IntroductionRNA-protein interactions are central to cellular metabolism,and they are likewise essential for viruses, most obviously forthose replicating through RNA intermediates. Hepatitis B virus(HBV), the causative agent of B-type hepatitis in humans(Blumberg, 1997), is a small enveloped virus whose 3.2 kbDNA genome is generated by reverse transcription. Thetemplate is a viral transcript called pregenomic RNA (pgRNA)that also serves as mRNA for the virus capsid protein and thereverse transcriptase. Binding of the enzyme to a 5′-proximalstem-loop structure, ε, on the RNA mediates its selectivepackaging into capsids and initiation of DNA synthesis(Nassal, 1999; Nassal, 2000). Previous evidence for cellularproteins that bind to HBV RNA near the ε signal (Perri andGanem, 1996; Perri and Ganem, 1997) or to the similarlystructured HIV-1 trans-activation response (TAR) element, forexample, the TAR RNA-binding protein TRBP (Daher et al.,2001; Gatignol et al., 1991), prompted us to screen, usingNorthwestern assays, a human liver cDNA expression libraryfor proteins that bind to an HBV-ε-containing RNA probe.

A 3′ terminally truncated 2.1 kb cDNA clone obtained bythis procedure was completed by fusion to an overlappingclone identified by molecular hybridization. This assembledsequence contained a continuous large open reading frame(ORF) for a 1208 amino acid protein with a calculated

molecular mass of 138 kDa. On the basis that it was isolatedas an RNA-binding protein and has ubiquitin-protein ligase(E3) activity (see below), it was termed hRUL138 (humanRNA-binding ubiquitin ligase of 138kDa). The relevance ofthe assembled clone was independently confirmed by thepublication, during the course of this study, of a continuouscDNA in the HUGE (human unidentified gene-encodedlarge proteins) collection (Kikuno et al., 2000), KIAA0675(Ishikawa et al., 1998), that encodes an identical protein. Thecorresponding gene is located on chromosome 3 but thefunction of its predicted gene product is unknown.

Database searches using the full-length sequence confirmedthat hRUL138 represents a hitherto unknown human proteinwith no homologs in the completely sequenced genomes ofS. cerevisiae, C. eleganor A. thaliana; however, highly relatedmouse cDNAs were recently identified (S. G. Kreft, PhDthesis, University of Heidelberg, 2000). No known RNA-binding motif (Perez-Canadillas and Varani, 2001) wasdetectable. Motif searches indicated a central coiled-coildomain and the potential presence of one to threetransmembrane regions; however, all had a weak score. Theonly really informative similarity to other known proteinsinvolved a short sequence stretch close to the C terminus ofhRUL138 that contains all of the essential Cys- and His-residues of an intact RING (‘really interesting new gene’)

605

Cellular as well as viral RNAs are usually found complexedwith proteins. In an attempt to identify proteins thatinteract with transcripts of hepatitis B virus (HBV), a DNAvirus that replicates through reverse transcription, apartial cDNA was isolated from a human cDNA expressionlibrary whose gene product bound to an HBV-derivedRNA. Using an overlapping clone from a molecularhybridization screen a full-length cDNA was assembled. Itcontained a large open reading frame for a 1208 amino-acidprotein of 138 kDa identical to the hypothetical product ofthe KIAA0675 clone. Closely related sequences are presentin mouse cDNA libraries but not in the genomes of lowerorganisms. The protein sequence contained no knownRNA-binding domain and, apart from a probable coiled-coil domain, the only significant homology involved acomplete RING-H2 motif. This suggested that the proteinmight be a novel RNA-binding RING-dependent ubiquitin-protein ligase or E3 enzyme. A motif critical for RNA

binding was experimentally mapped to a central Lys-richregion. Binding specificity is either broad or the protein hasas yet unknown physiological targets; hence, at present, apotential importance for HBV biology remains open. TheRING-H2 domain was functional in and essential for self-and trans-ubiquitylation in vitro and for proteasome-mediated turnover of the protein in vivo. We thereforetermed it hRUL138 for human RNA-binding ubiquitinligase of 138 kDa. hRUL138 mRNAs are expressed at lowlevels in most tissues. GFP-tagged hRUL138 derivativeswere found associated with cytoplasmic structures, possiblythe ER, but excluded from the nucleus. The combinedpresence of RNA binding and E3 activity in hRUL138raises the possibility that both are mechanistically linked.

Key words: E3 enzyme, Ubiquitin ligase, Ubiquitylation, RINGfinger

Summary

hRUL138, a novel human RNA-binding RING-H2ubiquitin-protein ligaseStefan G. Kreft and Michael Nassal*University Hospital Freiburg, Department of Internal Medicine II, Molecular Biology, Hugstetter Str. 55, D-79106 Freiburg, Germany*Author for correspondence (e-mail: nassal2@ukl.uni-freiburg.de)

Accepted 6 November 2002Journal of Cell Science 116, 605-616 © 2003 The Company of Biologists Ltddoi:10.1242/jcs.00261

Research Article

606

domain (Borden and Freemont, 1996) of the RING-H2type [consensus: CX2CX9-39CX1-3HX2-3H/CX2CX4-48CX2C;C=Cys, H=His, X=any aa; H/C=His in RING-H2 and Cys inRING-HC fingers (Borden, 2000)]. The recognizable featureswithin the hRUL138 protein are schematically shown in Fig.1B (see Results for further details).

RING domains are Zn2+-binding structures that were firstrecognized for their ability to mediate protein-proteininteractions (Saurin et al., 1996). Recently, they have beenshown to be integral parts of a second major class, in additionto HECT-domain proteins, of ubiquitin ligases (or E3 enzymes)in which the RING is essential for function (Borden, 2000;Freemont, 2000; Lorick et al., 1999). Ubiquitylation is a keyevent in proteasome-dependent protein degradation (Pickart,2000; Pickart and VanDemark, 2000); in addition it is

important in many other processes that do not necessarilyinvolve proteolysis (Pickart, 2001) such as signal transduction(Wang et al., 2001), endocytosis (Hicke, 2001) and virusbudding (Vogt, 2000). The mechanism of ubiquitylation(Pickart, 2001) involves energy-dependent activation of the 76amino acid ubiquitin protein with an ubiquitin-activating (orE1) enzyme and its transfer to a ubiquitin-conjugating (or E2)enzyme. Covalent ubiquitin attachment to the ε-amino groupof lysines, or sometimes to the N terminus, of target proteinsis mediated by E3 enzymes; further ubiquitins are usually,but not always, attached to the first one, leading topolyubiquitylation. E3s are the main determinant for targetspecificity; consistently, there is a greater variety of E3 than E2enzymes, whereas only one E1 is known in most eukaryotesincluding humans.

Together these considerations suggested that hRUL138might be a novel RING-H2 E3 that can bind RNA. Thefunctional analysis of hRUL138 reported here fully supportsthis notion.

Materials and MethodsIdentification of hRUL138 cDNAsThe initial cDNA clone was obtained by Northwestern screening of ahuman liver cDNA expression library in lambda Uni-ZAP XR phages(Stratagene) using an in-vitro-transcribed digoxygenin (DIG)-labeledRNA probe comprising the first 184 nt of HBV (subtype ayw) pgRNA,including the 60 nt ε signal. Phages were plated on E. coli XL1-BlueMRF′ cells and screened essentially as previously described (Gatignolet al., 1991), except that positive plaques were identified using an anti-DIG antibody (DIG Nucleic Acid detection kit, Roche). After plaquepurification a pBluescript SK(–) plasmid form of the 2.1 kb cDNAwas derived by excision of the cDNA insert from the lambda phage;the plasmid was termed pBls-NIII. Five additional human cDNAlibraries were screened using a 32P-labeled DNA probe encompassingnt 1497-1998 of the hRUL138 ORF (performed at RZPD screeningservice, German Cancer Research Center, Heidelberg, Germany;URL: http://www.rzpd.de). Four positive clones were obtained, asconfirmed by sequencing: ICRFp512 H14173Q2 and B2269Q2 (bothfetal liver); DKFZ586 C1754Q3 (uterus); and IMAGp998 N0865Q6(derived from the same Stratagene liver cDNA library as NIII). CloneICRFp512 H14173Q2 comprised an apparently intact 3′ end and wasused to reconstitute the full-length hRUL138 ORF. The generalstructures of the individual cDNAs are outlined in Fig. 1A.

Plasmid constructsA full-length hRUL138 cDNA was assembled from the cDNAfragments contained in plasmids pBlsNIII and ICRFp512H14173Q2using an AvrII restriction site located in the overlap region (nt position1704 of the hRUL138 ORF). Both fragments were cloned into thevector pT7AMVpol16 (Weber et al., 1994). The resulting plasmid,pT7-hRUL138, comprises the complete hRUL138 ORF under controlof the bacteriophage T7 promoter. It also served as starting materialfor various hRUL138 derivatives. For expression in E. coli, hRUL138or wildtype and mutant fragments thereof were fused to the maltose-binding protein (MBP) using the pMal-c2 vector (New EnglandBiolabs); His-tagged variants were obtained using the pET-30a(+)vector (Novagen). In some constructs a FLAG epitope sequence(DYKDDDDK) was inserted using oligonucleotide-mediatedmutagenesis. Point mutations in the Lys-rich motif (amino acids 662to 666 KKKTK changed to SGSTA) and in the RING-H2 domain(Cys1187 changed to Ser, i.e. C1187S) were generated by PCR-mediated mutagenesis. Eukaryotic hRUL138 expression vectors werebased on plasmid pTR-UF5 (Zolotukhin et al., 1996), which encodes

Journal of Cell Science 116 (4)

1 1208

120 140 390 410 1000 1020656 674 1148 1187

794 852 TM?TM?TM?

coiled coil

K-rich region RING-H2

CVI CHENLSPENLSVLPCAHKFHAQCI RPWLMQQGTCPTC

B

hRUL138 ORF (3624 nt)

IMAGp 998

DKFZp586

ICRFp512

ICRFp512

NIII

(A)n

∆14 bp

∆14 bp

∆140 bp

***

*

1497 1998

A 1 3624

1998

1667

905

301

816

2200

H14173Q2

C1754Q3

N0865Q6

B2269Q2

Fig. 1. Schematic representation of the hRUL138 ORF and protein.(A) cDNAs identified in this study. The line on the top represents thecomplete hRUL138 ORF, the bars below indicate cDNA sequences.Numbers are nt positions relative to the hRUL138 ORF. Non-translated regions are depicted by hatched bars and regions encodingidentical amino-acid sequences as in hRUL138 by dark grey bars.Light grey stippled bars correspond to identical nucleotide sequenceswhich are, however, out of frame owing to small, splicing-relateddeletions (shown as n) that lead to early translational stops (*); (A)ndenotes a poly-A tail. White bars indicate introns that have not beenspliced out in the corresponding cDNAs. NIII is the cDNA initiallyisolated by Northwestern screening, the other cDNAs were identifiedby molecular hybridization using an NIII derived DNA probe (blackline above NIII). (B) Predicted motifs within the hRUL138 primarysequence. Numbers are amino acid positions. The lysine-rich (K-rich) motif and the RING-H2 domain are indicated by black boxes;the RING-H2 sequence is shown below with the conserved Cys andHis residues highlighted in bold face. The prediction for the coiled-coil motif is strong between amino acids 794 to 852 and extends withweaker scores as indicated by the hatched bars. The weakly andvariably predicted putative transmembrane regions are denoted byTM?

607RNA-binding ubiquitin-protein ligase hRUL138

a codon-optimized version of the enhanced green fluorescent protein(eGFP) under control of the CMV-IE enhancer/promoter and an SV40intron plus poly-adenylation signal. Appropriate hRUL138-encodingfragments were inserted so as to generate C-terminal fusions witheGFP.

All plasmids were named according to the following scheme: thefirst letters indicate the parental plasmid (pT7, pMBP, pET, pTR-UF);the numbers following the term RUL give the amino-acid positionsof the full-length hRUL138 present in the construct; amino acidsubstitutions are indicated in parentheses and the presence of tags byan acronym either in front of (N-terminal tags) or after (C-terminaltags) the RUL term. For instance, pMBP-RUL681-1128-FLAG standsfor a plasmid encoding hRUL138 amino acids 681 to 1128 with a N-terminal MBP and a C terminal FLAG tag; the same names withoutthe plasmid-specific term denote the corresponding proteins. As amarker for the endoplasmic reticulum (ER), the dsRed2gene fromplasmid pdsRed2 (Clontech) was used to replace the eYFPgene inpEYFP-ER (Clontech), thus maintaining the original calreticulin ER-targeting and KDEL ER retrieval signals. The corresponding plasmidwas named pRFP2-ER. Detailed outlines of all cloning procedures areavailable from the authors upon request.

In vitro translationIn vitro translations were performed in rabbit reticulocyte lysateusing the TNT T7 Quick Coupled Transcription/Translation system(Promega) programmed with appropriately linearized pT7 plasmids.For 35S-labeling, [35S]methionine (specific activity 1,000 Ci/mmol;Amersham/Pharmacia) was used at 25 µCi per 50 µl reaction.

Expression of recombinant proteinsMBP fusion proteins were expressed in E. coli strain Top10(Invitrogen), pET constructs in E. coli BL21(DE3) cells (Novagen).MBP-tagged proteins were enriched by using amylose resin(New England Biolabs) and His-tagged variants by using Ni2+

nitrilotriacetate (Ni-NTA) agarose (Qiagen). For co-expression,BL21(DE3) cells were double-transformed with ampicillin-resistance-mediating pMBP and kanamycin-resistance-mediatingpET vectors and propagated in the presence of both antibiotics.Protein concentrations were estimated by comparing, on Coomassieblue stained SDS-PAGE gels, the corresponding band intensities withthose of serial dilutions of a bovine serum albumin (BSA) standardof known concentration.

Cell culture and DNA transfectionHuh7 human hepatoma cells were maintained in Dulbecco’s modifiedEagle medium (DMEM) supplemented with 10% fetal bovine serumand antibiotics (Nassal, 1992). DNA transfections were carried outusing FuGENE 6 transfection reagent (Roche) as suggested by themanufacturer.

Northern blot analysisA human multiple tissue Northern blot (Clontech) was hybridizedwith a [32P]dATP random primed probe (High Prime DNA labelingkit, Roche) corresponding to nt 1497-1998 of the hRUL138 ORF ora full-length actin control probe (Clontech) using standard conditions;hybridizing bands were visualized on a BAS-1500 Phosphoimager(Fuji).

Western blot analysesSodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE)-resolved proteins were transferred onto PVDF membranes(Amersham/Pharmacia) and incubated with the appropriate primary

antibody followed by horseradish-peroxidase-conjugated secondaryantibodies (Dianova) as described previously (König et al., 1998).Immunoreactive bands were visualized using the ECL-plus system(Amersham/Pharmacia). The following antibodies were used: anti-FLAG monoclonal antibody M2 (Sigma); anti-His monoclonalantibody anti-Tetra-His (Qiagen) and a polyclonal rabbit antiserumrecognizing the His-tag-containing linker sequence encoded by thepET-30 vector.

RNA-binding assaysDIG-labeled RNA probes were obtained by in vitro transcription (DIGRNA labeling kit; Roche) of an appropriately linearized T7 promotercontaining plasmids as suggested by the manufacturer. The parentalplasmid for the HBV-RT and AMV RNA probes was pT7HAMV-RT(provided by J. Beck) which contains HBV nucleotides 938-1130, thatis, part of the reverse transcriptase ORF, behind the 49 nt AMV leader.The HBV-RT probe consisted of both segments, the AMV probe ofonly the AMV part. Labeled RNA was purified by gel filtration (G-25 Sephadex Quick Spin columns, Roche) and quantified via dotblotting (DIG Nucleic Acid detection Kit, Roche). 32P-labeled RNAswere obtained by in vitro transcription in the presence of 32P-CTP(specific activity: 800 Ci/mmol; Amersham/Pharmacia) andquantified by Cerenkov scintillation counting.

Northwestern blotting using DIG-labeled RNA probes wasperformed essentially as described previously (Gatignol et al., 1991).Briefly, crude bacterial lysate (equivalent to 75 µl bacterial culture)or partially purified bacterially expressed proteins (100 ng to 2 µg)were resolved by SDS-PAGE and blotted onto a PVDF membrane.Membranes were sequentially immersed, at room temperature, inbinding buffer 1 (BB1; 50 mM Tris-HCl, pH 7.5, 50 mM NaCl, 1 mMEDTA, 1 mM dithiothreitol) containing decreasing concentrations ofguanidine-HCl (5 minutes each in 6 M, 3 M, 1.5 M, 0.75 M, 0.37 Mand 0.18 M). Thereafter the membranes were rinsed in BB1, blockedfor 1 hour in 2.5% skim milk powder in BB1 and rinsed twice in BB1;in some experiments an additional wash with 1 mg/ml of heparin inBB1 was included. RNA binding was carried out in BB1 containing50-100 ng/ml of the corresponding DIG-labeled RNA probe and 10µg/ml each of yeast RNA and herring sperm DNA for 1 hour at roomtemperature. After three washes in BB1, bound DIG-labeled RNA wasdetected via chemiluminescence (DIG Nucleic Acid detection kit;Roche) on X-ray film.

For in-solution RNA-binding assays, approximately 2 µg (about 40pmol; 400 nM) of the corresponding bacterially expressed His-taggedhRUL138 fragment, enriched by Ni-NTA agarose chromatographyand dialysed against 20 mM Tris-HCl (pH 7.5), was incubated with2×106 cpm HBV-ε-containing RNA [6×107 cpm/pmol; 40 fmol (300pM) in binding buffer 2 (BB2; 50 mM Tris-HCl pH 7.5, 350 mMNaCl, 10 mM β-mercaptoethanol, 10 µg/ml yeast tRNA)] containing0.4 U/µl RNasin (Promega) in a total volume of 100 µl for 1 hour at30°C. RNP complexes were separated from free RNA by adding tothe samples 40 µl (gel bed) Ni-NTA agarose in 400 µl BB2 containing20 mM imidazol (pH 7.9), followed by a 30 minute incubation at 4°Cwith gentle agitation. After four washes with 1 ml each of cold BB2containing 20 mM imidazol-bound 32P-labeled RNAs were quantifiedby Cerenkov counting.

In vitro ubiquitylation assaysFor self-ubiquitylation assays, hRUL138 and its variants were in vitrotranslated in the presence of [35S]methionine. 2 µl of the reaction mixwas then incubated in a total volume of 50 µl of ubiquitylation buffer(UB; 25 mM Tris-HCl, pH 7.5, 50 mM NaCl, 2 mM dithiothreitol,5.6 mM MgCl2, 4 mM ATP) with 4 µg ubiquitin (Sigma) ormethylated ubiquitin (Calbiochem) and approximately 10 ng of E1and 100 ng of the respective E2 (all kindly provided by MartinScheffner). As E1-purified His-tagged wheat germ E1 expressed in

608

insect cells was used, lysates from E. coli cells overexpressing eitherUbcH5 or UbcH7 served as a source for E2 (Nuber et al., 1996). Fortrans-ubiquitylation assays E. coli lysates containing co-expressedMBP-RUL681-1208-FLAG and His-RUL681-1128-His were passedthrough amylose resin and the proteins bound to 6 µl of amylose gelwere incubated in ubiquitylation buffer as described above, except that10 µg of methylated ubiquitin were used. His-RUL681-1128-His-onlyubiquitylation reactions were performed with crude bacterial lysates.Reactions were allowed to proceed for 90 minutes at 30°C and wereterminated by boiling for 5 minutes in SDS sample buffer. Sampleswere analyzed by SDS-PAGE and autoradiography (when 35S labeled)or by western blotting.

Proteasome inhibitionHuh7 cells were transfected with expression constructs for GFP- orFLAG-tagged hRUL138 derivatives. 62 hours post-transfection, thecells were incubated with a final concentration of 50 µM MG132(Calbiochem) in DMSO for 6 hours. Thereafter, they were washedtwice with PBS, scraped off the plate and lysed in SDS sample buffer.Equal aliquots from the lysates were analyzed by western blottingusing GFP- or FLAG-specific antibodies. Lysates from nontransfecteduntreated cells and from cells incubated with the same concentrationof DMSO only served as controls.

ImmunofluorescenceTransfected Huh7 cells were grown on glass coverslips. 48 hours post-transfection GFP and/or RFP derivatives were examined directly witha confocal laser scanning microscope (Zeiss LSM410). For nuclearcounter-staining with TOTO-3 iodide (Molecular Probes) the cellswere fixed with 2% paraformaldehyde in phosphate-buffered saline(PBS), permeabilized for 10 minutes using 0.05% saponin in PBS andincubated for 1 hour with a mix of two monoclonal anti-GFPantibodies (Roche) at a 1:1000 dilution in PBS containing 3% bovineserum albumin and 0.05% saponin. After washing and incubation withAlexa-488-conjugated goat-anti-mouse IgG (Molecular Probes) for 1hour, the cells were incubated with PBS containing 200 µg/ml ofRNase A for 30 minutes, washed with PBS containing 0.05% saponinfor 10 minutes and incubated with 1 µM TOTO-3 iodide in PBS. Afterrinsing with PBS, the coverslips were mounted with Vectashield H-1000 (Vector Laboratories). Image files were processed using AdobePhotoshop 5.5 software.

ResultsIdentification of human RUL138 cDNAsAs HBV is a liver-specific virus we initially screened a humanliver cDNA expression library for HBV RNA-binding proteins.Out of about 100,000 recombinant λ phage plaques (of whichone third would be expected to yield in-frame translationproducts) one gave a clearly positive signal when incubatedwith the HBV-ε-containing RNA probe in the presence of anabout 100-fold excess of yeast tRNA and herring sperm DNA.Its cDNA insert comprised 2166 nt and contained a 168 nt 5′-untranslated region followed by one large ORF extending tothe very 3′ end, without a stop codon or a poly-adenylationsignal or sequence. Screening of additional human cDNAlibraries, some derived from other tissues, by molecularhybridization with a DNA probe corresponding to nt positions1497 to 1998 of the hRUL138 ORF yielded four independentrelated clones (see schematic representation in Fig. 1A). All ofthese shared extensive sequence homologies with the originalcDNA; on the basis of the recently published draft sequence ofthe human genome, the gaps and/or insertions in some of the

cDNAs can all be explained by alternative or incompletesplicing events. Two clones, for instance, had a 14 nt deletionafter position 905, causing a frameshift and generating atranslational stop after 303 amino acids. Since both wereisolated from independent libraries they are probably derivedfrom authentic mRNAs, implying the existence of a shortform of hRUL138. One cDNA (ICRFp512 H14173Q2) thatoverlapped with the original clone by 332 nt had all the featuresof a complete 3′ end. It was used to assemble a continuoussequence of 4590 bp containing an ORF of 3624 nt encodinga 138 kDa protein identical to the hypothetical product of theKIAA0675 clone [GenBank Accesssion Number AB014575(Ishikawa et al., 1998)].

A tBLASTn search (NCBI; URL: http://www.ncbi.nlm.nih.gov:80/BLAST/) with the full-length hRUL138 proteinsequence in the H. sapienstranslated RNA database showed,besides the KIAA0675 product, only two hypothetical ORFswith extended similarity, corresponding to amino acids 128-339 and the region from amino acid 740 to near the C-terminus.One is the predicted tetratricopeptide repeat protein 3 (or TPRprotein D, TPRD; GenBank Accession Number D83077) onchromosome 21 (Ohira et al., 1996), the other a similar,hypothetical gene on the X chromosome (GenBank AccessionNumber XM_032867). The function of neither of the twoproteins is known, and the region of similarity between TPRDand hRUL138 does not involve the four TPR repeats (Smalland Peeters, 2000) that gave TPRD its name. The central regionof hRUL138, approximately from amino acids 600 to 950,showed some similarity to a variety of proteins containingmyosin heavy chain tail-like potential coiled-coil sequences(Lupas, 1996); the coiled-coil prediction is strongest for thesequence between position 794 to 852, extending with weakerscores in both directions from there (T. Doerks and P. Bork,personal communication). Different search programs fortransmembrane regions (PHDhtm, SOSUI, TMAP, TMHMM,TMpred) predicted one, two or three transmembrane helicesaround amino acids 120 to 140, 390 to 410, and 1001 to 1019;however, all had a low score that varied from program toprogram.

Although these sequence analyses did not offer any distinctclues to the potential function of hRUL138, the C-terminalRING-H2 domain contained all of the absolutely conservedCys- and His-residues that are typical for this motif, whichoccurs as a module in various proteins of otherwise unrelatedprimary sequence (Borden, 2000); functionally, however, thereis accumulating evidence that most RING finger proteins maybe protein-ubiquitin ligases (Freemont, 2000). We thereforewent on to experimentally address the RNA-binding potential,ubiquitin ligase activity, expression profile and intracellularlocalization of hRUL138.

A lysine-rich region in the center of hRUL138 is criticalfor RNA bindingThe hRUL138 sequence did not contain a known RNA-binding motif; however, because the truncated proteinencoded by the initial cDNA had been identified via its RNA-binding ability, the region responsible had to be present withinthe first 666 amino acids. Deletion mapping combined withNorthwestern blotting was used to further define this region.A candidate was a stretch of basic residues (amino acid

Journal of Cell Science 116 (4)

609RNA-binding ubiquitin-protein ligase hRUL138

positions 656 to 674, with 11 Lys plus 1 Arg residue within atotal of 19 amino acids; see Fig. 2) that remotely resembledthe Arg-rich RNA-binding part within the HIV-1 Tat protein(Cheng et al., 2001; Smith et al., 2000). In a first series ofexperiments hRUL138 fragments corresponding to residues 1to 205, 1 to 499 and 1 to 666 were expressed in E. coli asmaltose-binding protein (MBP) fusions. Crude bacteriallysates were subjected to Northwestern blotting with the HBV-ε-containing RNA probe in the presence of 10 µg/ml of yeastRNA and herring sperm DNA as nonspecific competitors.Specific signals were obtained only with longest protein (datanot shown). As this included, at its C-terminus, part of thesuspected basic region, three variants lacking 3, 6 or 9 aminoacids from the C-terminus were analysed in parallel to the 1to 666 derivative (Fig. 2A). Again, only the latter generatedsignals in the Northwestern blot, one at the 120 kDa positionexpected for the intact fusion protein and a second band atabout 50 kDa, most probably a cleavage product lacking theMBP part. Total protein loading on all lanes was similar, asshown by Coomassie blue staining of the gel used to generatethe blot (right panel in Fig. 2A); this also excluded thepossibility that the signals arose from RNA binding to E. coliproteins. Hence amino acids 664 to 666 (KTK) are importantfor RNA binding.

RNA binding was independent of the C-terminaldisposition of the basic motif and of the fusion to MBP, asshown by the variant K4–. In this pET30-vector-derived His-tagged construct, the four central Lys residues betweenpositions 662 to 666 were replaced by neutral amino acids

(KKKTK>SGSTA) in the context of hRUL138 amino acids510 to 878; hence it still contained the rest of the Lys-richmotif including four further lysines, that is, the total numberof basic residues was maintained. The corresponding wild-type construct, and an additional one coding for a C-proximalhRUL138 fragment of similar size (amino acids 832 to 1176)served as controls. The E. coli-expressed proteins wereenriched by Ni-NTA agarose chromatography and analyzedby Northwestern blotting (Fig. 2B, left panel). No signalswere observed for the C-proximal fragment and for themutated 510 to 878 protein; by contrast, a series of bands wasgenerated with the wild-type fragment 510-878; the largestwas present at the 50 kDa position expected for the intactfusion protein. Reprobing the same blot with an antibodyagainst the N-terminal His-tag (Fig. 2B, right) revealedsimilar amounts of the two 510 to 878 proteins plus severalsmaller, probably proteolytically derived products of similarmobility to the faster migrating bands on the Northwesternblot. The nonbinding C-proximal fragment was present inmuch larger amounts, indicating that the Northwesternsignals did not originate from unspecific binding of the probe.In some experiments, and upon long exposure of theautoradiograms, weak signals were also observed with theK4– mutant 510 to 878 protein. Hence mutating the fourcentral Lys-residues in the Lys-rich motif led to a drasticreduction, though possibly not complete abolishment, ofRNA binding.

This was further corroborated by in-solution RNA-bindingassays. The wild-type and the mutated 510 to 878 fragments

Fig. 2. Mapping of the RNA-binding region in hRUL138. (A) Deletion mapping. Crude lysates from bacteria expressing MBP-fused hRUL138fragments truncated at the indicated positions (scheme above the gels; stippled bars outline hRUL138 parts not present) were tested for bindingto a DIG-labeled HBV-ε-containing RNA probe by Northwestern blotting. Bound RNA was visualized using anti-DIG antibody followed bychemiluminescent detection (left panel). The band at about 50 kDa is most likely a proteolytic hRUL138 fragment. Equal loading wasconfirmed by Coomassie blue staining of the SDS-PAGE gel (right panel) used to generate the blot. (B) Mutagenesis of the K-rich motif.Internal hRUL fragments amino acids 510 to 878 containing the authentic K-rich motif or the variant K4– (amino acids 662 to 666=SGSTA)and amino acids 832 to 1176 were expressed in E. colias fusions with the pET30 encoded linker, and processed as in A. For Northwesternanalysis the blot was incubated with DIG-labeled HBV RNA (left panel); the bands below the 46 kDa marker position are probably degradationproducts. To control for loading, the same blot was reprobed with an antiserum recognizing the His-tag containing pET30 linker. (C) In-solutionRNA binding. The same proteins as in B, except that a fragment comprising amino acids 1 to 205 was used as negative control, were incubatedwith 32P-labeled HBV ε RNA, immobilized on Ni-NTA beads, and the radioactivity remaining on the beads was determined by scintillationcounting. Results are shown as a percentage of the cpm measured for the authentic 510-878 fragment.

610

were expressed as fusions with the pET30 linker; an N-proximal hRUL138 fragment (amino acids 1 to 205) fusedin the same way served as a control. The proteins wereincubated with 32P-labeled HBV RNA in the presence of 10µg/ml of yeast tRNA and 350 mM NaCl, bound to Ni-NTAagarose, and after several washing steps the radioactivityremaining on the beads was determined by liquid scintillation

counting; the presence of similar amounts of immobilizedprotein was confirmed by SDS-PAGE (data not shown).Compared to the wild-type protein, about 10-fold less RNAwas retained by the K4– variant and about 20-fold less by the1-205 fragment (Fig. 2C). These data confirmed that, also insolution, the Lys-rich motif is crucially involved in RNAbinding.

In order to test whether hRUL138 binds only to HBV-ε-containing RNA, similar Northwestern experiments wereperformed using other RNA probes. These included sequencesfrom another part of the HBV genome (HBV-RT, without theε stem-loop) and an A-U-rich unstructured RNA derived fromthe alfalfa mosaic virus (AMV) leader (Weber et al., 1994). OnNorthwestern blots loaded with the same wild-type and K4–

mutant 510-878 proteins as in Fig. 2, all RNAs gave signals,with the wild-type protein, of similar intensity to the originalprobe (Fig. 3). These data did therefore not reveal a significantpreference of hRUL138 for certain RNA sequences orstructures. Additional in-solution experiments (data notshown), in which RNA homopolymers were used ascompetitors, showed that poly-A and poly-U, at 10 µg/ml,reduced binding of the labeled HBV ε probe to theimmobilized wild-type 510 to 878 fragment by about 80%,whereas 100 µg/ml of poly-G and poly-C were required toachieve a similar reduction; double-stranded poly-IC wasineffective even at this concentration. These data showed thathRUL does not indiscriminately bind to all RNAs but, becausethe in vitro assays may not faithfully mimic physiologicalconditions, it remains an open question whether RNA bindingspecificity is naturally broad or whether specific natural targetsexist that we currently do not know.

Journal of Cell Science 116 (4)

Fig. 3.RNA-binding specificity of hRUL138. The same wild-typeand K4– variant hRUL138 fragments as in Fig. 2 were expressed aspET30 linker fusions and processed as before for Northwesternblotting, except that RNA probes derived from a different part of theHBV genome (panel HBV-RT) or an unrelated RNA derived fromthe alfalfa mosaic virus leader RNA (panel AMV) were used asprobes. Loading was controlled by reprobing the blots with an anti-His antibody. Note that weak but detectable signals were alsoobserved in the Northwestern blot lanes containing the K4– mutantproteins.

Fig. 4. Self-ubiquitylation of hRUL138. (A) Polyubiquitylation of full-length hRUL138. hRUL138 was in vitro translated in the presence of35S-Met and incubated with ATP and E1 plus ubiquitin (Ub) and UbcH5 or UbcH7 as indicated, and the reaction products were analyzed bySDS-PAGE and autoradiography. The borders of the stacking gel are indicated; material accumulating at the upper edge of the stacking gel ismarked poly-Ub RUL. (B) Effect of methyl ubiquitin. hRUL138, and the RING containing fragment 681-1208 were in vitro translated andprocessed as in A, except that methyl ubiquitin (MeUb) was used in two reactions. Both proteins were efficiently modified except that thereaction products were smaller (smear extending to the top of the separating gel; lanes 9 and 12). (C) Requirement for an intact RING-H2domain. RUL681-1208 and its C1187S mutant were in vitro translated and subjected to ubiquitylation assays as in A.

611RNA-binding ubiquitin-protein ligase hRUL138

The hRUL138 RING-H2 domain is functionally active inself- and trans-ubiquitylationMany, if not all, RING proteins may be ubiquitin ligases, withan intact RING being essential for ubiquitylation activity. E3activity can be reconstituted in vitro by providing to an E3enzyme ubiquitin, an ubiquitin-activating E1 and a conjugatingE2 enzyme plus ATP and the target protein; successfulreaction results in an increased molecular weight caused bypolyubiquitylation. In the absence of a trans-target, various E3shave been shown to self-ubiquitylate (Bays et al., 2001; Fanget al., 2000; Lorick et al., 1999; Nuber et al., 1998). As a firsttest for E3 activity we investigated whether hRUL138 iscapable of self-ubiquitylation. The protein was in vitrotranslated in rabbit reticulocyte lysate (RRL) in the presenceof 35S-Met and incubated with a reaction mix containing ATP,ubiquitin and recombinant E1 plus either UbcH5 or UbcH7 asthe E2. In control reactions, individual components wereomitted. Incubation with the complete mix containing UbcH5dramatically decreased the mobility of hRUL138, preventingmost of it from entering the stacking gel (Fig. 4A, lane 4); onlya small proportion remained at the original position. UbcH7,by contrast, had only minor, if any, effects. In the UbcH5reaction without added ubiquitin, some decrease in theintensity of the unmodified hRUL138 band and a concomitantaccumulation of labeled material at the border betweenstacking and separating gel were observed (Fig. 4A, lane 6),most probably due to the endogenous ubiquitylationcomponents known to be present in RRL (Mastrandrea et al.,1999). To prove that the observed mobility shift was caused bypolyubiquitylation, similar assays were performed usingmethylated ubiquitin [MeUb (Hershko and Heller, 1985)];chemical blocking of the lysine ε-amino groups preventsformation of polyubiquitin chains (but not multiple mono-ubiquitylation of a target containing more than one modifiablelysine residue). Again, a marked mobility shift was induced,but this time, expectedly, the products did not migrate as highup in the gel (Fig. 4B, compare lanes 9 and 10). The efficiencyof modification and the differences between ubiquitin andMeUb were even more obvious when a C-terminal RING-containing fragment, RUL681-1208, was used (Fig. 4B).Hence the observed mobility shifts are due to auto-ubiquitylation of hRUL138.

To prove that an intact RING-domain was required for auto-ubiquitylation the last of the essential Cys-residues in theRING motif was replaced by Ser (variant C1187S); acorresponding mutation in the established RING-E3 ligaseHrd1p abolished its activity (Bays et al., 2001). No reactionwas detected using the hRUL138 mutant whereas, in a parallelassay with the wild-type protein (Fig. 4C, lane 3), the above-described results were fully reproduced. These data confirmedthat hRUL138 is capable of E2- and RING-dependent auto-ubiquitylation.

A more rigorous proof of genuine E3 activity is the abilityto mediate ubiquitylation of a target protein in trans. Duringpurification of bacterially expressed MBP derivatives ofhRUL138 we had repeatedly observed that proteolyticfragments lacking the MBP part co-eluted with the full-lengthprotein from the amylose affinity matrix. This suggested thathRUL138 was able to dimerize, or oligomerize. Such ahomomeric interaction would allow one to bring aubiquitylation-deficient hRUL138 variant lacking the RING-

H2 motif in close proximity to a self-ubiquitylation-competenthRUL138 molecule, potentially leading to trans-ubiquitylation.Because full-length hRUL138 was poorly expressed inbacteria, RUL681-1208, known to be active after in vitrotranslation (Fig. 4B), was fused with an N-terminal MBP anda C-terminal FLAG tag (MBP-RUL681-1208-FLAG) and co-expressed, in E. coli, with a His-tagged fragment lacking theRING-H2 motif (His-RUL681-1128-His). Binding of bothproteins to the amylose resin was confirmed by westernblotting (data not shown). The immobilized complexes werethen subjected to ubiquitylation assays with MeUb. Thereaction products were separated by SDS-PAGE and visualizedby western blots using anti-His (for the trans-target) or anti-FLAG antibodies. The anti-His blot at 0 minutes incubationtime showed the expected 58 kDa product plus a second anti-His reactive band slightly above the 97 kDa marker, possiblya dimer. Irrespective of the exact identity of this additionalproduct (which was observed only in the co-expressionexperiments), it was evident that upon incubation with theubiquitylation-proficient derivative (Fig. 5, α-His, lane 2), butnot in its absence (lane 6), a series of new products with lowermobility was generated at the cost of the unmodified proteins.Interestingly, only a weak reaction was observed when bothproteins were separately expressed and then mixed (lane 8); weassume this difference is mainly caused by a more efficienthetero-oligomerization when the two proteins are co-expressed, whereas subunit exchange between separatelyexpressed homo-oligomers is slow. Anti-FLAG blots (Fig. 5,α-FLAG) confirmed that MBP-RUL681-1208-FLAGexpressed in E. coli was competent for auto-ubiquitylation.These data demonstrated that hRUL138 is able to mediatetrans-ubiquitylation and therefore has genuine E3 ligaseactivity. We note that these experiments were performed withRUL fragments lacking the Lys-rich region; RNA binding was

Fig. 5. Trans-ubiquitylation by hRUL138. MBP-RUL681-1208-FLAG and His-RUL681-1128-His, lacking the RING domain(schematically depicted on the top; RING+ and RING–, respectively)were co-expressed in E. coliand the material immobilized onamylose resin was subjected to ubiquitylation assays with methylubiquitin. Alternatively, the separately expressed proteins wereanalyzed individually or in a mixture (lanes mix). Reactions wereseparated by SDS-PAGE, blotted and incubated with either anti-Hisantibody (left panel; RING– proteins) or with anti-FLAG antibody(right panel; RING+ proteins). The asterisk denotes a probable dimerof the RING– protein.

612

therefore not required for the auto-ubiquitylation and thespecial setting of the trans-ubiquitylation assays.

The RING-H2 domain mediates proteasome-dependentdegradation of hRUL138 in vivoTo address the potential physiological relevance of the in vitroauto-ubiquitylation reaction, we examined the influence onsteady-state levels of hRUL138 in transfected Huh7 cells ofthe established inhibitor of proteasome-dependent proteindegradation, MG132 (Palombella et al., 1994; Rock et al.,1994). For detection by western blotting (and for fluorescencemicroscopy; see below) the full-length hRUL138 protein andthe RING-deleted hRUL1-1128 variant were fused to GFP,against which highly specific antibodies are available. Becausethe inhibitor is used dissolved in DMSO, control cells weretreated with solvent alone. Aliquots from SDS lysates ofMG132 and DMSO-only treated cells containing equalamounts of total protein were then analyzed by using amonoclonal anti-GFP antibody (Fig. 6). For the full-lengthfusion protein, the signals at the expected position weremarkedly enhanced by the inhibitor; the RING-deleted fusion,by contrast, produced much stronger signals without theinhibitor, and the signals remained unchanged in its presence.To exclude the possibility that GFP mediated the effects,hRUL138 was, in addition, fused to the FLAG peptide insteadof GFP and analyzed correspondingly using an anti-FLAGantibody (Fig. 6, right panel). Again, the presence of MG132led to a strong signal increase; specificity of detection wasconfirmed by the absence of signals at the relevant position innon-transfected MG132-treated Huh7 cells. Therefore, theRING-H2 domain in hRUL138-mediated in vitro auto-ubiquitylation is also involved in regulating, in a proteasome-dependent fashion, the steady-state levels of hRUL138 in intactcells. The stabilizing effect of deleting the RING-H2 domainwas also observed when the GFP fusions were analyzed byfluorescence microscopy (see below).

The rul138 gene is expressed at low levels in mosttissues and slightly higher in skeletal muscle, heart andkidneyTo analyze the tissue specificity of hRUL138 RNA expression,premade Northern blots containing polyA+ RNA from eightdifferent tissues were hybridized with the above-described 502bp random-primed hRUL138 cDNA probe. A 2 day exposureon the phosphoimager was required to generate theautoradiogram shown in Fig. 7. This revealed, in all tissuestested, a specific band at a position of about 4.8 kb; its intensitywas highest in skeletal muscle and decreased from heart andkidney, over the brain and pancreas to the lung, liver andplacenta. In addition, weaker bands at about 7.5 kb and, stillweaker, at about 5.5 kb were detectable in the tissues with thehighest levels of the 4.8 kb transcript. In the other tissues, thepresence of these larger products could not unambiguouslybe demonstrated. The 4.8 kb mRNA is large enough toaccommodate the entire hRUL138 ORF plus untranslated 5′and 3′ regions. The other bands probably represent alternativesplicing products. To control for equal RNA loading on theindividual lanes, and for a semiquantitative comparison, thesame blot was reprobed using a β-actin probe of similar specificactivity specific. After a 3 hour exposure, strong bands werevisible at the 2.0 kb position expected for β-actin, and inaddition at 1.8 kb for α-actin in heart and muscle. Although theweak hRUL138-specific bands precluded more accuratequantifications we estimate that they were at least ten times, andprobably hundred times, less abundant than β-actin mRNA.Hence hRUL138 mRNA is expressed at low levels. Directconfirmation of these data on the protein level will have to awaitthe availability of specific, high-affinity antibodies to hRUL138.

Journal of Cell Science 116 (4)

Fig. 6. Upon proteasome inhibition, hRUL138 steady-state levels inintact cells increase in a RING-dependent fashion. Transientlytransfected Huh7 cells expressing GFP fusions of hRUL138 or theRING-deleted variant RUL1-1128 were treated for 6 hours with theproteasome inhibitor MG132, or DMSO alone, and equal aliquotsfrom total cell lysates were analyzed by western blotting using amonoclonal anti-GFP antibody (left panel). The stabilizing effect ofMG132 was also observed when the FLAG epitope instead of GFPwas fused hRUL138; no signal at the corresponding position wasdetectable in nontransfected MG132-treated Huh7 cells (right panel).

Fig. 7.Tissue-specific expression of hRUL138-related mRNAs. Amultiple tissue Northern blot containing poly-A+ RNA from theindicated human tissues was probed with a 32P-labeled hRUL138DNA (nt positions 1497-1998 of the hRUL ORF) and analyzed byphosphorimaging. For control, the same blot was reprobed with anactin probe (lower panel). Numbers on the left show the positions ofRNA size markers (in kb); the major hRUL transcripts are indicatedby the arrowheads on the right.

613RNA-binding ubiquitin-protein ligase hRUL138

hRUL138 is excluded from nucleus and associated witha structured cytoplasmic compartmentTo investigate the intracellular distribution of hRUL138,plasmids encoding the above-described GFP fusions plusseveral additional derivatives were transiently transfectedinto Huh7 human hepatoma cells. Confocal laser scanningfluorescence microscopy showed, for the full-length fusionprotein, a weak fluorescence concentrated in some cytoplasmicstructures but excluded from the nucleus (Fig. 8A). This wasseen in various experiments but always only very few cells perdish exhibited clearly visible signals, suggesting that the full-length protein was poorly expressed, or unstable, or itsexpression led to a preferential loss of transfected cells. On thebasis of the increased steady-state levels of full-length hRULupon proteasome inhibition demonstrated by western blotting(see above) we incubated the transfected cells for 6 hours withMG132. This increased the fraction of detectably positive cells,as monitored by visual inspection; however fluorescenceintensities were still very low. Substantially more and strongerfluorescent cells were reproducibly observed with a C-terminally truncated construct lacking the RING domain(RUL1-1128-GFP), consistent with the anti-GFP western blotdata shown above (Fig. 6). Its intracellular distributionappeared very similar to that of the full-length fusion protein(Fig. 8B), indicating that the RING domain is not responsiblefor the distinct localization. Exclusion from the nucleus wasconfirmed by nuclear counterstaining using TOTO-3 iodide(Fig. 8B, right panel); hence the K-rich motif, despite somesequence similarity, is not active as a nuclear localizationsignal. Suspecting that the compartment in question might bethe ER, we generated a variant of an improved version of thered fluorescent protein with ER targeting and retention signals(RFP2-ER) on the basis of the commercially available yellowfluorescent protein (YFP)-ER derivative (Clontech). Huh7 cellstransfected with this construct showed cytoplasmic structures,sometimes accumulating around the nuclei, that wereindistinguishable from those stained with the YFP-ER variant(data not shown). Overall the fluorescence distributionappeared very similar to that observed with the hRUL138derivatives. For confirmation we sought to demonstratecolocalization of the proteins by co-transfection. The patternslooked very similar upon separate inspection of the green andred channels and regions of apparent co-staining were revealed(Fig. 8C). However, there were always regions withpreferential red or green staining; hence the localization of thetwo proteins was not completely overlapping. Such a partialbut incomplete overlap was observed in numerousexperiments. We also noted that in some cells, and increasingwith time in culture, the hRUL138-derived fluorescenceappeared to accumulate in the nuclear periphery, whereas theRFP2-ER signals remained more dispersed in the cytoplasm,suggesting a segregation of the two proteins. Preliminaryexperiments aimed at defining the localization signal(s) withinthe RUL138 sequence using variously truncated derivativesshowed similar patterns even for a variant consisting of onlyamino acids 510 to 878 (Fig. 8D). This suggests that none ofthe putative transmembrane regions is crucial for associationwith the compartment; possibly, association is mediated by thecoiled-coil region in hRUL138 but the exact nature of thatcompartment as well as the localization mechanism remains tobe determined.

DiscussionIn this study we have identified a large human RNA-bindingprotein, hRUL138, that is associated with intracytoplasmicstructures and contains a RING-H2 domain mediating self-and trans-ubiquitylation. These data indicate that hRUL138represents a novel type of RNA-binding E3 enzyme. Belowwe discuss this possibility in the light of the data obtained sofar.

Fig. 8. Intracellular localization of hRUL138. Huh7 cells weretransfected with expression constructs for the GFP fusion proteinsindicated on the top of each panel. For colocalization studies, theindicated hRUL-GFP construct was co-transfected with plasmidpRFP2-ER, encoding an ER-targeted red fluorescent protein.Unfixed cells were observed by confocal laser scanning microscopy,except for the TOTO-3 counterstaining in panel B; there GFP wasdetected indirectly using mouse anti-GFP antibodies and afluorescently labeled anti-mouse IgG antibody. (A) Localizeddistribution of hRUL138-GFP versus even distribution of GFPalone. (B) Extranuclear staining by hRUL1-1128-GFP versus bluenuclear staining by TOTO-3. (C) Coexpression of hRUL1-1128-GFP and RFP2-ER. (D) Coexpression of hRUL510-878-GFP andRFP2-ER.

614

RNA binding by hRUL138In the absence of a known RNA recognition motif withinthe hRUL138 sequence we used deletion mapping andNorthwestern blotting to identify a central, basic region (aminoacids 656 to 674) as important for RNA binding. The first sevenlysines, present in the product of the initial cDNA clone, butnot the first five, were sufficient for RNA binding. Binding wasalso detected when the basic motif was disposed internallyrather than at the C-terminus and was drastically reduced,though not completely abolished, when the central KKKTKsequence was mutated to SGSTA; this was also observed insolution assays. Because of the poly-cationic nature of the K-rich motif it might be argued that the RNA binding detected inthese assays is merely electrostatic. However, several lines ofevidence argue strongly against this view. First, all assays wereperformed in the presence of more than a hundred-fold excessof tRNA and herring sperm DNA. This included the initialexpression library screen, which was based on a procedure thatallowed successful isolation of another RNA-binding protein,TRBP (Gatignol et al., 1991). Further, during the solutionassays NaCl was present at the rather high concentration of 350mM. Second, the total number of basic residues (7 Lys and 1Arg) in the K-rich region of the K4- mutants with drasticallyreduced RNA binding is the same as in the strongly binding C-terminally truncated hRUL1-666 variant; this was furthercorroborated by the absence of a signal on the Nortwesternblots when a fragment from an unrelated protein containing thesequence K2NK4EKSK (i.e. 8 Lys in 11 amino acids in total)was run in parallel. Finally, the Northwestern signal from thehRUL510-878 was at least partially resistant to a washing stepwith a solution of 1 mg/ml of heparin (Konarska, 1989; Pinol-Roma et al., 1988), and the extent of resistance was similar tothat observed for the core protein of HBV, which is anestablished nucleic-acid-binding protein (Hatton et al., 1992;Nassal, 1992). Hence the central KKKTK element of the K-rich motif appears to be a critical part of a larger RNA-bindingdomain.

Regarding binding specificity, our Northwestern assaysusing three different RNA probes did not reveal a preferencefor HBV-ε-containing RNA, at least not under the conditionsused. Although database searches revealed stretches ofmultiple lysines in various predicted gene products, includingputative RNA helicases, no protein containing the identicalmotif, let alone one with characterized RNA-binding capacity,was found that would have allowed for inferences fromthe hRUL138-RNA interaction. Remotely, however, thepreponderance of Lys-residues resembles that of the argininesin the arginine-rich RNA-binding motif present, for instance,in the HIV-1 Tat protein. Peptides derived from this sequencebind specifically to TAR RNA, mainly mediated by a singleArg-residue surrounded by other basic residues that recognizea bulge; however, with about 10-fold lower affinity they alsointeract with other RNAs (Calnan et al., 1991). Possibly, ourassays measure a similar basal RNA-binding activity ofhRUL138. More quantitative binding studies using the sameprobe RNAs might reveal more explicit differences. To thisend, we have also tested, using the in-solution binding assay,the ability of homopolymeric RNAs to compete with bindingof the labeled probe RNA as described in Fig. 2C. Poly-A andpoly-U at 10 µg/ml reduced binding by about 80% whereasten-fold higher concentrations of poly-G and poly-C were

required for a similar reduction; double-stranded poly-IC wasineffective even at this concentration (data not shown). Hencethere is some discrimination between different kinds of RNA,which again argues against mere electrostatic binding; atpresent, however, these data do not allow for clear-cutconclusions on the actual RNA targets of hRUL138. Given itslarge size, the fragments used in the assays may only beimperfect mimics of the entire protein; furthermore, otherfactors associating with hRUL138 could affect its bindingspecificity in vivo. Altogether we envisage two alternatives forphysiological RNA binding by hRUL138: either it has a truelybroad selectivity or, more likely, there are specific RNA targetsthat remain to be identified, possibly by similar approaches tothose recently used for finding target RNAs of the FMRPprotein whose absence causes fragile X syndrome (Brown etal., 2001; Darnell et al., 2001).

Ubiquitin ligase activity of hRUL138When complemented with ubiquitin (or methyl ubiquitin), E1,E2 and ATP, both in vitro translated hRUL138 and bacteriallyexpressed fragments containing the RING-H2 domain werecapable of self-ubiquitylation. Efficient reaction required all ofthe aforementioned components and an intact RING domain,as demonstrated by the inactivity of variants that lacked theRING domain or carried a Cys to Ser mutation in the RINGmotif (Bays et al., 2001). hRUL138 is also capable of trans-ubiquitylation, as shown by the efficient ubiquitylation of aRING-deleted variant upon co-expression in E. coli. Plausibly,the spatial proximity between the E2-E3 complex and the targetprotein required for trans-ubiquitylation was brought about bythe ability of hRUL138 to undergo homomeric interactions,and these might be mediated by the coiled-coil domain in thecenter of hRUL138. Whether homomeric interactions betweenhRUL138 molecules exist in vivo and, if so, whether trans-ubiquitylation from one to another molecule within thecomplex occurs cannot be answered by these in vitroexperiments. However, preliminary data show that a non-related RNA-binding protein, selenocysteine-insertion-sequence (SECIS)-RNA-binding protein 2 [SBP2 (Copeland etal., 2001)] can also be ubiquitylated by hRUL138. Togetherthese data demonstrated that hRUL138 meets the criteria of agenuine E3 enzyme. The physiological relevance of the auto-ubiquitylation activity of hRUL138 is strongly supported by itsmarked accumulation in transfected cells upon inhibition ofproteasome-dependent turnover, which, by contrast, had noeffect on a RING-deficient variant. Since the latter wasexpressed much better even without proteasome inhibition, itis likely that the authentic protein is subject to RING-mediatedautoregulation as, for example, reported for the RING E3Mdm2 (Fang et al., 2000).

Expression and intracellular localization of hRUL138Northern blotting revealed a weak hRUL138-specific band ofabout 4.8 kb in all tissues tested, with the highest intensity inskeletal muscle followed by heart, kidney and brain, and finallypancreas, lung, liver and placenta. These data agree well withRT-PCR-derived semiquantifications for KIAA0675 RNA(available at: http://www.kazusa.or.jp/huge/). In that study, thehighest levels (10-50 fg/ng specific plasmid DNA equivalents

Journal of Cell Science 116 (4)

615RNA-binding ubiquitin-protein ligase hRUL138

per ng poly-A+ RNA) were found in the heart, skeletal muscle,kidney and testis, lower amounts (1-10 fg/ng poly-A+ RNA)were found in the brain and very low levels (<1 fg/ng poly-A+

RNA) in the lung, liver, pancreas, spleen and ovary. HencehRUL138 mRNAs are expressed at very low levels; the sameis probably true for the protein. Whether the higher levels ofhRUL138 mRNA in skeletal muscle and heart are related tothe low proliferative activity of these tissues is unclear; notably,two other ubiquitin ligases, MuRF1 and MAFbx, are alsopredominantly expressed in these tissues and have been shownto be critically involved in muscle atrophy (Bodine et al.,2001). However, hRUL138 mRNA expression may also beinducible in these and in other tissues and, in addition, beregulated on the protein level (see above). Two larger cross-reactive bands of about 5.5 kb and 7.5 kb were detected in sometissues. Given that several of the cDNA clones obtained bymolecular hybridization contained intronic sequences, theselarger mRNAs are probably derived from alternative splicingevents. In two of the clones, for instance, a 805 bp intron is notspliced out between positions 1962 and 1963 of the hRUL138ORF (see Fig. 1A); the larger mRNAs might correspond tothese cDNAs. Although the full range of potential splicingvariants remains to be explored it is highly likely from theappearance of more than one hRUL mRNA species thatalternative splicing does occur physiologically, implying theexistence of various isoforms.

Regarding the intracellular distribution of hRUL138, theweakly expressed full-length GFP fusion protein was clearlylocated outside the nucleus in what appeared to be a structuredcytoplasmic compartment. The RING-deleted variant hRUL1-1128 gave much stronger signals with a very similardistribution. On the basis of the similar appearance of cellstransfected with an ER-targeted RFP2 variant these structuresmight represent the ER, or a subcompartment thereof, becausethe two proteins did not fully colocalize and appeared tosegregate with time. This localization was apparentlyindependent of the RING-H2 domain, which was not toosurprising given that examples for both RING-dependent andRING-independent localization of RING proteins are known(Hu and Fearon, 1999; Shin et al., 2001). We note that thisconclusion rests on the similar appearance of the truncatedhRUL-GFP proteins and that of the very few detectably GFP-positive cells transfected with full-length hRUL138-GFP. Wecannot rule out the possibility that these few cells might bedefective in ubiquitylation or proteasome-mediated turnover.Hence it remains formally possible that, in a normal cellularenvironment, an active hRUL138 RING domain does affect theprotein's localization. However, even a GFP fusion withhRUL510-878 had a very similar intracellular distribution;hence, neither of the three weakly predicted transmembranehelices is essential for localization. This suggests thatcompartment association occurs, different from that of thetransmembrane-domain-containing ubiquitin ligases Hrd1p(Bays et al., 2001) or Tul1 (Reggiori and Pelham, 2002), in anindirect fashion, possibly via interactions mediated by thepredicted coiled-coil region (amino acids 600 to 950), whichis part of the localizing variant 510 to 878.

What is the physiological role of hRUL138?As shown above, hRUL138 is a non-abundant RNA-binding

RING-H2 ubiquitin-protein ligase that localizes to anintracytoplasmic compartment, possibly the ER. At presentwe can only speculate about its physiological function. Theabsence of homologs in lower eukaryotes not only preventsstraightforward genetic experiments but also indicates afunction restricted to higher organisms; together with its lowabundance this makes a specialized function more likelythan a role in general cellular metabolism. In this light, arelationship with ER-associated degradation (ERAD), amechanism involved in cellular quality control and regulationof normal ER-resident proteins (Bays et al., 2001; Fang et al.,2001), does not seem likely. That hRUL138 cooperates, invitro, productively with UbcH5 but not UbcH7 indicates thespecificity of the E2-E3 interaction. However, each E2 enzymecan interact with more than one E3 and, in this respect, UbcH5appears to be relatively promiscuous, not permitting directconclusions on the physiological function of hRUL138.Eventually the natural RNA and protein ubiquitylation targetswill have to be identified; both are probably major endeavours,especially when considering that the 138 kDa protein isprobably only one out of several splicing variants. Perhaps themost intriguing and at the same time tractable question iswhether the RNA-binding and the ubiquitylation activity ofhRUL138 are functionally related. If so, hRUL138 may be anE3 enzyme that achieves target specificity using RNA as amediator. In the auto-ubiquitylation and homo-oligomer-basedtrans-ubiquitylation assays described above we did not findan influence of RNA on ubiquitylation activity; however, inpreliminary studies hRUL138-mediated ubiquitylation ofSBP2 was markedly enhanced by the presence of RNA.Although the underlying mechanism remains to be elucidatedthis suggests that RNA-binding proteins or ribonucleoproteinparticles (RNPs) might be physiological targets of hRUL138;whether this includes RNPs derived from HBV, or otherviruses, remains to be determined.

We thank Martin Scheffner for providing purified E1 and E2enzymes and expert advice on the ubiquitylation assays, Jürgen Beckfor advice on RNA-protein interactions, Tobias Doerks and Peer Borkfor the coiled-coil predictions and Mary Follo for help with confocalmicroscopy. The RZPD screening service (German Cancer ResearchCenter, Heidelberg) is acknowledged for identifying related cDNAclones. This work was supported in part by grants from the theDeutsche Forschungsgemeinschaft (DFG), the Bundesministerium fürBildung und Forschung, and by the Fonds der Chemischen Industrie.S.G.K. is grateful for a predoctoral fellowship from the Boehringer-Ingelheim Foundation.

ReferencesBays, N. W., Gardner, R. G., Seelig, L. P., Joazeiro, C. A. and Hampton,

R. Y. (2001). Hrd1p/Der3p is a membrane-anchored ubiquitin ligaserequired for ER- associated degradation. Nat. Cell Biol.3, 24-29.

Blumberg, B. S. (1997). Hepatitis B virus, the vaccine, and the control ofprimary cancer of the liver. Proc. Natl. Acad. Sci. USA94, 7121-7125.

Bodine, S. C., Latres, E., Baumhueter, S., Lai, V. K., Nunez, L., Clarke,B. A., Poueymirou, W. T., Panaro, F. J., Na, E., Dharmarajan, K. et al.(2001). Identification of ubiquitin ligases required for skeletal muscleatrophy. Science294, 1704-1708.

Borden, K. L. (2000). RING domains: master builders of molecular scaffolds?J. Mol. Biol.295, 1103-1112.

Borden, K. L. and Freemont, P. S.(1996). The RING finger domain: a recentexample of a sequence-structure family. Curr. Opin. Struct. Biol.6, 395-401.

616

Brown, V., Jin, P., Ceman, S., Darnell, J. C., O’Donnell, W. T., Tenenbaum,S. A., Jin, X., Feng, Y., Wilkinson, K. D., Keene, J. D. et al. (2001).Microarray identification of FMRP-associated brain mRNAs and alteredmRNA translational profiles in fragile X syndrome. Cell 107, 477-487.

Calnan, B. J., Tidor, B., Biancalana, S., Hudson, D. and Frankel, A. D.(1991). Arginine-mediated RNA recognition: the arginine fork. Science252,1167-1171.

Cheng, A. C., Calabro, V. and Frankel, A. D.(2001). Design of RNA-binding proteins and ligands. Curr. Opin. Struct. Biol.11, 478-484.

Copeland, P. R., Stepanik, V. A. and Driscoll, D. M.(2001). Insight intomammalian selenocysteine insertion: domain structure and ribosomebinding properties of Sec insertion sequence binding protein 2. Mol. Cell.Biol. 21, 1491-1498.

Daher, A., Longuet, M., Dorin, D., Bois, F., Segeral, E., Bannwarth, S.,Battisti, P. L., Purcell, D. F., Benarous, R., Vaquero, C. et al. (2001). Twodimerization domains in the trans-activation response RNA-binding protein(TRBP) individually reverse the protein kinase R inhibition of HIV-1 longterminal repeat expression. J. Biol. Chem.276, 33899-33905.

Darnell, J. C., Jensen, K. B., Jin, P., Brown, V., Warren, S. T. and Darnell,R. B. (2001). Fragile X mental retardation protein targets G quartet mRNAsimportant for neuronal function. Cell 107, 489-499.

Fang, S., Ferrone, M., Yang, C., Jensen, J. P., Tiwari, S. and Weissman,A. M. (2001). The tumor autocrine motility factor receptor, gp78, is aubiquitin protein ligase implicated in degradation from the endoplasmicreticulum. Proc. Natl. Acad. Sci. USA98, 14422-14427.

Fang, S., Jensen, J. P., Ludwig, R. L., Vousden, K. H. and Weissman, A.M. (2000). Mdm2 is a RING finger-dependent ubiquitin protein ligase foritself and p53. J. Biol. Chem.275, 8945-8951.

Freemont, P. S.(2000). RING for destruction? Curr. Biol. 10, R84-R87.Gatignol, A., Buckler-White, A., Berkhout, B. and Jeang, K. T.(1991).

Characterization of a human TAR RNA-binding protein that activates theHIV-1 LTR. Science251, 1597-1600.

Hatton, T., Zhou, S. and Standring, D. N.(1992). RNA- and DNA-bindingactivities in hepatitis B virus capsid protein: a model for their roles in viralreplication. J. Virol. 66, 5232-5241.

Hershko, A. and Heller, H. (1985). Occurrence of a polyubiquitin structurein ubiquitin-protein conjugates. Biochem. Biophys. Res. Commun.128,1079-1086.

Hicke, L. (2001). A new ticket for entry into budding vesicles - ubiquitin. Cell106, 527-530.

Hu, G. and Fearon, E. R.(1999). Siah-1 N-terminal RING domain is requiredfor proteolysis function, and C-terminal sequences regulate oligomerizationand binding to target proteins. Mol. Cell. Biol.19, 724-732.

Ishikawa, K., Nagase, T., Suyama, M., Miyajima, N., Tanaka, A., Kotani,H., Nomura, N. and Ohara, O.(1998). Prediction of the coding sequencesof unidentified human genes. X. The complete sequences of 100 new cDNAclones from brain which can code for large proteins in vitro. DNA Res.5,169-176.

Kikuno, R., Nagase, T., Suyama, M., Waki, M., Hirosawa, M. and Ohara,O. (2000). HUGE: a database for human large proteins identified in theKazusa cDNA sequencing project. Nucleic Acids Res.28, 331-332.

Konarska, M. M. (1989). Analysis of splicing complexes and small nuclearribonucleoprotein particles by native gel electrophoresis. Methods Enzymol.180, 442-453.

König, S., Beterams, G. and Nassal, M.(1998). Mapping of homologousinteraction sites in the hepatitis B virus core protein. J. Virol. 72, 4997-5005.

Lorick, K. L., Jensen, J. P., Fang, S., Ong, A. M., Hatakeyama, S. andWeissman, A. M. (1999). RING fingers mediate ubiquitin-conjugatingenzyme (E2)-dependent ubiquitination. Proc. Natl. Acad. Sci. USA96,11364-11369.

Lupas, A. (1996). Coiled coils: new structures and new functions. TrendsBiochem. Sci.21, 375-382.

Mastrandrea, L. D., You, J., Niles, E. G. and Pickart, C. M.(1999). E2/E3-mediated assembly of lysine 29-linked polyubiquitin chains. J. Biol. Chem.274, 27299-27306.

Nassal, M. (1992). The arginine-rich domain of the hepatitis B virus coreprotein is required for pregenome encapsidation and productive viralpositive- strand DNA synthesis but not for virus assembly. J. Virol.66, 4107-4116.

Nassal, M. (1999). Hepatitis B virus replication: novel roles for virus-hostinteractions. Intervirology42, 100-116.

Nassal, M. (2000). Macromolecular interactions in hepatitis B virusreplication and particle assembly. In DNA virus replication, vol. 26 (ed. A.J. Cann), pp. 1-40. Oxford, UK: Oxford University Press.

Nuber, U., Schwarz, S., Kaiser, P., Schneider, R. and Scheffner, M.(1996).Cloning of human ubiquitin-conjugating enzymes UbcH6 and UbcH7 (E2-F1) and characterization of their interaction with E6-AP and RSP5. J. Biol.Chem.271, 2795-2800.

Nuber, U., Schwarz, S. E. and Scheffner, M.(1998). The ubiquitin-proteinligase E6-associated protein (E6-AP) serves as its own substrate. Eur. J.Biochem.254, 643-649.

Ohira, M., Ootsuyama, A., Suzuki, E., Ichikawa, H., Seki, N., Nagase, T.,Nomura, N. and Ohki, M. (1996). Identification of a novel human genecontaining the tetratricopeptide repeat domain from the Down syndromeregion of chromosome 21. DNA Res.3, 9-16.

Palombella, V. J., Rando, O. J., Goldberg, A. L. and Maniatis, T.(1994).The ubiquitin-proteasome pathway is required for processing the NF- kappaB1 precursor protein and the activation of NF-kappa B. Cell 78, 773-785.

Perez-Canadillas, J. M. and Varani, G.(2001). Recent advances in RNA-protein recognition. Curr. Opin. Struct. Biol.11, 53-58.

Perri, S. and Ganem, D.(1996). A host factor that binds near the termini ofhepatitis B virus pregenomic RNA. J. Virol. 70, 6803-6809.

Perri, S. and Ganem, D.(1997). Effects of mutations within and adjacent tothe terminal repeats of hepatitis B virus pregenomic RNA on viral DNAsynthesis. J. Virol. 71, 8448-8455.

Pickart, C. M. (2000). Ubiquitin in chains. Trends Biochem. Sci.25, 544-548.Pickart, C. M. (2001). Mechanisms underlying ubiquitination. Annu. Rev.

Biochem.70, 503-533.Pickart, C. M. and VanDemark, A. P. (2000). Opening doors into the

proteasome. Nat. Struct. Biol.7, 999-1001.Pinol-Roma, S., Choi, Y. D., Matunis, M. J. and Dreyfuss, G.(1988).

Immunopurification of heterogeneous nuclear ribonucleoprotein particlesreveals an assortment of RNA-binding proteins. Genes Dev.2, 215-227.

Reggiori, F. and Pelham, H. R.(2002). A transmembrane ubiquitin ligaserequired to sort membrane proteins into multivesicular bodies. Nat. CellBiol. 4, 117-123.

Rock, K. L., Gramm, C., Rothstein, L., Clark, K., Stein, R., Dick, L.,Hwang, D. and Goldberg, A. L.(1994). Inhibitors of the proteasome blockthe degradation of most cell proteins and the generation of peptidespresented on MHC class I molecules. Cell 78, 761-771.

Saurin, A. J., Borden, K. L., Boddy, M. N. and Freemont, P. S.(1996).Does this have a familiar RING? Trends Biochem. Sci.21, 208-214.

Shin, M. E., Ogburn, K. D., Varban, O. A., Gilbert, P. M. and Burd, C. G.(2001). FYVE domain targets Pib1p ubiquitin ligase to endosome andvacuolar membranes. J. Biol. Chem.276, 41388-41393.

Small, I. D. and Peeters, N.(2000). The PPR motif – a TPR-related motifprevalent in plant organellar proteins. Trends Biochem. Sci.25, 46-47.

Smith, C. A., Calabro, V. V. and Frankel, A. D.(2000). An RNA-bindingchameleon. Mol. Cell 6, 1067-1076.

Vogt, V. M. (2000). Ubiquitin in retrovirus assembly: actor or bystander? Proc.Natl. Acad. Sci. USA97, 12945-12947.

Wang, C., Deng, L., Hong, M., Akkaraju, G. R., Inoue, J. and Chen, Z. J.(2001). TAK1 is a ubiquitin-dependent kinase of MKK and IKK. Nature412, 346-351.

Weber, M., Bronsema, V., Bartos, H., Bosserhoff, A., Bartenschlager, R.and Schaller, H.(1994). Hepadnavirus P protein utilizes a tyrosine residuein the TP domain to prime reverse transcription. J. Virol. 68, 2994-2999.

Zolotukhin, S., Potter, M., Hauswirth, W. W., Guy, J. and Muzyczka, N.(1996). A “humanized” green fluorescent protein cDNA adapted for high-level expression in mammalian cells. J. Virol. 70, 4646-4654.

Journal of Cell Science 116 (4)