Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M312842200 on March 23, 2004

J. Biol. Chem., Vol. 279, Issue 25, 26433-26444, June 18, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
279/25/26433    most recent
M312842200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bondos, S. E.
Right arrow Articles by Matthews, K. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bondos, S. E.
Right arrow Articles by Matthews, K. S.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Hox Transcription Factor Ultrabithorax Ib Physically and Genetically Interacts with Disconnected Interacting Protein 1, a Double-stranded RNA-binding Protein*

Sarah E. Bondos{ddagger}, Daniel J. Catanese, Jr.{ddagger}§, Xin-Xing Tan{ddagger}, Alicia Bicknell||, Likun Li**, and Kathleen S. Matthews{ddagger}{ddagger}

From the Department of Biochemistry and Cell Biology, Rice University, Houston, Texas 77005

Received for publication, November 24, 2003 , and in revised form, March 4, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
The Hox protein family consists of homeodomain-containing transcription factors that are primary determinants of cell fate during animal development. Specific Hox function appears to rely on protein-protein interactions; however, the partners involved in these interactions and their function are largely unknown. Disconnected Interacting Protein 1 (DIP1) was isolated in a yeast two-hybrid screen of a 0–12-h Drosophila embryo library designed to identify proteins that interact with Ultrabithorax (Ubx), a Drosophila Hox protein. The Ubx ·DIP1 physical interaction was confirmed using phage display, immunoprecipitation, pull-down assays, and gel retardation analysis. Ectopic expression of DIP1 in wing and haltere imaginal discs malforms the adult structures and enhances a decreased Ubx expression phenotype, establishing a genetic interaction. Ubx can generate a ternary complex by simultaneously binding its target DNA and DIP1. A large region of Ubx, including the repression domain, is required for interaction with DIP1. These more variable sequences may be key to the differential Hox function observed in vivo. The Ubx ·DIP1 interaction prevents transcriptional activation by Ubx in a modified yeast one-hybrid assay, suggesting that DIP1 may modulate transcriptional regulation by Ubx. The DIP1 sequence contains two dsRNA-binding domains, and DIP1 binds double-stranded RNA with a 1000-fold higher affinity than either single-stranded RNA or double-stranded DNA. The strong interaction of Ubx with an RNA-binding protein suggests a wider range of proteins may influence Ubx function than previously appreciated.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
During animal development, differential transcription regulation is key to subdivision and specification of a diverse array of tissues. Position along the anterior-posterior axis in developing metazoans is specified by the Hox transcription factor family (1–7). Alterations in expression patterns of Hox transcription factors can rearrange appendages, a dramatic example of their capacity to instigate developmental programs (1). However, Hox transcription factors bind short, partially degenerate DNA binding sites via the highly conserved homeodomain, suggesting that DNA recognition solely by the homeodomain is insufficient to achieve the precise regulatory control required for Hox function in vivo (2). Further, members of the Hox family can activate or repress transcription, often of the same gene, depending on the cellular context (3, 4). Despite this variability, genetic and molecular studies demonstrate that Hox proteins specifically regulate a large number of genes throughout the genetic hierarchy (1, 5). Thus, Hox proteins require a mechanism to assimilate external information to produce differential DNA binding and transcription regulation functions.

The function of many transcription factor families relies on a complex web of protein interactions to generate the requisite specificity, diversity, and reliability of transcriptional outcomes (6–8). Indeed, the transcriptional activity of Hox proteins is influenced by homomeric and heteromeric cooperative DNA binding (9–13) and by specific interactions with components of the transcription apparatus (14). Hox proteins bind cooperatively to DNA with Extradenticle (Exd)1 in arthropods or with Pbx in vertebrates via their YPWM motifs (10, 13). Residues flanking this YPWM motif stabilize the protein complex and also make contacts with the DNA (13, 15). Thus, interactions with heterologous proteins may provide the functional specificity requisite for Hox function in vivo. Despite the potential importance of these interactions, very little is known about the range of Hox protein partners or the role of these partners in development.

Ultrabithorax (Ubx) is a Hox protein that orchestrates development of the posterior thorax of Drosophila melanogaster. Ubx specifies parasegments 5 and 6 and contributes to the differentiation of more posterior segments (6, 16–20). Within these regions, Ubx influences midgut, central nervous system, peripheral nervous system, leg, and haltere development (6, 18–20). Several functional domains have been identified in Ubx (Fig. 1). Our previous work has shown that a transcription activation domain resides between residues 159 and 242, and transcription activation levels are further enhanced by inclusion of amino acids 68–158 (21). Immediately following the activation domain is the YPWM motif and the homeodomain. The C terminus of the protein contains a glutamine/alaninerich region that contributes to transcription repression (22, 23). Although aspects of the transcription regulatory domains have been identified, the mechanisms, including protein interactions, that mediate these functions remain unknown. To identify potential protein partners that may modulate the function of Ubx, we have performed a yeast two-hybrid screen.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 1.
Schematic of the protein sequences for Ubx Ib and DIP1-c. A, a schematic of the Ubx Ib sequence, showing the position of various functional domains: homeodomain (HD), the B, mI, and mII microexons subject to alternative splicing, the YPWM motif (Y) (site of Exd interaction), a polyglycine region (Poly G), the activation domain (AD) that has a predicted {alpha}-helix (H) and activation enhancing region (AER) (21), and a glutamine- and alanine-rich region (QA motif) that participates in transcriptional repression (22, 23). Also depicted is the region of Ubx Ib that interacts with DIP1, as determined by yeast two-hybrid experiments. B, a schematic of the DIP1-c sequence, showing regions that exhibit homology to other proteins, including the PEST and nuclear localization (NLS) sequences. Two predicted dsRNA-binding domains (dsRBD), a sequence homologous to the DEAD box (D); and four clustered PXXP motifs (P). Also depicted is the region of DIP1-c that interacts with Ubx Ib (solid line), as determined by deletion mutants, as well as a region N-terminal (dashed line) that enhances the Ubx Ib ·DIP1-c interaction.

 
A partner prominently identified in our two-hybrid cDNA library screen was Disconnected Interacting Protein 1 (DIP1), which contains a putative nuclear localization signal and two double-stranded RNA-binding domains (dsRBDs). Global overexpression of DIP1 causes embryonic lethality (24). However, specific expression of DIP1 in the eye-antennal imaginal disc generates duplications of antenna and mouth structures and loss of the arista, reflecting aberrant expression of homothorax and spalt major and duplication of the distalless expression domain (24). DIP1 also interacts with other transcription regulators, including Disco and Su(var)3–9 (24, 25). Thus DIP1, like Ubx, can alter transcription regulation to impact developmental programs, although the molecular mechanism is unknown.

Herein, we detail a physical interaction between Ubx and DIP1 using modified yeast two-hybrid methods, phage display, immunoprecipitation, GST pull-down assays, and gel retardation supershifts. In addition, a genetic interaction between ubx and DIP1 is established. Formation of a DIP1 ·Ubx ·DNA ternary complex and the demonstration that DIP1 represses transcription activation by Ubx in a modified yeast one-hybrid assay suggest a physiological role for the interaction. Intriguingly, DIP1 is shown to bind dsRNA with extremely high affinity. The strong interaction of a Hox protein with an RNA-binding protein implies that influences on Hox activity may be wider than previously appreciated.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Yeast Two-hybrid Screen
The "interaction trap" method developed by the Brent Laboratory for yeast two-hybrid screens (26) has been utilized in these experiments. Matchmaker LexA two-hybrid system (Clontech) was used, and a 0–12 h D. melanogaster embryonic cDNA library (27) was a gift from Roger Brent (The Molecular Sciences Institute, Berkeley, CA). The plasmid pLexA-Ubx{Delta}N216 (21) contains coding sequences for amino acids 216–389 of the Ubx Ib isoform fused to the LexA DNA binding domain, a construct initially used to eliminate transcriptional activation by Ubx Ib alone. Using standard methods (28, 29), the yeast strain EGY48 was stably co-transformed with the plasmid pLexA-Ubx{Delta}N216 and the reporter plasmid p8op-lacZ prior to the introduction of the D. melanogaster embryonic cDNA library fused to the B42 activation domain in the pJG42AD vector. Transformants were plated on SD minimal medium lacking uracil, histidine, and tryptophan and containing galactose as the carbon source. Positive colonies were selected by blue color, and DNA was isolated and sequenced. Full-length constructs of Ubx Ib and DIP1-c isoforms were also utilized in this assay. The Ubx Ib ·DIP1-c interaction was reassayed in the yeast two-hybrid system using Ubx proline mutant 1 (21), a full-length Ubx Ib construct incapable of transcription activation. Finally, the yeast two-hybrid screen was repeated using pLexA-DIP1 and pJG42AD-Ubx. In this reverse yeast two-hybrid assay, yeast were plated on medium containing glucose to inhibit expression of the toxic Ubx-B42 chimera. Once the colonies were well formed, they were replica-plated onto galactose plates to induce protein expression.

Sequence Analysis
The overlapping cDNA and expressed sequence tags from the Berkeley FlyBase were aligned and combined using DNA Strider 1.2. Homology searches of DIP1 were completed using the Prosite Data base (30) and the Blast program (31).

Phage Display Assay
The T7Select Phage Display System (Novagen) was used according to the manual. DIP1-c, the longest isoform, was cloned into T7Select 1–1b vector between the EcoRI and XhoI restriction sites and was fused to the region encoding the C terminus of the 10B capsid protein gene of T7 phage to display, on average, ≤1 DIP1-c protein on each product phage as a fusion protein. Phage were amplified, and a specific amount of phage lysate containing either 500 or 1,000 phages was applied to the surface of a Petri dish upon which cell extracts containing biotinylated Ubx Ib had been immobilized (see below). Controls utilized cell extracts with vector lacking the Ubx Ib coding region. The dish was washed five times with 100 mM Tris-HCl, pH 7.5, 100 mM NaCl, 0.1% Tween 20. Phage that had been captured by Ubx Ib were eluted with 1% SDS. Two 20-min elutions at room temperature were collected and were diluted 20-fold with LB medium. The phage lysate was titered with Escherichia coli BLT5615 both before and after biopanning. These data were used to calculate the percentage of phage retained.

Protein Expression and Purification
Ubx Ib, cloned into pET3c between the NdeI and BamHI restriction sites, was a gift from Philip A. Beachy (Johns Hopkins University). Ubx Ib was expressed in E. coli BL21(DE3)pLysS and purified as described previously (32).

To produce biotinylated Ubx Ib for phage display assays, Ubx Ib was cloned into the pDW363 vector (33) between the EcoRI and BamHI restriction sites. Biotinylated Ubx Ib was expressed in E. coli AR120. Cell growth and protein induction were performed as described (32). Cell pellets from a 200-ml growth were resuspended in 2 ml of 10% sucrose, 50 mM Tris-HCl, pH 7.4, with 0.5 mM PMSF and 0.8 mg/ml lysozyme and frozen at -80 °C. The frozen cell pellet was thawed on ice and diluted with an equal volume of 10% sucrose, 50 mM Tris-HCl, pH 7.4, 0.8 M NaCl, 20 mM EDTA, 4 mM DTT, and 10 µl of 100 mM PMSF. The cells were incubated on ice for 45 min. Subsequently, 4 µl of 10 mg/ml DNase I was added before a further 30-min incubation at room temperature. The slurry was centrifuged for 30 min at 11,000 x g, and the supernatant was applied to avidin-coated Petri dishes to immobilize biotinylated Ubx Ib. The plates were washed with 100 mM Tris-HCl, pH 7.5, 100 mM NaCl, 0.1% Tween 20, to remove all bacterial proteins prior to the addition of phage.

DIP1-c was cloned between the NdeI and XhoI restriction sites in the pET28a vector (Novagen) to express an N-terminally His6-tagged DIP1-c in E. coli BL21(DE3)pLysS. Overnight Luria broth culture (8 ml) was used to inoculate each of 12 LB cultures (1 liter) and grown at 37 °C. At midlog phase, DIP1-c expression was induced with 1 mM isopropyl-{beta}-D-thiogalactopyranoside, and the cells were harvested 2 h later. Cells collected from 2-liter growths and stored at -20 °C were thawed on ice and lysed in lysis buffer (50 mM NaH2PO4, pH 8.0, 300 mM NaCl, 1 mM imidazole, 5% glucose, 10 mM DTT). After lysis, DNase I and RNase A (each 20 mg/ml) were added. The lysate was centrifuged at 13,000 x g for 15 min, and the supernatant was filtered before loading onto a 5-ml HiTRAP heparin column (Amersham Biosciences) at 4 °C on an Akta fast protein liquid chromatography apparatus (Amersham Biosciences). The heparin column was pre-equilibrated with Buffer A (50 mM NaH2PO4, pH 8.0, 100 mM NaCl, 5% glucose, 10 mM DTT, passed through 0.2-µm filter, and degassed) and washed with Buffer A after loading the protein. DIP1-c was eluted with an 8-column volume gradient of Buffer A to Buffer A with 1 M NaCl, and fractions were analyzed by SDS-PAGE. DIP1-c fractions were pooled and loaded onto a ~2.5-ml Ni2+-nitrilotriacetic acid-agarose (Qiagen) column pre-equilibrated with lysis buffer at 4 °C. The column was washed with lysis buffer with 20 mM imidazole, and DIP1-c was eluted in lysis buffer with 100 mM imidazole. Fractions were analyzed by SDS-PAGE. DIP1-c fractions were dialyzed twice at 4 °C against 1-liter exchanges (1 h each) of 20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 2.5 mM CaCl2, 5% glucose, 10 mM DTT. Biotinylated thrombin (Novagen) was added at a 1:100 dilution, mixed well by inverting the tube, and allowed to cleave His6-DIP1-c for 16 h at 4 °C. The protein mixture was mixed with 100 µl of avidin (Novagen) and 500–1000 µl of Ni2+-nitrilotriacetic acid-agarose resin slurries and incubated at 4 °C for 30 min. Centrifugation at 2,200 x g for about 10 s removed the resins along with biotinylated thrombin and His6 tag released from DIP1-c. The resulting DIP1-c was dialyzed at 4 °C against two 1-liter exchanges (1 h each) of 50 mM NaH2PO4, pH 8.0, 300 mM NaCl, 5% glucose, 10 mM DTT and stored at 4 °C for up to 1 week because freezing abrogates protein function. Every other day, solid DTT was introduced to about 100 µM to prevent oxidation effects on protein function.

Immunoprecipitation
Equimolar aliquots of purified Ubx Ib, purified DIP1-c, 2H.10"7" anti-Ubx antibody, and 10H.7 anti-Ubx antibody were combined and incubated with gentle rocking at 4 °C overnight. Hybridoma cells lines producing these antibodies were a gift from A. Javier Lopez (The John Hopkins School of Medicine) (34). Control incubations included either Ubx Ib or DIP1-c alone with the anti-Ubx antibodies. Each mixture was centrifuged at 16,000 x g. The supernatant was removed, and the pellet was washed with 50 µl of phosphate-buffered saline buffer with Tween 20 and recentrifuged for 5 min to dilute any remaining supernatant. The supernatant, wash, and pellet fractions were analyzed by SDS-PAGE and Western blotting.

GST Pull-down Assay
The coding region of full-length Ubx Ib was subcloned into the pGEX-6P-2 vector (Amersham Biosciences, Inc.) via the EcoRI and XhoI restriction sites to produce pGEX-Ubx. The vector was transformed into E. coli BL21(DE3) cells. Cells were grown in 1 liter of LB at 37 °C to an absorbance at 600 nm of 0.3–0.4. Expression of GST-Ubx was induced with the addition of 1 mM isopropyl-{beta}-D-thiogalactopyranoside followed by growth for an additional 2 h. Cells (200 ml) were harvested at 2,500 x g for 5 min, and the cell pellets were resuspended in 2 ml of 10% (w/v) sucrose, 50 mM Tris-HCl, pH 7.4, 0.8 mg/ml lysozyme, with 10 µl of 100 mM PMSF. Cell pellets were stored at -20 °C. Cell pellets thawed on ice were diluted with an equal volume of 10% (w/v) sucrose, 50 mM Tris-HCl, pH 7.4, 0.8 M NaCl, 20 mM EDTA, 4 mM DTT, with 10 µlof100 mM PMSF. After 45 min, 4 µl of 10 mg/ml DNase I were added, and the lysate was incubated at room temperature for 30 min. The slurry was centrifuged at 3,000 x g for 30 min, and the supernatant was used immediately.

DIP1-c was either added as 200 µg of purified protein (full-length only, with or without His6 tag) or as an in vitro transcription/translation reaction (full-length and DIP1-c deletions, TNT T7 PCR Quick Master Mix; Promega). Template DNA for the TNT reaction was amplified by PCR in which the 5' primer contained the sequence 5'-GGATCCTAATACGACTCACTATAGGAACAGCC(C/A)CCATGG-3' required to initiate transcription and translation. The resulting products were cloned into the pGEM-T vector (Promega), and 2 µg of purified DNA digested with BamHI was used in two reactions for each DIP1-c deletion. The TNT products were tested for expression levels, and then divided between GST pull-down experiments with cell lysates expressing GST-Ubx or lysates from cells without the pGEX-Ubx vector as a negative control. The GST pull-down protocol was followed as described by Amersham Biosciences using Glutathione SepharoseTM 4b beads.

Drosophila Genetics
The Drosophila line containing the UAS-DIP1-c insertion was created by Alain Pélisson (CNRS, France) and obtained from Ana Campos (McMaster University, Hamilton, Canada). The Df(1) LB6/Dp(1;Y) line was provided by Bloomington Stock Center (stock number 5999). The P{GawB}BxMS1096 Gal4 flies were acquired from Sean Carroll (University of Wisconsin, Madison, WI), and flies containing the Ubxbx-34e gypsy insertion were received from Bloomington Stock Center (stock number 457). The S880 wild-type fly line was obtained from Michael Stern (Rice University, Houston, Texas). DIP1-c was ectopically expressed in the wing and haltere using the Gal4-UAS system (35) by crossing MS1096-Gal4 virgins to UAS-DIP1-c males, producing progeny heterozygous for both elements. Progeny exhibited a fully penetrant shriveled wing phenotype, which was lethal in most males, probably due to strong expression of DIP1-c in males, since MS1096-Gal4 is on the X chromosome. To evaluate a genetic interaction, heterozygous female virgins from the previous cross were mated to males homozygous for Ubxbx-34e, although the same results are obtained when heterozygous males and Ubxbx-34e virgins are used. All progeny therefore harbored the Ubxbx-34e insertion, and progeny that also contained the Gal4 and UAS elements were identified by the shriveled wing phenotype generated by DIP1-c ectopic expression (see "Results"). Genetic interactions were also evaluated in crosses between Ubxbx-34e homozygote virgins and Dp(1;Y) males. Pictures of halteres remaining on the fly and dissected wings and halteres were acquired with a Leica MZF1111 microscope fitted with a Zeiss AxioCam MRC digital camera and using Axiovision software. Flies were submerged in ethanol to dissect wings and halteres, which were mounted in 80% glycerol. Images of dissected halteres were also acquired using a Zeiss AxioPlan 2 microscope and Metamorph software.

Northern Blotting and Reverse Transcriptase-PCR
RNA purification and Northern blotting of the S880 line of D. melanogaster followed previously described procedures (36) using the entire DIP1 gene as a probe. The Ambion RETROscript kit was used for reverse transcriptase PCR, and random decamers were used for priming first strand synthesis. Primers directed at the full-length DIP1-c sequence were used for all subsequent rounds of PCR. Otherwise, protocols in the Ambion manual were followed.

