-RNF4 + nickel beads; lane 4, His-PSF + GST-RNF4 + nickel beads; lane 5, His-SPBP + GST-RNF4 + nickel beads. The RNF moiety of GST-RNF4mut carries the following amino acid substitutions: C158A, H160L, C163A, C166A.

The Role of the PHD/LAP Domain-- The data in Fig. 5A reveal an interesting feature of the interaction between RNF4 and SPBP. An SPBP bait construct containing amino acids 1108-1965 (the entire carboxyl-terminal half of the protein, including the interaction region) failed to interact, while deletion of just the PHD/LAP domain at the extreme COOH terminus (maintaining 1108-1912) restored interaction. This indicates that the PHD/LAP domain counteracts RNF4 binding. To substantiate this, two other splice variants of SPBP were tested: one that only deviates from SPBP by having a slightly different PHD/LAP domain (SPBP1) and one that contains no PHD/LAP domain (SPBP2; see sequence comparison in Fig. 5B). Fig. 5A shows that the PHD/LAP-less variant interacted efficiently, while interaction with SPBP1 was markedly reduced. The repression exerted by the alternative PHD/LAP domain in SPBP1 was less severe than that observed in SPBP. The function of the PHD/LAP domain was further studied by site-directed mutagenesis. Three cysteines and one histidine residue in the PHD/LAP domain were altered to alanine and leucine, respectively, aiming to inactivate both of the putative zinc-coordinating centers. These point mutations alleviated the inhibitory effect of the PHD/LAP domains in both SPBP and SPBP1. RNF4 did not interact directly with the LAP domain (data not shown).

View larger version (19K): [in this window] [in a new window]   Fig. 5.   The PHD/LAP domain restrains the interaction with RNF4. A, two-hybrid assays testing RNF4-VP16AD against bait constructs containing various segments of SPBP. All bait constructs except the Gal4DB negative control contain the region of SPBP necessary to bind RNF4. -Galactosidase reporter activities were determined as in Fig. 3. The representation of the PHD/LAP domain as two fingers in tandem is tentative. LAPmut has the following site-specific substitutions: C1926A, C1931A, H1936L, and C1939A. B shows the amino acid sequences at the COOH termini of the three isoforms of SPBP tested in A.

To test directly whether the PHD/LAP domain interact with the RNF4 binding region in SPBP, the PHD/LAP was fused to VP16 AD and tested against various fragments of SPBP fused to Gal4 DB in two-hybrid assays. As evidenced in Table I, interaction in trans between the PHD/LAP and the RNF4 target region in SPBP was readily detectable. Elimination of the zinc coordination capability by substitutions of three cysteines and one histidine (LAPmut) totally abolished binding. Table I further shows that interaction with the RING finger and the PHD/LAP domain responded similarly to truncations in SPBP. The point mutations in SPBP shown above to obliterate SPBP-RNF4 interaction also prevented the trans-binding of the PHD/LAP domain. RNF4 and the PHD/LAP domain thus appear to compete for the same target in SPBP. Thereby, the internal zinc finger (the PHD/LAP) may restrain the access of the external zinc finger (the RING finger of RNF4).

View this table: [in this window] [in a new window]   Table I The PHD/LAP-domain interacts in trans with the RNF4 binding region in SPBP VP16AD fused to either RNF4, the PHD/LAP domain (SPBP1812-1965) or a PHD/LAP domain containing the site-specific substitutions C1926A, C1931A, H1936L, and C1939A was tested against Gal4DB fused to SPBP1108-1812 in two-hybrid assays. -Galactosidase reporter gene activities were assessed as described in the legend to Fig. 3.

RNF4 Causes an Increased Accumulation of SPBP-SPRE Complexes-- Binding of SPBP to its cognate DNA target, SPRE, is readily detectable in mobility shift assays (2). To assess the effect of RNF4 on SPBP-SPRE complexes, radiolabeled SPRE was incubated with recombinant SPBP_1300-1812 (Fig. 6). Subsaturating amounts of SPBP (lanes 3 and 5) resulted in low levels of complex formation. Addition of RNF4 to binding reactions containing the same amounts of SPBP as in lanes 3 and 5 caused an increase in the amounts of bandshifted complex (lanes 4 and 6). The position of the shifted band was not altered in the presence of RNF4. No bandshift was seen with RNF4 alone (lane 2), and the presence of an excess of unlabeled SPRE-oligonucleotide prevented the bandshift (compare lanes 1 and 6). In summary, RNF4 stimulates accumulation of SPBP-SPRE complexes but does not seem to enter a ternary complex sufficiently robust to be detectable as a supershift.

View larger version (42K): [in this window] [in a new window]   Fig. 6.   RNF4 stimulated accumulation of protein-DNA complexes. Approximately 0.75 pmol of duplex 32P-labeled SPRE-oligonucleotide (2) was incubated with the indicated amounts of recombinant protein. Complex formation was monitored by mobility shift analysis using native polyacrylamide gel electrophoresis. SPBP, 6 His-SPBP1300-1812; RNF4, GST-RNF4. Lane 1, 200 ng of SPBP + 300 ng of RNF4 + 20 pmol of unlabeled SPRE; lane 2, 300 ng of RNF4; lane 3, 100 ng of SPBP; lane 4, 100 ng of SPBP + 300 ng of RNF4; lane 5, 200 ng of SPBP; lane 6, 200 ng of SPBP + 300 ng of RNF4.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

This report provides evidence that SPBP interacts directly with the RING finger protein RNF4. Both proteins are confined to the nuclear compartment and appear to be generally expressed. Nuclear localization of RNF4 was recently reported for the human and rat orthologs (26, 40). Binding of RNF4 to SPBP takes place in a ~350 amino acids region containing the putative DNA binding domain but not the leucine zipper previously proposed to be involved in heterodimer formation (1). Single amino acid substitutions preventing binding map to a region immediately carboxyl-terminal to the basic putative DNA-binding motif. Sequence conservation between SPBP and related proteins can be noted in this region, for instance, the LVCCLC motif around 1735, where inhibitory mutations map, is conserved in the human ortholog AR1 (U19345), the retinoic acid induced protein GT1 (42), and a hypothetical human protein (AL133649).

We find that the endogenous zinc finger in SPBP, the atypical PHD/LAP domain, is a protein-protein interaction module capable of mediating intra-chain interactions in SPBP. This particular subtype of the PHD/LAP domain is also present in AR1 (GenBank^TM accession number U19345), trithorax (5), the human trithorax, ALL (43), and a hypothetical human protein (GenBank^TM accession number AL133649). Deletions or point mutations preventing binding of RNF4 also block interaction with the PHD/LAP domain, and the presence of the PHD/LAP domain in cis reduces the binding of RNF4. The simplest explanation for this is that the binding sites for the two zinc finger motifs fully or partially overlap, although more complex conformational mechanisms cannot be excluded. The net result is that the PHD/LAP domain functions as a negative modulator of cofactor binding. In fact, the observed restraint on cofactor binding is sufficiently strong that the system is likely to be regulated, i.e. by secondary modifications or other cellular factors.

SPBP-mediated transcriptional activation is enhanced in the presence of RNF4. It was recently reported that RNF4 functions as a positive cofactor with steroid receptors, notably the androgen and estrogen receptors (26). In addition, Sp1 can utilize RNF4 as a coactivator (26, 44), and a potential role in POZ family member-mediated repression has been reported (45). Taken together, this suggests that RNF4 is able to interact with a broad variety of transcription factors. Our observation that RNF4 is more abundantly expressed than SPBP, at least at the mRNA level, is consistent with this view. Distinct modes of interaction are utilized by RNF4 to associate with different transcription factors. Thus, binding to SPBP and Sp1 is mediated through the RING finger (this report and Ref. 44), while steroid receptors are contacted by the non-RING moiety of RNF4 (26). The mechanism of RNF4-mediated coactivation is not understood. Given the interplay between RNF4 and the PHD/LAP domain it seems plausible that binding of the cofactor can induce or stabilize a distinct conformational state in SPBP increasing the transactivation potential of the latter. Alternatively, RNF4 itself may mediate contact to the general transcription machinery. Indeed, the non-RING moiety of RNF4 interacts avidly with the TATA-binding protein, TBP, in vitro (although the protein shows no endogenous transactivation activity when tested as a Gal4DB-fusion in transient expression assays) (26). The observation that the RING finger alone is sufficient to stimulate SPBP-mediated transactivation may imply that direct interaction with TBP is not mechanistically relevant in our case. However, we cannot exclude the possibility that the truncated RNF4 can form functional multimers with endogenous wild type RNF4 and thereby restore the ability to interact with TBP (RNF4 self-interacts).²

Some transcriptional coactivators can promote complex formation between transcription factors and DNA without entering a stable ternary complex. For instance, the coactivator Jab1 stabilizes complexes between c-Jun or JunD and the Ap1 site (46), and MBF exerts a similar effect on BmFTZ-F1 (47). Likewise, RNF4 causes both Sp1 (26) and SPBP to bind their cognate DNA targets more strongly in vitro. This may contribute mechanistically to the stimulatory effect on transactivation. Like in the case of MBF and BmFTZ-F1, RNF4 promoted SPBP-SPRE complex formation occurred with recombinant proteins. The effect on Sp1 was only detectable in cell extracts (44), implying that additional factors may be required in this case. In vivo experiments reveal an additional difference, in that 20 amino acids at the extreme amino terminus of RNF4 are required for coactivation with Sp1 (44), while they are dispensable for enhancement of SPBP-mediated transactivation. Further analysis is required to clarify whether RNF4 interacts with multiple downstream targets or mediate convergence at a single molecular level.

	ACKNOWLEDGEMENTS

We thank S. Hollenberg for providing two-hybrid libraries, P. James and C. Bendixen for providing yeast strains, and J. Moscat for providing the plasmid PALCAT. Claus Bus and Lene Langfeldt are akcnowledged for expert technical assistance.

	FOOTNOTES

^* This work was supported by Grants 97141149132 (to C. L.) and 9810032 (to P. J.) from the Danish Cancer Society, Grant 9901846 from the Danish Natural Science Research Council, and by the Karen Elise Jensen Foundation.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

These authors contributed equally to this work.

^§ To whom correspondence should be addressed. Tel.: 45-89-423188; Fax: 45-86-196500; E-mail: bjb@mbio.aau.dk.

Published, JBC Papers in Press, June 9, 2000, DOI 10.1074/jbc.M003405200

² C. Lyngsø, Bouteiller, C. K. Damgaard, D. Ryom, S. Sanchez-Muñoz, P. L. Nørby, B. J. Bonven, and P. Jørgensen, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: NTA, nitrilotriacetic acid; PCR, polymerase chain reaction; CAT, chloramphenicol acetyltransferase; GST, glutathione S-transferase; GFP, green fluorescent protein.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				E. Sjottem, C. Rekdal, G. Svineng, S. S. Johnsen, H. Klenow, R. D. Uglehus, and T. Johansen The ePHD protein SPBP interacts with TopBP1 and together they co-operate to stimulate Ets1-mediated transcription Nucleic Acids Res., October 8, 2007; 35(19): 6648 - 6662. [Abstract] [Full Text] [PDF]

				A. Kosoy, T. M. Calonge, E. A. Outwin, and M. J. O'Connell Fission Yeast Rnf4 Homologs Are Required for DNA Repair J. Biol. Chem., July 13, 2007; 282(28): 20388 - 20394. [Abstract] [Full Text] [PDF]

				G. Li, N. Chen, R. L. Roper, Z. Feng, A. Hunter, M. Danila, E. J. Lefkowitz, R. M. L. Buller, and C. Upton Complete coding sequences of the rabbitpox virus genome J. Gen. Virol., November 1, 2005; 86(11): 2969 - 2977. [Abstract] [Full Text] [PDF]

				S. Landais, R. Quantin, and E. Rassart Radiation Leukemia Virus Common Integration at the Kis2 Locus: Simultaneous Overexpression of a Novel Noncoding RNA and of the Proximal Phf6 Gene J. Virol., September 1, 2005; 79(17): 11443 - 11456. [Abstract] [Full Text] [PDF]

				V. Gburcik, N. Bot, M. Maggiolini, and D. Picard SPBP Is a Phosphoserine-Specific Repressor of Estrogen Receptor {alpha} Mol. Cell. Biol., May 1, 2005; 25(9): 3421 - 3430. [Abstract] [Full Text] [PDF]

				S. J. Hirvonen-Santti, V. Sriraman, M. Anttonen, S. Savolainen, J. J. Palvimo, M. Heikinheimo, J. S. Richards, and O. A. Janne Small Nuclear RING Finger Protein Expression during Gonad Development: Regulation by Gonadotropins and Estrogen in the Postnatal Ovary Endocrinology, May 1, 2004; 145(5): 2433 - 2444. [Abstract] [Full Text] [PDF]

				F. J. Kaiser, T. Moroy, G. T. G. Chang, B. Horsthemke, and H.-J. Ludecke The RING Finger Protein RNF4, a Co-regulator of Transcription, Interacts with the TRPS1 Transcription Factor J. Biol. Chem., October 3, 2003; 278(40): 38780 - 38785. [Abstract] [Full Text] [PDF]

				V. V. Kapitonov and J. Jurka The Esterase and PHD Domains in CR1-Like Non-LTR Retrotransposons Mol. Biol. Evol., January 1, 2003; 20(1): 38 - 46. [Abstract] [Full Text] [PDF]

				E. Kalkhoven, H. Teunissen, A. Houweling, C. P. Verrijzer, and A. Zantema The PHD Type Zinc Finger Is an Integral Part of the CBP Acetyltransferase Domain Mol. Cell. Biol., April 1, 2002; 22(7): 1961 - 1970. [Abstract] [Full Text] [PDF]

				R. Pero, F. Lembo, E. A. Palmieri, C. Vitiello, M. Fedele, A. Fusco, C. B. Bruni, and L. Chiariotti PATZ Attenuates the RNF4-mediated Enhancement of Androgen Receptor-dependent Transcription J. Biol. Chem., January 25, 2002; 277(5): 3280 - 3285. [Abstract] [Full Text] [PDF]

				R. Pero, F. Lembo, D. Di Vizio, A. Boccia, P. Chieffi, M. Fedele, G. M. Pierantoni, P. Rossi, R. Iuliano, M. Santoro, et al. RNF4 Is a Growth Inhibitor Expressed in Germ Cells but Not in Human Testicular Tumors Am. J. Pathol., October 1, 2001; 159(4): 1225 - 1230. [Abstract] [Full Text]

				M. Hakli, U. Karvonen, O. A. Janne, and J. J. Palvimo The RING Finger Protein SNURF Is a Bifunctional Protein Possessing DNA Binding Activity J. Biol. Chem., June 22, 2001; 276(26): 23653 - 23660. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/34/26144    most recent M003405200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar