Ubiquibodies, Synthetic E3 Ubiquitin Ligases Endowedwith Unnatural Substrate Specificity for Targeted ProteinSilencing*

Received for publication, December 21, 2013, and in revised form, January 22, 2014 Published, JBC Papers in Press, January 28, 2014, DOI 10.1074/jbc.M113.544825

Alyse D. Portnoff‡, Erin A. Stephens§, Jeffrey D. Varner‡¶, and Matthew P. DeLisa‡¶1

From the ‡Department of Biomedical Engineering, §Graduate Field of Biochemistry, Molecular, and Cell Biology, and ¶School ofChemical and Biomolecular Engineering, Cornell University, Ithaca, New York 14853

Background: Techniques that harness the power of the ubiquitin-proteasome pathway (UPP) for protein knock-out arelimited to a narrow set of protein targets.Results: Engineered “ubiquibodies” specifically and systematically removed exogenous target proteins.Conclusion: Diverse protein targets can be redirected to the UPP using this new protein silencing method.Significance: Ubiquibodies offer a simple, reproducible, and customizable technique for selectively and controllably depletingcellular proteins.

The ubiquitin-proteasome pathway (UPP) is the main route ofprotein degradation in eukaryotic cells and is a common mech-anism through which numerous cellular pathways are regulated.To date, several reverse genetics techniques have been reportedthat harness the power of the UPP for selectively reducing thelevels of otherwise stable proteins. However, each of theseapproaches has been narrowly developed for a single substrateand cannot be easily extended to other protein substrates ofinterest. To address this shortcoming, we created a generaliz-able protein knock-out method by engineering protein chime-ras called “ubiquibodies” that combine the activity of E3 ubiq-uitin ligases with designer binding proteins to steer virtually anyprotein to the UPP for degradation. Specifically, we repro-grammed the substrate specificity of a modular human E3 ubiq-uitin ligase called CHIP (carboxyl terminus of Hsc70-interact-ing protein) by replacing its natural substrate-binding domainwith a single-chain Fv (scFv) intrabody or a fibronectin type IIIdomain monobody that target their respective antigens withhigh specificity and affinity. Engineered ubiquibodies reliablytransferred ubiquitin to surface exposed lysines on target pro-teins and even catalyzed the formation of biologically relevantpolyubiquitin chains. Following ectopic expression of ubiqui-bodies in mammalian cells, specific and systematic depletion ofdesired target proteins was achieved, whereas the levels of a nat-ural substrate of CHIP were unaffected. Taken together, engi-neered ubiquibodies offer a simple, reproducible, and customi-zable means for directly removing specific cellular proteinsthrough accelerated proteolysis.

Explanation of the physiological function of a cellular proteinoften requires assessing the consequences of its removal. Tech-niques for reverse genetics such as antisense deoxyoligonucle-otides and RNA interference have been developed to interruptprotein expression at the DNA or RNA level; however, theseapproaches act at various stages of biosynthesis and rely on theendogenous protein degradation apparatus to remove any pre-existing target protein. By operating at the post-translationallevel, protein knock-out techniques have the potential to dis-sect complicated protein functions at a higher resolution thanapproaches functioning at the level of DNA or RNA.

The forerunners of these post-translational knock-out tech-niques include antibodies and their derivatives, whose highselectivity and modularity are ideal for inactivating intracellularproteins (1). However, antibodies and their fragments are lim-ited by the need to form disulfide bonds for proper folding;hence, they do not always function in the reducing environ-ment inside cells. This has led to the concept of intracellularantibody fragments (intrabodies) and antibody mimics such asthe human fibronectin type III domain (FN3)2 and DARPins(designed ankyrin repeat proteins), all of which can be engi-neered to bind their antigens inside living cells (2– 4). Thesedesigner binding proteins (DBPs) exhibit the high specificityand affinity of conventional monoclonal antibodies, but aremuch smaller, can fold efficiently in a reducing environment,and can be manipulated and delivered as genes, which makesthem particularly useful as drug discovery tools and next-gen-eration therapeutics (5–7).

These developments notwithstanding, DBPs have severaldrawbacks for intracellular protein knock-out. For instance,because there are no natural elimination pathways for DBP-target complexes, the intracellular level of a DBP must exceedthe expression level of its target, which can be challenging dueto the inefficiency of most existing delivery methods. This can

* This work was supported by the National Science Foundation Career AwardCBET-0449080 (to M. P. D.), National Institutes of Health Grant CA132223A(to M. P. D.), New York State Office of Science, Technology, and AcademicResearch Distinguished Faculty Award (to M. P. D.), National Institutes ofHealth Chemical Biology Training Grant T32 GM008500 (fellowship toA. D. P.), National Science Foundation GK-12 DGE-0841291 (fellowship toA. D. P.), a Cornell Presidential Life Science Fellowship (to E. A. S.), and theNational Science Foundation Graduate Research Fellowship ProgramGrant DGE-1144153 (to E. A. S.).

1 To whom correspondence should be addressed: School of Chemical andBiomolecular Engineering, Cornell University, Ithaca, NY 14853. Tel.: 607-254-8560; Fax: 607-255-9166; E-mail: md255@cornell.edu.

2 The abbreviations used are: FN3, fibronectin type III domain; DBP, designerbinding protein; UPP, ubiquitin proteasome pathway; scFv, single-chainFv; uAb, ubiquibody; TPR, tetratricopeptide repeat; �-gal, �-galactosidase;CHIP, C terminus of Hsc70-interacting protein.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 289, NO. 11, pp. 7844 –7855, March 14, 2014© 2014 by The American Society for Biochemistry and Molecular Biology, Inc. Published in the U.S.A.

7844 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 289 • NUMBER 11 • MARCH 14, 2014

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

be further compounded if the association between the DBP andits target protein is short-lived, leading to “escape” of the targetand thus ineffective inhibition. Even when high-affinity bindersare available, target inactivation is not guaranteed because not allDBPs are endowed with “intrinsic-neutralizing” properties. Thus,we sought to develop a protein silencing strategy that links a DBPwith the natural degradation machinery of the cell, the ubiquitinproteasome pathway (UPP), such that the steady-state levels of theintended target of the DBP are systematically reduced.

The UPP is the main route of protein degradation in eukary-otic cells and is a common mechanism through which numer-ous cellular pathways are regulated (8). Most proteins are tar-geted for proteasomal degradation after being covalentlymodified with a polyubiquitin chain, which is attached to eitherinternal lysine residues or the N-terminal residue on target pro-teins. This polyubiquitin serves as the signal for recognition andproteolytic destruction by the 26S proteasome or, in somecases, by lysosomes. The canonical cascade of ubiquitinationinvolves three enzymes termed E1, E2, and E3. The substratespecificity of a given ubiquitin-proteolytic pathway is conferredby the E3 (9), whose intrinsic structural flexibility enablesaccommodation of substrates with different sizes and struc-tures (10). This conformational flexibility along with the factthat “consensus” ubiquitination sites are not required for ubiq-uitin transfer (11) raises the distinct possibility that E3s may beengineered for accelerating the degradation of unnatural targetproteins of interest. Indeed, several methods have beenreported for steering cellular proteins to the UPP for degrada-tion. Most of these have been based on a multimeric E3 com-plex called SCF, which consists of Skp1 (S-phase kinase-associ-ated protein 1), Cul-1, F-box-containing substrate receptor,and RING proteins. The F-box domain of the substrate recep-tor binds to Skp1 to connect to the core SCF complex, whereasa protein-protein domain (such as the WD40 domain or leucinerich repeat) of the receptor is responsible for substrate recruit-ment. Protein knock-out using the SCF system has been devel-oped by introducing established protein-protein interactionmodules to the F-box-containing substrate receptor for specificrecruitment of target proteins (12, 13) or using peptide-small-molecule hybrids, known as Protacs, that bridge the interactionbetween the intended target and the F-box (14). Although thesestudies demonstrate that SCF-type E3s can be used to eliminatespecific protein targets, the approach requires the endogenousSCF core machinery to carry out ubiquitination. It is possible,therefore, that the core ligase machinery might become over-whelmed upon high level expression of engineered F-box proteins,thereby suppressing the degradation of both the intended target aswell as native substrates (15). A related concern is the complexityof SCF complex formation, which requires the presence of numer-ous protein components in a very precise molecular ratio.

A viable alternative is the use of “stand-alone” E3s that bindtheir substrates and transfer ubiquitin to these substrates with-out the need for additional accessory factors. For example, thesingle-chain U-box type E3 called CHIP (carboxyl terminus ofHsc70-interacting protein) has been redesigned for ubiquitina-tion of specific targets (10, 16, 17). However, nearly all engi-neered E3s described so far, including both U-box and SCF-types, rely on a naturally existing binding partner to interact

with the desired target protein. As a result, engineered E3s haveonly been developed for a limited set of substrates and cannotbe easily extended to other protein substrates of interest. Caus-sinus et al. (13) recently addressed this issue by engineering achimera comprised of the SCF-type E3 F-box domain and asingle-domain antibody fragment (i.e. nanobody) specific forGFP, which was capable of depleting target GFP fusions.Although the engineered chimera still targeted just a single sub-strate, namely the GFP domain of a target fusion protein, thedevelopment of “GFP protein trap” methods in cultured cells(18) and even whole, live organisms (19) makes it possible totarget a wide array of GFP-tagged proteins. However, becausesilencing requires the targets to be recombinantly fused to GFP,this method is not able to deplete natural targets in the absenceof the GFP domain. The F-box-nanobody fusions were alsounable to process GFP on its own, suggesting that engineeredSCF systems might be constrained by a size limit below whichsilencing may not be not possible.

To create a method for depletion of virtually any target pro-tein, we engineered chimeras whereby DBPs were fused to theC-terminal U-box ligase domain of human CHIP. We choseCHIP for several reasons. First, unlike SCF-based systems,CHIP does not rely on other subunits for its functionality. Sec-ond, CHIP has a broad substrate diversity (20), suggesting thatits specificity can be altered without affecting its ubiquitintransfer activity. Third, CHIP is modular in nature, and itsU-box domain is known to remain active in fusions (10, 16, 17).Our idea was that a target protein recognized by the DBPdomain would be specifically degraded upon ectopic expres-sion of the engineered fusion; hence, we call these fusionsubiquibodies (uAbs) because of their potential for antibody-mimetic binding and ubiquitination of target proteins.

EXPERIMENTAL PROCEDURES

Plasmid Construction—Full-length human CHIP (a gift fromCam Patterson) was PCR amplified for cloning into pET28a(�)using 5� NcoI and 3� SalI restriction sites. DNA encoding adouble tag of FLAG-His6 was created by dimerizing primerswith a 5� SalI overhang and a 3� HindIII overhang. Double liga-tion was performed to insert the CHIP PCR product and primerdimer between NcoI and HindIII sites in pET28a(�), yieldingplasmid pET28a-CHIP. To create truncated CHIP�TPR, DNAcorresponding to amino acids 128 –303 of human CHIP wasPCR amplified with introduction of a 5� NcoI site and a 3� SalIsite. Double ligation was performed as above to insert thisproduct along with the primer dimer into pET28a(�), yield-ing pET28a-CHIP�TPR. The genes encoding scFv13 andscFv13-R4 (a gift from Pierre Martineau) were PCR-amplified,and each was double ligated into pET28a(�) with the aboveprimer dimer, yielding the control plasmids pET28a-scFv13and pET28a-scFv13-R4, respectively. To create CHIP�TPRfusions, PCR was used to introduce an EcoRI site followed by ashort, flexible linker of GSGSG to the 5� end of CHIP�TPR. Inparallel, each of the DBPs including scFv13, scFv13-R4, and thescFv D10 (a gift from Andreas Plückthun) was PCR-amplifiedwith a 5� NcoI site and 3� EcoRI site. Double ligation was thenused to insert each single-chain Fv (scFv) along with theGSGSG linker-CHIP�TPR product between the NcoI and SalI

Selective Protein Knock-out Using Engineered Ubiquibodies

MARCH 14, 2014 • VOLUME 289 • NUMBER 11 JOURNAL OF BIOLOGICAL CHEMISTRY 7845

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

sites of pET28a-CHIP, yielding pET28a-scFv13-uAb, pET28a-R4-uAb, and pET28a-D10-uAb. Due to an internal EcoRI site inthe FN3 monobody YS1 (a gift from Shohei Koide), overlapextension PCR was used to add the GSGSG linker and N termi-nus of CHIP�TPR to the YS1 PCR product, until it reached aunique BstBI site within CHIP�TPR. This overlap extensionPCR product was then ligated between NcoI and BstBI sitesof pET28a-R4-uAb, yielding pET28a-YS1-uAb. The R272Apoint mutation was introduced into pET28a-R4-uAb using aQuikChange site-directed mutagenesis kit (Stratagene). Forexpression in eukaryotic cells, the above constructs were PCR-amplified from their pET28a(�) backbones using primers thatintroduced a Kozak sequence at the start codon as well as a 5�HindIII site and a 3� XbaI site. The resulting PCR products werethen cloned between the HindIII and XbaI sites of plasmidpcDNA3. Due to an internal SalI site, the YS1 gene was directlycloned into pcDNA3 using overlap extension to add a HindIIIsite and Kozak sequence to the 5� end and FLAG tag, His6 tagand XbaI site to the 3� end. The target substrate proteins �-galand maltose-binding protein (MBP) were PCR-amplified usingprimers that introduced a Kozak sequence at the start codon aswell as 5� HindIII site for MBP or 5� XhoI for �-gal and a 3� XbaIsite for both and cloned in the corresponding sites of pcDNA3.

Protein Expression and Purification—All purified proteinswere obtained from cultures of Escherichia coli BL21(DE3) cellsgrown in 500 ml of Luria-Bertani medium. Expression wasinduced with 0.1 mM IPTG when the culture density (A600)reached 0.6 – 0.8 and proceeded at 30 °C for 6 h, after whichcells were harvested by centrifugation at 4,000 � g for 20 min at4 °C. The resulting pellets were stored at �80 °C overnight.Thawed pellets were resuspended in 15 ml buffer A (20 mM

sodium phosphate, 0.5 M NaCl and 20 mM imidazole, pH 7.4)and lysed with a high-pressure homogenizer (Avestin Emulsi-Flex C5). Lysates were cleared by centrifugation at 20,000 � gfor 20 min at 4 °C and then subjected to Ni2�-affinity purifica-tion with an ÅKTA FPLC using a 1-ml HisTrap column (GEHealthcare). Samples were washed with 10% buffer B (20 mM

sodium phosphate, 0.5 M NaCl and 500 mM imidazole, pH 7.4)before elution with 50% buffer B. Purified proteins weredesalted over a 5-ml HiTrap column equilibrated with ubiquiti-nation reaction buffer (20 mM MOPs, 100 mM KCl, 1 mM DTT,5 mM MgCl2, pH 7.2).

In Vitro Ubiquitination Assay—Ubiquitination assays wereperformed as described previously (10) in the presence of 0.1�M purified human recombinant UBE1 (Boston Biochem), 2�M human recombinant UbcH5�/UBE2D1 (Boston Biochem),3 �M uAb (or equivalent control protein), 3 �M E. coli �-gal(Sigma), 50 �M human recombinant ubiquitin (BostonBiochem), 4 mM ATP, and 1 mM DTT in 20 mM MOPs, 100 mM

KCl, 5 mM MgCl2, pH 7.2. Reactions were carried out at 37 °Cfor 2 h (unless otherwise noted) and stopped by boiling in 2�Laemmli loading buffer for analysis by immunoblotting.

Mass Spectrometry Analysis—For LC-MS/MS sample prep-aration, ubiquitination assays were performed as described pre-viously (21) but with 0.1 �M UBE1, 20 �M UbcH5�, 20 �M

R4-uAb, 20 �M �-gal, and 500 �M ubiquitin. Reactions wereresolved by SDS-PAGE and stained by Coomassie prior to gelexcision. The protein bands were cut from an SDS-PAGE gel

and cut into �1-mm cubes. The gel bands were washed in 200�l of deionized water for 5 min, followed by 200 �l of 100 mM

ammonium bicarbonate/acetonitrile (1:1) for 10 min, andfinally 200 �l of acetonitrile for 5 min. The acetonitrile wasdiscarded, and the gel bands were dried in a speed-vac for 10min. The gel pieces were rehydrated with 70 �l of 10 mM DTTin 100 mM ammonium bicarbonate and incubated for 1 h at56 °C. The samples were allowed to cool to room temperature,after which 100 �l of 55 mM iodoacetamide in 100 mM ammo-nium bicarbonate was added and the samples were incubated atroom temperature in the dark for 60 min. Following incubation,the gel slices were again washed as described above. The gelslices were dried and rehydrated with 50 �l of trypsin at 50ng/�l in 45 mM ammonium bicarbonate and 10% acetonitrileon ice for 30 min. The gel pieces were covered with an addi-tional 25 �l of 45 mM ammonium bicarbonate and 10% aceto-nitrile, and incubated at 37 °C for 19 h. The digested peptideswere extracted twice with 70 �l of 50% acetonitrile, 5% formicacid (vortexed 30 min and sonicated 10 min) and once with 70�l of 90% acetonitrile, 5% formic acid. Extracts from each sam-ple were combined and lyophilized.

The lyophilized in-gel tryptic digest samples were reconsti-tuted in 20 �l of nanopure water with 0.5% formic acid fornanoLC-ESI-MS/MS analysis, which was carried out by a LTQ-Orbitrap Velos mass spectrometer (Thermo-Fisher Scientific)equipped with a CorConneX nano ion source device (CorSolu-tions LLC). The Orbitrap was interfaced with a nano HPLCcarried out by an UltiMate3000 UPLC system (Dionex). Thegel extracted peptide samples (2– 4 �l) were injected onto aPepMap C18 trap column-nano Viper (5 �m, 100 �m � 2 cm,Thermo Dionex) at 20 �l/min flow rate for online desalting andthen separated on a PepMap C18 RP nanocolumn (3 �m,75 �m � 15 cm, Thermo Dionex) which was installed in the “Plugand Play” device with a 10-�m spray emitter (NewObjective). Thepeptides were then eluted with a 90-min gradient of 5% to 38%acetonitrile in 0.1% formic acid at a flow rate of 300 nl/min. TheOrbitrap Velos was operated in positive ion mode with nanosprayvoltage set at 1.5 kV and source temperature at 275 °C. Internalcalibration was performed with the background ion signal at m/z445.120025 as the lock mass. The instrument was operated in par-allel data-dependent acquisition mode using FT mass analyzer forone survey MS scan for precursor ions followed by MS/MS scanson top 7 highest intensity peaks with multiple charged ions abovea threshold ion count of 7500 in both LTQ mass analyzer andhigh-energy collision dissociation (HCD)-based FT mass analyzerat 7,500 resolution. Dynamic exclusion parameters were set atrepeat count 1 with a 15-s repeat duration, exclusion list size of500, 30-s exclusion duration, and �10 ppm exclusion mass width.HCD parameters were set at the following values: isolation widthof 2.0 m/z, normalized collision energy of 35%, activation Q at 0.25,and activation time of 0.1 ms. All data were acquired using Xcali-bur operation software (version 2.1, Thermo-Fisher Scientific).

All MS and MS/MS raw spectra were processed and searchedusing Proteome Discoverer 1.3 (PD1.3, Thermo-Fisher Scien-tific) against databases downloaded from the NCBI database.The database search was performed with two-missed cleavagesite by trypsin allowed. The peptide tolerance was set to 10ppm, and MS/MS tolerance was set to 0.8 Da for collision-

Selective Protein Knock-out Using Engineered Ubiquibodies

7846 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 289 • NUMBER 11 • MARCH 14, 2014

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

induced dissociation and 0.05 Da for HCD. A fixed carbam-idomethyl modification of cysteine, variable modifications onmethionine oxidation, and ubiquitin modification of lysinewere set. The peptides with low confidence score (with anXcorr score 2 for doubly charged ion and 2.7 for triplycharged ion) defined by PD1.3 were filtered out, and theremaining peptides were considered for the peptide identifica-tion with possible ubiquitination determinations. All MS/MSspectra for possibly identified ubiquitination peptides from ini-tial database searching were manually inspected and validatedusing both PD (version 1.3) and Xcalibur (version 2.1) software.

Cell Culture, Transfection, and Lysate Preparation—All celllines were obtained from ATCC and cultured in standard

medium at 37 °C with 5% CO2. HEK293T and COS-7 cells werecultured in DMEM with 10% heat inactivated FBS and 1% anti-biotic-antimycotic (Cellgro). BHK21 cells were cultured inEMEM with 10% heat-inactivated FBS and 1% antibiotic-anti-mycotic (Cellgro). Cells were transfected in six-well dishes at60 – 80% confluency with 2 �g of total plasmid DNA (293T andBHK21) or 1 �g of total plasmid DNA (COS7) using emptypcDNA3 plasmid to balance all transfections. For jetPRIME�transfection (Polyplus Transfection), a 1:2 (293T) or 1:3(BHK21 and COS7) ratio of jetPRIME® to DNA was used (w/v),and at 4 h post-transfection, the growth medium was refreshed.At 24 h post-transfection, cells were harvested by trypsiniza-tion, washed with PBS, and frozen at �20 °C until analyzed by

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 2.5 5 7.5 10 12.5

scFv13-R4

CHIPΔTPR

R4-uAb

D10-uAb

R4-uAbR272A

Abs

orba

nce

(492

nm

)

[Protein] (ng)

d

cCHIP

R4-uAbR272A

CHIPΔTPR

R4-uAb

D10-uAb

scFv1

3-uAb

5037

25

scFv1

3-R4

anti-FLAGkDa

YS1-uAb

20

10075

MW

E3* T

Proteasome

Ubiquitin

ATP

SE3

K48K48

K48

+ATP

Proteasome

AMP + PPi

K48K48

K48

+ATP

DBP

a

b

T

E1

uAb pathway

E2

N1

C303

TPR repeats U-boxCoiled-coil

176

436scFv13-R4

CHIP

Hsp70/Hsp90 E2

CHIPΔTPR

R4-uAb

128

GS

FIGURE 1. Remodeling CHIP for ubiquitination of unnatural target proteins. a, schematic of natural and engineered UPP systems. Normally, E1, E2, and E3mediate the transfer of ubiquitin to native substrates (S). Ubiquitin-tagged substrates are then degraded by the proteasome. In the engineered uAb pathway,a truncated E3 (E3*) is fused to a DBP, which remodels the specificity of E3* for an unnatural target protein (T). b, linear representation of CHIP, CHIP�TPR, andR4-uAb. Numbers refer to amino acid positions from N terminus (N) to C terminus (C). The proteins are aligned vertically with the coiled-coil and U-box domainsof CHIP. CHIP�TPR is a truncated version of CHIP lacking the TPR domain. R4-uAb was designed with an additional Gly-Ser (GS) linker connecting the scFv13-R4intrabody to CHIP�TPR. c, Western blot analysis of cell lysates derived from E. coli strain BL21(DE3) expressing full-length CHIP, CHIP�TPR, scFv13-R4, andseveral different uAb constructs. The uAbs were comprised of DBPs specific for E. coli �-gal (scFv13-uAb, R4-uAb, and R4-uAbR272A), bacteriophage gpD(D10-uAb) and E. coli MBP (YS1-uAb). An equivalent amount of total protein was loaded in each lane. Anti-FLAG antibodies were used to detect the differentproteins. d, binding activity of purified R4-uAb measured by ELISA with �-gal as antigen. The intrabody scFv13-R4 served as a positive control, whereasCHIP�TPR and D10-uAb served as negative controls. Also tested was R4-uAbR272A, a derivative of R4-uAb carrying a point mutation in the U-box domain thatis known to block the interaction between CHIP and the E2 enzyme, UbcH5�.

Selective Protein Knock-out Using Engineered Ubiquibodies

MARCH 14, 2014 • VOLUME 289 • NUMBER 11 JOURNAL OF BIOLOGICAL CHEMISTRY 7847

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

immunoblotting. Thawed cells were lysed in Nonidet P-40 lysisbuffer (150 mM NaCl, 1% Nonidet P-40 and 50 mM TrisHCl, pH7.4) by pipetting and mixing at 4 °C for 30 min. Soluble fractionswere obtained by centrifugation of lysed cells at 18,000 � g at4 °C for 20 min. The insoluble pellets were then washed with 50mM Tris-HCl and 1 mM EDTA, pH 8, followed by solubilizationin an equal volume of 2% SDS in PBS by boiling for 10 min.Cooled samples were centrifuged at room temp for 10 minat 18,000 � g to remove the remaining cellular debris, andthe lysates were collected as the insoluble fractions. Both solu-ble and insoluble fractions were boiled in 2� Laemmli samplebuffer for analysis by immunoblotting. Soluble lysates werenormalized using a detergent compatible total protein assay(Bio-Rad), and 10 �g of total protein was loaded with equivalentvolume of insoluble fractions for immunoblotting comparison.

Protein Analysis—SDS-PAGE and immunoblotting of pro-teins was performed according to standard procedures. Bio-Rad Coomassie G-250 stain was used to visualize proteins inSDS-PAGE (Bio-Rad, Mini-PROTEAN� TGX). The followingprimary antibodies were utilized for immunoblotting: rabbitanti-�-gal (Abcam, ab616), mouse anti-ubiquitin (Millipore,P4D1-A11), rabbit anti-Lys-48 (Millipore, Apu2), rabbit anti-His6-HRP (Abcam, ab1187), mouse anti-GAPDH (Millipore,6C5), mouse anti-FLAG®-HRP (Sigma, M2), mouse anti-Hsp70(Enzo Life Sciences, C92F3A), mouse anti-MBP-HRP (NewEngland Biolabs, E8038). Secondary antibodies goat anti-rabbitand anti-mouse IgGs with HRP conjugation (Promega) wereutilized as needed. For ELISA, a previously established protocolwas used to detect binding to �-gal (22) with slight modification.Briefly, a 96-well enzyme immunoassay plate was coated with 100�l of �-gal (Sigma) at 10 �g/ml overnight at 4 °C. The platewas then washed three times with 200 �l of PBST (1�PBS�0.1% Tween 20) per well for 5 min with shaking and blockedwith 250 �l of PBS with 3% milk per well at room temperature,slowly mixing for 3 h. Following three washes as above, purifiedprotein samples were introduced in blocking buffer as serialdilutions with 60 �l per well and incubated at room tempera-ture, slowly mixing for 1 h. Three washes were used to removenon-bound protein before introducing 50 �l of anti-His6-HRP(diluted 1:10,000 in PBST � 1% milk) and incubating at roomtemperature with slow mixing for 1 h. Three final washes wereperformed before incubation with 200 �l of o-phenylenedi-amine dihydrochloride (SIGMAFASTTM tablets) in the dark for30 min. The reaction was then quenched with 50 �l of 3 N

H2SO4, and absorbance was read at 492 nm.Pulldown Assays—Thawed cells were lysed as above; then, clar-

ified lysates were normalized by a detergent compatible total pro-tein assay (Bio-Rad) and diluted to contain 0.75 mg/ml proteinwith 10 mM imidazole to reduce nonspecific binding. 200 �l ofdiluted lysates were incubated with 30 �l of nickel-nitrilotri-acetic acid magnetic agarose beads (Qiagen, 5% solution)mixing at 4 °C for 1–2 h, and washed with 20 mM imidazole in

100

150

250

75

50

100

150250

75

100

150

250

UbE1E2

R4-uAb +++_

++

+_

+

++

_ +++

_ ++++

scFv

13-R

4CHIP

TPR

D10-u

Ab

R4-uA

bR27

2A

50

37

75

25

anti- -gal

*

anti-K48

anti-Ub

anti-His6 kDa

MW

-gal-Ub conjugates

n

β-gal

Time (min): 0 5 15 30 60 120 MW

250

150

100

75

β-gal-Ub conjugates

n

anti-β-gal

kDa

a

b

FIGURE 2. In vitro ubiquitination of �-gal by engineered ubiquibodies. a,in vitro ubiquitination of �-gal in the presence of R4-uAb. At the indicatedtimes, reactions were stopped by boiling and immunoblotted with anti-�-galantibodies. b, ubiquitination of �-gal was evaluated in the presence (�) orabsence (�) of each ubiquitin pathway component, namely ubiquitin (Ub),E1, E2, and R4-uAb as E3. Controls included scFv13-R4, CHIP�TPR,

R4-uAbR272A, and D10-uAb. An equivalent amount of total protein was addedto each lane. Immunoblots were probed with anti-�-gal, anti-ubiquitin, anti-Lys-48, and anti-His6 antibodies. Protein bands corresponding to �-gal,mono-ubiquitinated �-gal (*), and �-gal-ubiquitin conjugates as well as themolecular weight of the marker bands (MW) are indicated. The results arerepresentative of at least three replicate experiments.

Selective Protein Knock-out Using Engineered Ubiquibodies

7848 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 289 • NUMBER 11 • MARCH 14, 2014

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

lysis buffer. Bound proteins were eluted with 250 mM imid-azole (50 mM Tris-HCl, pH 7.9, and 50 mM NaCl) andboiled in 2� Laemmli sample buffer for analysis byimmunoblotting.

�-Gal Activity Assay—�-Gal activity was determined using a�-gal assay kit (Invitrogen) according to the manufacturer’sinstructions for the microtiter plate format. Briefly, cell pelletswere resuspended in 1� lysis buffer (0.25 M Tris, pH 8.0) and lysedwith three freeze-thaw cycles before clarification at 18,000 � g for5 min at 4 °C. Then, 2.5 �l of HEK293T lysate (or 10 �l of COS7 orBHK21 lysate) was added to a well containing 50 �l of 1� cleavagebuffer (0.1 M sodium phosphate, 10 �M KCl, 1 �M MgSO4-7H2O,pH 7) with �-mercaptoethanol and 17 �l of o-nitrophenyl-�-D-galactopyranoside (4 mg/ml). Reactions proceeded at 37 °C for 30min and were stopped with 125 �l of stop buffer (1 M sodiumcarbonate) before measuring absorbance at 420 nm. The amount

of o-nitrophenyl-�-D-galactopyranoside hydrolyzed was calcu-lated using the following formula: nmol o-nitrophenyl-�-D-galac-topyranoside hydrolyzed (A420)(1.92 � 105 nl)/(4500 nl/nmol-cm)(1 cm). Specific activity was determined according to thefollowing formula: specific activity o-nitrophenyl-�-D-galacto-pyranoside hydrolyzed/t/mg protein, where t is the reaction timein minutes and mg is the amount of total protein assayed. For eachsample, the background activity from untransfected cell lysateswas subtracted, and specific activities within biological sampleswere normalized to cells transfected with �-gal alone.

RESULTS

Remodeling the Substrate Specificity of CHIP Using Intra-bodies—CHIP is a constitutively active E3 ubiquitin ligase, con-sisting of an N-terminal tetratricopeptide repeat (TPR)domain, a coiled-coil domain, and a C-terminal U-box ligase

a

250

150

10075

50

Time (min): MW 0 15 30 45 60

R4-uAb

β-gal-Ub conjugates

R4-uAb-Ubconjugates

β-gal

250

150

10075

50

Ubiquitination in theabsence of β-gal

0 15 30 45 60 MW

n

n

Ubiquitination of β-gal

kDakDa

K662

K348

K518

K774 K775

200 400 600 800 1000m/z

0

10

20

30

40

50

60

70

80

90

100 712.48

385.38

486.34

599.39

698.43 371.33

354.31 840.50 175.14 680.39 467.38 581.37 908.49

KGG Q L L T P L R

Rel

ativ

e ab

unda

nce

b2

y2

y3

y4

y5

y7y1b7

b5

288.35

y6

484.41 b3

597.37 b4

261.20 b1

b 1 b 2 b 3 b 4 b 5 b 6 b 7

y 7 y 6 y 5 y 4 y 3 y 2 y 1

FIGURE 3. Isolation of ubiquitinated �-gal. a, Coomassie-stained SDS-PAGE analysis of R4-uAb-mediated ubiquitination reactions in the presence and absence ofE. coli �-gal at various times after initiation. Protein bands corresponding to unmodified �-gal, R4-uAb, various ubiquitin conjugates, and the molecular weight of themarker bands (MW) are indicated. Trypsin digest and subsequent LC-MS/MS analysis was performed on the protein bands delineated by the red box. b, mapping ofubiquitinated lysines (red spheres) onto crystal structure of a single �-gal monomer (blue). All identified lysines are solvent exposed in the homotetramer (gray) modelgenerated in PyMOLTM using Protein Data Bank code 1DP0. c, MS/MS spectra of a representative �-gal peptide 775KQLLTPLR782, containing a Gly-Gly modified(ubiquitinated) lysine residue. The modified lysine residue is labeled with GG and corresponds to Lys-775 in �-gal. Fractionation of the peptides into b and yions was performed, and the corresponding peaks are labeled on the spectra. The y axis, relative abundance, was normalized to the most abundant identifiedpeptide fragment.

