DOI: 10.1002/ijch.201300063

Effective Molarity Redux : Proximity as a Guiding Force inChemistry and BiologyElissa M. Hobert,[a] Amy E. Doerner,[b] Allison S. Walker,[b] and Alanna Schepartz*[b, c]

1. Introduction

The interior of a cell is not unlike Grand Central Stationat 5 pm. Hordes of people (or molecules) in all shapesand sizes, each one (well, almost) guided by a defineddestination and departure time. Ensuring that everyonereaches their destination requires sophisticated mecha-nisms to impose temporal and spatial specificity on bothtransient interactions and eventual outcomes. Two strat-egies employed to regulate macromolecular interactionsin a cellular context are colocalization and compartmen-talization. Macromolecular interactions can be promotedand specified by localizing the partners within the samesubcellular compartment or by holding them in proximitythrough covalent or non-covalent interactions with pro-teins, lipids, or DNA (Figure 1). The net result of bothstrategies is an increase in effective molarity between sig-naling partners. In this review, we chose specific examplesto illustrate one mechanism employed by nature toensure the faithful passage of information: the power ofproximity to accelerate, guide, or otherwise influence thereactivity of signaling proteins and the information thatthey encode.

2. Increasing Effective Molarity Using NaturalProtein Domains

Many applications of effective molarity, especially thosethat employ natural protein domains, fall within the bur-

geoning field of synthetic biology, where researchers aimto gain insight into the molecular logic of signaling sys-tems by engineering novel pathways and studying theirbehavior. This sort of molecular engineering demandsboth a toolkit of modular parts as well as an understand-ing of how these elements can be combined reliably intofunctional units.[1] Once a detailed understanding of eachpart has been obtained, perturbations can be introducedthat illuminate the influence of factors like architecture,affinity, dynamics, and allostery on pathway output.Beyond understanding how signaling pathways are con-trolled, engineered pathways may have direct applications

Abstract : The cell interior is a complex and demanding envi-ronment. An incredible variety of molecules jockey to identi-fy the correct position – the specific interactions that pro-mote biology, which are often hidden among countless un-productive options. Ensuring that the business of the cell issuccessful requires sophisticated mechanisms to imposetemporal and spatial specificity – both on transient interac-tions and their eventual outcomes. Two strategies employedto regulate macromolecular interactions in a cellular contextare colocalization and compartmentalization. Macromolecu-lar interactions can be promoted and specified by localizing

the partners within the same subcellular compartment, orby holding them in proximity through covalent or non-cova-lent interactions with proteins, lipids, or DNA – themes thatare familiar to any biologist. The net result of these strat-egies is an increase in effective molarity: the local concen-tration of a reactive molecule near its reaction partners. Wewill focus on this general mechanism, employed by natureand adapted in the lab, which allows delicate control incomplex environments: the power of proximity to accelerate,guide, or otherwise influence the reactivity of signaling pro-teins and the information that they encode.

Keywords: chemical inducers of dimerization · effective molarity · protein-protein interactions · signal transduction

[a] E. M. HobertDepartment of Biological Chemistry and Molecular Pharmacolo-gyHarvard Medical SchoolBoston, MA 02115 (USA)

[b] A. E. Doerner, A. S. Walker, A. SchepartzDepartment of ChemistryYale UniversityNew Haven, CT 06511 (USA)e-mail: alanna.schepartz@yale.edu

[c] A. SchepartzDepartments of Chemistry and Molecular, Cellular and Develop-mental BiologyYale UniversityNew Haven, CT 06511 (USA)phone: (+1) 2034325094

Isr. J. Chem. 2013, 53, 567 – 576 � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 567

Review

to improve human health[2] and the production of com-mercial chemicals such as biofuels.[3]

2.1 Early Work

Early progress in this area made extensive use of natural-ly modular protein domains to rewire signaling path-ways.[1] Simply tagging a catalytic domain (often a kinase)with a novel targeting domain (an FKBP, SH3, or PDZdomain, for example) is often sufficient to redirect cata-lytic activity, effectively rewiring the signaling pathway.An early materialization of this idea was reported byHoward et al. , who engineered a novel adaptor protein toredirect signaling through the epidermal growth factor re-ceptor (EGFR) (Figure 2).[4] Normally, activated EGFRrecruits the adaptor protein Grb2 through an interactionbetween the SH2 domain of Grb2 and phosphotyrosine1068 on the EGFR C-terminal tail.[5] Grb2 then recruitsdownstream signaling proteins via its SH3 domains toinduce proliferation. To redirect the response to EGFRactivation, Howard et al. expressed a chimeric adaptor inwhich the SH2 domain of Grb2 was linked to the deatheffector domain (DED) of the FADD adaptor in place ofboth SH3 domains of Grb2. The FADD receptor normal-ly serves to link activation of the Fas receptor to down-stream apoptotic proteins. Indeed, expression of this chi-meric adaptor protein in mammalian cells rewired thecell signaling pathways so that activation of EGFR led toapoptosis rather than proliferation. Notably, this strategydid not require modification of the endogenous signalingprotein, EGFR.

2.2 Effective Molarity as a Tool in Metabolic Engineering

Harnessing key concepts found in nature, scaffold pro-teins have been engineered to increase the production of

Alanna Schepartz received her B.S. inChemistry from the State University ofNew York-Albany in 1982 and thenearned her Ph.D. for work with RonaldBreslow at Columbia University. From1986–1988 she was an NIH postdoc-toral fellow with Peter Dervan at Cal-tech. She has received a number ofawards, including the Frank H. West-heimer Prize Medal, the ACS ChemicalBiology Prize, and the ACS Ronald Bre-slow Award for Achievement in Bio-mimetic Chemistry. Her laboratory de-velops and applies chemical tools to study, manipulate, and under-stand protein�protein and protein�DNA interactions inside the cell.She is especially interested in how these interactions regulate infor-mation transfer and intracellular trafficking.

Elissa Hobert received her B.S. inChemistry in 2006 from the Universityof Wisconsin-Madison, where sheworked in the lab of Laura Kiessling.After obtaining her undergraduatedegree, she began graduate studies atYale in the laboratory of Alanna Sche-partz. Currently she is an NIH postdoc-toral fellow working with Jon Clardy inthe Department of Biological Chemistryand Molecular Pharmacology at Har-vard Medical School.

Amy Doerner received her B.A. inChemistry in 2010 from Whitman Col-lege. After obtaining her undergraduatedegree, she began graduate studies atYale in the laboratory of Alanna Sche-partz. Currently she is carrying out re-search on the mechanism of informa-tion transfer through receptor tyrosinekinases and on the design of allosteri-cally regulated protein adaptors.

Allison Walker received her B.S. inChemistry in 2013 from Brown Univer-sity. After obtaining her undergraduatedegree, she began graduate studies atYale in the laboratory of Alanna Sche-partz where she is an NSF PredoctoralFellow. Currently she is carrying out re-search on repurposing the ribosome tocatalyze the synthesis of unnaturalpolymers.

Figure 1. A slow bimolecular reaction, characterized by the rateconstant kintermolecular (units of M�1 sec�1) can be accelerated whenthe two reactants are held in close proximity. In this case, a unimo-lecular reaction ensues that is characterized by the rate constantkintramolecular (units of sec�1). The effective molarity of one substraterelative to the other in the latter case is defined by the ratio of thetwo rate constants and possesses units of M.

568 www.ijc.wiley-vch.de � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Isr. J. Chem. 2013, 53, 567 – 576

Review A. Schepartz et al.

desirable metabolites in microorganisms. Several naturalenzymes employ substrate channeling to increase metabo-lite production, including tryptophan synthase, carbamoylphosphate synthase and polyketide synthases.[6] Colocaliz-ing the necessary enzymes on one scaffold increases theeffective concentration of metabolic intermediates whilealso preventing their accumulation to toxic levels. Ina landmark example, Dueber and Keasling engineeredEscherichia coli to express a scaffold that organizes theenzymes necessary for the production of mevalonate, anisoprenoid precursor.[3] The chimeric scaffold possessedthree domains: SH3, PDZ and a GTPase-binding domain,and the peptide ligands for these domains were fused tothree kinetically relevant enzymes in the mevalonate bio-synthetic pathway. E. coli expressing the optimized scaf-fold and reaction components produced 77-fold more me-valonate than those lacking the scaffold. This early suc-cess established a strong foundation for the design andimplementation of synthetic systems capable of metabo-lite production.

Scaffolds can also be generated to assemble hydrolyticenzyme complexes capable of processing cellulose.[7] Innature, the enzymes responsible for cellulose hydrolysisare organized into the cellulosome, a multienzyme com-plex located on the bacterial cell surface.[8] The cellulo-some is organized around scaffoldin, a scaffold proteincomposed of a cellulose-binding module and several co-hesin domains. The cohesin domains bind to the dockerindomains of cellulases via a high-affinity (<10�9 M) inter-action.[7a,b] When compared to non-complexed cellulases,hydrolytic enzymes organized onto this scaffold increasethe rate of cellulose hydrolysis, particularly in the case ofrecalcitrant cellulose.[7c] The modular architecture of thecellulosome has enabled the engineering of synthetic cel-lulosomes that hydrolyze cellulose, with the eventual goalof producing biofuels.[7d,e] In one example, yeast that ex-press scaffolds for two copies each of endoglucanase andb-glucosidase exhibited a 4.2-fold enhancement in cellu-

lose hydrolysis when compared to yeast lacking the scaf-fold. Furthermore, cells displaying the engineered cellulo-some exhibited a twofold increase in ethanol productioncompared to cells displaying only single copies of endo-glucanase and b-glucosidase. The use of adaptor scaffol-dins is a flexible strategy to control the number of en-zymes present on the cell surface that could be generallyapplied toward other reactions. Engineering systemsbased around the dockerin-cohesin interaction has beenused to template the formation of other cascade enzymes,including those involved in glycolysis and gluconeogene-sis.[7f]

2.3 Directing Bacterial Two-Component Systems through NovelAdaptors

Bacterial two-component systems differ from eukaryoticsignaling pathways in that they generally do not relyupon adaptor and scaffold proteins to relay signals. Bac-terial two-component systems are organized around twoproteins: a histidine kinase and a response regulator. Thehistidine kinase senses an environmental signal, under-goes autophosphorylation, and then phosphorylates theresponse regulator, which usually responds by alteringtranscriptional activity.[9] Recently Whitaker et al. de-scribed an adaptor protein that exploited induced proxim-ity to redirect a histidine kinase to particular responseregulators (Figure 3).[10] The Taz histidine kinase wastagged with an SH3 domain and the response regulatorCpxR was tagged with a mutually orthogonal leucinezipper segment. A series of adaptor proteins were de-signed to associate with both the SH3 domain and the or-thogonal leucine zipper, effectively recruiting non-nativeresponse regulators to the histidine kinase. In bacterialacking native two-component pathways, the authors ob-served a 17-fold increase in gene activation over that ob-served with an adaptor lacking the SH3 ligand. The au-thors extended their preliminary results to create an auto-

Figure 2. Howard et al. reengineered EGFR signaling by creating a chimeric adaptor protein. A) In both EGFR and Fas signaling, a modularadaptor protein links activated receptor proteins to downstream signaling events. B) A chimeric adaptor protein was created that linkedEGFR activation to apoptosis.[4]

Isr. J. Chem. 2013, 53, 567 – 576 � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.ijc.wiley-vch.de 569

Proximity as a Guiding Force in Chemistry and Biology

inhibited histidine kinase that is functionalized with bothan SH3 domain for binding the adaptor and an SH3ligand that binds the SH3 domain in the absence of theadaptor. Association of the adaptor with the autoinhibit-ed histidine kinase releases the autoinhibitory interaction,allowing phosphotransfer to the response regulator. Thesuccess of this design demonstrates that colocalizing thetwo-component signaling proteins is sufficient to controlphosphotransfer specificity. Future experiments couldfocus on engineering two-component systems to senseand respond to non-natural signals, such as toxins.

2.4 Creating Novel Signaling Proteins through DomainRecombination

As an alternative to creating novel adaptor and scaffold-ing proteins to control protein-protein interactions, signal-ing enzymes themselves can be altered to rewire signalingpathways.[11] In a landmark example, Dueber et al. engi-neered the actin regulatory switch neuronal Wiskott�Al-drich syndrome protein (N-WASP) so that it was con-trolled by novel factors.[11a] Normally, N-WASP is held inan inactive conformation by two intramolecular interac-tions that both must be displaced to induce activity. Tochange the signaling properties of N-WASP, the endoge-nous autoinhibitory interactions were replaced witha PDZ domain and its peptide ligand and an SH3 domainand its ligand. The engineered N-WASP was activated inthe presence of both a PDZ peptide ligand and an SH3peptide ligand, recapitulating the AND gate that existsnaturally. The authors varied the affinities between thepeptide ligands and domains as well as the linker lengthsbetween domains to gain insight into the features neces-sary for repression and activation.

Similar concepts were applied to reprogram cell mor-phology by engineering synthetic guanine nucleotide ex-change factors (GEFs) that respond to unnatural signals,such as protein kinase A (PKA) phosphorylation.[11b]

GEFs promote the exchange of bound GDP to GTP to

activate Rho family members that are essential in main-taining the actin cytoskeleton. GEFs possess a modularstructure in which the catalytic domain is adjacent to anautoinhibitory domain. Yeh et al. took advantage of thismodularity by flanking the GEF catalytic domain withboth a PDZ domain and a PDZ peptide ligand. In the ab-sence of phosphorylation by PKA, the catalytic motif washeld in an autoinhibited form by the intramolecular inter-action between the PDZ domain and its ligand. Phos-phorylation of the PDZ ligand by PKA releases the intra-molecular interaction, activating the catalytic domain.Both of these studies take advantage of the natural mod-ularity of proteins and signaling pathways to engineer cel-lular signaling proteins capable of responding to unnatu-ral inputs. Engineering signaling proteins by domain re-combination has been particularly useful in understandingenzyme mechanisms and how enzymes function in signal-ing pathways.

2.5 Maximizing Precision

Researchers in the burgeoning field of synthetic biologyhave made good use of natural protein domains to exploiteffective molarity in the creation of novel synthetic sys-tems. As these natural domains are encodable, they canbe evolved to display new or improved signaling proper-ties. Because the interactions between natural protein do-mains and their ligands are often well studied, imple-menting these protein-ligand pairs in synthetic biologycontexts is straightforward and predictable. However, be-cause the parts they contain are natural, their specificitiesare in general not unique, leading to interactions withmany other signaling proteins inside the cell. These �off-target� interactions can result in undesirable and/or un-predictable outcomes because of interaction of proteinmodules with endogenous cellular proteins.[12] Further-more, this method requires the modification of the pro-tein of interest. As stated previously, the substitution ofendogenous protein domains with other natural proteindomains presents the possibility for unintended cross-talk.[12–13] One strategy for minimizing cross-talk makesuse of modular protein domains in organisms that natu-rally lack these domains. Other strategies, discussed inthe sections that follow, make use of small molecules, nu-cleic acid templates, and wholly unnatural protein do-mains.

3. Inducing Specificity with Small Molecules

3.1 Chemical Inducers of Dimerization

One strategy for introducing greater or alternative specif-icity into an engineered signaling pathway exploits smallmolecules – natural or designed – that bind two proteinpartners simultaneously. In classic work that ushered inan era of chemical biology research, Schreiber and Crab-

Figure 3. Bacterial two-component signaling can be directed withadaptor proteins designed to promote the interaction betweena histidine kinase and an unnatural response regulator. Ternarycomplex formation mediated by the adaptor promotes phospho-transfer from the histidine kinase to the response regulator, leadingto transcriptional activation.[10]

570 www.ijc.wiley-vch.de � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Isr. J. Chem. 2013, 53, 567 – 576

Review A. Schepartz et al.

tree developed this idea in a cellular context using mole-cules referred to as �chemical inducers of dimerization�(abbreviated as CID). In the initial report, the syntheticligand FK1012, containing two copies of the FKBP-bind-ing moiety FK506, effectively dimerized and thereby acti-vated T antigen cell surface receptors fused to FKBPmodules on the surface of living cells.[14] More recent de-velopments based on the interaction of rapamycin or itsanalogs with FKBP and FRB increased versatility by ex-panding the scope to include heterodimeric interactions.Myriad chemical inducers of dimerization have been usedto illuminate the role of proximity in biology and to engi-neer a diverse array of unnatural interactions and pro-cesses.[15] More recently, CID has been used to localizea protein of interest to a specific organelle (Figure 4).[16]

Chemical inducers of dimerization are attractive scaffoldsto increase effective molarity because they are cell per-meable, bind tightly to target proteins, and can diffusemore rapidly than larger biomolecules making it easier tocontrol cellular availability. However, like scaffolds basedon natural proteins, some small-molecule dimerizers canengage in crosstalk and off-target effects; one exampleare the inhibitory interactions of rapamycin with endoge-nous mTOR.[17] Use of a mutant FRB fusion protein anda rapalog that selectively binds to the mutant FRB overmTOR can minimize cross-talk.[18] Alternatively, the useof orthogonal fusion proteins, such as bacterial dihydrofo-late reductase (DHFR), or ligands that lack endogenousbinding partners can help.

3.1.1 Initiating Transcription

In a related way, the yeast three-hybrid system exploitsa designed, bifunctional small molecule to colocalize theDNA-binding and activation regions of a transcriptionfactor and thereby turn on transcription. The DNA-bind-ing domain is fused to one of the small-molecule partnerswhile the activation domain is fused to the other. Addi-tion of the small molecule guides the activation domainin proximity of the DNA-binding domain, bound toDNA, promoting transcription. In its first incarnation, thethree-hybrid system exploited an FK506-dexamethasonehybrid small molecule to identify FKBP12 from withina Jurkat cDNA library.[19] Since that time, the yeast three-hybrid system has been applied in many contexts, includ-ing the identification of novel enzyme catalysts[20] and asa screen to identify protein-ligand-protein interactionsuseful for small molecule�induced dimerization.[18a,21]

3.1.2 Probing Histone Modifications

A recent application of chemical inducers of dimerizationin the context of epigenetics focused on the posttransla-tional trimethylation of histone H3K9 (H3K9me3). Tocontrol H3K9me3, the chromatin shadow domain ofHP1a (csHP1a), a domain that recruits H3K to specifichistone methylases, was fused to FRB while a DNA-bind-ing domain was fused to FKB. This arrangement ensuresthat csHP1a is only recruited to chromatin when rapamy-cin is present. Using this system, it was discovered thatrecruitment of csHP1a represses gene expression and in-creases production of H3K9me3. The CID also facilitatedanalysis of how removing rapamycin and therefore stop-ping the recruitment of csHP1a affected repression ofgenes across generations of cells.[22]

3.2 Facilitating Complex Biology

Chemically induced dimerization has also been used togenerate orthogonal Boolean logic gates in living cells. Inrecent work, this goal was accomplished using two or-thogonal chemical dimerizers, rapamycin and a gibberellinanalog. Gibberellin works as a dimerizer by binding tothe protein gibberellin-insensitive dwarf1 (GID1) andcausing a change in conformation that allows GID1 tobind a second protein, gibberellin insensitive (GAI). Two�proteins of interest� can be fused to either GAI or GID1;dimerization is induced upon addition of gibberellin or anappropriate analog. The gibberellin analog used (GA3-AM) binds GID1 only after the acetoxymethyl group iscleaved off by an endogenous esterase. OR gates werecreated by expressing both GAI-effector and FRB-effec-tor fusion proteins; in this way, the cellular reponse canbe controlled by addition of rapamycin or GA3-AM.AND gates were created by fusing GAI to a localizationdomain, fusing FKBP to GID1, and fusing FRB to an ef-

Figure 4. Chemical inducers of dimerization. A) The interaction ofrapamycin with FKBP and FRB brings ‘protein of interest 1’ (POI1)into close proximity with ‘protein of interest 2’ (POI2) to initiatea cellular response. B) Application of CID to localize a protein of in-terest to a specific organelle.[16b]

Isr. J. Chem. 2013, 53, 567 – 576 � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.ijc.wiley-vch.de 571

Proximity as a Guiding Force in Chemistry and Biology

fector protein; in this way, both rapamycin and GA3-AMare required to localize the effector protein to the mem-brane and produce a cellular response.[23]

Induced proximity also provides the basis for a small-molecule strategy to guide the proteosome to specific –presumably undesirable – protein substrates. In earlywork, �proteolysis-targeting chimeric molecules� were de-signed to approximate a protein of interest and an ubiqui-tin ligase (an enzyme that mediates the ubiquitination ofa target protein), bringing about protein ubiquitinationand subsequent degradation.[24] These bifunctional smallmolecules are composed of a target protein�bindingligand and an E3 ubiquitin ligase ligand. In theory, thesemolecules effectively reprogram the substrate specificityof the E3 ubiquitin ligase to any target for which a small-molecule ligand exists.

In an ambitious manifestation of proximity-induced re-actions, Spiegel and Barbas, building on the early prece-dent of Schultz,[25] made use of proximity to specify a so-phisticated immune response as opposed to degradationof a single protein.[26] Using antibody-recruiting molecules(ARMs) possessing an antibody-binding motif anda target cell�binding motif, it has been possible to targetHIV positive cells,[27] prostate cancer cells,[28] and meta-static cancer cells.[29] This approach goes beyond simplyrewiring signaling enzyme specificity, but it utilizes thesame concepts, such as ternary complex formation tobring biomolecules in proximity to change biological

output. An analytical treatment of three-component equi-libria that should prove useful for optimizing occupancyof the ternary complex has also recently been reported.[30]

4. DNA/RNA-Templated Reactions

Nucleic acids can also be used as scaffolds to increase theeffective molarity of chemical and biochemical reactionsin solution (Figure 5). Indeed, the natural processes oftranscription, translation, replication, recombination,splicing, and DNA repair all rely on reactions promotedby nucleic acid templates. In the context of unnatural pro-cesses, substrates and reagents can be colocalized to spe-cific nucleic acid regions either by hybridization or specif-ic protein� or small molecule�nucleic acid interfaces. Be-cause engineering these interactions is often straightfor-ward and predictable, utilization of DNA and RNA tobring two molecules into proximity in a programmablemanner is particularly versatile.

4.1 Novel Reactions

Early examples of unnatural reactions templated by se-quence-specific nucleic acid interactions include phospho-diester bond formation,[31] duplex- and triplex-mediatedalkylation,[32] and DNA cleavage[33] reactions. More re-cently, the increased effective molarity provided by colo-

Figure 5. A) Three strategies by which substrates and reagents can be colocalized to an oligomeric nucleic acid scaffold. B) DNA templatescan template reactions and lead to the discovery of new reactions.[36] C) Complex RNA architectures can organize more efficient metabolicpathways.[50]

572 www.ijc.wiley-vch.de � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Isr. J. Chem. 2013, 53, 567 – 576

Review A. Schepartz et al.

calized DNA hybridization has been exploited to facili-tate difficult chemical reactions, discover novel reactivity,and template the assembly of unnatural biopolymers.[34]

For example, Liu and coworkers have harnessed DNAhybridization to promote entropically unfavorable reac-tions, such as macrocyclization, which are impeded by un-desirable intermolecular side reactions.[35] These previous-ly detrimental synthetic obstacles are overcome by lowsolution reactant concentrations (nM�mM), in which onlythe templated reaction proceeds to any significant extentbecause of the high local concentration induced by thenucleic acid. This technology has also been applied ina discovery mode to identify novel carbon-carbon bondforming reactions, such as in the case of a palladium-cata-lyzed reaction between an alkyne and alkene that occursonly because of the increased effective molarity betweenreactants.[36]

4.2 Complex Molecule and Library Syntheses

For many DNA-templated reactions, base pairing – notchemistry – is rate limiting. In these cases the rate of thetemplated reaction is independent of the distance (alongthe DNA strand) connecting the binding sites of the tworeactants,[37] allowing multiple synthetic steps to be tem-plated on a single oligomer. This versatility provides forthe construction of molecules with significant syntheticcomplexity as well as diverse libraries.[35a] Complex mole-cules can also be synthesized by templating each step ofthe reaction separately.[38] Because the reactions do notproceed unless templated, this process can take place ina single reaction vessel, greatly simplifying the synthe-sis.[38] In the most recent manifestation of this idea, Liuand coworkers have used DNA-templated synthesis tocouple b-peptide building blocks in a sequence-pro-grammed manner.[34] It should be noted, however, thatproduct inhibition is a genuine challenge for all DNA-templated reactions (as well as all proximity-aided reac-tions). While methods have been developed to disruptthe product-template interaction,[39] product inhibition re-mains a significant limitation and highlights that specifici-ty, not turnover, is the true advantage of these templatedreactions.

4.3 Templated Reactions Inside the Cell

The utility of DNA and RNA for inducing proximity ofligands both large and small is not restricted to reactionsperformed in the test tube. Ma and Taylor have devel-oped a technology that utilizes a DNA-templated reac-tion to potentially target a drug to cancerous cells.[40] Inan early example, a DNA template programmed as an en-dogenous oncogene templates the colocalization to a pro-drug-DNA conjugate and a uniquely functionalized oligo-nucleotide – one modified with an imidazole group capa-ble of catalyzing release of the drug.[41] Although much

effort has been put into the development of DNA-tem-plated reactions for drug release, problems of biocompati-bility and delivery of oligonucleotide-conjugated drugshave thwarted progress.[41] Recently, Winssinger and cow-orkers have overcome some of these challenges by utiliz-ing an azide reduction reaction that potentially could beused to release any functionalized small molecule(Figure 6).[42]

4.4 More Efficient Metabolic Pathways

Nucleic acid scaffolds can also promote multistep trans-formations and improve the flux through metabolic path-ways. The broad toolkit of programmable oligomericstructures makes DNA and RNA invaluable scaffolds.[43]

Colocalization of enzymes in a metabolic pathway canimprove reaction efficiency by increasing the effectivemolarity of intermediates (see Section 2) and, if the pro-teins are close enough, intermediate transfer is even moreefficient as diffusion is dimensionally restricted along theprotein hydration shells.[44]

Reaction optimization on these scaffolds is idiosyncrat-ic: the properties of the system are dependent on a largenumber of difficult-to-predict factors, including the detailsof scaffold structure and secondary scaffold-enzyme inter-actions. For example, the activities of some enzymes thatare colocalized to a scaffold have been observed tochange,[45] while others have found the enzymatic activityto remain constant.[44] Therefore, careful examination isnecessary to confirm the system behaves as engineered.While this technology has been successfully applied invitro,[46] modeling of the diffusion of intermediates ina scaffolded enzyme pair has revealed that the DNA scaf-fold persists only on the timescale of minutes before a suf-ficient concentration of intermediate has been built up.[47]

While this may be true for in vitro experiments in simplebuffered solutions, one could imagine a greater benefit ofscaffolds in the complex milieu of a cell, and further in-

Figure 6. DNA-templated activation of a prodrug. In this work, anoncogenic DNA sequence serves to colocalize a phosphine reduc-ing agent and a prodrug in order to potentially deliver a small-mol-ecule therapeutic to a diseased cell.[42]

Isr. J. Chem. 2013, 53, 567 – 576 � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.ijc.wiley-vch.de 573

Proximity as a Guiding Force in Chemistry and Biology

vestigations into the benefits of such template reactionsin vivo must be conducted.

This technology has also been successfully implementedin live cells. However, the chemical attachment of nucleo-tides to the proteins of interest is not feasible for cellularexperiments; therefore, encodable methods must be uti-lized such as RNA aptamer�protein[48] or zinc finger�DNA[49] interactions. Aldaye and coworkers were able tocolocalize proteins involved in hydrogen production in anorganelle-like area inside the cell by engineering aptamerdomains into the scaffold and fusing aptamer-binding pro-teins to the proteins of interest.[50] This scaffold complexi-ty, which protein scaffolds lack, significantly improved hy-drogen production in cells in comparison to RNA scaf-folds that were more dimensionally simple.[50] Additional-ly, other labs have used DNA templates and zinc fingerprotein fusions to engineer metabolic pathways.[51] Amajor setback to the use of oligomer scaffolds to altercellular pathways is the requirement that the protein ofinterest be genetically tagged, thus not allowing for thereorganization of endogenous signaling pathways. In sum-mary, the precision of base pairing, the diversity of oligo-nucleotide nanostructures and the ease of design havemade DNA and RNA templates a helpful tool in scaf-folding reactions from the molecular scale up to the or-ganization of complex metabolic pathways, regardless ofthe obstacles associated with biocompatibility.

5. Unnatural Domains

A final potential solution to guide macromolecular associ-ations in a crowded environment and decrease cellularcross-talk exploits adaptors or scaffolds composed ofwholly unnatural domains with prescribed (and presuma-bly optimized) recognition properties. More than 50 dif-ferent protein domains possessing unique and orthogonalstructures and binding properties have been developedover the past 15 years. Examples include affibodies,[52]

monobodies,[53] designed ankyrin repeat proteins (DAR-Pins),[54] and tetratricopeptide repeats (TPRs),[55] as wellas the pancreatic fold�based miniature proteins that wehave studied.[56] One advantage of adaptors composed ofwholly unnatural recognition domains is that they can beevolved on the basis of one function only – specificity –and thus have the potential for fewer off-target interac-tions than a natural protein domain whose evolution hasbeen guided by multiple, often antagonistic ideals. Anoth-er advantage of a synthetic adaptor approach is that theirfunction does not require the modification of signalingproteins in the pathway being manipulated.

5.1 A Toolkit of Coiled Coil Domains

One benefit of 25 years of research on coiled coil interac-tions is a clear understanding of the rules governing

strand orientation, stoichiometry, and preference for ho-modimeric or heterodimeric interactions. Woolfson andcoworkers have now made use of this information to de-velop a toolkit of mutually interacting protein parts basedon de novo designed coiled-coil peptides.[57] A set of parts(peptides) were reported, whose lengths span 21, 24, or28 residues, and which interact in a heterodimeric mannerwith equilibrium dissociation constants in the micromolarto sub-nanomolar range (Figure 7). The designs incorpo-

rate a buried asparagine residue whose hydrogen-bondingpreferences specify uniqueness in both stoichiometry andaxial orientation. Notably, despite the sophistication ofthe design, the sequences of the peptide parts are notfully defined and can be further modified at a number ofresidues (including those that are solvent exposed). Thislevel of flexibility is particularly desirable for applicationsother than just simple oligomerization: for instance, inmore complex association/dissociation processes involvingcomplex dynamics and subunit exchange.

5.2 Wholly Unnatural Adaptor Proteins

We recently reported a wholly unnatural adaptor proteinthat acts as a catalyst to redirect the Src family kinaseHck to phosphorylate hDM2, a negative regulator of thep53 tumor suppressor and a naturally poor Hck substrate(Figure 8).[58] The design of this adaptor began with two

Figure 7. A toolkit of mutually interacting protein parts based onde novo designed coiled-coil peptides.[57]

Figure 8. A miniature protein�based adaptor protein that redirectsthe Src family kinase Hck to phosphorylate hDM2, a negative regu-lator of the p53 tumor suppressor and a naturally poor Hck sub-strate.[58]

574 www.ijc.wiley-vch.de � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Isr. J. Chem. 2013, 53, 567 – 576

Review A. Schepartz et al.

previously reported miniature proteins, YY2 and 3.3.Miniature protein YY2 uses residues within its PPII helixto interact with the SH3 domains of certain Src familykinases, such as Hck. This interaction disrupts the intra-molecular SH3-domain interaction that down-regulateskinase activity and results in kinase activation. Miniatureprotein 3.3 uses residues within its a-helix to bind tohDM2, thereby inhibiting the association of hDM2 withp53. A series of adaptor proteins varying in the linkerlength were constructed; one was shown to exploit tem-plated catalysis to redirect the Src family kinase Hck tophosphorylate hDM2. Phosphorylation occurs with multi-ple turnover and at a single site targeted by the non-Srcfamily kinase c-Abl kinase in the cell. Notably, miniatureproteins are encodable and evolvable and facilitate in-duced proximity of endogenous (unmodified) cellularproteins using orthogonal domains with less potential forunintended cross-talk.

6. Summary and Outlook

In this review, we have described several methods thatchemists and synthetic biologists have used to rewire sig-naling pathways. These methods all exploit effective mo-larity as a way to promote the reaction between two sig-naling partners. Recent work in the field of synthetic biol-ogy has drawn on classic concepts that utilize increasedeffective molarity to promote unnatural reactions, in par-ticular chemically induced dimerization and templatedcatalysis. The former benefits from temporal control andthe confidence that the requisite domains will perform asadvertised. The latter benefits from the ability to functionwith natural, as opposed to modified, signaling domains.

Acknowledgements

The authors are grateful to members of the Schepartzlaboratory for valuable discussions and to JonathanMiller (Yale University) for an astute editing of themanuscript. Some of the work discussed in this reviewwas carried out in the authors� laboratory and was sup-ported by grants from the US National Institutes ofHealth and the National Foundation for Cancer Re-search. Allison S. Walker was supported by the NationalScience Foundation under Grant No. DGE-1122492.

References

[1] R. B. Bhattacharyya, A. Rem�nyi, B. J. Yeh, W. A. Lim,Annu. Rev. Biochem. 2006, 75, 655–680.

[2] P. Wei, W. W. Wong, J. S. Park, E. E. Corcoran, S. G. Peisa-jovich, J. J. Onuffer, A. Weiss, W. A. Lim, Nature 2012, 488,384–388.

[3] J. E. Dueber, G. C. Wu, G. R. Malmirchegini, T. S. Moon,C. J. Petzold, A. V. Ullal, K. L. J. Prather, J. D. Keasling,Nat. Biotech. 2009, 27, 753–759.

[4] P. L. Howard, M. C. Chia, S. Del Rizzo, F.-F. Liu, T.Pawson, Proc. Natl. Acad. Sci. USA 2003, 100, 11267 –11272.

[5] A. G. Batzer, D. Rotin, J. M. UreÇa, E. Y. Skolnik, J. Schles-singer, Mol. Cell Biol. 1994, 14, 5192 –5201.

[6] E. W. Miles, S. Rhee, D. R. Davies, J. Biol. Chem. 1999, 274,12193 –12196.

[7] a) F. Schaeffer, M. Matuschek, G. Guglielmi, I. Miras, P. M.Alzari, P. B�guin, Biochemistry 2002, 41, 2106 –2114; b) I.Miras, F. Schaeffer, P. B�guin, P. M. Alzari, Biochemistry2002, 41, 2115 –2119; c) C. You, X.-Z. Zhang, N. Sathitsuka-noh, L. R. Lynd, Y.-H. P. Zhang, Appl. Environ. Microbiol.2012, 78, 1437 –1444; d) L.-H. Fan, Z.-J. Zhang, X.-Y. Yu,Y.-Y. Xue, T.-W. Tan, Proc. Natl. Acad. Sci. USA 2012, 109,13260 –13265; e) S.-L. Tsai, N. A. DaSilva, W. Chen, ACSSynth. Biol. 2013, 2, 14–21; f) C. You, S. Myung, Y.-H. P.Zhang, Angew. Chem. Intl. Ed. 2012, 51, 8787–8790.

[8] a) E. A. Bayer, J.-P. Belaich, Y. Shoham, R. Lamed, Annu.Rev. Microbiol. 2004, 58, 521 –554; b) C. M. G. A. Fontes,H. J. Gilbert, Annu. Rev. Biochem. 2010, 79, 655 –681.

[9] A. M. Stock, V. L. Robinson, P. N. Goudreau, Annu. Rev.Biochem. 2000, 69, 183–215.

[10] W. R. Whitaker, S. A. Davis, A. P. Arkin, J. E. Dueber,Proc. Natl. Acad. Sci. USA 2012, 109, 18090 –18095.

[11] a) J. E. Dueber, B. J. Yeh, K. Chak, W. A. Lim, Science2003, 301, 1904–1908; b) B. J. Yeh, R. J. Rutigliano, A. Deb,D. Bar-Sagi, W. A. Lim, Nature 2007, 447, 596 –600.

[12] W. A. Lim, Nat. Rev. Mol. Cell Biol. 2010, 11, 393–403.[13] A. Zarrinpar, S.-H. Park, W. A. Lim, Nature 2003, 426, 676 –

680.[14] D. M. Spencer, T. J. Wandless, S. L. Schreiber, G. R. Crab-

tree, Science 1993, 262, 1019–1024.[15] A. Fegan, B. White, J. C. T. Carlson, C. R. Wagner, Chem.

Rev. 2010, 110, 3315–3336.[16] a) T. Komatsu, I. Kukelyansky, J. M. McCaffery, T. Ueno,

L. C. Varela, T. Inoue, Nat. Meth. 2010, 7, 206–208; b) J. J.Kohler, C. R. Bertozzi, Chem. Biol. 2003, 10, 1303 –1311.

[17] L. M. Ballou, R. Z. Lin, J. Chem. Biol. 2008, 1, 27 –36.[18] a) J. H. Bayle, J. S. Grimley, K. Stankunas, J. E. Gestwicki,

T. J. Wandless, G. R. Crabtree, Chem. Biol. 2006, 13, 99 –107; b) S. D. Liberles, S. T. Diver, D. J. Austin, S. L. Schreib-er, Proc. Natl. Acad. Sci. USA 1997, 94, 7825–7830.

[19] E. J. Licitra, J. O. Liu, Proc. Natl. Acad. Sci. USA 1996, 93,12817 –12821.

[20] a) K. Baker, C. Bleczinski, H. Lin, G. Salazar-Jimenez, D.Sengupta, S. Krane, V. W. Cornish, Proc. Natl. Acad. Sci.USA 2002, 99, 16537–16542; b) H. Lin, H. Tao, V. W. Cor-nish, J. Am. Chem. Soc. 2004, 126, 15051 –15059.

[21] J. L. Czlapinski, M. W. Schelle, L. W. Miller, S. T. Laughlin,J. J. Kohler, V. W. Cornish, C. R. Bertozzi, J. Am. Chem.Soc. 2008, 130, 13186 –13187.

[22] N. A. Hathaway, O. Bell, C. Hodges, E. L. Miller, D. S.Neel, G. R. Crabtree, Cell 2012, 149, 1447 –1460.

[23] T. Miyamoto, R. DeRose, A. Suarez, T. Ueono, M. Chen, T.Sun, M. J. Wolfgang, C. Mukherjee, D. J. Meyers, T. Inoue,Nat. Chem. Biol. 2012, 8, 465–470.

[24] J. S. Schneekloth Jr. , F. Fonseca, M. Koldobskiy, A. Mandal,R. Deshaies, K. Sakamoto, C. M. Crews, J. Am. Chem. Soc.2004, 126, 3748–3754.

[25] K. M. Shokat, P. G. Schultz, J. Am. Chem. Soc. 1991, 113,1861–1862.

Isr. J. Chem. 2013, 53, 567 – 576 � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.ijc.wiley-vch.de 575

Proximity as a Guiding Force in Chemistry and Biology

[26] M. Popkov, B. Gonzalez, S. C. Sinha, C. F. Barbas, Proc.Natl. Acad. Sci. USA 2009, 106, 4378–4383.

[27] C. G. Parker, R. A. Domaoal, K. S. Anderson, D. A. Spie-gel, J. Am. Chem. Soc. 2009, 131, 16392–16394.

[28] a) R. P. Murelli, A. X. Zhang, J. Michel, W. L. Jorgensen,D. A. Spiegel, J. Am. Chem. Soc. 2009, 131, 17090 –17092;b) A. X. Zhang, R. P. Murelli, C. Barinka, J. Michel, A. Co-cleaza, W. L. Jorgensen, J. Lubkowski, D. A. Spiegel, J. Am.Chem. Soc. 2010, 132, 12711–12716.

[29] C. E. Jakobsche, P. J. McEnaney, A. X. Zhang, D. A. Spie-gel, ACS Chem. Biol. 2012, 7, 316 –321.

[30] E. F. Douglass Jr, C. J. Miller, G. Sparer, H. Shapiro, D. A.Spiegel, J. Am. Chem. Soc. 2013, 135, 6092 –6099.

[31] R. Naylor, P. T. Gilham, Biochemistry 1966, 5, 2722–2728.[32] K. B. Grant, P. B. Dervan, Biochemistry 1996, 35, 12313 –

12319.[33] P. G. Schultz, J. S. Taylor, P. B. Dervan, J. Am. Chem. Soc.

1982, 104, 6861–6863.[34] J. Niu, R. Hili, D. R. Liu, Nat. Chem. 2013, 5, 282 –292.[35] a) Z. J. Gartner, B. N. Tse, R. Grubina, J. B. Doyon, T. M.

Snyder, D. R. Liu, Science 2004, 305, 1601 –1605; b) C. Per-civalle, J.-F. Bartolo, S. Ladame, Org. Biomol. Chem. 2013,11, 16–26.

[36] M. W. Kanan, M. M. Rozenman, K. Sakurai, T. M. Snyder,D. R. Liu, Nature 2004, 431, 545–549.

[37] Z. J. Gartner, D. R. Liu, J. Am. Chem. Soc. 2001, 123, 6961 –6963.

[38] M. L. McKee, P. J. Milnes, J. Bath, E. Stulz, R. K. O’Reilly,A. J. Turberfield, J. Am. Chem. Soc. 2012, 134, 1446 –1449.

[39] a) C. Dose, S. Ficht, O. Seitz, Angew. Chem. Int. Ed. 2006,45, 5369 –5373; b) Y. He, D. R. Liu, J. Am. Chem. Soc. 2011,133, 9972–9975; c) J. Michaelis, A. Maruyama, O. Seitz,Chem. Comm. 2013, 49, 618 –620.

[40] Z. Ma, J. S. Taylor, Proc. Natl. Acad. Sci. USA 2000, 97,11159 –11163.

[41] M. F. Jacobsen, E. Clo, A. Mokhir, K. V. Gothelf, Chem-medchem 2007, 2, 793 –799.

[42] K. Gorska, A. Manicardi, S. Barluenga, N. Winssinger,Chem. Comm. 2011, 47, 4364 –4366.

[43] a) K. A. Afonin, E. Bindewald, A. J. Yaghoubian, N. Voss,E. Jacovetty, B. A. Shapiro, L. Jaeger, Nat. Nano. 2010, 5,676–682; b) J. L. Fu, M. H. Liu, Y. Liu, H. Yan, Acc. Chem.Res. 2012, 45, 1215–1226.

[44] J. Fu, M. Liu, Y. Liu, N. W. Woodbury, H. Yan, J. Am.Chem. Soc. 2012, 134, 5516–5519.

[45] J.-L. Lin, I. Wheeldon, ACS Catalysis 2013, 3, 560 –564.[46] a) O. I. Wilner, Y. Weizmann, R. Gill, O. Lioubashevski, R.

Freeman, I. Willner, Nat. Nano. 2009, 4, 249–254; b) T.Nojima, H. Konno, N. Kodera, K. Seio, H. Taguchi, M.Yoshida, PloS one 2012, 7, e52534; c) M. Erkelenz, C. H.Kuo, C. M. Niemeyer, J. Am. Chem. Soc. 2011, 133, 16111 –16118.

[47] O. Idan, H. Hess, Nat. Nano. 2012, 7, 769–770.[48] D. H. J. Bunka, P. G. Stockley, Nat. Rev. Microbiol. 2006, 4,

588–596.[49] E. Nakata, F. F. Liew, C. Uwatoko, S. Kiyonaka, Y. Mori, Y.

Katsuda, M. Endo, H. Sugiyama, T. Morii, Angew. Chem.Int. Ed. 2012, 51, 2421 –2424.

[50] C. J. Delebecque, A. B. Lindner, P. A. Silver, F. A. Aldaye,Science 2011, 333, 470 –474.

[51] a) R. J. Conrado, G. C. Wu, J. T. Boock, H. Xu, S. Y. Chen,T. Lebar, J. Turnsek, N. Tomsic, M. Avbelj, R. Gaber, T. Ko-privnjak, J. Mori, V. Glavnik, I. Vovk, M. Bencina, V.Hodnik, G. Anderluh, J. E. Dueber, R. Jerala, M. P. DeLisa,Nucl. Acids Res. 2012, 40, 1879–1889; b) J. H. Lee, S.-C.Jung, L. M. Bui, K. H. Kang, J.-J. Song, S. C. Kim, Appl. En-viron. Microbiol. 2013, 79, 774–782.

[52] P.-�. Nygren, FEBS Journal 2008, 275, 2668–2676.[53] A. Koide, S. Koide, in Protein Engineering Protocols,

Vol. 352 (Eds.: K. Arndt, K. M�ller), Humana Press, 2007,pp. 95 –109.

[54] M. T. Stumpp, H. K. Binz, P. Amstutz, Drug Discov. Today2008, 13, 695 –701.

[55] T. Z. Grove, M. Hands, L. Regan, Protein Eng. Des. Sel.2010, 23, 449 –455.

[56] a) N. J. Zondlo, A. Schepartz, J. Am. Chem. Soc. 1999, 121,6938–6939; b) J. W. Chin, R. M. Grotzfeld, M. A. Fabian,A. Schepartz, Bioorg. Med. Chem. Lett. 2001, 11, 1501 –1505; c) J. W. Chin, A. Schepartz, Angew. Chem. Intl. Ed.Engl. 2001, 40, 3806 –3809; d) J. W. Chin, A. Schepartz, J.Am. Chem. Soc. 2001, 123, 2929 –2930; e) J. K. Montclare,A. Schepartz, J. Am. Chem. Soc. 2003, 125, 3416 –3417.

[57] F. Thomas, A. L. Boyle, A. J. Burton, D. N. Woolfson, J.Am. Chem. Soc. 2013, 135, 5161 –5166.

[58] E. M. Hobert, A. Schepartz, J. Am. Chem. Soc. 2012, 134,3976–3978.

Received: June 12, 2013Accepted: July 7, 2013

Published online: August 16, 2013

576 www.ijc.wiley-vch.de � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Isr. J. Chem. 2013, 53, 567 – 576

Review A. Schepartz et al.