Northwestern Blotting
Poly(I)-poly(C) dsRNA and poly(C) single-stranded RNA (Amersham Biosciences) were hydrolyzed according to Ref. 37 and end-labeled with [{gamma}-32P]ATP (MP Biomedicals, Inc.). The Northwestern blotting protocol (38) was adapted as follows. His6-DIP1-c and BSA were applied to a nitrocellulose membrane (0.45-µm pore size; Schleicher & Schuell) in separate rows containing 0.43, 0.87, 1.74, 3.04, 4.35, 8.69, 13.0, and 17.4 µg of protein using dot blot microfiltration. Membranes were then incubated in 10 mM Tris-HCl, pH 8.0, 25 mM KCl, 10 mM NaCl, 10% glycerol, 0.5 mM DTT, 0.1 mM EDTA, 0.04% BSA for 1 h at room temperature with constant agitation, followed by a 2-h incubation in a rotating hybridization chamber with 1 x 105 cpm/ml 32P-labeled poly(I)-poly(C) at 2.24 x 10-10 M or poly(C) at 4.95 x 10-9 M. After three washes in buffer (10 min of incubation/wash), membranes were air-dried and exposed to a Fuji phosphorimaging plate for 16 h. Data were analyzed using MacBAS version 2.0 software.

RNA Template Preparation
Several RNAs, used for gel retardation experiments, were produced by in vitro transcription.

Human Immunodeficiency Virus (HIV) TAR RNA—The DNA template contained the sequence of the TAR RNA (5'-GGCAGAUCUGAGCCUGGGAGCUCUCUGCC-3') behind the T7 RNA polymerase (RNAP) promoter (5'-TAATACGACTCACTATAG-3'). This template (1 µM), resuspended in water, was annealed with the T7 primer oligonucleotide (1.2 µM).

Adenovirus VA1 RNA—The template was amplified by PCR from pVA1, a gift from Goran Akusjarvi (Uppsala University, Sweden). The primers, 5'-ATTAATACGACTCACTATAGGGGCACTCTTCCGTGGTCTGGTG-3' and 5'-AAAAGGAGCGCTCCCCCGTTGTC-3', were used to generate a 182-base pair product that contains the VA1 sequence under the T7 RNAP control. The PCR product was purified using the Qiagen PCR Clean-up kit, and the DNA template was resuspended in water.

In Vitro Transcription—Each transcription reaction consisted of 30 mM Tris-HCl, pH 8.1, 2 mM spermidine, 0.01% Triton X-100, 25 mM MgCl2, each NTP at 4 mM, 10 mM DTT, 1 µM DNA template (HIV TAR or adenovirus VA1), and 50 µg/ml T7 RNAP (a gift from Yousif Shamoo, Rice University). The reactions were incubated at 37 °C for 3 h. To stop the reactions, an equal amount of 2x formamide solution (90% (v/v) formamide, 1x TBE (39), 25 mM EDTA, 0.02% bromphenol blue, 0.01% xylene cyanol) was added, and the samples were boiled for 1 min. The reactions were loaded on a 20% acrylamide (19:1) gel containing 7 M urea and 1x TBE. The RNA products were excised, crushed through a plastic syringe, and soaked in 0.3 M NaCl overnight. The RNA solution was syringe-filtered through a membrane with 0.22-µm pore size (GeneMate), precipitated with ethanol, and resuspended in water treated with diethyl pyrocarbonate.

Dephosphorylation—Transcribed RNA was dephosphoryated for 1 h at 37 °C using 10 units of alkaline phosphatase (Sigma) in 50 mM Tris-HCl, pH 7.9, 100 mM NaCl, 10 mM MgCl2, 1 mM DTT. The dephosphorylated products were precipitated with ethanol and resuspended in water.

Radiolabeling—RNA was radiolabeled with [{gamma}-32P]ATP (MP Biomedicals, Inc.) using polynucleotide kinase as described by Promega. Radiolabeled RNA was purified in 20 mM Tris-HCl, pH 7.5, or TE buffer, pH 8.0 (39), with a NICKTM column (Amersham Biosciences). The column was washed in 400 µl of buffer, and the RNA eluted in an additional 400 µl of buffer. The RNA was aliquoted and stored at -20 °C.

RNA Gel Retardation—DIP1-c ·RNA binding was measured in 20 mM Tris-HCl, pH 7.5, 100 mM KCl, 100 µg/ml BSA, 5 mM DTT, 10% glycerol. The final concentration of RNA in each reaction was ~4 x 10-11 M (TAR) and ~1 x 10-12 M (VA1), more than 10-fold below the dissociation constant for these RNA interactions. The concentration of DIP1-c ranged from 1.6 x 10-12 to 6.3 x 10-9 M for VA1 RNA and from 5 x 10-10 to 2 x 10-6 M for dsDNA and TAR RNA with no DIP1-c present in the first reaction. Reactions were incubated at room temperature for 20 min. Bound complexes and free RNA were separated on 4% acrylamide (37.5:1) gels that contained 3% glycerol in 0.5x TBE (39). The gels were blotted on filter paper, dried on a vacuum slab gel dryer, and exposed to a Fuji phosphorimaging plate for 16 h. Data were analyzed using MacBAS version 2.0 and Igor Pro version 2.02 software as previously described (40).

DNA Labeling and DNA Gel Retardation
Complementary DNA oligonucleotides (topstrand, 5'-GTCTGATCAGTAGAGCCTTAATGGCCGTAG-3') were annealed, and the dsDNA was labeled for 1.5 h in reactions containing 0.3 nM DNA, 10 units of polynucleotide kinase, and 1x kinase buffer (Promega), 2 mM spermidine, and 100 µCi of [{gamma}-32P]ATP (MP Biomedicals, Inc.) in a final volume of 20 µl. The labeling reaction was terminated by the addition of 2 µl of 0.5 M EDTA, and the labeled DNA was loaded onto a NICKTM column (Amersham Biosciences) pre-equilibrated in TE buffer, pH 8.0 (39). The column was washed in 400 µl of TE, and the DNA was eluted in an additional 400 µl of TE buffer. DNA was aliquoted and stored at -20 °C.

DNA gel retardation was performed as previously described (41, 42) with minor variations. Protein and DNA were diluted in DNA binding buffer: 50 mM Tris-HCl, pH 7.5, 100 or 200 mM KCl, 10% glycerol, 5 mM DTT, 100 µg/ml BSA. Ubx Ib activity was measured as previously described (41, 42). All Ubx Ib preparations were ~90–100% active. DIP1-c supershift assays contained 3 x 10-12 M DNA (100–300 cpm/µl). DIP1-c concentration ranged from 4 x 10-8 to 2.5 x 10-7 M. In a duplicate set of lanes, Ubx Ib was present at 6.3 x 10-8 M.

Yeast One-hybrid Analysis of DIP1-c Effect on Transactivation by Ubx Ib
The Matchmaker LexA Two-hybrid System (Clontech) was used in these experiments with modification of the pJG vector. The coding region of full-length, wild-type Ubx Ib was subcloned into the pLexA vector via the EcoRI and BamHI restriction enzyme sites to produce pLexA-Ubx. The plasmid pJG was modified by removing the coding region of the B42 activation domain with digestion of XhoI and EcoRV to produce vector pJG{Delta}B42AD. The DIP1-c coding region was subcloned into the vector pJG{Delta}B42AD using EcoRI and XhoI, and the resulting plasmid, pJG{Delta}B42AD-DIP1, was transformed into the yeast strain EGY48 carrying pLexA-Ubx and the reporter plasmid p8op-lacZ, with pJG{Delta}B42AD as a negative control. The vector pCL1, which allows for constitutive activation of the lacZ reporter gene using wild-type Gal4, was used as a control to ensure DIP1-c effects on Ubx were specific. The transformants were plated on SD minimal medium plates lacking uracil, histidine, and tryptophan and containing galactose as the carbon source. The plates were incubated at 30 °C until colonies became visible. X-Gal (40 mg/ml) was added to the plates to monitor transactivation of the lacZ reporter gene by Ubx Ib.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Identification of DIP1 as a Partner for Ubx Ib—A yeast two-hybrid screen was used to search for proteins that interact with the Drosophila Hox protein Ubx Ib. The Ubx Ib isoform was selected to include all possible amino acid sequences (Fig. 1). Because Ubx Ib contains an activation domain, the LexA-Ubx fusion is constitutively active, precluding detection of partners (21). A mutant in which the N-terminal 216 amino acids are deleted to abrogate transcription, Ubx{Delta}N216, was therefore generated for the initial yeast two-hybrid assays (21). Ubx{Delta}N216 was fused to the LexA DNA binding domain, and a 0–12-h Drosophila melanogaster cDNA library was fused to the B42 activation domain. Isolation and sequencing of cDNA from colonies that exhibited {beta}-galactosidase production using Ubx{Delta}N216 as bait yielded the sequence of the 3'-terminal portion of the DIP1 gene. The entire 5' region of the gene was identified by searching the Berkeley Drosophila Genome Project Database (available on the World Wide Web at www.fruitfly.org) for sequences that overlapped with the DIP1 cDNA sequence. Full-length DIP1-c, the longest isoform, fused with the B42 activation domain, yielded a product able to interact with Ubx{Delta}N216 in yeast two-hybrid assays, an indication that the interaction is not an artifact caused by the missing DIP1-c 5' sequence in the cDNA library. To ensure that the DIP1-c interaction does not require buried residues that are exposed by the Ubx N-terminal deletion, yeast two-hybrid assays were repeated using full-length Ubx Ib proline mutant 1, in which transactivation was abrogated by the A223P/A226P/Q233D mutations (21). DIP1-c still interacts with full-length proline mutant 1 Ubx, indicating that the Ubx Ib ·DIP1-c interaction does not require exposure of interior domains due to deletion of the N terminus. Interaction in the yeast two-hybrid system indicates these full-length proteins have higher affinity for each other than for endogenous yeast proteins.

Verification of the Ubx Ib ·DIP1-c Interaction—The interaction of Ubx Ib and DIP1-c was first confirmed by switching their positions in the yeast two-hybrid assay, such that Ubx Ib was fused to the B42 activation domain and DIP1-c was fused to the LexA DNA binding domain. This experiment detects whether either protein interacts with LexA or B42 to generate a false positive. Because the fusion of full-length wild-type Ubx Ib with the B42 activation domain resulted in a toxic protein, a special "colony lift" procedure was developed for this assay. Yeast colonies were first grown on a glucose-containing plate, repressing expression of the toxic Ubx-B42 fusion protein, and then the colonies were transferred to a plate containing galactose to induce fusion protein expression. The interaction of full-length Ubx Ib and DIP1-c was confirmed in these "reverse" yeast two-hybrid experiments.

The Ubx Ib ·DIP1-c interaction was further verified using phage display methods. Two advantages of phage display are that wild-type, full-length Ubx Ib can be utilized, and there are no yeast proteins present that might bridge Ubx Ib and DIP1-c. Full-length biotinylated Ubx Ib was bound to a Petri dish coated by avidin, and the Petri dish was incubated with phage expressing DIP1-c on the surface. After five washes, 77% ± 4% (four experiments) of the phage was retained on the plate, indicating a strong Ubx Ib ·DIP1-c interaction. The positive control for this system employed interaction of Ubx Ib and Exd. The Ubx Ib ·Exd partnership (two experiments) resulted in 74 and 76% phage retention, values comparable with Ubx Ib ·DIP1-c. Negative controls yielded <10% phage retention. The experiment was also repeated with identical results using purified Ubx Ib without a biotin tag (data not shown). Therefore, biotin does not cause structural rearrangement in Ubx Ib that enhances interaction with DIP1-c. These assays demonstrate that full-length Ubx Ib can interact effectively with full-length DIP1-c in the absence of yeast or bacterial proteins.

Finally, the full-length proteins were shown to interact by immunoprecipitation and GST pull-down assays. DIP1-c was cloned into the plasmid pET28a, which contains sequences for adding a His6 tag to the N terminus of the protein. This modification allows rapid purification and detection of the protein. All buffers contained 5% glucose to minimize precipitation of DIP1-c as established by an aggregation assay (32). DIP1-c runs at an anomolously high molecular weight on SDS-PAGE, a trait previously reported for other proteins containing dsRNA-binding domains (38, 43). Ubx Ib can be precipitated out of solution using two antibodies to Ubx directed at opposite ends of the protein (epitopes approximately located at residues 70–131 and 226–297) (34). DIP1-c co-precipitated with Ubx Ib and the antibodies but was not precipitated by the antibodies in the absence of Ubx Ib (Fig. 2A).



View larger version (45K):
[in this window]
[in a new window]
  		FIG. 2.
Verification of the Ubx Ib ·DIP1-c interaction. A, Ubx Ib and His6-DIP1-c were co-immunoprecipitated by two anti-Ubx monoclonal antibodies, 10H.7 and 2H10"7", for the experiments in the first four lanes. His6-DIP1-c was combined with these antibodies for the three right lanes as a control. Shown are Western blots of tagged and untagged DIP1-c. Lanes 1 and 2 were probed with the 10H.7 antibody to Ubx, and the last five lanes were probed with the {alpha}-tetra-His antibody (Qiagen), an antibody to the His6 tag on DIP1-c. S, supernatant; P, pellet; W, wash. B, either His6-DIP1-c alone or GST-Ubx lysate and His6-DIP1-c were combined with glutathione-Sepharose 4B beads (Amersham Biosciences). Following separation, the beads were washed (W) and then eluted (E) with reduced glutathione. The Western blot was probed with the {alpha}-tetra-His antibody (Ab). S, supernatant; W, three washes; E, eluate. C, the GST pull-down assay was repeated with DIP1-c treated with thrombin to remove the His6 tag. The Western blot was probed with rabbit polyclonal antibodies to DIP1.

 
The high concentration of purified proteins used in coimmunoprecipitation might force nonspecific Ubx Ib ·DIP1-c interaction. Therefore, GST pull-down assays using E. coli protein extracts were also performed to demonstrate specific interaction with Ubx Ib. Using Glutathione SepharoseTM beads, GST-Ubx was precipitated from a bacterial lysate. DIP1-c alone, either with or without the histidine tag, does not precipitate in the presence of Glutathione SepharoseTM. When GST-Ubx is included in the reaction, however, both versions of DIP1-c are removed from solution (Fig. 2, B and C). The histidine tag on DIP1-c apparently does not abrogate the interaction with Ubx Ib. These data also demonstrate that DIP1-c preferentially interacts with Ubx Ib in the presence of competing bacterial proteins. In summary, the Ubx Ib ·DIP1-c interaction has been confirmed by three different yeast two-hybrid, two-phage display, and immunoprecipitation and GST pull-down experiments.

Ectopic Expression of DIP1 in Wings and Halteres—Expression of ubx in the wing or haltere imaginal discs promotes haltere formation from the default developmental programs (44). Because Hox proteins play a causal role in appendage specification in Drosophila and DIP1 is expressed in wing and haltere imaginal discs (24), we determined the effects of DIP1 overexpression on development of these appendages using the Gal4-UAS system (35). P{GawB}BxMS1096 Gal4 drives transcription in the wing and haltere imaginal discs, predominantly on the dorsal side (45), and the UAS-DIP1-c responder expresses well in Drosophila (24). Ectopic expression of DIP1-c in wing imaginal discs resulted in a dramatic reduction in wing size and a thicker, shriveled appearance (Fig. 3, A and B), a phenotype that does not occur with either element in isolation. Because individual cells were visible, cell density could be measured. The change in wing size was due to a change in cell number, not cell size. The wing pictured in Fig. 3B is one of the larger ones observed, but the smaller wings were concave and could not lie flat for microscopy without folding or tearing. On rare occasions (<2%), bifurcated wings were observed. Bifurcations were always on the smallest, most concave wings and therefore may occur in flies with the highest levels of ectopic DIP1-c expression. In addition, both wing veins and the line of bristles along the margin were lost, although some heavy bristles are clustered proximal to the notum and at the most distal edge. Ectopic expression of DIP1-c in haltere imaginal discs resulted in a range of deformed halteres (Fig. 3, D and E) compared with the wild-type haltere (Fig. 3C). Most halteres contained a misshapen capitellum, and ~10% were also bifurcated. Like the wing overexpression data, ectopic expression phenotypes demonstrate that DIP1-c is a functional protein in vivo that influences growth and patterning.



View larger version (59K):
[in this window]
[in a new window]
  		FIG. 3.
Ectopic expression of DIP1-c in wing and halteres. A, wild-type Drosophila wing. B, ectopic expression of DIP1-c in wing and haltere imaginal discs using the MS1096-Gal4 driver in the Gal4-UAS system results in a shriveled wing (B) with three clusters of bristles, indicated by the arrows. A wild-type haltere (C) is shaped like a balloon, whereas overexpression of DIP1-c generates misshapen halteres (D), ~10% of which are bifurcated (E). Black scale bars, 1 mm; white scale bars, 0.25 mm.

 
Genetic Interaction of Ubx Ib and DIP1-c—Ectopic expression of ubx in the wing imaginal disc also generates small shriveled wings with excess bristles (45). The similarity of the DIP1 and ubx ectopic expression phenotypes in the wing suggests a potential for a genetic interaction between the two genes. To explore this possibility, we utilized the Ubxbx-34e gypsy insertion line, which causes a reduction of ubx gene expression in the haltere imaginal disc (44). Homozygotes have an enlarged haltere, with long, dark costal bristles (Fig. 4A), whereas halteres in the heterozygotes appear wild-type (Fig. 4B). Initial efforts examined whether alterations in DIP1 expression could modulate the Ubxbx-34e heterozygote phenotype. Indeed, flies containing a single copy of Ubxbx-34e and a duplication of the region of the X chromosome containing the DIP1 gene onto the Y chromosome had a punctate line of short, dark bristles on the capitellum with a denser cluster of bristles at the distal tip (Fig. 4C). To ensure that this effect was caused by the DIP1 gene, DIP1-c was ectopically expressed in the wing and haltere imaginal discs of Ubxbx-34e heterozygous flies using the MS1096-Gal4 driver. Ubx is naturally expressed across the entire haltere imaginal disc with highest levels in the pouch (46). Therefore, Ubx and ectopic DIP1-c domains will overlap. These flies had long bristles across the entire surface of the haltere, an enhancement of the Ubxbx-34e heterozygous pheno-type that is not visible in wild-type flies or flies with only the Ubxbx-34e and MS1096-Gal4 elements (Fig. 4, D–L). For halteres mounted in the same orientation, negative controls heterozygous for just MS1096-Gal4 and Ubxbx-34e had twice as many bristles as wild type, whereas flies heterozygous for MS1096-Gal4, UAS-DIP1-c, and Ubxbx-34e had four times as many bristles as wild type. Thus, ectopic expression of DIP1-c, either by duplication or by specific overexpression, enhances the Ubxbx-34e effect. This genetic interaction between the ubx and DIP1 genes provides in vivo evidence for a physiological role of the demonstrated physical association between the two proteins.



View larger version (103K):
[in this window]
[in a new window]
  		FIG. 4.
Genetic interaction between ubx and DIP1-c. A reduction in Ubx expression in flies homozygous for Ubxbx-34e results in an enlarged haltere with a thick row of dark costal bristles (A) (44), which are absent in the heterozygote (B). Overexpression of DIP1 using a duplication of the region of the X chromosome containing the DIP1 gene in Ubxbx-34e heterozygotes partially restores the ectopic bristle phenotype (C). Smaller ectopic bristles can also be produced by expression of DIP1-c using the Gal4-UAS system (D), one of which is marked by the arrow. To view these effects more clearly, dissected halteres were examined at higher resolution. E and F, wild-type haltere in two different focal planes, showing the distribution of elongated bristles on the sides but not across the face of the haltere. G and H, haltere from a fly heterozygous for MS1096-Gal4 and Ubxbx-34e, showing some increase in the number of bristles, and similar to controls containing Ubxbx-34e and UAS-DIP1-c (data not shown). I and J, haltere from a fly heterozygous for MS1096-Gal4, UAS-DIP1-c, and Ubxbx-34e. DIP1-c overexpression produces an increased number of bristles both along the margins (white arrows) and across the surface of the haltere compared with wild-type and the negative control. The line of bristles along the edge of the haltere in I extends beyond the arrows in other focal planes. Note the unusual haltere shape due to DIP1-c overexpression. K and L, enlarged view of distal surface of halteres from J and H, respectively. Ectopic bristles indicated by green arrows can be distinguished from a linear arrangement of smaller hairs by the circular structures at the base of the bristles. Whereas an occasional bristle occurs (L, green arrow) in heterozygous flies of MS1096-Gal4 and Ubxbx-34e, the density of ectopic bristles is higher (K) in flies harboring all three genetic elements. All halteres in D–L were dissected from female flies. White scale bars, 0.25 mm.

 
Identification of Functional Motifs within the DIP1 Sequence—Although DIP1 is not homologous to another known protein throughout its entire length, the sequence does contain regions similar to known motifs that indicate potential biological function (Fig. 1). Three short motifs, a nuclear localization signal (residues 142–159) (47), a DEAD box (residues 260–263) potentially involved in Mg2+ binding (48), and a PEST motif (residues 107–138) (49), were found in DIP1 (Fig. 1B), along with multiple potential phosphorylation, glycosylation, and myristoylation sites (data not shown). These short motifs, if actually employed in vivo, may serve to regulate the function of DIP1. Importantly, two dsRBDs (residues 171–237 and 335–403) (50) were also identified in DIP1. dsRBDs are 65–68-amino acid motifs that are divided into two categories: type A domains, which are conserved throughout the motif in sequence and secondary structure of {alpha}-{beta}-{beta}-{beta}-{alpha} (where {alpha} denotes {alpha}-helix and {beta} is {beta}-sheet), and type B domains, which are only conserved in the C-terminal helix (37, 50). Interestingly, the first dsRBD of DIP1 is a type A domain whereas the second dsRBD is a type B domain (data not shown) (24). Yeast two-hybrid experiments indicate that DIP1-c self-associates (data not shown), a characteristic common to proteins containing dsRBDs (38, 51). Although these domains are generally only loosely conserved, the highest sequence similarity was to two domains found within RED1, an RNA editase (52). DIP1-c is 29% (87/292) identical and 43% (127/292) similar to human RED1, in which there is 41% (28/68) identity and 50% (34/68) similarity between the first dsRBD of each protein.

Nucleic Acid Binding by DIP1-c—The ability of purified DIP1-c to bind RNA was initially tested using Northwestern blotting (Fig. 5). DIP1-c was able to bind poly(I)-poly(C) at subnanomolar concentrations, indicating a strong affinity for dsRNA. In contrast, a 22-fold higher concentration of poly(C) was required to observe weak, but significant, binding in the same range of protein concentrations. Also, DIP1-c binds a 30-base pair dsDNA duplex with low affinity (Fig. 6 and Table I). Strong interaction with dsRNA and weak interaction with single-stranded RNA and dsDNA is characteristic of proteins containing double-stranded RNA-binding domains (37, 38, 50). Therefore, the double-stranded RNA-binding domains identified through sequence similarity are functionally active.



View larger version (105K):
[in this window]
[in a new window]
  		FIG. 5.
Northwestern blot to determine RNA binding of His6-DIP1-c. Nitrocellulose membranes to which protein had been applied were incubated with a solution containing 1 x 105 cpm/ml 32P-labeled RNA. Poly(I)-poly(C) dsRNA at 2.24 x 10-10 M was used for the experiment on the left; Poly(C) single-stranded RNA at 4.95 x 10-9 M was utilized for the blot shown on the right. The first column on both membranes was filtered with water, the second column with solutions containing the indicated amounts of DIP1-c, and the final column with solutions containing identical amounts of BSA. The RNA binding in the DIP1-c column is due solely to the presence of the DIP1-c protein.

 



View larger version (86K):
[in this window]
[in a new window]
  		FIG. 6.
DIP1-c binds dsRNA with high affinity. The ability of DIP1-c to bind dsDNA, TAR RNA, and Adenovirus VA1 RNA was investigated using gel retardation. A, a gel retardation assay for dsDNA; B, an assay for TAR RNA; C, an assay for VA1 RNA. The final RNA or DNA concentration in each reaction for A and B was ~4 x 10-11 M and ~1 x 10-12 M for C. B, bound species; F, free DNA or RNA. Lane 1 of each gel contained no protein. For A and B, DIP1-c was present at concentrations of 5 x 10-10 to 2 x 10-6 M in lanes 2–20, whereas in C, DIP1-c was added at concentrations of 1.6 x 10-12 to 6.3 x 10-9 M in lanes 2–20. To the right of each gel is the binding curve that is representative of the data. For A (four gels) and B (three gels), the resulting curves are based on the disappearance of the free DNA or RNA species, whereas for C (four gels), the curves show both disappearance of free RNA (open circles) and the appearance of DIP1-c ·RNA complexes (closed circles). Table I summarizes the calculated binding constants for the three curves.

 


View this table:
[in this window]
[in a new window]
  		TABLE I
Comparison of nucleic acid binding of DIP1-c with other dsRNA-binding proteins Reported Kd values are shown for Staufen (53) and interferon-inducible dsRNA-dependent kinase (PKR) (54–56), and Kapp is shown for DIP1-c. Interactions identified by Northwestern blotting techniques are denoted by a plus sign. Staufen and PKR, as well as other dsRNA-binding proteins, are reported to not bind dsDNA on the basis that dsDNA is not able to compete with dsRNA using Northwestern blotting techniques, as indicated by a minus sign. ND, the experiment has not been reported. The value of ≤100 nm for DIP1-c with poly(I)-poly(C) dsRNA is estimated from the area of protein coverage on the filter in the Northwestern blot (Fig. 5), assuming that the RNA concentration at the surface is equivalent to solution and a thickness of the protein layer of ~0.1 nm. The binding constants reported for DIP1-c are calculated from the data presented in Fig. 6, in which three gels were used for TAR RNA, four gels were used for dsDNA, and 12 gels were used for VA1 RNA.

 
The affinity of DIP1-c for commonly used dsRNAs was compared with other dsRBD proteins to determine whether the DIP1-c domains are sufficiently active for in vivo function. Two dsRNA ligands were transcribed in vitro and tested as ligands in gel retardation assays (Fig. 6). dsRNA binding proteins are reported to bind dsRNA ligands with high affinity (Kd ~10-9M), regardless of sequence or physiological relevance (53–56). In these assays, DIP1-c bound with high affinity to HIV TAR RNA (50 nM ≥ Kapp ≥ 1 nM) but bound with much higher affinity to the adenovirus VA1 RNA (Kapp ~ 50 pM). The binding affinities of DIP1-c for dsRNA targets are comparable with or greater than those reported for other dsRNA-binding proteins (Table I).

Alternative Splicing of DIP1—Alternative splicing of the DIP1 gene was indicated by the presence of four major bands on a Northern blot of mRNA isolated from 0–16 h embryos from the S880 wild-type D. melanogaster strain. The existence of multiple DIP1 RNA isoforms was confirmed by reverse transcriptase PCR (Fig. 7). Sequences of these DNAs indicated differences that are derived from alternative splicing. Several DIP1 isoforms have previously been deposited at GenBank.2 We have identified a novel isoform, DIP1-d, in which the putative nuclear localization signal is not present (Fig. 8). A version of DIP1 potentially confined to the cytoplasm suggests the protein may have multiple functions (24).



View larger version (38K):
[in this window]
[in a new window]
  		FIG. 7.
Evidence for alternative splicing of DIP1. The left panel depicts a Northern blot of embryonic and adult D. melanogaster total RNA. Four main bands in embryonic total RNA indicate four alternatively spliced products of DIP1. Three of these bands are also present in mRNA from adult flies. Right panel, agarose gel depicting the reverse transcriptase PCR. This gel is 1% agarose, and lane 1 shows a negative control in which no reverse transcriptase was added to the first strand synthesis reaction. Lane 2 shows the experimental product with four prominent bands, and lanes 3 and 4 show molecular weight markers. The three isoforms identified by sequencing in the reverse transcriptase PCR reaction are noted.

 



View larger version (35K):
[in this window]
[in a new window]
  		FIG. 8.
Six isoforms of DIP1. There are five isoforms of DIP1 that have been deposited in GenBankTM by other laboratories, two of which were also identified in our experiments. These include DIP1-c/Klett-c and DIP1-b/Klett-d, identified in our experiments, and DIP1-a, Klett-a, and Klett-b, identified by others. We also report an additional isoform, DIP1-d, which has the nuclear localization signal (NLS) spliced out. The dsRNA-binding domains (dsRBD1 and -2) are also indicated as well as the length of each isoform in amino acids (AA). The amino acids included in each isoform are indicated by the small numbers below the sequence.

 
Identification of the Protein-Protein Interaction Regions of Ubx Ib and DIP1-c—Comparison of the Ubx Ib ·DIP1-c interaction domains with known functional domains provides a means to predict activities that may be altered or precluded by the Ubx Ib ·DIP1-c interaction. Deletion mutants of each protein were used in the yeast two-hybrid system to identify the sequences of each protein required for the Ubx Ib ·DIP1-c inter-action. A series of deletion mutants were made in which sequences were successively removed from the N- and C-terminal ends of DIP1-c (Fig. 9). Residues 1–266 are not required for interaction, since the interaction was identified originally for an N-terminal truncated form of DIP1-c. Decreased interaction with Ubx Ib can be detected when residues 267–307 are removed, as indicated by pale blue colonies. From the C terminus, residues 366–410 can be removed with no adverse effect, but removing residues 327–365 abolishes the Ubx Ib ·DIP1-c interaction. Thus, the DIP1-c sequence required for interaction with Ubx Ib resides in amino acids 327–365, and residues in the 267–307 region can enhance this interaction (Fig. 1B). These data are in agreement with GST pull-down experiments for key DIP1-c deletions (Fig. 9). This large region may be required to stabilize the tertiary structure supporting the interacting residues.



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 9.
Yeast two-hybrid and GST pull-down assays to determine DIP1-c sequences required for Ubx Ib binding. The constructs indicated were fused to the B42 activation domain in pB42AD and utilized in yeast two-hybrid assays with Ubx{Delta}N216 fused to the LexA DNA binding domain. Blue, yeast colonies exhibiting high levels of {beta}-galactosidase activity, reflective of partner formation with Ubx Ib; Pale Blue, colonies with reduced enzyme activity and therefore reduced Ubx Ib ·DIP1-c complex formation; White (no enzyme activity), Ubx Ib ·DIP1-c binding has been abolished. To confirm these results, full-length DIP1-c and key deletions were expressed in vitro and tested for interaction with GST-Ubx Ib. ++, strong interaction in the GST pulldown assay; +, a weak interaction; -, no interaction. The protein motifs with DIP1-c identified through homology searches are also depicted.

 
For Ubx Ib, the interaction domain must reside in the C-terminal 173 amino acids, since this protein was the original "bait" in the yeast two-hybrid experiments. All deletions made to Ubx{Delta}N216 on the C-terminal end or on the N-terminal end abolish the Ubx Ib ·DIP1-c interaction (Fig. 1A). Therefore, either multiple regions of Ubx Ib are required for this interaction, or a specific conformation of the protein supported by these regions is necessary to bind DIP1-c. The homeodomain alone is insufficient for this interaction. The DIP1-c interaction domain of Ubx Ib includes part of the activation domain, the Exd interaction motif, the three microexons, the homeodomain, and the repression domain. These Ubx Ib functions may be modulated by DIP1-c binding. Likewise, Ubx Ib binding may affect the affinity of the DIP1-c ·RNA interaction due to the overlap of the region of Ubx Ib interaction and the second dsRBD. Preliminary results suggest that DIP1-c can bind RNA in the presence of Ubx Ib (data not shown); however, specific RNA sequences or structures may influence the effect of Ubx Ib on DIP1-c ·RNA binding.

Ubx Ib and DIP1-c Form a Ternary Complex on Ubx Target DNA—For DIP1-c to influence transcription regulation by Ubx Ib in vivo, it must be able to participate in a ternary DIP1-c ·Ubx Ib ·DNA complex. To determine whether Ubx Ib can simultaneously bind both DIP1-c and its target DNA, a gel retardation assay was employed. As shown in Fig. 10 (lanes 12 and 13), Ubx Ib binds DNA with a single consensus binding site of 5'-TAATGG-3' (57) to generate a single shifted band. Under the same conditions, only a small degree of DIP1-c ·DNA binding was observed, consistent with the general ability of dsRBDs to weakly interact with dsDNA (37, 38, 50). However, when DIP1-c and Ubx Ib are combined, a novel DNA complex is observed. This "supershift" is only present when both proteins are added to the reaction. Western blot analysis detected both proteins in the supershifted band (data not shown). The supershift occurs at protein concentrations at which DIP1-c does not bind DNA, implying that DIP1-c binds directly to Ubx Ib to form the ternary complex. This complex was also observed in a buffer containing 200 mM KCl, which would be expected to disrupt nonspecific interactions. Although the homeodomain is encompassed by the region of Ubx Ib required for DIP1-c interaction, DIP1-c does not interfere with DNA binding by Ubx Ib and must leave the homeodomain binding surface exposed. This observation is consistent with the inability of the isolated homeodomain to bind DIP1-c. The binding of DIP1-c to Ubx Ib ·DNA does not introduce a significant alteration in the affinity of Ubx Ib for DNA (data not shown), suggesting that DIP1-c does not modulate single site DNA binding by Ubx Ib. Observation of this supershift demonstrates Ubx Ib ·DIP1-c interaction can occur at nanomolar protein concentrations.



View larger version (90K):
[in this window]
[in a new window]
  		FIG. 10.
Ternary complex between Ubx Ib, DIP1-c, and Ubx target DNA. A gel retardation assay was employed to investigate the interaction between Ubx Ib and DIP1-c in the presence of Ubx target DNA. The 30-base pair DNA contains one Ubx consensus binding site, 5'-TAATGG-3' (57). The final DNA concentration in all reactions was 10-12 M. No protein was present in lane 1. In lanes 2–6, DIP1-c was present at 4 x 10-9 to 4 x 10-7 M. Lanes 7–13 contained Ubx Ib at 6.3 x 10-8 M. Lanes 7–11 contained the same DIP1-c dilutions as lanes 2–6. The molar ratios of DIP1-c/Ubx Ib in lanes 7–11 were 0.6, 1.0, 1.6, 2.5, and 4.0, respectively. D, faint bands generated by DIP1-c ·DNA in lanes 2–6; C, a DIP1-c ·Ubx Ib ·DNA complex formed in lanes 7–11. U, Ubx Ib ·DNA band in lanes 12 and 13; W, wells; F, free DNA.

 
DIP1-c Specifically Represses Ubx Ib Activation—The ability to form a ternary complex, as well as the overlap between the DIP1-c interaction domain and Ubx Ib activation and repression domains, suggests that DIP1-c may alter the mode (activation or repression) or the strength of transcription regulation. To test this hypothesis, a modified yeast one-hybrid assay was employed to investigate the function of the Ubx Ib ·DIP1-c interaction (Table II). Full-length, wild-type Ubx Ib fused to the LexA DNA-binding domain can constitutively activate transcription in yeast, yielding blue colonies in the presence of X-gal (15). Transcriptional activation by Ubx Ib in this assay is independent of Ubx Ib ·DNA binding but relies on LexA affinity for its operator. When full-length, wild-type DIP1-c (no B42 fusion) was expressed in these yeast, the colonies remained white on screening with X-gal, indicating that {beta}-galactosidase was not expressed. A control experiment was performed with pCL1, a vector which allows constitutive activation of the reporter gene by expressing wild-type yeast Gal4. When DIP1-c was co-expressed with Gal4, the colonies still turned blue in the presence of X-gal. These data suggest that DIP1-c is not a general inhibitor of transcription or translation but instead must specifically interact with Ubx Ib to affect its transcriptional activation. The observed inhibition of transactivation of Ubx Ib by DIP1-c is not caused by blocking DNA binding by Ubx Ib, because a ternary complex is observed in vitro, and Ubx Ib is not required to bind DNA in the fusion with LexA used in the yeast one-hybrid assay. Further, this inhibition is not caused by blocking DNA binding by LexA, or the Ubx Ib ·DIP1-c interaction would not be observed in the yeast two-hybrid assays. Thus, an interesting possibility is that DIP1-c may regulate Ubx Ib function in vivo by modulating the balance between transcription activation and repression by Ubx.


View this table:
[in this window]
[in a new window]
  		TABLE II
Yeast one-hybrid assays to determine possible DIP1-c function The constructs indicated were expressed in yeast cells; DIP1-c was in pJG{Delta}B42AD (no B42 fusion), and full-length Ubx Ib was fused to the LexA DNA binding domain. Blue, the yeast colonies exhibited high levels of {beta}-galactosidase activity, reflective of the constitutive transcriptional activation by sequences within full-length Ubx Ib. Expressing full-length DIP1-c in the same yeast cells resulted in white colonies, indicating that Ubx Ib transactivation was inhibited. The observed inhibition is Ubx-specific, because yeast cells co-expressing DIP1-c and Gal4 turn blue in the presence of X-gal. The vector pCL1 allows for constitutive activation of the reporter gene using wild-type Gal4. Thus, DIP1-c is not a general inhibitor of transcription or translation.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
The Hox family plays a critical role in segmental specification and cell fate determination in Drosophila development (58, 59). Given the global consequences of Hox function, activity must be tightly regulated in a temporal-, spatial-, and gene-specific manner. However, the high homology of the homeodomain combined with relatively promiscuous DNA binding in vitro is incompatible with the requirement for specific, yet plastic, function in vivo. A range of protein-protein interactions appears to be key to solving this Hox specificity puzzle (1, 6, 60–62). In order to understand how Hox proteins achieve their exquisite discriminatory function in vivo, these interactions must be discovered, and their effects on function must be understood.

Potent transcriptional activation domains within Hox proteins and consequent protein toxicity to yeast cells have impeded identification of Hox protein partners by traditional yeast two-hybrid methods. By modifying this approach, confirming the interaction with phage display, immunoprecipitation, GST pull-down assays, and gel retardation supershifts, and by identifying a genetic interaction, we have identified DIP1 as a Ubx Ib ·interacting protein. The DIP1 expression domains in the central nervous system, imaginal discs, and embryo overlap with Ubx expression, and both proteins are localized to the nucleus (6, 18–20, 24). Importantly, the Ubx Ib ·DIP1-c interaction requires Ubx amino acid sequences outside of the highly conserved homeodomain. Protein interactions in these much more variable regions could provide a source of functional discrimination within the Hox family.

The DIP1 protein contains two regions homologous to dsRNA-binding domains, a sequence for nuclear localization, PEST motifs, and four clustered PXXP sequences. Although, PXXP motifs are commonly used for mediating protein-protein interactions (63, 64), Ubx Ib does not contain a corresponding Src homology 3 domain. DIP1-c activity could thus be regulated (or regulate) through binding to other proteins via these PXXP motifs in vivo.

Northwestern blotting and gel retardation experiments demonstrate that DIP1-c, the longest isoform, binds with very high affinity to dsRNA and with ~1000-fold lower affinity to single-stranded RNA and dsDNA, consistent with active dsRBDs (37, 38, 50). The existence of alternative splicing products for DIP1 (GenBankTM) (24), as also observed in our experiments, suggests regulated expression of different forms in spatial and/or temporal patterns in Drosophila. In particular, we identified an isoform in which the putative nuclear localization signal is removed, potentially creating a cytoplasmic version of DIP1. The Campos laboratory identified additional isoforms and presented evidence supporting the existence of a cytoplasmic DIP1 isoform in nurse cells (24).