Selective Protein Knock-out Using Engineered Ubiquibodies

MARCH 14, 2014 • VOLUME 289 • NUMBER 11 JOURNAL OF BIOLOGICAL CHEMISTRY 7849

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

domain (23). The TPR domain of CHIP binds to the molecularchaperones Hsc70-Hsp70 and Hsp90, facilitating ubiquitina-tion of chaperone-bound client proteins (24). In this fashion,CHIP uses the chaperone as an adaptor to target a variety ofsubstrates for ubiquitination and degradation by the protea-some. In the absence of substrates, CHIP is able to ubiquitinatethe chaperones directly (25). This substrate diversity suggestedto us that CHIP specificity could be remodeled using antigen-specific DBPs without affecting its E3 ubiquitin ligase activity(Fig. 1a). To test this hypothesis, we removed the entire TPRdomain of CHIP (CHIP�TPR) and replaced it with a scFvintrabody called scFv13-R4, which specifically binds �-ga-lactosidase (�-gal) (Fig. 1b) (4). The resulting R4-uAb fusionwas expressed and purified from E. coli cells. Immunoblottinganalysis revealed that R4-uAb was expressed as a stable fusionprotein, whereas expression of a fusion with the parental scFv13clone, which was not optimized for intracellular expression (4)could not be detected (Fig. 1c). Hence, intrabodies, but not con-ventional scFvs, are a suitable format for uAb development.Importantly, purified R4-uAb was observed to bind �-gal aseffectively as scFv13-R4 alone (Fig. 1d), indicating that fusion toCHIP�TPR does not affect antigen-binding activity of theintrabody domain. In contrast, no significant �-gal binding wasseen for CHIP�TPR or a control uAb comprised of the D10intrabody (D10-uAb), which is specific for the bacteriophagegpD protein (Fig. 1d) (26).

Next, we performed in vitro ubiquitination assays with puri-fied components, including R4-uAb as the E3 enzyme and �-galas the target (note that �-gal has 20 lysine residues that serve aspotential ubiquitin attachment sites). UbcH5� was used as theE2 enzyme because it has previously been shown to functionwith CHIP in vitro (10). High molecular weight bands corre-sponding to ubiquitinated �-gal were observed, especially dur-ing longer incubation times (Fig. 2a), which was characteristicof CHIP-mediated polyubiquitination of its native targets (10).These results confirm that the CHIP�TPR domain retained E3ligase activity in the context of the chimera. Ubiquitination onlyproceeded when all pathway components were included in thereaction (Fig. 2b), indicating that R4-uAb activity was depen-dent on a complete ubiquitination pathway. Similar polyubiq-uitination was detected with an antibody specific for Lys-48-linked polyubiquitin chains (Fig. 2b), confirming the presenceof ubiquitin linkages that are known to signal proteasomal deg-radation (27). Importantly, when reactions were performedwith scFv13-R4, CHIP�TPR, or the nonspecific D10-uAb,there was no detectable ubiquitination of �-gal (Fig. 2b). Like-wise, R4-uAbR272A, which carries an R272A substitution knownto inhibit E2 binding (28), failed to conjugate ubiquitin (Fig. 2b),even though it was capable of binding �-gal (Fig. 1d).

Identification of Lysine Residues Modified by Anti-�-galUbiquibody—We next investigated which lysine residues wereubiquitinated by our engineered R4-uAb fusion. In vitro ubiq-uitination of �-gal was monitored by SDS-PAGE, and the for-mation of higher molecular weight species was clearly evidentin the region of the gel where ubiquitinated �-gal would beexpected to resolve (Fig. 3a). We also detected R4-uAb-ubiqui-tin conjugates, consistent with earlier reports showing autou-biquitination of CHIP and CHIP fusions (10). However at early

time points, these were at lower molecular weight regions of thegel. Therefore, bands on the gel corresponding to modified�-gal after 15 min of ubiquitination, were excised, digested withtrypsin, and analyzed by liquid chromatography-tandem mass

150

100

50

37

25

37

R4-uAbD10

-uAb

empty

vecto

r

scFv1

3-R4

β-gal

only

MW

GAPDH

anti-GAPDH

β-gal

anti-β-gal

uAb

scFv

anti-His6

50

75Hsp70

anti-Hsp70

R4-uAbR

272A

kDa

150

100

anti-β-gal

insoluble β-gal

50

25

a

b

Nor

mal

ized

spe

cific

act

ivity

1 0.58 0.08 0.09 0.03 0.93 0.20 1.19 0.02 β-gal remaining

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4 COS7 BHK21 293T

β-gal only β-gal +R4-uAb

β-gal + scFv13-R4

β-gal + D10-uAb

FIGURE 4. R4-uAb-mediated proteolysis of �-gal in mammalian cells. a,immunoblots of soluble (top panels) or insoluble (bottom panel) fractions pre-pared from HEK293T cells transfected with pcDNA3-�-gal alone at 0.05 �g ofplasmid DNA per well (�-gal only) or co-transfected with pcDNA3-�-galalong with one of the following: pcDNA3-R4-uAb, pcDNA3-D10-uAb,pcDNA3-R4-uAbR272A, or pcDNA3-scFv13-R4 (each transfected at 1.25 �gof plasmid DNA per well). The triangle indicates increasing amounts ofpcDNA3-R4-uAb plasmid DNA (0.05, 0.25, 0.75, and 1.25 �g of plasmid DNAper well) used to transfect cells. The percentages of �-gal remaining in eachsample were quantitated by densitometry scanning and are indicated. Blotswere probed with antibodies specific for �-gal, His6, GAPDH, and Hsp70 asindicated. An equivalent amount of total protein was loaded in each lane, asconfirmed by immunoblotting with anti-GAPDH antibodies. The immunoblotresults are representative of at least three replicate experiments. b, �-galactivity measured in samples described in a and in Fig. 5. The activity reportedfor pcDNA3-R4-uAb corresponds to the highest co-transfection level in 293Tand the lowest co-transfection level in COS7 and BHK21. Data were normal-ized to the signal for the �-gal-only control and is expressed as the mean �S.D. of biological triplicates.

Selective Protein Knock-out Using Engineered Ubiquibodies

7850 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 289 • NUMBER 11 • MARCH 14, 2014

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

spectrometry (LC-MS/MS). Peptides corresponding to 80% ofthe �-gal sequence were identified using Mascot software.Trypsin digestion of a ubiquitinated protein leaves the C termi-nus of ubiquitin, Gly-Gly, attached to the ubiquitinated lysineresidue. Therefore, we searched the MS data for such modifi-cation of �-gal lysines and found five modified residues: Lys-348, Lys-518, Lys-662, Lys-774, and Lys-775. All five lysines aresolvent accessible (Fig. 3b), consistent with the location of ubiq-uitin attachment sites on native CHIP substrates Hsp70 andHsp90 (21). A representative MS/MS spectrum of a �-gal pep-tide that includes the identified ubiquitination site Lys-775 isdepicted in Fig. 3c.

Peptides corresponding to 95% of the ubiquitin sequencewere also identified in the MS experiments. Specifically, Lys-6,Lys-11, Lys-48, and Lys-63 residues were found to be ubiquiti-nated in peptides isolated from our high molecular weight sam-ples, suggesting these chain linkages in the polyubiquitinationof �-gal by R4-uAb (data not shown). These same chain link-ages were observed previously on natural substrates that hadbeen ubiquitinated by full-length CHIP in vitro (21). It is worthnoting that although Lys-48-linked chains are considered thecanonical linkage associated with targeting proteins for protea-somal degradation, all linkages identified here, including Lys-6,

Lys-11, and Lys-63, have been implicated as targeting signalsfor the 26S proteasome (29, 30).

Ectopic Expression of R4-uAb Fusion Mediates ProteasomalDegradation—We next investigated whether R4-uAb-medi-ated ubiquitination would result in �-gal depletion by the UPPin mammalian cells. First, HEK293T cells were transiently co-transfected with pcDNA3-based plasmids encoding E. coli�-gal and the R4-uAb, where each construct was under controlof the strong human CMV promoter. Next, cellular �-gal levelswere measured by immunoblotting and by determining �-galactivity 24 h post-transfection. When HEK293T cells were co-transfected with pcDNA3-�-gal and increasing amounts ofpcDNA3-R4-uAb, the �-gal levels were systematically reducedto as low as 3% of the steady-state levels measured in cells trans-fected with only the pcDNA3-�-gal plasmid (Fig. 4a). In con-trast, no reduction in �-gal expression or activity was observedfollowing co-transfection with the pcDNA3-D10-uAb orpcDNA3-scFv13-R4 plasmids (Fig. 4, a and b). Interestingly,cells co-transfected with R4-uAbR272A exhibited an intermedi-ate level of �-gal expression (Fig. 4a), indicating that this pointmutation in the U-box domain of CHIP inhibits some but notall E2 interactions in vivo. Importantly, the levels of a house-keeping protein, GAPDH, and a native binding partner of full-

50

37

150

100

25

37

kDa

150

100

50

25

50

37

150

100

25

37

kDa

150

100

50

25

R4-uAbD10

-uAb

empty

vecto

r

scFv1

3-R4

β-gal

only

MW

GAPDHanti-GAPDH

β-gal

uAb

scFv

anti-His6

anti-β-gal

insoluble β-gal

D10-uA

b

empty

vecto

r

scFv1

3-R4

β-gal

only

MW

GAPDHanti-GAPDH

β-gal

uAb

scFv

anti-His6

anti-β-gal

insoluble β-gal

BHK21 cells COS-7 cells

1 0.09 0.17 0.13 0.99 0.96 0.09 β-gal remaining 1 0.05 0.22 0.09 1.53 1.41 0.02 β-gal remaining

anti-β-gal

R4-uAb

anti-β-gal

FIGURE 5. R4-uAb-mediated proteolysis of �-gal in different cell lines. Immunoblots of extracts prepared from BHK21 (left) and COS-7 (right) cells trans-fected with pcDNA3-�-gal alone at 0.05 �g of plasmid DNA per well (�-gal only) or co-transfected with pcDNA3-�-gal along with one of the following:pcDNA3-R4-uAb, pcDNA3-scFv13-R4, or pcDNA3-D10-uAb (each transfected at 1.25 �g of plasmid DNA per well). The triangle indicates increasing amounts ofpcDNA3-R4-uAb plasmid DNA (0.05, 0.25, and 0.75 �g of plasmid DNA per well) used to transfect cells. The percentages of �-gal remaining in each sample werequantitated by densitometry scanning and are indicated. Blots were probed with antibodies specific for �-gal, His6, and GAPDH as indicated. An equivalentamount of total protein was loaded in each lane, as confirmed by immunoblotting with anti-GAPDH antibodies. The immunoblot results are representative oftwo replicate experiments.

Selective Protein Knock-out Using Engineered Ubiquibodies

MARCH 14, 2014 • VOLUME 289 • NUMBER 11 JOURNAL OF BIOLOGICAL CHEMISTRY 7851

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

length CHIP, Hsp70, were not affected by co-transfections (Fig.4a). Evaluation of insoluble fractions confirmed that �-galdepletion was not due to partitioning into aggresomes,aggresome-like induced structures (31, 32), or inclusion bodies(Fig. 4a). Similar silencing results for co-transfection ofpcDNA3-�-gal and pcDNA3-R4-uAb were obtained in BHK21and COS-7 cells (Fig. 4b and 5), indicating that target proteinknock-out by engineered ubiquibodies is transferable betweendifferent cell lines.

To confirm that R4-uAb specifically binds and ubiquitinates�-gal in vivo, interactions involving �-gal were determinedusing a pulldown assay. HEK293T cells were transiently trans-fected with pcDNA3-R4-uAb or pcDNA3-�-gal alone, orco-transfected with pcDNA3-�-gal and pcDNA3-R4-uAb,pcDNA3-scFv13-R4, or pcDNA3-D10-uAb, which each con-tained C-terminal His6 tags. As expected, both R4-uAb andscFv13-R4, but not D10-uAb, co-precipitated �-gal (Fig. 6). Fur-thermore, high molecular weight proteins co-precipitated byR4-uAb were observed to cross-react with an anti-ubiquitin anti-body (Fig. 6), suggesting that ubiquitinated �-gal was present incells co-transfected with pcDNA3-R4-uAb. These results suggestthat �-gal depletion in mammalian cells results from specific tar-get binding and ubiquitination by the engineered R4-uAb.

An FN3-based Ubiquibody Promotes Degradation of Malt-ose-binding Protein—To demonstrate the generality and mod-ularity of our approach, we sought to create a new ubiquibodyagainst a different target antigen by remodeling the substratespecificity of CHIP with a small antibody mimetic domainknown as a monobody. Specifically, we fused an FN3 mono-body called YS1 (MBP-74) that is specific for E. coli MBP (33) tothe N terminus of CHIP�TPR, resulting in the ubiquibody YS1-uAb. Co-transfection of HEK293T cells with pcDNA3-MBPand increasing amounts of pcDNA3-YS1-uAb systematicallyreduced MBP levels to as low as 7% of the steady-state levelsmeasured in cells transfected with only the pcDNA3-MBP plas-mid (Fig. 7, a and b). For the lowest level of MBP gene dosage(0.01 �g of plasmid DNA per well), YS1-uAb-mediated silenc-ing was equally effective over all dosages of ubiquibody plasmidDNA (Fig. 7a). As the MBP dosage was increased (0.05 and 0.1 �gof plasmid DNA per well), the extent of knockdown correlatedlinearly with ubiquibody gene dosage, ranging from 80% to below10% of the target protein remaining in cells (Fig. 7c). Thus, variabledegrees of knockdown were achieved by simply changing theratio of pcDNA3-MBP plasmid to pcDNA3-YS1-uAb plasmid,demonstrating the potential for tuning ubiquibody knockdownbased on gene dosage. In contrast, the Hsp70 levels wereunchanged in the presence of YS1-uAb (Fig. 7), demonstratingthe remodeled specificity of CHIP�TPR. When HEK293T cellswere co-transfected with control plasmids expressing unfusedYS1, R4-uAb, or D10-uAb, MBP depletion was not observed(Fig. 7), confirming that YS1-uAb-mediated proteolysis ishighly specific and CHIP�TPR-dependent.

DISCUSSION

We have shown that engineered ubiquibodies are a general-izable platform for directing otherwise stable proteins to theUPP for degradation. Ubiquibodies exploit the modular archi-tecture of human CHIP, an E3 ubiquitin ligase with exquisite

structural flexibility that accommodates substrates with differ-ent sizes and structures (10). Specifically, we swapped theN-terminal TPR domain of CHIP, which is responsible for

β-ga

l only

β-ga

l + R

4-uA

b

β-ga

l + sc

Fv13

-R4

β-ga

l + D

10-u

Ab

MW R4-uA

b only

100

150

anti-β-gal

β-gal

100

150

250

75

pull-

dow

n : 6

x-H

isanti-Ub

50

37

anti-His6

uAb

scFv

100

150

50

37

37

kDa

extra

cts

GAPDH

anti-β-gal

β-gal

anti-GAPDH

uAb

scFv

β-gal-Ubnconjugates

anti-His6

FIGURE 6. Co-precipitation of ubiquitinated �-gal in HEK293T cells. Immu-noblots of pulldown samples or extracts prepared from HEK293T cells trans-fected with pcDNA3-R4-uAb alone (R4-uAb only), pcDNA3-�-gal alone (�-galonly), or co-transfected with pcDNA3-�-gal along with one of the following:pcDNA3-scFv13-R4, pcDNA3-R4-uAb, or pcDNA3-D10-uAb. Pulldown was per-formed by subjecting extracts to nickel-nitrilotriacetic acid magnetic agarosebeads followed by immunoblotting with antibodies specific for �-gal, ubiquitin,GAPDH, and His6 as indicated. An equivalent amount of total protein was loadedin each lane, as confirmed by immunoblotting with anti-GAPDH antibodies.Immunoblots are representative of three replicate experiments.

Selective Protein Knock-out Using Engineered Ubiquibodies

7852 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 289 • NUMBER 11 • MARCH 14, 2014

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

R = 0.83732

R = 0.7595 0.0

0.2

0.4

0.6

0.8

1.0

0 5 10 15 20 25 30 35 40

50

37

kDa

50

37

25

25

50

37

YS1-uAbR4-u

Ab

empty

vecto

r

YS1MBP on

ly

MW

37 GAPDHanti-GAPDH

MBP

anti-MBP

YS1-uAb

YS1anti-His6

50

75Hsp70

anti-Hsp70

anti-MBP

insoluble

1 0.08 0.17 0.23 0.07 0.86 0.38 0.05 MBP remaining

R4-uAb

kDa37

50

37

50

37 MBP

control uAbs

YS1-uAb

YS1-uAbD10

-uAb

no tra

nsfec

tion

R4-uAb

MBP only

GAPDH

anti-MBP

anti-6x-His

anti-GAPDH

MW

1 0.42 0.05 0.09 0.90 0.73 0.99 0 MBP remaining

50

75 Hsp70

anti-Hsp70

YS1

25

20

YS1

anti-His6

Frac

tion

MB

P R

emai

ning

Ratio of YS1-uAb DNA to MBP DNA

c

a

0.05 μg MBP plasmid0.10 μg MBP plasmid

b

( )( )

FIGURE 7. YS1-uAb-mediated proteolysis of MBP in mammalian cells. a and b, representative immunoblots of soluble or insoluble (bottom panel of a)fractions prepared from HEK293T cells transfected with pcDNA3-MBP alone (MBP only at 0.01 (a) or 0.1 �g (b) of plasmid DNA per well) or co-transfected withpcDNA3-MBP along with one of the following: pcDNA3-YS1-uAb, pcDNA3-YS1, pcDNA3-R4-uAb, or pcDNA3-D10-uAb (each transfected at 1.75 �g of plasmidDNA per well). The triangles indicate increasing amounts of pcDNA3-YS1-uAb plasmid DNA (0.25, 0.75, 1.25, and 1.75 �g of plasmid DNA per well in a and 0.5,1.5, and 1.75 �g of plasmid DNA per well in b) used to transfect cells. The percentages of MBP remaining in each sample were quantitated by densitometryscanning and are indicated. Blots were probed with antibodies specific for MBP, His6, Hsp70, and GAPDH as indicated. An equivalent amount of total proteinwas loaded in each lane, as confirmed by immunoblotting with anti-GAPDH antibodies. The immunoblot results are representative of at least two replicateexperiments. c, scatter plot of densitometry analyses from five independent MBP knockdown experiments in HEK293T cells, where the level of MBP transfectionwas 0.05 or 0.1 �g of plasmid DNA per well as indicated. Linear regression analysis was performed on the data corresponding to the two different MBPtransfection levels and R2 values are shown.

Selective Protein Knock-out Using Engineered Ubiquibodies

MARCH 14, 2014 • VOLUME 289 • NUMBER 11 JOURNAL OF BIOLOGICAL CHEMISTRY 7853

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

native substrate binding, for a DBP of known specificity. Theresulting ubiquibodies were highly specific for their intendedtargets while avoiding Hsp70, a natural binding partner andpotential substrate of wild-type CHIP. Owing to the catalyticnature of ubiquitination, a single ubiquibody molecule has thepotential for elimination of numerous copies of its target pro-tein. This represents a major advantage over target inactivationusing binding proteins alone, which require a one-to-one stoi-chiometric ratio (or greater) with their target because there areno elimination pathways for binding protein-target complexes.Moreover, ubiquibodies based on CHIP do not rely on othersubunits for their functionality, making them self-sufficientcompared with targeted proteolysis methods that exploit mul-timeric SCF-type E3 ligases (12–14).

In this work, we attained significant knockdown in three dif-ferent cell lines (HEK293T, BHK21, and COS7) using transientco-transfection to deliver target and ubiquibody proteins. Theextent of silencing correlated fairly well with the dosage of theubiquibody plasmid DNA, with knockdown ranging from 80%to as little as 3% of the target protein remaining in cells. Never-theless, under the conditions tested here, complete target abla-tion was not observed. We speculate that this was due to (i)poor transfection efficiency associated with transient co-trans-fection (12) and (ii) the relatively high level of plasmid-basedtarget protein expression. We are optimistic that future silenc-ing of endogenous targets may result in complete target abla-tion given the lower expression level for most chromosomallyencoded gene products as well as the potential for improvedtransfection efficiency when the target protein plasmid is omit-ted. Transfection efficiency could be even further improved byinfection with different doses of recombinant adenovirusesexpressing ubiquibody genes. Along similar lines, better controlof knockdown might be possible through the use of tightlyregulated, externally inducible promoters to activate ubiqui-body expression. Alternatively, fine-tuning of silencingactivity might be possible through the creation of ubiqui-body “switches” whereby target binding or ubiquitin conjuga-tion activity is engineered to be dependent on a small moleculeligand (34) or even light (35). Such allosteric ubiquibodiescould potentially promote faster degradation because theirresponse would only require changes in proteins alreadypresent in the cell and would not be slowed by having to waitfor transcription/translation.

It should be pointed out that our approach was reminiscentof a previous study in which targeted silencing of the oncopro-tein c-Myc was accomplished by replacing the TPR domain ofCHIP with Max, a binding partner of c-Myc (16). In a notabledeparture from this earlier work, we devised a more generalstrategy for depleting target proteins that does not requireknowledge of the interaction network of a protein but insteadleverages the growing repertoire of pre-existing DBPs or theavailability of techniques for their rapid isolation (e.g. phagedisplay, cell surface display). In light of this design, a notableconstraint is the requirement that the DBP fold properly inthe intended subcellular compartment (e.g. cytoplasm).Indeed, when a traditional scFv antibody fragment was fused toCHIP�TPR, no detectable expression was observed. This isbecause correct folding and function of a prototypical scFv

requires disulfide bond formation (4), which is disfavored in thereducing cytoplasmic compartment. As demonstrated here,affinity domains that are engineered to function in the absenceof their native disulfide bonds (e.g. scFv-based intrabodies) orthat do not contain disulfide bonds such as small, highly stableantibody mimics (e.g. FN3, DAPRins, etc.) can easily overcomethis limitation. Another issue worth considering is the knownautoubiquitination of CHIP (10), which could result inunwanted depletion of the ubiquibody. However, even thoughpolyubiquitination of R4-uAb was consistently observed invitro, stable accumulation of all ubiquibodies was observed inmammalian cells, with expression levels increasing in a plasmiddose-dependent fashion. Nonetheless, it might be possible toeliminate autoubiquitination by mutating acceptor lysines inCHIP to residues that do not adversely affect the ability of CHIPto interact with cognate E2 and transfer ubiquitin to desiredsubstrates.

Domain swapping with different DBPs (e.g. scFv intrabody,FN3 monobody) confirmed the striking conformational flexi-bility of CHIP in ubiquitin transfer (10), targeting multipleacceptor lysines on two structurally distinct substrates withvaried distances and of continuously increasing size (due to thegrowing ubiquitin chain). These results also confirm the poten-tial of CHIP for customizable target degradation. Indeed, giventhe plethora of existing DBPs against known cellular targets andthe availability of robust technologies for on-demand isolationof new DBPs that function inside cells (26, 36, 37), ubiquibodiesare likely to become a powerful tool for reverse genetics. Inaddition to the wide array of endogenous proteins that can betargeted with newly designed ubiquibodies, the existingR4-uAb construct described here could be used in conjunctionwith �-gal protein trapping (38) to silence virtually any �-gal-tagged protein expressed from its endogenous loci in culturedcells and whole organisms. Furthermore, because ubiquibodiesoperate at the post-translational level and their DBP domainscan be created to bind specific protein conformations or post-translational modifications such as phosphorylation (39), theubiquibody technique has the potential for depleting certainprotein isoforms while sparing others. This is significantbecause post-translational modification-specific silencing ofproteins is not possible using current knockdown methods thatfunction at the level of DNA or RNA. Thus, ubiquibodies offera simple, reproducible, and customizable technique for selec-tively and controllably accelerating the turnover of numerousproteins and protein isoforms in somatic cells.

Acknowledgments—We thank Cam Patterson, Pierre Martineau,Andreas Plückthun, and Shohei Koide for kindly providing plasmidsencoding genes used in this study. We thank Wei Chen and ShengZhang at the Cornell Biotechnology Resource Center for performingthe mass spectrometry experiments and database searches.

REFERENCES1. Carlson, J. R. (1988) A new means of inducibly inactivating a cellular

protein. Mol. Cell. Biol. 8, 2638 –26462. Kawe, M., Forrer, P., Amstutz, P., and Plückthun, A. (2006) Isolation of

intracellular proteinase inhibitors derived from designed ankyrin repeatproteins by genetic screening. J. Biol. Chem. 281, 40252– 40263

Selective Protein Knock-out Using Engineered Ubiquibodies

7854 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 289 • NUMBER 11 • MARCH 14, 2014

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

3. Koide, A., Abbatiello, S., Rothgery, L., and Koide, S. (2002) Probing proteinconformational changes in living cells by using designer binding proteins:application to the estrogen receptor. Proc. Natl. Acad. Sci. U.S.A. 99,1253–1258

4. Martineau, P., Jones, P., and Winter, G. (1998) Expression of an antibodyfragment at high levels in the bacterial cytoplasm. J. Mol. Biol. 280,117–127

5. Bloom, L., and Calabro, V. (2009) FN3: a new protein scaffold reaches theclinic. Drug Discov. Today 14, 949 –955

6. Stocks, M. (2005) Intrabodies as drug discovery tools and therapeutics.Curr. Opin. Chem. Biol. 9, 359 –365

7. Stumpp, M. T., Binz, H. K., and Amstutz, P. (2008) DARPins: a new gen-eration of protein therapeutics. Drug Discov. Today 13, 695–701

8. Ciechanover, A. (1998) The ubiquitin-proteasome pathway: on proteindeath and cell life. EMBO J. 17, 7151–7160

9. Kerscher, O., Felberbaum, R., and Hochstrasser, M. (2006) Modificationof proteins by ubiquitin and ubiquitin-like proteins. Annu. Rev. Cell Dev.Biol. 22, 159 –180

10. Qian, S. B., Waldron, L., Choudhary, N., Klevit, R. E., Chazin, W. J., andPatterson, C. (2009) Engineering a ubiquitin ligase reveals conformationalflexibility required for ubiquitin transfer. J. Biol. Chem. 284, 26797–26802

11. Cardozo, T., and Pagano, M. (2004) The SCF ubiquitin ligase: insights intoa molecular machine. Nat. Rev. Mol. Cell Biol. 5, 739 –751

12. Zhang, J., Zheng, N., and Zhou, P. (2003) Exploring the functional com-plexity of cellular proteins by protein knockout. Proc. Natl. Acad. Sci.U.S.A. 100, 14127–14132

13. Caussinus, E., Kanca, O., and Affolter, M. (2012) Fluorescent fusion pro-tein knockout mediated by anti-GFP nanobody. Nat. Struct. Mol. Biol. 19,117–121

14. Sakamoto, K. M., Kim, K. B., Kumagai, A., Mercurio, F., Crews, C. M., andDeshaies, R. J. (2001) Protacs: chimeric molecules that target proteins tothe Skp1-Cullin-F box complex for ubiquitination and degradation. Proc.Natl. Acad. Sci. U.S.A. 98, 8554 – 8559

15. Zhou, P. (2005) Targeted protein degradation. Curr. Opin. Chem. Biol. 9,51–55

16. Hatakeyama, S., Watanabe, M., Fujii, Y., and Nakayama, K. I. (2005) Tar-geted destruction of c-Myc by an engineered ubiquitin ligase suppressescell transformation and tumor formation. Cancer Res. 65, 7874 –7879

17. Ma, Y., Gu, Y., Zhang, Q., Han, Y., Yu, S., Lu, Z., and Chen, J. (2013)Targeted degradation of KRAS by an engineered ubiquitin ligase sup-presses pancreatic cancer cell growth in vitro and in vivo. Mol. CancerTher. 12, 286 –294

18. Zheng, X. H., and Hughes, S. H. (1999) An avian sarcoma/leukosis virus-based gene trap vector for mammalian cells. J. Virol. 73, 6946 – 6952

19. Morin, X., Daneman, R., Zavortink, M., and Chia, W. (2001) A protein trapstrategy to detect GFP-tagged proteins expressed from their endogenousloci in Drosophila. Proc. Natl. Acad. Sci. U.S.A. 98, 15050 –15055

20. Cyr, D. M., Höhfeld, J., and Patterson, C. (2002) Protein quality control:U-box-containing E3 ubiquitin ligases join the fold. Trends Biochem. Sci27, 368 –375

21. Kundrat, L., and Regan, L. (2010) Identification of residues on Hsp70 andHsp90 ubiquitinated by the cochaperone CHIP. J. Mol. Biol. 395, 587–594

22. Contreras-Martínez, L. M., and DeLisa, M. P. (2007) Intracellular ribo-some display via SecM translation arrest as a selection for antibodies withenhanced cytosolic stability. J. Mol. Biol. 372, 513–524

23. Zhang, M., Windheim, M., Roe, S. M., Peggie, M., Cohen, P., Prodromou,C., and Pearl, L. H. (2005) Chaperoned ubiquitylation– crystal structures

of the CHIP U box E3 ubiquitin ligase and a CHIP-Ubc13-Uev1a complex.Mol. Cell 20, 525–538

24. Connell, P., Ballinger, C. A., Jiang, J., Wu, Y., Thompson, L. J., Höhfeld, J.,and Patterson, C. (2001) The co-chaperone CHIP regulates protein triagedecisions mediated by heat-shock proteins. Nat. Cell Biol. 3, 93–96

25. Qian, S. B., McDonough, H., Boellmann, F., Cyr, D. M., and Patterson, C.(2006) CHIP-mediated stress recovery by sequential ubiquitination ofsubstrates and Hsp70. Nature 440, 551–555

26. Koch, H., Gräfe, N., Schiess, R., and Plückthun, A. (2006) Direct selectionof antibodies from complex libraries with the protein fragment comple-mentation assay. J. Mol. Biol. 357, 427– 441

27. Pickart, C. M. (1997) Targeting of substrates to the 26S proteasome.FASEB J. 11, 1055–1066

28. Xu, Z., Devlin, K. I., Ford, M. G., Nix, J. C., Qin, J., and Misra, S. (2006)Structure and interactions of the helical and U-box domains of CHIP, theC terminus of HSP70 interacting protein. Biochemistry 45, 4749 – 4759

29. Xu, P., Duong, D. M., Seyfried, N. T., Cheng, D., Xie, Y., Robert, J., Rush, J.,Hochstrasser, M., Finley, D., and Peng, J. (2009) Quantitative proteomicsreveals the function of unconventional ubiquitin chains in proteasomaldegradation. Cell 137, 133–145

30. Saeki, Y., Kudo, T., Sone, T., Kikuchi, Y., Yokosawa, H., Toh-e, A., andTanaka, K. (2009) Lysine 63-linked polyubiquitin chain may serve as atargeting signal for the 26S proteasome. EMBO J. 28, 359 –371

31. Lelouard, H., Gatti, E., Cappello, F., Gresser, O., Camosseto, V., and Pierre,P. (2002) Transient aggregation of ubiquitinated proteins during dendriticcell maturation. Nature 417, 177–182

32. Kirkin, V., McEwan, D. G., Novak, I., and Dikic, I. (2009) A role for ubiq-uitin in selective autophagy. Mol Cell 34, 259 –269

33. Koide, A., Gilbreth, R. N., Esaki, K., Tereshko, V., and Koide, S. (2007)High-affinity single-domain binding proteins with a binary-code inter-face. Proc. Natl. Acad. Sci. U.S.A. 104, 6632– 6637

34. Wright, C. M., Wright, R. C., Eshleman, J. R., and Ostermeier, M. (2011) Aprotein therapeutic modality founded on molecular regulation. Proc. Natl.Acad. Sci. U.S.A. 108, 16206 –16211

35. Strickland, D., Moffat, K., and Sosnick, T. R. (2008) Light-activated DNAbinding in a designed allosteric protein. Proc. Natl. Acad. Sci. U.S.A. 105,10709 –10714

36. Waraho, D., and DeLisa, M. P. (2009) Versatile selection technology forintracellular protein-protein interactions mediated by a unique bacte-rial hitchhiker transport mechanism. Proc. Natl. Acad. Sci. U.S.A. 106,3692–3697

37. Karlsson, A. J., Lim, H. K., Xu, H., Rocco, M. A., Bratkowski, M. A., Ke, A.,and DeLisa, M. P. (2012) Engineering antibody fitness and function usingmembrane-anchored display of correctly folded proteins. J. Mol. Biol. 416,94 –107

38. Friedel, R. H., Plump, A., Lu, X., Spilker, K., Jolicoeur, C., Wong, K., Ven-katesh, T. R., Yaron, A., Hynes, M., Chen, B., Okada, A., McConnell, S. K.,Rayburn, H., and Tessier-Lavigne, M. (2005) Gene targeting using a pro-moterless gene trap vector (“targeted trapping”) is an efficient method tomutate a large fraction of genes. Proc. Natl. Acad. Sci. U.S.A. 102,13188 –13193

39. Kummer, L., Parizek, P., Rube, P., Millgramm, B., Prinz, A., Mittl, P. R.,Kaufholz, M., Zimmermann, B., Herberg, F. W., and Plückthun, A. (2012)Structural and functional analysis of phosphorylation-specific binders ofthe kinase ERK from designed ankyrin repeat protein libraries. Proc. Natl.Acad. Sci. U.S.A. 109, E2248 –E2257

Selective Protein Knock-out Using Engineered Ubiquibodies

MARCH 14, 2014 • VOLUME 289 • NUMBER 11 JOURNAL OF BIOLOGICAL CHEMISTRY 7855

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

Alyse D. Portnoff, Erin A. Stephens, Jeffrey D. Varner and Matthew P. DeLisaSpecificity for Targeted Protein Silencing

Ubiquibodies, Synthetic E3 Ubiquitin Ligases Endowed with Unnatural Substrate

doi: 10.1074/jbc.M113.544825 originally published online January 28, 20142014, 289:7844-7855.J. Biol. Chem. 

  10.1074/jbc.M113.544825Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/289/11/7844.full.html#ref-list-1

This article cites 39 references, 19 of which can be accessed free at

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from