Previously, DIP1-b overexpression in the eye-antennal imaginal disc was shown to induce transformation phenotypes of adult head structures (24). Herein, we demonstrate the DIP1-c ectopic expression also alters the development of wings and halteres. Thus, DIP1-c can function in ubx-expressing cells. A genetic interaction was identified between underexpression of Ubx in the haltere and overproduction of DIP1, both by a duplication of the DIP1 gene and by the Gal4-UAS system. The enhancement of the Ubx mutant phenotype produced by increased DIP1 expression may be caused by interactions in vivo that hinder the normal function of Ubx in these cells.

Indeed, DIP1-c specifically blocks transcription activation by Ubx Ib in yeast cells. Differential transcriptional regulation by Hox proteins, presumably mediated by protein interactions, is required for specific Hox function (1, 6, 21–23, 62, 65). Although a portion of the activation domain of Ubx Ib (amino acids 159–242) is required for DIP1-c interaction, neither deletion of amino acids 1–216 nor mutations between 223–233 abrogate the interaction, suggesting that these residues are not required for the DIP1-c partnership. However, interaction with general transcription factors may be sterically occluded by DIP1-c interaction, or DIP1-c binding may lock Ubx Ib into a conformation that inhibits transcriptional activation.

Gel retardation assays demonstrate ternary complex formation between Ubx Ib, DIP1-c, and DNA that would provide a mechanism by which DIP1-c could modulate transcription regulation in a Ubx Ib ·specific manner. The small impact of DIP1-c on Ubx Ib ·DNA affinity suggests that this regulation would derive from influence on other Ubx Ib functions, consistent with the loss of transcriptional activation by the Ubx Ib ·DIP1-c complex in yeast. DIP1 expression coincides with the onset of zygotic transcriptional activity in the embryo, further suggesting a role for DIP1 in transcriptional regulation (24). Interestingly, the region of Ubx Ib that is required to interact with DIP1-c contains the C-terminal glutamine/alanine-rich region, which mediates transcription repression in arthropods (22, 23). In vivo, Ubx represses both distalless and spalt in the absence of Extradenticle, a Hox co-factor required for Ubx-mediated repression of several genes (66). It will be of interest to determine if this repression is influenced by DIP1.

Conversely, Ubx Ib interaction may modulate DIP1 function. The Ubx Ib interaction domain on DIP-c includes part of the C-terminal dsRNA binding domain and hence may impact interactions with RNA. Several homeodomain proteins bind RNA in vivo. The fold of the RNA-binding domain of ribosomal protein L11 was shown to be similar in structure to the homeodomain (67). Further, Bicoid, a homeodomain protein that establishes the anterior-posterior gradient in the Drosophila embryo, binds the Bicoid response element in the 3'-untranslated region of caudal mRNA via its homeodomain, thus repressing translation (68, 69). Other transcription factors and DNA-binding proteins have been shown to bind RNA as well (70). The Hox protein Antennepedia interacts with Split Ends, a protein with RNA recognition motifs (71). The interaction of DIP1 in vitro with a zinc finger transcription factor required for eye development, Disco, and the histone methyltransferase, Su(var)3–9, supports the hypothesis that DIP1 acts as a bridge between regulation of DNA transcription and RNA processing/transport/editing and suggests that this "bridge" is not exclusive to the Hox family of proteins (24, 25, 72–74). Indeed, overexpression of DIP1 in the eye-antennal disc alters the expression of homothorax and spalt major and generates an ectopic distalless expression domain, ultimately causing homeotic transformations of head structures (24). Global overexpression of DIP1 is embryonic lethal (24), also consistent with a general function in transcription regulation and/or RNA metabolism.

The growing data demonstrating interaction of DNA- and RNA-binding proteins suggest that Hox proteins may participate in more diverse functions than previously indicated. Identification of other Hox proteins capable of DIP1 interaction will be of considerable interest, since this interaction provides potential for significant functional differences within this highly homologous class of proteins. Such studies will refine our understanding of the mechanisms by which Hox proteins regulate the complex cell specification process.


   FOOTNOTES
 
* This work was supported by National Institutes of Health (NIH) Grant GM22441 (to K. S. M.) and Robert A. Welch Foundation Grant C-576. Spectroscopic facilities utilized were provided by the Keck Center for Computational Biology and the Lucille P. Markey Charitable Trust. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

The on-line version of this article (available at http://www.jbc.org) contains two additional figures. Back

{ddagger} These three authors contributed equally to this work. Back

§ Was supported in part by NIH Molecular Biophysics Training Grant 5T32-GM08208. Back

Present address: CytoGenix, Inc., 9881 S. Wilcrest, Houston, TX 77099. Back

|| Present address: UCSD Biology Student Affairs 038, 9500 Gilman Dr., La Jolla, CA 92093-0348. Back

** Present address: Scott Dept. of Urology, Baylor College of Medicine, 6560 Fannin, Suite 2100, Houston, TX 77030. Back

{ddagger}{ddagger} To whom correspondence should be addressed: Dept. of Biochemistry and Cell Biology, Rice University, 6100 S. Main Street, Houston, TX 77005. Tel.: 713-348-4871; Fax: 713-348-6149; E-mail: ksm{at}rice.edu.

1 The abbreviations used are: Exd, Extradenticle; Pbx, vertebrate homolog of Exd; Ubx, ultrabithorax; Ubx Ib, longest Ubx isoform; DIP1, Disconnected Interacting Protein 1; dsRBD, double-stranded RNA-binding domain; DIP1-c, longest DIP1 isoform; Disco, disconnected; Su(var)3–9, suppressor of variegation 3–9; RED1, RNA editase 1; BSA, bovine serum albumin; RNAP, RNA polymerase; GST, glutathione S-transferase; dsRNA, double-stranded RNA; PMSF, phenylmethylsulfonyl fluoride; DTT, dithiothreitol; X-gal, 5-bromo-4-chloro-3-indolyl-{beta}-D-galactopyranoside; HIV, human immunodeficiency virus. Back

2 The nucleotide sequences for the DIP1 isoforms have previously been deposited in the GenBankTM data base with the following accession numbers: AF175713 [GenBank] (DIP1-a), AF182154 [GenBank] (DIP1-b), AF218310 [GenBank] and AF175711 [GenBank] (DIP1-c), AY217028 [GenBank] (DIP1-d), AJ250866 [GenBank] (KLETT-a), and AJ250867 [GenBank] (KLETT-b). Back


   ACKNOWLEDGMENTS
 
We thank the following: Roger Brent for the D. melanogaster embryonic cDNA library, Philip Beachy for the Ubx vector, Yousif Shamoo for the T7 RNAP construct and for technical assistance, Javier Lopez for Ubx monoclonal antibodies, Michael Stern for the S880 fly line, Sean Carroll for the MS1096-Gal4 driver fly line, Goran Akusjarvi for the VA1 vector, and Ana Campos and colleagues for UAS-DIP1-c and Dp(1;Y) fly lines and for sharing data prior to publication. We thank Erin Engelhardt for assistance with Ubx activation assays in yeast. We also thank the members of the Matthews laboratory for scientific discussions and suggestions.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Biggin, M. D., and McGinnis, W. (1997) Development 124, 4425-4433[Abstract]
  2. Manak, J. R., and Scott, M. P. (1994) Development 120, (suppl.) 61-71
  3. Beachy, P. A., Krasnow, M. A., Gavis, E. R., and Hogness, D. S. (1988) Cell 55, 1069-1081[CrossRef][Medline] [Order article via Infotrieve]
  4. Vigano, M. A., Rocco, G. D., Zappavigna, V., and Mavilio, F. (1998) Mol. Cell. Biol. 18, 6201-6212[Abstract/Free Full Text]
  5. Weatherbee, S. D., Halder, G., Kim, J., Hudson, A., and Carroll, S. (1998) Genes Dev. 12, 1474-1482[Abstract/Free Full Text]
  6. Bondos, S. E., and Tan, X.-X. (2001) Crit. Rev. Eukaryotic Gene Expression 11, 145-171[Medline] [Order article via Infotrieve]
  7. Wolberger, C. (1998) Curr. Opin. Genet. Dev. 8, 552-559[CrossRef][Medline] [Order article via Infotrieve]
  8. Massagué, J., and Wotton, D. (2000) EMBO J. 19, 1745-1754[CrossRef][Medline] [Order article via Infotrieve]
  9. Piper, D. E., Batchelor, A. H., Chang, C.-P., Cleary, M. L., and Wolberger, C. (1999) Cell 96, 587-597[CrossRef][Medline] [Order article via Infotrieve]
  10. Chan, S.-K., Pöpperl, H., Krumlauf, R., and Mann, R. S. (1996) EMBO J. 15, 2476-2487[Medline] [Order article via Infotrieve]
  11. Beachy, P. A., Varkey, J., Young, K. E., von Kessler, D. P., Sun, B. I., and Ekker, S. C. (1993) Mol. Cell. Biol. 13, 6941-6956[Abstract/Free Full Text]
  12. van Dijk, M. A., and Murre, C. (1994) Cell 78, 617-624[CrossRef][Medline] [Order article via Infotrieve]
  13. Passner, J. M., Ryoo, H. D., Shen, L., Mann, R. S., and Aggarwal, A. K. (1999) Nature 397, 714-719[CrossRef][Medline] [Order article via Infotrieve]
  14. Zhu, A., and Kuziora, M. A. (1996) J. Biol. Chem. 271, 20993-20996[Abstract/Free Full Text]
  15. Shanmugam, K., Featherstone, M. A., and Saragovi, H. U. (1997) J. Biol. Chem. 272, 19081-19087[Abstract/Free Full Text]
  16. Lewis, E. B. (1978) Nature 276, 565-570[CrossRef][Medline] [Order article via Infotrieve]
  17. Beachy, P. A., Helfand, S. L., and Hogness, D. S. (1985) Nature 313, 545-551[CrossRef][Medline] [Order article via Infotrieve]
  18. Christen, B., and Bienz, M. (1992) Mech. Dev. 39, 73-80[CrossRef][Medline] [Order article via Infotrieve]
  19. Graba, Y., Aragnol, D., Laurenti, P., Garzino, V., Charmot, D., Berenger, H., and Pradel, J. (1992) EMBO J. 11, 3375-3384[Medline] [Order article via Infotrieve]
  20. Castelli-Gair, J., Greig, S., Micklem, G., and Akam, M. (1994) Development 120, 1983-1995[Abstract]
  21. Tan, X.-X., Bondos, S., Li, L., and Matthews, K. S. (2002) Biochemistry 41, 2774-2785[CrossRef][Medline] [Order article via Infotrieve]
  22. Ronshaugen, M., McGinnis, N., and McGinnis, W. (2002) Nature 415, 914-917[CrossRef][Medline] [Order article via Infotrieve]
  23. Galant, R., and Carroll, S. B. (2002) Nature 415, 910-913[CrossRef][Medline] [Order article via Infotrieve]
  24. De Sousa, D., Mukhopadhyay, M., Pelka, P., Zhao, X., Dey, B. K., Robert, V., Pélisson, A., Bucheton, A., and Campos, A. R. (2003) J. Biol. Chem. 278, 38040-38050[Abstract/Free Full Text]
  25. Krauss, V., O'Grady, M., Hoffmann, J., Schotta, G., Fischer, A., Kuhfittig, S., Dorn, R., Grigliatti, T. A., and Reuter, G. (2001) 17th European Drosophila Research Conference in Edinburgh (UK), F8, September 1–5, 2001, Edinburgh, UK
  26. Golemis, E. A., Serebriiskii, I., Finley, R. L., Jr., Kolonin, M. G., Gyuris, J., and Brent, R. (1999) Current Protocols in Molecular Biology, pp. 20.1.1-20.1.40, John Wiley & Sons
  27. Finley, R. L., Jr., Thomas, B. J., Zipursky, S. L., and Brent, R. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 3011-3015[Abstract/Free Full Text]
  28. Gietz, R. D., and Schiestl, R. H. (1991) Yeast 7, 253-263[CrossRef][Medline] [Order article via Infotrieve]
  29. Chen, D.-C., Yang, B.-C., and Kuo, T.-T. (1992) Curr. Genet. 21, 83-84[CrossRef][Medline] [Order article via Infotrieve]
  30. Hofmann, K., Bucher, P., Falquet, L., and Bairoch, A. (1999) Nucleic Acids Res. 27, 215-219[Abstract/Free Full Text]
  31. Altschul, S. F., Madden, T. L., Schäffer, A. A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D. J. (1997) Nucleic Acids Res. 25, 3389-3402[Abstract/Free Full Text]
  32. Bondos, S. E., and Bicknell, A. (2003) Anal. Biochem. 316, 223-231[CrossRef][Medline] [Order article via Infotrieve]
  33. Tsao, K.-L., DeBarbieri, B., Michel, H., and Waugh, D. S. (1996) Gene (Amst.) 169, 59-64[CrossRef][Medline] [Order article via Infotrieve]
  34. Lopez, A. J., and Hogness, D. S. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 9924-9928[Abstract/Free Full Text]
  35. Brand, A., and Perrimon, N. (1993) Development 118, 401-415[Abstract]
  36. Soehnge, H., Huang, X., Becker, M., Whitley, P., Conover, D., and Stern, M. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 13262-13267[Abstract/Free Full Text]
  37. Krovat, B. C., and Jantsch, M. F. (1996) J. Biol. Chem. 271, 28112-28119[Abstract/Free Full Text]
  38. Ho, C. K., and Shuman, S. (1996) Virology 217, 272-284[CrossRef][Medline] [Order article via Infotrieve]
  39. Sambrook, J., Fritsch, E. F., and Maniatis, F. (1989) Molecular Cloning: A Laboratory Manual, 2nd Ed., A.2.5-A.2.6, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY
  40. Moraitis, M. I., Xu, H., and Matthews, K. S. (2001) Biochemistry 40, 8109-8117[CrossRef][Medline] [Order article via Infotrieve]
  41. Li, L., von Kessler, D., Beachy, P. A., and Matthews, K. S. (1996) Biochemistry 35, 9832-9839[CrossRef][Medline] [Order article via Infotrieve]
  42. Li, L., and Matthews, K. S. (1997) Biochemistry 36, 7003-7011[CrossRef][Medline] [Order article via Infotrieve]
  43. Bass, B. L., Hurst, S. R., and Singer, J. D. (1994) Curr. Biol. 4, 301-314[CrossRef][Medline] [Order article via Infotrieve]
  44. Martinez-Laborda, A., Serrano-Cartagena, J., and Micol, J. L. (1996) Mol. Gen. Genet. 250, 540-546[Medline] [Order article via Infotrieve]
  45. Grenier, J. K., and Carroll, S. B. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 704-709[Abstract/Free Full Text]
  46. Shashidhara, L. S., Agrawal, N., Bajpai, R., Bharathi, V., and Sinha, P. (1999) Dev. Biol. 212, 491-502[CrossRef][Medline] [Order article via Infotrieve]
  47. Kallio, P. J., Okamoto, K., O'Brien, S., Carrero, P., Makino, Y., Tanaka, H., and Poellinger, L. (1998) EMBO J. 17, 6573-6586[CrossRef][Medline] [Order article via Infotrieve]
  48. Fry, D. C., Kuby, S. A., and Mildvan, A. S. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, 907-911[Abstract/Free Full Text]
  49. Rogers, S., Wells, R., and Rechsteiner, M. (1986) Science 234, 364-368[Abstract/Free Full Text]
  50. St. Johnston, D., Brown, N. H., Gall, J. G., and Jantsch, M. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 10979-10983[Abstract/Free Full Text]
  51. Patel, R. C., Stanton, P., McMillan, N. M. J., Williams, B. R. G., and Sen, G. C. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 8283-8287[Abstract/Free Full Text]
  52. Melcher, T., Maas, S., Herb, A., Sprengel, R., Seeburg, P. H., and Higuchi, M. (1996) Nature 379, 460-463[CrossRef][Medline] [Order article via Infotrieve]
  53. Wickham, L., Duchaîne, T., Luo, M., Nabi, I. R., and DesGroseillers, L. (1999) Mol. Cell. Biol. 19, 2220-2230[Abstract/Free Full Text]
  54. Fierro-Monti, I., and Mathews, M. B. (2000) Trends Biochem. Sci. 25, 241-246[CrossRef][Medline] [Order article via Infotrieve]
  55. Green, S. R., Manche, L., and Mathews, M. B. (1995) Mol. Cell. Biol. 15, 358-364[Abstract]
  56. Tian, B., and Mathews, M. B. (2001) J. Biol. Chem. 276, 9936-9944[Abstract/Free Full Text]
  57. Ekker, S. C., Young, K. E., von Kessler, D. P., and Beachy, P. A. (1991) EMBO J. 10, 1179-1186[Medline] [Order article via Infotrieve]
  58. Laughon, A. (1991) Biochemistry 30, 11357-11367[CrossRef][Medline] [Order article via Infotrieve]
  59. McGinnis, W., and Krumlauf, R. (1992) Cell 68, 283-302[CrossRef][Medline] [Order article via Infotrieve]
  60. Li, X., and McGinnis, W. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 6802-6807[Abstract/Free Full Text]
  61. Graba, Y., Aragnol, D., and Pradel, J. (1997) BioEssays 19, 379-388[CrossRef][Medline] [Order article via Infotrieve]
  62. Chariot, A., Gielen, J., Merville, M.-P., and Bours, V. (1999) Biochem. Pharmacol. 58, 1851-1857[CrossRef][Medline] [Order article via Infotrieve]
  63. Kim, H.-H., Tharayil, M., and Rudd, C. E. (1998) J. Biol. Chem. 273, 296-301[Abstract/Free Full Text]
  64. Venot, C., Maratrat, M., Dureuil, C., Conseiller, E., Bracco, L., and Debussche, L. (1998) EMBO J. 17, 4668-4679[CrossRef][Medline] [Order article via Infotrieve]
  65. Li, X., Murre, C., and McGinnis, W. (1999) EMBO J. 18, 198-211[CrossRef][Medline] [Order article via Infotrieve]
  66. Galant, R., Walsh, C. M., and Carroll, S. B. (2002) Development 129, 3115-3126[Abstract/Free Full Text]
  67. Xing, Y., GuhaThakurta, D., and Draper, D. E. (1997) Nat. Struct. Biol. 4, 24-27[CrossRef][Medline] [Order article via Infotrieve]
  68. Dubnau, J., and Struhl, G. (1996) Nature 379, 694-699[CrossRef][Medline] [Order article via Infotrieve]
  69. Rivera-Pomar, R., Neissing, D., Schmidt-Ott, U., Gehring, W. J., and Jäckle, H. (1996) Nature 379, 746-749[CrossRef][Medline] [Order article via Infotrieve]
  70. Cassiday, L. A., and Maher, L. J., III. (2002) Nucleic Acids Res. 30, 4118-4126[Abstract/Free Full Text]
  71. Wiellette, E. L., Harding, K. W., Mace, K. A., Ronshaugen, M. R., Wang, F. Y., and McGinnis, W. (1999) Development 126, 5373-5385[Abstract]
  72. Steller, H., Fischbach, K. F., and Rubin, G. M. (1987) Cell 50, 1139-1153[CrossRef][Medline] [Order article via Infotrieve]
  73. Campos, A. R., Lee, K. J., and Steller, H. (1995) J. Neurobiol. 28, 313-329[CrossRef][Medline] [Order article via Infotrieve]
  74. Tschiersch, B., Hofmann, A., Krauss, V., Dorn, R., Korge, G., and Reuter, G. (1994) EMBO J. 13, 3822-3831[Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
Y. Liu, K. S. Matthews, and S. E. Bondos
Multiple Intrinsically Disordered Sequences Alter DNA Binding by the Homeodomain of the Drosophila Hox Protein Ultrabithorax
J. Biol. Chem., July 25, 2008; 283(30): 20874 - 20887.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
279/25/26433    most recent
M312842200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bondos, S. E.
Right arrow Articles by Matthews, K. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bondos, S. E.
Right arrow Articles by Matthews, K. S.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2004 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement