Protein Conformational Switches: From Nature to Design

Jeung-Hoi Ha and Stewart N. Loh*[a]

� 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2012, 18, 7984 – 79997984

DOI: 10.1002/chem.201200348

Introduction

Protein conformational switches are polypeptides that un-dergo a significant change in structure upon receiving aninput signal. Input stimuli can consist of covalent modifica-tions, such as phosphorylation or cleavage of the polypep-tide backbone, or more commonly, molecular recognitionevents. Examples of the last type include absorbinga photon, binding a drug, or engaging an entire cell via a sur-

face receptor. The resulting conformational change estab-lishes the output response. For instance, the structuralchange may modulate enzymatic activity or expose new sur-face for the protein to interact with other molecules. Manyproteins bind other molecules and/or become covalentlymodified; however, it is the property of stimulus-inducedstructural change that defines the more select cohort of con-formational switches.

Switchable proteins are of special interest for two reasons.First, they are of fundamental importance to biology. Theycan regulate cellular signaling pathways, including those in-volved in growth control, development, sensory signal medi-ation, and pathogen infection. For example, the b2 adrener-gic receptor (b2AR) is a member of the large family ofG protein coupled receptors (GPCRs) which respond toa variety of signals, such as light, odorants, hormones, andneurotransmitters. When an agonist binds to the extracellu-lar face of b2AR, two of the seven membrane-spanning a-helices are displaced toward the cytoplasm by as much as14 �.[1] This movement generates a binding site at the cyto-plasmic face of b2AR for the G protein heterotrimer(Gsabg). Interaction with b2AR causes Gsabg to exchangeits bound GDP for GTP then dissociate into Ga-GTP andGbg subunits. Ga-GTP and Gbg then go on to bind and ac-tivate their downstream targets.

Another class of functionally important switches is theviral protein “harpoons” that mediate fusion of the viralmembrane with that of the host. Influenza haemagglutinin(HA) is the protein responsible for anchoring the virus tothe cellular receptor. Once the docked virus is endocytosedinto the cell, the pH inside the endosome decreases anda proton binding event triggers a dramatic conformationalchange in HA.[2] The three lobes of HA splay apart toreveal a buried triple-stranded coiled-coil stalk. A loop-to-helix folding transition causes the coiled-coil to extend andflip outwards, thrusting the hydrophobic fusion peptide(which resides at the tip of the newly lengthened stalk) intothe vacuolar membrane. In this conformation HAs are pro-posed to distort and ultimately bridge the viral and vacuolarmembranes, although details remain unclear. Similar confor-mational changes are thought to occur in gp41 (HIV),[3]

spike protein (SARS),[4] GP2 (Ebola virus),[5] and function-ally homologous proteins from other viruses. The triggeringconditions differ (decrease in pH, binding to receptor, pro-teolytic cleavage, redox change), but the basic harpooningaction is likely to be conserved.

The second reason why switchable proteins are of interestis that their switching mechanisms, once understood, can bemodified by the tools of protein engineering to create mole-cules with new functionalities. Research in this area is large-ly motivated by two related applications: functional regula-tion and biosensing. The goal of functional regulation is tobe able to turn on, turn off, or alter the biological functionof a target protein by application of a desired stimulus. Byimbuing such regulatory capability into, for example, tran-

Abstract: Protein conformational switches alter theirshape upon receiving an input signal, such as ligandbinding, chemical modification, or change in environ-ment. The apparent simplicity of this transformation—which can be carried out by a molecule as small asa thousand atoms or so—belies its critical importance tothe life of the cell as well as its capacity for engineeringby humans. In the realm of molecular switches, proteinsare unique because they are capable of performing a va-riety of biological functions. Switchable proteins aretherefore of high interest to the fields of biology, bio-technology, and medicine. These molecules are begin-ning to be exploited as the core machinery behinda new generation of biosensors, functionally regulatedenzymes, and “smart” biomaterials that react to theirsurroundings. As inspirations for these designs, re-searchers continue to analyze existing examples of allo-steric proteins. Recent years have also witnessed the de-velopment of new methodologies for introducing con-formational change into proteins that previously hadnone. Herein we review examples of both natural andengineered protein switches in the context of four basicmodes of conformational change: rigid-body domainmovement, limited structural rearrangement, global foldswitching, and folding–unfolding. Our purpose is tohighlight examples that can potentially serve as plat-forms for the design of custom switches. Accordingly,we focus on inducible conformational changes that aresubstantial enough to produce a functional response(e.g., in a second protein to which it is fused), yet arerelatively simple, structurally well-characterized, andamenable to protein engineering efforts.

Keywords: allostery · biosensors · protein design · pro-tein engineering · protein folding

[a] J.-H. Ha, S. N. LohDepartment of Biochemistry & Molecular BiologyState University of New York Upstate Medical University750 East Adams StreetSyracuse, NY 13210 (USA)Tel: (315) 464-8731Fax: (315) 464-8750E-mail : lohs@upstate.edu

Chem. Eur. J. 2012, 18, 7984 – 7999 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 7985

REVIEW

scription factors or signaling proteins, it might be possible toprogram developmental pathways or rewire cellular signal-ing circuitry. Biosensors operate on a similar principle, butthe emphasis is placed on detecting the stimulus. According-ly, the conformational change itself is of primary interest. Itis designed to transduce the input event, which typicallytakes place on a recognition domain, to a reporter elementfor detection. The reporter can be an electronic device ora separate biological element, such as a fluorescent proteinor enzyme. Use of protein switches allows one to achievethis communication in a number of creative ways, includingby integrating recognition and reporting functions ina single molecule.

As in many areas of engineering, our attempts to buildswitches often draw heavily from nature�s designs. Severalstudies of structures in the Protein Data Bank (PDB) havesurveyed the types of structural changes that occur in re-sponse to external perturbation.[6] These changes can begrouped into the following broad and sometimes overlap-ping categories: 1) rigid-body domain movement (most com-monly hinge or shear-type motions), 2) limited structural re-arrangement (in which a segment of the structure moves orbecomes ordered/disordered), and 3) large-scale structuralrearrangement in which the protein can switch global folds.A fourth category, binding-induced folding of intrinsicallydisordered proteins (IDPs), is now recognized as a switchingmechanism used by many regulatory and signaling pro-teins.[7] IDPs illustrate that changes in structure and dynam-ics often go hand-in-hand, but this is not always the case.NMR studies of retroviral matrix proteins,[8] PDZ domain,[9]

eglin c,[10] and others[11] have shown that ligand binding candecrease or even increase dynamic motions at the active siteas well as at distant locations, in the absence of significantstructural changes. The same phenomenon has been ob-served as a result of post-translational modifications, such asphosphorylation.[12] These proteins represent a class of “en-tropic switches” the potential of which for design remainsuntested.

Herein we review current efforts to develop protein con-formational switches. The number of existing switches anddesign strategies is too large to discuss in a single treatise;the reader is referred to other articles for additional per-spectives.[13] Moreover, the size and complexity of naturalswitches such as b2AR and HA (notwithstanding the factthat they are membrane bound) restricts their practical useas templates for design. At the opposite end of the spectrumare proteins that undergo functional switching with onlyminor changes in conformation. The engineering potentialof these molecules is limited as well. For instance, EF-handdomains mediate calcium binding in a large family of signal-ing and transport proteins. The simplest family members, ex-emplified by calbindin D9k, contain two EF-hands that bindCa2+ cooperatively. The binding mechanism involves alloste-ric coupling of conformational substates,[14] but the differ-ence between the bound and free structures of calbindin D9k

is too small to be useful. The conformational change mustfirst be amplified before the EF-hand can serve as a calci-

um-triggered functional switch. Natural as well as artificialmeans for doing so are discussed in the sections below. Asa rule of thumb, for a conformational change to be consid-ered exploitable it should involve unfolding, rigid-body reor-ientation, or otherwise substantial movement of at least onesecondary structural element. We therefore focus on switch-ing mechanisms that are simple enough to be amenable toengineering efforts, yet are of sufficient magnitude to war-rant such endeavors. We concentrate on some well-charac-terized cases from each of the four categories of conforma-tional change.

Rigid-Body Domain Movement

Open-to-closed motions are exemplified by rigid-body dis-placement of two or more domains with respect to each

Jeung-Hoi Ha was born in Seoul (SouthKorea), and received her B.Sc. in Bio-chemistry at Yonsei University. She ob-tained her Ph.D. in Biochemistry at theUniversity of Wisconsin-Madison underthe supervision of Professor M. ThomasRecord, Jr. Her Ph.D. work focused on thehydrophobic effect and thermodynamics ofthe lac repressor–operator interaction. Shereceived postdoctoral training first fromProfessor Ronald Raines at the Universityof Wisconsin-Madison, then from Profes-sor David McKay in the Department ofStructural Biology at Stanford University.At Stanford, she investigated the structure and mechanism of the HSC70molecular chaperone. She moved to the State University of New York Up-state Medical University in 1996. She is a Senior Research Scientist in theDepartment of Biochemistry & Molecular Biology. She and Dr. Loh joint-ly manage Dr. Loh�s research group and are known as the left and rightbrains of the laboratory.

Stewart N. Loh received his B.Sc. inChemistry from the University of Utah andhis Ph.D. in Biochemistry from the Uni-versity of Wisconsin-Madison, under thesupervision of Professor John Markley.His Ph.D. training involved applying nu-clear magnetic resonance techniques, in-cluding amide hydrogen exchange and hy-drogen/deuterium isotope fractionationmeasurements, to investigate folding ofstaphylococcal nuclease. He was a DamonRunyon-Walter Winchell PostdoctoralFellow in the laboratory of ProfessorRobert L. Baldwin in the BiochemistryDepartment at Stanford University, where he studied the kinetic foldingmechanisms of apomyoglobin and ribonuclease A. He joined the StateUniversity of New York Upstate Medical University as an Assistant Profes-sor in 1996. He and Dr. Ha have continued to study the thermodynamicsand kinetics of protein folding since then, and have specifically concentrat-ed on folding-based switching mechanisms for the past ten years. Hebecame an Associate Professor in 2002 and a Full Professor of Biochem-istry and Molecular Biology in 2008. Dr. Loh�s current research focus is toapply principles of protein folding and unfolding to convert ordinary pro-teins into molecular switches for use in biology, biotechnology, and medi-cine.

www.chemeurj.org � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2012, 18, 7984 – 79997986

S. N. Loh and J.-H. Ha

other, often by a hinge-type action. Secondary structure istypically not affected. In a comprehensive study of 839 non-redundant pairs of ligand-free and ligand-bound proteins inthe PDB, Amemiya et al.[6e] found that 325 exhibited struc-tural changes that were linked to ligand binding. Of these325 proteins, half underwent domain and/or open-to-closedmovements while the rest displayed local changes (e.g., rear-rangement of a surface loop).

Domain movement is particularly agreeable to biosensordevelopment. The general idea is to attach reporter groups,for example, environment or distance-sensitive fluorophores,at positions that detect the movement. Periplasmic bindingproteins (PBPs) embody this design. PBPs are a superfamilyof nonenzymatic receptors the members of which are foundin all kingdoms of life. They are versatile scaffolds for bio-sensor development, because they bind a variety of smallmolecules with high specificity and dissociation constants(Kd) in the nm to mm range.[15] Substrates for PBPs includeribose,[16] maltose,[17] glucose/galactose,[18] allose,[19] lactate,[20]

cellulose,[21] amino acids,[22] vitamin B12,[23] iron sidero-phores,[24] nickel metallophores,[25] copper,[26] phosphonate,[27]

and glutathione.[28] Figure 1 illustrates the “venus flytrap”action of E. coli ribose binding protein (RBP). The structureconsists of N- and C-terminal lobes that are predominantlyin an open conformation in the ligand-free state (Figure 1).The substrate binding pocket is located in the hinge regionbetween the domains. Ribose binding causes the two do-mains to close by about 308, clamping down on the substrateand almost completely occluding it from the solvent.

Establishing the conformational change goes only halfway toward constructing a functional switch. In biosensorapplications it is necessary to couple the change to a detec-tion event. Fluorescence is often the method of choice as itoffers sensitive and potentially ratiometric output. Intrinsicprotein fluorescence is seldom useful, because tryptophanand tyrosine absorb and emit in the crowded ultravioletrange, have relatively low quantum yields, photobleacheasily, and may not report on the binding event in theirnative locations. Thus, two general approaches have

emerged for engendering a conformation-dependent fluores-cence response. The first is to introduce a chemical fluoro-phore at a site at which the environment is perturbed by thedomain movement. Although choosing dyes and locationscan be guided by computational approaches (and structuralintuition), it is still an empirical exercise. To illustrate, DeLorimer et al. placed eight different dyes at a total of 67 po-sitions among ten PBPs.[29] Of these, 24 of the locations re-sulted in a robust change in intensity and/or a ratiometricresponse, and only one or a few of the dyes could producethat result at a given site. A more recent study of phospho-nate binding protein found that the IANBD fluorophorewas able to report on domain closure when placed at two of31 positions. In RBP, one of the best combinations wasfound to be 4-fluoro-7-aminosulfonylbenzofurazan attachedto Cys265,[16d] which is located in the hinge region(Figure 1).

The second strategy for visualizing conformational changeis to employ donor and acceptor groups that give rise to dis-tance-dependent fluorescence phenomena such as quench-ing, excimer formation, and Fçrster resonance energy trans-fer (FRET). The first two processes require that donor andacceptor come into contact, whereas FRET response is sig-moidal over a relatively large distance centered at the Fçr-ster radius (R0). The R0 values of most donor/acceptor pairs(30–80 �) are comparable to the diameters of small tomedium-sized proteins, making FRET suitable for detectingrigid-body domain movements. Additional advantages ofFRET are that the output is often ratiometric, and FRETpairs can be genetically encoded in the form of fluorescentproteins (FPs). The disadvantages are that it can be techni-cally difficult to introduce donor and acceptor dyes into thesame protein, and the change in donor–acceptor distancemust be substantial to be detected by FRET. In the case ofPBPs, the location of the N- and C-termini near the tips ofthe respective domains allows FP donors and acceptors tobe appended to ends of the protein in order to detectdomain closure, as was demonstrated for RBP,[16a] maltosebinding protein (MBP),[30] glucose/galactose binding pro-tein,[31] and phosphonate binding protein.[27] For glucosebinding protein, two halves of the protein aequorin can re-place the FPs to provide bioluminescent output.[32] Glucosebinding brings the two aequorin fragments together whichallows them to bind and reconstitute the functional mole-cule.

The bacterial actin analogue ParM is another example ofa protein that undergoes a rigid-body, open-to-closed move-ment on interacting with its cognate ligands (ADP orGDP).[33] The nucleotide binding site is again located at theinterface between two domains, which rotate by about 258to close down on the substrate. Cysteine was introduced atsix locations, and subsequent labeling with a single diethyla-minocoumarin dye revealed that the fluorescence of oneconstruct increased by three fold in the presence of ADP.[34]

The same researchers subsequently improved their designby introducing two tetramethylrhodamine groups so thatthey stack and quench each other�s fluorescence when ParM

Figure 1. Rigid-body domain movement exemplified by E. coli RBP in itsopen (ribose-free; left) and closed (ribose-bound; right) states.[122] Aminoand carboxy domains are colored blue and red, respectively, and ribose isshown in purple. Residue 265, located in the hinge region and a site ofenvironmentally sensitive fluorophore attachment, is depicted as a greysphere. PDB codes are 1URP (left) and 2DRI (right).

Chem. Eur. J. 2012, 18, 7984 – 7999 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 7987

REVIEWProtein Conformational Switches

is in the open state.[33b] Domain closure separates the dyes,causing fluorescence to increase by 15-fold.

The Fc region of immunoglobulin E (IgEFc) engagesa high-affinity receptor (FceRI) on the surface of mast cellsand basophils. IgEFc is a homodimer of three subdomains:Ce2, Ce3, and Ce4. X-ray structural analysis revealed thatIgEFc adopts a crooked conformation in its free state, with(Ce2)2 bending back against (Ce3)2 and (Ce4)2, and thatbending increases upon binding to FceRI.[35] To visualize thischange Hunt et al. fused FP donor and acceptor groups tothe N- (Ce2) and C-terminus (Ce4) of IgEFc.[36] FRET effi-ciency was substantial in the absence of receptor and in-creased significantly in its presence. The IgEFc biosensor isuseful for investigating the molecular bases of antigen rec-ognition and allergic response. The complexity of the con-formational change, however, suggests that exploiting it asa general sensing mechanism would be challenging.

The cis/trans isomerization about X-Pro peptide bondscan effect conformational switching by altering the directionof the polypeptide chain. The Crk family of adaptor proteinsemploys Pro isomerism to toggle between open and closedconformational states. Crks mediate formation of a numberof protein complexes by binding to specific phospho-Tyr-containing motifs (via SH2 domains) and Pro-rich peptides(via SH3 domains) on target proteins. Cellular Crk consistsof an SH2 domain followed by two SH3 domains (SH3N andSH3C). The SH3 domains are connected by a 50 amino acidlinker, in which the highly conserved Pro238 abuts SH3C.cis-Pro238 orients the n-src loop so that Pro238, Phe239,and Ile270 interact with SH3N.[37] The resulting intramolecu-lar interaction between SH3N and SH3C inhibits Crk frombinding its ligands. When Pro238 isomerizes to trans, the n-src loop rearranges, Phe239 and Ile270 swing toward SH3C,and Crk is free to engage its targets. Similarly, a Pro switchin the SH2 domain of interleukin-2 tyrosine kinase (Itk) re-models the SH2 binding surface, such that it has a greateraffinity for phospho-Tyr peptides or the Itk SH3 domain inthe trans and cis states, respectively.[38] As a final example,cis/trans isomerization of a single Pro acts as a molecularhinge for gating the 5-hydroxytryptamine type 3 (5-HT3) re-ceptor[39] and probably other members of the Cys-loopfamily of ion channels.[40] Pro8* lies in the all-helical trans-membrane domain of 5-HT3 at the interface with the extra-cellular domain. It is at the tip of a loop that connects thechannel-lining M2 helix to another transmembrane helix.The trans-to-cis isomerization of Pro8* is thought to shiftthe M2 helix and open the pore. Pro isomerization consti-tutes a modest conformational rearrangement; it affects thegeometry of just one bond. Nevertheless, because Pro8* issituated at the end of an a-helix, and because a-helices arerelatively rigid, isomerization establishes a pivoting actionthat results in substantial movement of the other end of theM2 helix.

With respect to the engineering potential of Pro switches,it is clear that peptide bond isomerization can provide lever-age for transducing an input signal to a structural change.Peptidylprolyl cis/trans isomerases (PPIases) including cyclo-

philins, FK506 binding proteins, and parvulins add furtherinterest to such designs. PPIases can exert control at thelevel of switching rate and phosphorylation status of theamino acid preceding Pro (the Pin1 PPIase only catalyzesisomerization of phospho-Ser and phospho-Thr prolylbonds).[41] It is less clear how one can link Pro isomerizationto a specific stimulus. For example, the neurotransmitterbinding site in 5-HT3 is distant from Pro8* and the mecha-nism by which binding leads to isomerization is not known.One promising approach is to modify the polypeptide back-bone with a photoresponsive chemical group that undergoescis/trans isomerization upon absorption of light, as we willdiscuss in the section on limited structural rearrangementbelow.

Domain motions of a receptor protein are not only usefulfor biosensing, but also for modulating the function of an-other protein. To do so requires a physical interaction be-tween the two proteins. Indeed, the principle challenge insuch efforts is to determine how to best connect the mole-cules to maximize interplay of their conformational respons-es. End-to-end fusion with a long peptide linker is themethod of choice when the goal is to completely decoupletheir functions. Increasing the points of contact, for example,by inserting one protein into the middle of a second proteinusing short linkers, increases the likelihood that structural ordynamic changes in one domain will be propagated to theother (Figure 2). With that objective in mind the followingconsiderations can be helpful.

1) Interdomain communication can be maximized by insert-ing the receptor protein into the reporter near the re-porter�s active site, or inserting the reporter into the re-ceptor at a location near the receptor�s binding site.

2) To avoid perturbing structure and stability it is generallyadvisable that the insertions take place in surface loopsrather than in interior locations or secondary structureelements.

3) Circularly permuting the guest protein (see Figure 3)prior to inserting it into the host provides a more gentleand versatile means for effecting coupling (Figure 2 A).Circular permutation exchanges the position of the origi-nal N- and C-termini with that of a site elsewhere in theprotein, usually chosen to be a surface loop. This rear-rangement serves two purposes. First, it guarantees thatthe new N- and C-termini of the guest will be close inspace, thereby increasing the number of host sites intowhich it can be inserted (including secondary structureelements) without disrupting the structure of the host.Second, being able to choose the permutation site intro-duces a valuable combinatorial aspect to the process.

4) The length of the peptides used to fuse the two proteinsis a critical variable. Very long linkers decouple any po-tential interaction, whereas excessively short linkers maylead to the “mutually exclusive folding” (MEF) effect inwhich the guest protein stretches the host protein at thepoint of insertion, the host simultaneously compressesthe guest, and one unfolds the other (Figure 2 B).[42]

www.chemeurj.org � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2012, 18, 7984 – 79997988

S. N. Loh and J.-H. Ha

MEF represents a special case of interdomain fusion andit is discussed in a separate section below. One can there-fore vary the junction points on both receptor and re-porter proteins, along with the lengths and sequences oflinker peptides, to optimize interdomain coupling.

As an illustration of this procedure, Ostermeier and co-workers took a library-based approach by inserting randomcircular permutants of TEM1 b-lactamase (cpBLA) intorandom positions in MBP.[43] Their goal was to create

a hybrid protein in which cata-lytic activity of the cpBLAdomain is regulated by sugarbinding to the MBP domain.They screened a library of ap-proximately 106 variants byplating transformed E. coli onmedia containing maltose anddifferent concentrations of am-picillin. Several switches werefound in which maltose acted aseither a positive or negative ef-fector.[43a] (They also discoveredone enzyme that was non-com-petitively inhibited by Zn2+ , de-spite the fact that neitherparent protein possesses zinc-binding activity.[44] This seren-dipitous result demonstratesthat unanticipated structuralchanges can give rise to novelswitching properties, particular-ly when fusing proteins ina combinatorial fashion.) Thecatalytic efficiency of one posi-tive effector variant, RG13, wassimilar to that of wild-type(WT) BLA in the presence ofsaturating maltose and de-creased by 25-fold in the ab-sence of sugar.[43b]

A recent NMR study ofRG13 provides clues to thestructural basis for MBP-cpBLA coupling.[45] Chemicalshift analysis suggests that theMBP domain is unperturbed bythe cpBLA domain except forthe a-helix into which cpBLAwas inserted. Resonances in thecpBLA active site were largelyunchanged in the maltose-bound and maltose-free pro-teins except for Glu166. Theauthors speculated that Glu166(believed to be the general basefor b-lactam acylation)[46] is

properly positioned in the presence of maltose and displacedin its absence. However, the pattern of maltose-dependentchemical shift changes in the cpBLA domain does not im-mediately suggest a mechanism by which movement ofGlu166 is coupled to domain closure of the MBP domain.

With respect to functional switches, one of the more ver-satile stimuli is light.[47] Phytochromes are photoswitchesthat regulate a variety of signal transduction pathways inplants, fungi, and bacteria.[48] Phytochrome switches are re-versible: transformation from the inactive (Pr) to the active

Figure 2. Domain-in-domain insertion schemes for effecting conformational coupling. A) Domain insertionwith circular permutation. The guest is first circularly permuted so that its new N- and C-termini are near itsbinding/active site. The permutant is then inserted into a surface loop near the binding/active site of the host.This placement maximizes the chance that binding-induced conformational changes of one domain will betransduced to functional changes in the other domain. B) Mutually exclusive folding. A guest with a long N-to-C terminal distance is inserted into a surface loop of a host (left-most structure). If the linkers used to jointhe two proteins are sufficiently short (second structure from left), the guest stretches and unfolds the host, orthe host compresses and unfolds the guest (second structure from right). The host domains of some MEF con-structs are able to refold by domain swapping with identically unfolded hosts to generate dimers and higherorder oligomers (right-most structure).

Figure 3. Circular permutation changes the topology of a protein but frequently not its structure or function.The left structure represents the WT fold of a 100-amino acid protein. Circular permutation can be concep-tualized by first joining residue 100 to residue 1 with a flexible peptide linker (green), generating a hypotheticalfully circularized protein (center structure). Permutation is completed by clipping the protein at a location ofchoice, typically chosen to be a surface loop. Cleaving the protein at position 40 (right structure) shifts the N-and C-termini of the protein to positions 40 and 39, respectively, and effectively exchanges the positions of thetermini with that of a surface loop without significantly perturbing the overall fold.

Chem. Eur. J. 2012, 18, 7984 – 7999 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 7989

REVIEWProtein Conformational Switches

(Pfr) structure is triggered by absorption of red light by Pr,and reversed by absorption of far-red light by Pfr. The pho-tosensory core consists of three domains [Per-ARNT-Sim(PAS), GAF, and PHY]. The bilin chromophore is bound tothe central GAF domain. Absorption of a red photon indu-ces isomerization about a double bond in bilin. Recentstructures suggest that chromophore isomerization alters therelative orientations of GAF and PHY, a motion possiblymediated by rearrangement of the helices that connectthem.[49] The photosensory core of phytochrome has not yetbeen fused to another protein to achieve functional regula-tion in a manner akin to MBP and BLA. However, severaldesigns have linked phytochrome-interacting factors (pro-teins that bind specifically to Pfr) to other molecules to gen-erate photoswitchable protein-protein interactions[50] thatmodulate gene expression,[51] actin polymerization,[52] pro-tein splicing,[53] and protein localization.[54]

Limited Structural Rearrangement

Continuing with the theme of light-activated proteins, sever-al photoswitches exhibit localized conformational changesupon photon absorption. Photosensory domains of the light-oxygen-voltage (LOV) class transduce photon absorption tophototropic and other blue light responses in plants andalgae. The structure of LOV2 from oat phototropin 1 con-sists of a PAS core domain to which an approximate 20-resi-due C-terminal helix (Ja) is loosely associated (Fig-ure 4 A).[55] When LOV2 absorbs blue light, the noncova-lently bound flavin mononucleotide (FMN) chromophoreforms a covalent adduct with a Cys residue. Structuralchanges originating from FMN are transmitted to Ja causingit to unfold (Figure 4 A). A recent paper reported that Ja

unfolding is mediated by unfolding of the N-terminal helix,

but the latter event has yet to be capitalized upon.[56] ThePAS core of LOV2 remains largely unperturbed.

The location of Ja at the very end of LOV2 proved to beprovidential, because Ja can be used to transduce conforma-tional changes between LOV2 and an effector protein fusedto its C-terminus. Sosnick, Moffat, and co-workers were thefirst to employ this design.[57] Their approach was to attachan effector protein that contains an N-terminal a-helix (theDNA-binding protein Trp repressor; TrpR) to the C-termi-nus of LOV2, such that the two proteins share Ja and com-pete for its binding. The system is thermodynamically tunedsuch that LOV2 has a greater affinity for Ja in the darkstate. The effector protein is rendered inactive by the ab-sence of its N-terminal helix. Photon absorption then shiftsthe relative binding affinity, allowing the target protein tosteal Ja from LOV2. Some structural (and amino acid se-quence) overlap between LOV2 and TrpR is needed to es-tablish this competition. To that end they progressively trun-cated the N-terminal helix of TrpR. The construct thatshowed the best photoactivation joined Ja of the LOV2domain to the middle of the N-terminal a-helix of the TrpRdomain. This variant, named LovTAP, exhibited a 5.6-foldincrease in DNA binding affinity after illumination.

In a later study, the same researchers attempted to en-hance coupling between photon absorption and DNA bind-ing by rational design.[58] They considered mutations thatwere predicted to preferentially stabilize either Ja-LOV2docking in the dark (off) state or Ja undocking in the lit(on) state. Two mutations in the Ja helix, G528A andN538E, stabilized docking in the dark state and increasedthe dynamic range of photoswitching from 5.6 to about 70.

Photoactive yellow protein (PYP) is a photoreceptor usedby Halorhodospira halophilia to detect and avoid blue light.In the dark state, PYP is fully folded and the covalentlybound p-coumaric acid chromophore is in the trans configu-ration. Photon absorption triggers trans-to-cis isomerizationfollowed by transfer of a proton from Glu46 to the chromo-phore. The presence of strain in the chromophore and thenegative charge on the buried Glu46 causes the first 25 resi-dues of PYP to unfold, exposing Glu46 and p-coumaric acidto the solvent and relieving the unfavorable interactions(Figure 4 B).[59] Woolley and colleagues created a light-acti-vated DNA-binding switch conceptually analogous toLovTAP by fusing the GCN4 DNA binding domain to theN-terminus of PYP.[60] As with LovTAP, they were able toimprove photoactivation by rational stabilization of the darkstate.[61]

LovTAP and the GCN4-PYP chimera require somedegree of homology between the ends of the input andoutput domains. Localized structural unfolding can, howev-er, achieve coupling in other ways as well. Crosson et al. re-placed the oxygen-sensing PAS B domain of the histidinekinase FixL with the LOV domain from the YtvA photore-ceptor to generate YF1, a kinase that is regulated by lightinstead of oxygen.[62] Further analysis suggested that switch-ing occurs by means of a rotary mechanism in which Ja me-diates a 40–608 twist of a coiled-coil that links the input and

Figure 4. Photoswitching by limited structural rearrangement. Dark statesof A) LOV2[123] and B) PYP[124] are shown. Photon absorption causes theC-terminal Ja helix of LOV2 (green) and the N-terminal 25 residues ofPYP (orange) to lose structure and detach from the core domains (blue).Chromophores are depicted in purple. PDB codes of LOV2 and PYP are2V0U and 1NWZ, respectively.

www.chemeurj.org � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2012, 18, 7984 – 79997990

S. N. Loh and J.-H. Ha

output domains.[63] The authors proposed that Ja and the N-terminal helix of the kinase unite to form a continuous sig-naling helix. Because PAS domains are widespread andsense a wide range of stimuli, this mix-and-match strategycan potentially be used to introduce photoswitching intoother PAS-containing proteins or to integrate multiplesignal responses into a single protein.[64]

Rac1 is a member of the Rho family of signaling GTP-ACHTUNGTRENNUNGases. Wu et al. created photoactivatable Rac1 (PA-Rac1) byfusing LOV2 to the N-terminus of Rac.[65] In the dark state,the close proximity of the two domains appears to stericallyinhibit Rac1 from interacting with its effector (p21-activatedkinase). Ja unfolding releases this occlusion. Functionalphotoswitches such as PA-Rac are a boon for the in vivomanipulation of biochemical pathways (“optogenetics”), be-cause light can be applied to specific locations in the cellwith almost no side effects (excluding radiation damage atsub-visible wavelengths).[66]

LOV and PYP depend on bound cofactors to trigger pho-toswitching. Chemically modifying a protein�s backbonewith a photoresponsive element is an alternative approachthat can in principle be applied to proteins that do not con-tain natural chromophores. Several recent designs draw par-allels to the cis/trans Pro switches discussed in the previoussection. Originally generated by cleaving ribonuclease Ainto two fragments using subtilisin, ribonuclease S(RNase S) is a longstanding model protein for folding stud-ies. RNase S is the native complex formed by binding andfolding of the S-peptide (residues 1–20) and the S-protein(residues 21–124). Fischer and co-workers synthesized an S-peptide variant that contained a thioxylated peptide bond(C=S in place of C=O) between Ala4 and Ala5.[67] The thio-xylated S-peptide was >99 % trans in reconstitutedRNase S, and, being nearly isosteric with the peptide bond,did not significantly diminish activity of the enzyme. Illumi-nation with 254 nm light increased the cis-thioxopeptidebond content to about 30 % without causing the S-peptideto dissociate. Initial rates of substrate hydrolysis decreasedby a similar amount, suggesting that cis-thioxo-RNase S iscatalytically inert and that photocontrol of catalytic activityhad been partially achieved. A similar result was reportedfor the functional interaction between thioxylated kinin andits receptor.[68] Thioxo and selenoxo[69] modifications estab-lish valuable triggers for the proven cis/trans switchingmechanism. The principle limitation is that the isomerizinggroups must be introduced by synthetic methods and this re-stricts their placement to peptides or very small proteins. Inaddition, they can be destabilizing (particularly to a-heli-ces)[70] and they may only partially isomerize upon irradia-tion.

Perhaps the most successful examples of switches that un-dergo limited conformational rearrangements are the de-signs based on the calmodulin (CaM) scaffold. CaM is an in-termediary messenger protein that links the functions of nu-merous target enzymes, ion channels, and other proteins tointracellular calcium concentration. CaM is composed oftwo helical domains connected by a stretch of about 27

amino acids (Figure 5 A). Each domain contains two Ca2+

-binding EF-hands. In the absence of metal, the domainsand linker lend CaM a roughly dumb-bell shaped appear-ance. Calcium binding triggers a unique structural transfor-mation that extends the linker into a flexible a-helix andcreates a binding surface from portions of the linker andEF-hand domains. Chiefly hydrophobic in character, thispocket is capable of recognizing up to several hundred regu-latory peptide and protein sequences.[71] The two EF-handdomains wrap tightly around the target peptide, resulting ina compact, globular complex (Figure 5 A). Approaches thatcapitalized on CaM�s distinctive conformational change togenerate a family of fluorescent calcium sensors, includingcameleons (Figure 5 B),[72] pericams,[73] and camgaroos,[74]

were reviewed elsewhere.[75]

What if it were possible to alter the binding specificity ofproteins that possess natural switching mechanisms, or graftcompletely new binding sites onto them? If so, then onecould make new biosensors by modifying CaM or PBPs torecognize new ligands and simply recycling the detectionmethodologies outlined above. Similarly, if catalytic functioncan be engineered into an existing switch then one can hopeto create allosterically regulated enzymes. By combiningbasic physical principles with modern computational tools,DeGrado and colleagues did just that.[76] Their goal was toconvert CaM into an enzyme that catalyzes Kemp elimina-tion in response to calcium binding. Carbon acid deprotona-

Figure 5. Calcium-triggered structural rearrangement of A) CaM andB) the cameleon biosensor. Structures of the calcium-free[125] and Ca2+

/peptide-bound[126] proteins are depicted at left and right, respectively.The N-terminal EF-hand is in orange, the C-terminal EF-hand is in red,the linker region is in green, and calcium ions are grey spheres. TheCaM-binding M13 peptide is in purple. Cameleon was created by fusingthe M13 peptide to the C-terminus of CaM, then attaching cyan FP(CFP) and yellow FP (YFP) to the amino and carboxy termini of thefusion protein. CFP and YFP are far apart in calcium-free cameleon;consequently, exciting CFP produces mostly cyan photon emission. Whenthe CaM domain of cameleon binds calcium it wraps around the attachedM13 peptide and brings CFP and YFP into close proximity. Exciting CFPnow produces fewer cyan photons and more yellow photons due to effi-cient FRET between CFP and YFP. PDB codes are 1DMO (left) and2BBM (right).

Chem. Eur. J. 2012, 18, 7984 – 7999 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 7991

REVIEWProtein Conformational Switches

tion is facilitated by a carboxylate base in a non-aqueous en-vironment. To that end they introduced a Glu residue intothe mostly hydrophobic pocket formed by the EF-handdomain and the linker in the Ca2+-bound state. The result-ing enzyme, named AlleyCat, exhibited a kcat/KM value of5.6 m

�1 s�1 in 10 mm Ca2+ , and a 25-fold lower catalytic effi-ciency in the absence of metal. Recent successes in compu-tational design of enzyme active sites[77] indicate that it maybe feasible to expand the repertoire of existing allosteric en-zymes and binding proteins to include new functionalswitches.

Global Fold Switching

Most amino acid sequences fold to a single structure thatcorresponds to the global minimum of free energy. Membersof the newly recognized class of “metamorphic” proteins,[78]

however, have the singular ability to interconvert betweentwo alternate native conformations, the energy minima ofwhich are relatively closely spaced. The distinction betweenthe conformational changes exhibited by metamorphic pro-teins and those of PBP, LOV, PYP, CaM, and so forth, isthat the latter changes can be regarded as perturbations ofa single native structure, whereas the former involve conver-sion between two unrelated folds. Lymphotactin (Ltn) is thearchetypical example of a metamorphic protein. At low tem-peratures and in the presence of NaCl, Ltn adopts the can-onical a/b chemokine fold.[79] In this form its function is tobind its target GPCR. At high temperature and no salt, Ltntakes on a dimeric all-b structure, the purpose of which is tobind glycosaminoglycan groups found in the extracellularmatrix. Virtually all of the tertiary contacts are different inthe two conformations,[79] and the entire molecule needs tounfold in order to transform from one state to the other.[78a]

The finding that the two forms are equally populated atphysiological temperature and in the presence of physiologi-cal NaCl concentration suggests that Ltn is poised to carryout either function in the cell. Other examples of metamor-phic proteins include the spindle checkpoint proteinMad2,[80] Cro transcription factors,[81] and the sugar-bindingprotein tachylectin-2.[82]

Only a handful of proteins are known to possess naturalfold-switching behavior. Several studies, however, suggestthat it may be feasible to engineer this capability into otherproteins. Proteins can be induced to switch conformationsby changing a relatively small number of amino acids, asshown for Cro proteins,[81, 83] Arc and Cro repressors,[84]

metal-dependent switches,[85] and Streptococcus protein G.[86]

Recent work of Bryan and Orban with protein G stands outas an example of this phenomenon. Protein G contains twotypes of domains: GA and GB (Figure 6). GA (45 structuredamino acids) adopts a three-helical fold and binds humanserum albumin, while GB (56 structured amino acids) takeson a mixed a/b fold and binds the constant region of IgG.No significant sequence homology exists between GA andGB. Alexander et al. set out to determine the minimum

number of mutations that are sufficient to convert one foldto the other. In a series of studies they progressively in-creased sequence identity until they found that the GA andGB folds remained autonomous and stable (DGfold��3 kcalmol�1) when all but three (positions 20, 30, and 45) of the 56structured amino acids were made to be the same.[87] Inother words, two polypeptides that are 95 % identical canfold into two completely different and stable structures!Commonizing the residues at positions 20 and 30 produceda 98 % identical pair in which GA98 and GB98 still adoptedthe respective all-a and a/b folds (Figure 6), but each wasdestabilized to the point where 10 % of the molecules wereunfolded. This instability allowed the single remaining “mu-tation” at position 45 (Leu in GA98 and Tyr in GB98) to dic-tate the switch between folds—a remarkable result.

Changing the identity of an amino acid is not a practicaltrigger for a conformational switch. Nonetheless, the abovestudy creates an impact, because it demonstrates that a pro-tein can be tweaked so that it is poised between two folds.[88]

One can imagine that, instead of mutation, binding energyor a change in solution conditions (c.f. Ltn) could tip thebalance to the other fold.

Folding–Unfolding Switches

The most dramatic conformational change that a proteincan undergo is also one that all proteins experience: foldingand unfolding. Folding makes for a particularly effectiveswitching mechanism for several reasons. First, becausea protein�s native function is typically abolished upon dena-turation, the unfolded form constitutes a default off-state infunctional switches. Second, so many of a protein�s proper-ties change as a result of folding–unfolding (structure, dy-namics, charge distribution, etc.) that it is likely thata means to detect binding can be devised when designingbiosensors. Third, a protein constantly cycles through all itspossible conformations—from native to globally unfolded toall states in between—according to their Boltzmann distri-butions. Thus, even a protein, the structure of which appears

Figure 6. Fold-switching of protein G. GA98 (left) and GB98 (right) differonly by a single amino acid at position 45 (sphere), but they adopt twocompletely different folds. Amino acid sequences are colored identicallyin the two structures. PDB codes are 2LHC (left) and 2LHD (right).

www.chemeurj.org � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2012, 18, 7984 – 79997992

S. N. Loh and J.-H. Ha

immutable, is already a reversible switch, though it spendsmost of its time in one state.

It is now recognized that a significant fraction of proteinsin the cell, perhaps as high as one-half,[89] contain regions ofintrinsic disorder. An IDP lacks well-defined structure,either in part or in whole, until it engages its cognate li-gand(s). The advantages that “unstructure” confers to mo-lecular recognition are not clear; some possibilities includeincreasing binding promiscuity, increasing the rate of macro-molecular association, and offering multiple sites for post-translational modification (e.g. phosphorylation).[7a,b,d] Breastcancer type 1 susceptibility protein (BRCA1) is an 1863-amino acid protein that functions in DNA repair, cell cyclecheckpoint control, and other processes that help maintaingenome integrity.[90] The N-terminal RING domain (residues1–103) and the C-terminal BRCT domain (residues 1646–1863) are structured, but the central region of about 1500amino acids is intrinsically disordered.[91] The central regionbinds a number of proteins involved in the DNA damage re-sponse including p53.[92] Cissell et al. utilized the 219–498fragment of BRCA1 as an intrinsically disordered bindingelement to detect p53.[93] Binding of p53 induced theBRCA1 fragment to fold, which was then detected by stra-tegic placement of environmentally sensitive tetramethyl-ACHTUNGTRENNUNGrhodamine or dansyl dyes.

When attempting to design a conformational switch, onemay draw on the axiom that it is easier to induce a proteinto unfold than to change native conformations. There aretwo main methods by which one can tip the thermodynamicbalance of stability towards the unfolded state: substitutionof one or a few amino acids (frequently in the tightlypacked core of the protein) and truncation of amino acidsfrom either terminus. Truncation can be beneficial becauseit does not require prior knowledge of 3D structure, and thedegree of destabilization can be tuned to an extent by delet-ing amino acids one at a time.[94] Kohn and Plaxco demon-strated these principles by progressively truncating aminoacids from the C-terminus of the SH3 domain (57 aminoacids) from Fyn tyrosine kinase.[95] It unfolded after thefourth deletion (CD4). Binding the cognate peptide inducedCD4 to fold, which was monitored by fluorescence of the in-trinsic Trp residue or that of a BODIPY group previouslyattached to a Cys side chain.

“Alternate frame folding” (AFF) incorporates elements ofbinding-induced folding and fold switching.[96] It is a tech-nique for modifying an existing binding protein so that it un-dergoes a binding-induced (or cleavage-induced) conforma-tional change. We have applied AFF to convert barnase intoan artificial zymogen[97] and calbindin D9k and RBP into flu-orescent sensors for calcium[96,98] and ribose (unpublished),respectively. The main steps in this process are illustratedfor RBP in Figure 7. First, a segment from the carboxy ter-minus is duplicated and appended to the amino terminususing a peptide linker long enough to span the ends of theWT protein (alternately, a segment from the amino terminuscan be copied and joined to the carboxy terminus). Thelength of this segment is determined by two factors: 1) it

must contain at least one amino acid critical for ligand bind-ing, and 2) it should end at a surface loop. RBP containsmany surface loops, but it is generally recommended tochoose the one closest to the key binding residue (Asp218)in order to minimize length. The twin segment establishesa second “frame” of protein folding. RBP can fold to theWT native structure (N-fold) using amino acids 1–277. Itcan also fold using amino acids (210’–277’)-linker-(1–209) toyield the circularly permuted structure (N’-fold). The dupli-cate peptides are expected to extend from the amino or car-boxy termini (in the N- and N’-fold, respectively) as disor-dered tails. The protein cannot adopt N and N’ conforma-tions simultaneously, because they compete for residues 1–

Figure 7. Conformational switching by alternate frame folding. Schemat-ics of the N-fold and N’-fold of RBP-AFF are depicted at top andbottom, respectively. Amino acid sequences are shown below each struc-ture. The critical binding residue (D218) and permutation site (position210) are indicated. Residues 210–277 (red) are duplicated and appendedto the amino terminus using a flexible peptide linker (green). The dupli-cated amino acids (pink) are indicated by prime superscripts but are oth-erwise numbered identically to those of the parent sequence. The bind-ing-knockout mutation D218A is introduced into the red sequence. Inthe absence of ligand, the N-fold (residues 1–277) predominates over theN’-fold (residues 210’–277’, the linker, and residues 1–209). Addition ofribose (yellow), and the binding energy provided by D218’ (pink spheres)in the remodeled binding pocket, drives the shift from the N-fold to theN’-fold. The protein cannot simultaneously adopt N- and N’-folds be-cause they compete for a shared sequence (grey). The dashed linesdenote that the pink segment is unstructured in the N-fold and the redsegment is unstructured in the N’-fold. Placement of distance-sensitivefluorophores at the amino terminus and at position 210 links the confor-mational change to an optical output.

Chem. Eur. J. 2012, 18, 7984 – 7999 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 7993

REVIEWProtein Conformational Switches

209. It consequently interconverts between the two formswith the equilibrium distribution determined by their rela-tive stabilities. Asp218 is mutated to Ala in the more stableof the two forms (N-fold in this case) so that ligand bindingshifts the population to the N’-fold. Finally, the conforma-tional change is detected by placing fluorophores at the N-terminus and in the loop where the permutation took place.These sites are naturally suited to FRET because they arespatially close in N’ and are predicted to be more distant inN.

Calbindin D9k (75 amino acids) corresponds to just one ofthe EF-hand domains in CaM. It does not appreciablychange structure on calcium binding.[99] Nevertheless, Strat-ton et al. converted calbindin to a fluorescent calciumsensor (calbindin-AFF) by the AFF method.[96] SubsequentNMR structural studies verified the fold shift and demon-strated that the duplicate peptides in calbindin-AFF doindeed extend from N and N’ as disordered tails.[98b] Thisresult is significant, because it suggests that similar binding-induced fluorescence changes can be elicited from otherAFF-based sensors. AFF, however, is not limited to biosen-sor design. Mitrea et al. used it to convert barnase intoa functional switch (barnase-AFF) that is activated by pro-teolytic cleavage, that is, an artificial zymogen.[97] Barnasewas modified as in Figure 7, except a key catalytic residuewas mutated instead of a ligand-binding residue, and anHIV-1 protease recognition sequence was inserted into theloop that bore the permutation site. Protease cleavage trig-gered the shift from the catalytically inert N conformationto the active N’ form, increasing kcat/KM by about 250-fold.

Folding–unfolding switches appear to be major players incellular mechano-transduction pathways. Mechano-sensingproteins are stimulated by mechanical forces, for example,those that are generated by fluid flow, membrane deforma-tion, and stresses at cell-cell and cell-extracellular matrixjunctions. The giant protein titin is one of the most wellstudied examples of a mechano-sensor.[100] By integratingactin and myosin filaments at sarcomeres, titin is responsiblefor passive elasticity of the sarcomere as well as transducingmuscle stress into a host of signaling responses. Its spring-like properties are established by the linear array of repeat-ing Ig and fibronectin-like domains that unfold and refoldwith muscle load. It also contains a single titin kinasedomain (TK), the catalytic activity of which is controlled bya unique force-triggered mechanism. The C-terminal tail ofTK consists of two a-helices and a b-strand. In the un-stressed state, the tail inhibits TK by wrapping around thecatalytic domain and occluding the nucleotide bindingsite.[101] The structurally related myosin light-chain kinasesare activated when CaM binds to their tails and exposes theATP pocket. TK and twitchin kinase domains are not acti-vated by CaM binding, but rather by mechanical unfoldingof the inhibitory tail.[102] The approximate 30 pN gatingforce required to unfold the tail without unfolding the cata-lytic domain is consistent with the pulling force generatedby the six myosin heads in the crown of a myosin filament.Cadherins,[103] integrins,[104] filamin A,[105] PKD domains of

polycistin-1,[106] and the negative regulatory region of Notchreceptor[107] are other examples of mechano-sensors thatappear to operate by force-induced unfolding, whereas pro-teins such as the MscL ion channel[108] and the AT1 andother GPCRs[109] likely respond to mechanical stimuli withconformational changes that do not involve unfolding.

As with all of the switching mechanisms discussed here,a basic question is how to incorporate a natural phenomen-on—force-induced unfolding in this case—into new switchdesigns. Mutually exclusive folding is one such ap-proach.[42,110] MEF is a special class of protein-in-proteinfusion. Rather than attempting to make the insertion struc-turally compatible to both host and guest (e.g. by circularlypermuting the guest), a guest is chosen with an N-to-C dis-tance at least twice as long as the end-to-end distance of theloop in the host (Figure 2 B, left-most structure). The gueststretches the host at the point of insertion, while the host si-multaneously compresses the ends of the guest (Figure 2 B,second structure from left). The more thermodynamicallystable of the two domains unfolds the less stable domainand renders it inactive (Figure 2 B, second structure fromright). One can take advantage of the intrinsic coupling be-tween folding and binding to generate a protein thatswitches between active and inactive conformations depend-ing on ligand. Using the GCN4 DNA binding domain (75 �N-to-C distance) as the guest and the ribonuclease barnase(10 � end-to-end loop distance) as the host, Ha et al. creat-ed a chimera in which the RNase activity of the barnasedomain was regulated by DNA binding to the GCN4domain.[111]

More recently we showed that MEF can also be a switchfor triggering self-assembly.[112] When ubiquitin was insertedinto each of six different surface loops of barnase, the fusionproteins formed dimers, trimers, tetramers, and higher-orderoligomers. The X-ray structure of one of the insertion var-iants revealed a long, linear polymer in which the bindinginterface between protomers consisted of domain-swappedbarnase subunits (Figure 2 B, right-most structure). The mo-lecular stress imposed on the monomeric barnase domain byubiquitin was relieved by unfolding of the barnase domainfollowed by intermolecular refolding with identically dis-rupted barnase subunits. The structures of both ubiquitinand barnase domains in the swapped oligomer were virtuallyidentical to their WT counterparts, suggesting that it may bepossible to create self-assembling oligomers and polymersthat retain and integrate the functions of the parent mole-cules.

One distinction of the MEF design is that it may be mod-ular. Any stable guest protein with a moderately long N-to-C distance should be able to perform the same stretchingfunction as ubiquitin and GCN4. As to the host, there is noreason to expect that barnase is unique in its ability todomain swap upon forced unfolding. On the other hand,Peng and Li created an MEF chimera in which the 27th Igdomain of titin was inserted into a surface loop of the GB5protein.[113] They were able to directly witness unfolding ofGB5 by Ig in real time, but subsequent refolding/domain

www.chemeurj.org � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2012, 18, 7984 – 79997994

S. N. Loh and J.-H. Ha

swapping of GB5 was not observed. Therefore, the princi-ples that determine whether MEF generates a self-assemblyswitch (Figure 2 B, right-most structure) or a functionalswitch (Figure 2 B, second structure from right) are largelyunknown.

The main concerns with folding–unfolding-based switchesare that unfolded proteins tend to misfold and/or aggregate,particularly in complex environments such as the inside ofa cell, and degradation can be a concern. Natural IDPs haveevolved the ability to remain soluble when unfolded, likelydue (at least in part) to their atypically high ratio of chargedto hydrophobic amino acids.[89] Destabilization of native pro-teins, on the other hand, is implicated as one of the rootcauses of amyloidoses and other aggregation-related diseas-es.[114] SH3 CD4 remained monomeric and soluble at>100 mm concentration and it was not degraded when ex-pressed in bacterial cells. Calbindin-AFF and the MEF con-structs were similarly well behaved, but the extent to whichunfolded forms of large proteins retain these favorable char-acteristics remains to be determined.

Adjusting the Properties of ConformationalSwitches

The defining characteristics of any switch include reversibili-ty, response time, sensitivity (how much stimulus is requiredfor a response), dynamic range (over what range of stimulusconcentration the switch responds), and gain (signal changebetween on and off states). These properties can be under-stood—and perhaps manipulated—by recognizing that con-formational switching originates from the transitions be-tween structural substates. According to this view, the on-and off-states are part of the native ensemble of structures;they are local minima in a free-energy surface that are sepa-rated by a distinct energy barrier. Protein switches thuscycle between on and off states as one of their natural dy-namic modes. This notion leads to the following conclusions.

Reversibility and response time : Conformational switchesare usually reversible, though the forward and reverse ratesare highly variable. For example, apo-MBP interconverts be-tween the open state (95% populated at equilibrium) andthe closed state (5 % populated) on a timescale of ns–ms.[115]

PBPs therefore bind their ligands with rates approachingthe diffusion limit.[116] The overall response time (i.e., howfast a PBP sensor can respond to a change in ligand concen-tration) is limited by the off-rate, which can be slow forhigh-affinity PBPs. The ParM open-to-closed change, on theother hand, appears to be slower and ADP binds with anon-rate of 105

m�1 s�1.[117] Folding-based switches are likewise

inherently reversible as long as the unfolded proteins do notaggregate or become degraded. Unfolded SH3 CD4 refoldswith a rate constant of 102 s�1, which is comparable to therate at which it responds to its ligand.[95] In other instancesthe recognition event may be rapid, but the overall rate ofswitching is limited by a slower process that was introduced

by fusion with a reporter domain or other modification. Thefirst 25 amino acids of PYP unfold within a few millisecondsafter photoexcitation,[118] but the corresponding sequencefrom the PYP-GCN4 chimera does not fully unfold anddetach from the PYP domain in the absence of the DNAligand.[61] Though calbindin binds calcium with a diffusion-limited rate of 109

m�1 s�1,[119] calbindin-AFF switches confor-

mations on the timescale of seconds.[96] The difference is ap-parently due to an unfolding step being at least partiallyrate-limiting.

In some cases it has been possible to adjust responsetimes using rational or random mutagenesis. For switches inwhich the rates are limited by local or global unfolding, forexample, calbindin-AFF and Ltn,[78a] introducing a destabiliz-ing mutation generally increases unfolding rates and thisstrategy hastened the response time of calbindin-AFF.[98a]

Structural characterization of intermediates in the LOVphototransduction pathway allowed Moffat and colleaguesto make mutants in which the LOV2 photocyle was acceler-ated or decelerated by up to 80-fold.[47] The comparativelyslow rate of Pro cis/trans isomerization (typically 100–10�3 s�1) has been suggested to control the timing of variouscellular processes. PPIases accelerate this rate by up to 106-fold. Pro isomerization is among the only example ofa switching mechanism for which the response time can besped up by an enzyme.

Gain and sensitivity : According to the conformational sub-state view, gain and sensitivity are determined by the free-energy difference between on- and off-states. High gain re-quires that the latter be of much lower energy than theformer in the absence of stimulus so that nearly all mole-cules are initially in the off position. The trade-off is thatsensitivity may suffer: the amount of effector required toinvert the populations (i.e. turn on the switch) is proportion-al to their energetic difference. One can turn this relation-ship into an advantage by rationally adjusting energy levels(thermodynamic tuning). For example, many biosensor ap-plications demand that Kd match the concentration range ofanalyte to be detected. A PBP may bind glucose with micro-molar Kd, but it would fail as a blood sugar monitor for dia-betes, since it would always be saturated by the millimolarglucose content present in serum. One can attempt to raiseKd by mutating residues in the binding pocket, but thesesubstitutions often alter specificity. An alternate approach isto employ binding-induced folding. Because binding energyis used to invert the unfolded (off) and folded (on) popula-tions, the apparent Kd is weaker than the intrinsic Kd of thenative protein by a factor of (1+ Kfold)/Kfold, in which Kfold =

exp(�DGfold/RT). To illustrate, Kd of SH3 CD4 was 100-foldhigher than that of WT SH3,[95] and the calcium affinity ofcalbindin-AFF was tuned in both directions by introducinga destabilizing point mutation into either the N or N’frame.[98a] The same strategy was used to suppress cytotoxicRNase activity of barnase-AFF in its uncleaved form tonearly undetectable levels.[97] For the micromolar Kd glucosesensor mentioned above, one could achieve the requisite

Chem. Eur. J. 2012, 18, 7984 – 7999 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 7995

REVIEWProtein Conformational Switches

103-fold reduction in glucose affinity by destabilizing theWT PBP by about 4 kcal mol�1. The principle of thermody-namic tuning applies to the other examples presented hereas well, although in practice it is challenging for switches forwhich mechanisms are not well characterized (e.g. LOV2,PYP) or for which populations are not easily perturbed (Proswitches).

Finally, dynamic range remains a limitation of currentswitch designs. For biosensors, simple one-site binding dic-tates that the minimal useful signal change (which typicallymeans inverting the populations from >10:1 off:on to>10:1 on:off) occurs over less than a 102-fold change in ef-fector concentration. Many assays specify that the switch re-spond over broader or narrower concentration regimes.Vall�e-B�lisle et al. recently showed that dynamic range canbe extended many orders of magnitude by combining var-iants of the same switch that possess different binding affini-ties.[120] Although they demonstrated this principle by alter-ing stability (and therefore Kd) of a DNA aptamer by meansof base-pair substitution, the same thermodynamic tuningstrategy can in principle be applied to protein switches. Forapplications that demand a steep response curve overa narrow concentration range, they introduced a receptorthat binds the target with high affinity but does not undergoconformational switching. This “depletant” acts as a sink toprevent free ligand from accumulating until the total targetconcentration exceeds that of the depletant.

Summary and Outlook

Proteins are unique in nature in that they incorporatea large array of conformational change modes into an equal-ly diverse repertoire of biological functions. Cells and virus-es employ protein conformational switches for molecularrecognition, environmental sensing, and regulating signalingpathways involved in development, homeostasis, immunity,infection, and other cellular processes. Structural and bio-chemical elucidation of some of these switching mechanismshas enabled researchers to begin adapting them for use inbiosensing and functional regulation. As more structuresbecome available, the main challenges shift to altering spe-cificity of existing binding domains, and learning how totransduce natural conformational changes within a bindingdomain to a functional or optical output module. Thenumber of ways in which two (or more) proteins can bejoined is combinatorial. Existing studies, few as they are,suggest that subtle structural changes in one domain canalter function of an attached domain in a manner that is notanticipated or even understood. Thus, future efforts willlikely benefit from the collaboration of library-based screen-ing methods and computational simulations.

A parallel tack is to exploit the conformational changecommon to all proteins—the folding–unfolding reaction—toconvert ordinary proteins into switches. Natural IDPs thatfold upon binding provide a source of prospective biosen-sors as well as an inspiration for like designs. An altogether

new class of switch can be envisioned by once again fusingtwo (or more) proteins, then using the physics that linkbinding to folding (and perhaps folding to unfolding) to es-tablish communication between domains within the fusionprotein, as well as between the fusion protein and its envi-ronment. Hilser and colleagues recently presented a modelof allostery that recapitulates the intricate switching behav-ior shown by some natural systems, whereupon a givenligand can act as both an antagonist and an agonist.[121] Thephysical manifestation of their formalism is a multidomainprotein in which the domains interact favorably or unfavora-bly when in their high-affinity states. The allosteric responseis greatest when one or more of the domains folds uponbinding; no additional conformational change is needed.The possibility of creating complex switches by incorporat-ing this concept with some of the designs discussed herein isespecially captivating.

Acknowledgements

This work was supported by NIH. grant R01GM069755 to S.N.L.

[1] S. G. Rasmussen, B. T. DeVree, Y. Zou, A. C. Kruse, K. Y. Chung,T. S. Kobilka, F. S. Thian, P. S. Chae, E. Pardon, D. Calinski, J. M.Mathiesen, S. T. Shah, J. A. Lyons, M. Caffrey, S. H. Gellman, J.Steyaert, G. Skiniotis, W. I. Weis, R. K. Sunahara, B. K. Kobilka,Nature 2011, 477, 549 – 555.

[2] S. C. Harrison, Nat. Struct. Mol. Biol. 2008, 15, 690 –698.[3] W. Weissenhorn, A. Dessen, S. C. Harrison, J. J. Skehel, D. C.

Wiley, Nature 1997, 387, 426 –430.[4] D. R. Beniac, S. L. deVarennes, A. Andonov, R. He, T. F. Booth,

PLoS One 2007, 2, e1082.[5] a) V. N. Malashkevich, B. J. Schneider, M. L. McNally, M. A.

Milhollen, J. X. Pang, P. S. Kim, Proc Natl Acad Sci USA 1999, 96,2662 – 2667; b) W. Weissenhorn, A. Carfi, K. H. Lee, J. J. Skehel,D. C. Wiley, Mol. Cell 1998, 2, 605 – 616.

[6] a) M. Gerstein, W. Krebs, Nucleic Acids Res. 1998, 26, 4280 –4290;b) S. Flores, N. Echols, D. Milburn, B. Hespenheide, K. Keating, J.Lu, S. Wells, E. Z. Yu, M. Thorpe, M. Gerstein, Nucleic Acids Res.2006, 34, D296 –301; c) G. Qi, R. Lee, S. Hayward, Bioinformatics2005, 21, 2832 –2838; d) F. Seeliger, T. Leeb, M. Peters, M. Brug-mann, M. Fehr, M. Hewicker-Trautwein, Vet. Pathol. 2003, 40, 530 –539; e) T. Amemiya, R. Koike, S. Fuchigami, M. Ikeguchi, A.Kidera, J. Mol. Biol. 2011, 408, 568 –584.

[7] a) P. E. Wright, H. J. Dyson, Curr. Opin. Struct. Biol. 2009, 19, 31–38; b) M. M. Babu, R. van der Lee, N. S. de Groot, J. Gsponer,Curr. Opin. Struct. Biol. 2011, 21, 432 –440; c) C. K. Fisher, C. M.Stultz, Curr. Opin. Struct. Biol. 2011, 21, 426 –431; d) P. Tompa,Curr. Opin. Struct. Biol. 2011, 21, 419 –425.

[8] a) K. Chen, I. Bachtiar, G. Piszczek, F. Bouamr, C. Carter, N. Tjan-dra, Biochemistry 2008, 47, 1928 – 1937; b) C. Tang, E. Loeliger, P.Luncsford, I. Kinde, D. Beckett, M. F. Summers, Proc. Natl. Acad.Sci. USA 2004, 101, 517 –522.

[9] C. M. Petit, J. Zhang, P. J. Sapienza, E. J. Fuentes, A. L. Lee, Proc.Natl. Acad. Sci. USA 2009, 106, 18249 –18254.

[10] M. W. Clarkson, S. A. Gilmore, M. H. Edgell, A. L. Lee, Biochem-istry 2006, 45, 7693 – 7699.

[11] S. R. Tzeng, C. G. Kalodimos, Curr. Opin. Struct. Biol. 2011, 21,62– 67.

[12] B. F. Volkman, D. Lipson, D. E. Wemmer, D. Kern, Science 2001,291, 2429 – 2433.

[13] a) X. I. Ambroggio, B. Kuhlman, Curr. Opin. Struct. Biol. 2006, 16,525 – 530; b) M. Ostermeier, Curr. Opin. Struct. Biol. 2009, 19, 442 –

www.chemeurj.org � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2012, 18, 7984 – 79997996

S. N. Loh and J.-H. Ha

448; c) M. Ostermeier, Protein Eng. Des. Sel. 2005, 18, 359 –364;d) C. M. Wright, R. A. Heins, M. Ostermeier, Curr. Opin. Chem.Biol. 2007, 11, 342 –346; e) A. Vall�e-B�lisle, K. W. Plaxco, Curr.Opin. Struct. Biol. 2010, 20, 518 –526; f) S. Koide, Curr. Opin. Bio-technol. 2009, 20, 398 – 404; g) M. V. Golynskiy, M. S. Koay, J. L.Vinkenborg, M. Merkx, ChemBioChem 2011, 12, 353 –361;h) M. M. Stratton, S. N. Loh, Protein Sci. 2011, 20, 19– 29.

[14] L. M�ler, J. Blankenship, M. Rance, W. J. Chazin, Nat. Struct. Biol.2000, 7, 245 –250.

[15] M. A. Dwyer, H. W. Hellinga, Curr. Opin. Struct. Biol. 2004, 14,495 – 504.

[16] a) I. Lager, M. Fehr, W. B. Frommer, S. Lalonde, FEBS Lett. 2003,553, 85– 89; b) A. J. Bjçrkman, S. Mowbray, J. Mol. Biol. 1998, 279,651 – 664; c) K. P. Ravindranathan, E. Gallicchio, R. M. Levy, J.Mol. Biol. 2005, 353, 196 –210; d) N. C. Vercillo, K. J. Herald, J. M.Fox, B. S. Der, J. D. Dattelbaum, Protein Sci. 2007, 16, 362 – 368.

[17] a) I. L. Medintz, J. R. Deschamps, Curr. Opin. Biotechnol. 2006, 17,17– 27; b) J. D. Dattelbaum, L. L. Looger, D. E. Benson, K. M. Sali,R. B. Thompson, H. W. Hellinga, Protein Sci. 2005, 14, 284 –291.

[18] M. J. Borrok, Y. Zhu, K. T. Forest, L. L. Kiessling, ACS Chem.Biol. 2009, 4, 447 – 456.

[19] U. Magnusson, B. N. Chaudhuri, J. Ko, C. Park, T. A. Jones, S. L.Mowbray, J. Biol. Chem. 2002, 277, 14077 –14084.

[20] N. Akiyama, K. Takeda, K. Miki, J. Mol. Biol. 2009, 392, 559 – 565.[21] M. J. Cuneo, L. S. Beese, H. W. Hellinga, J. Biol. Chem. 2009, 284,

33217 – 33223.[22] a) S. Trakhanov, N. K. Vyas, H. Luecke, D. M. Kristensen, J. Ma,

F. A. Quiocho, Biochemistry 2005, 44, 6597 –6608; b) J. S. Sack,S. D. Trakhanov, I. H. Tsigannik, F. A. Quiocho, J. Mol. Biol. 1989,206, 193 –207; c) G. A. Bermejo, M. P. Strub, C. Ho, N. Tjandra,Biochemistry 2010, 49, 1893 – 1902; d) G. A. Bermejo, M. P. Strub,C. Ho, N. Tjandra, J. Am. Chem. Soc. 2009, 131, 9532 –9537; e) A.Scir�, A. Marabotti, M. Staiano, L. Iozzino, M. S. Luchansky, B. S.Der, J. D. Dattelbaum, F. Tanfani, S. D’Auria, Mol. BioSyst. 2010,6, 687 –698.

[23] a) N. K. Karpowich, H. H. Huang, P. C. Smith, J. F. Hunt, J. Biol.Chem. 2003, 278, 8429 –8434; b) K. P. Locher, E. Borths, FEBSLett. 2004, 564, 264 – 268.

[24] B. C. Chu, H. J. Vogel, Biol. Chem. 2011, 392, 39 –52.[25] C. Cavazza, L. Martin, E. Laffly, H. Lebrette, M. V. Cherrier, L.

Zeppieri, P. Richaud, M. Carriere, J. C. Fontecilla-Camps, FEBSLett. 2011, 585, 711 – 715.

[26] a) L. Zhang, M. Koay, M. J. Maher, Z. Xiao, A. G. Wedd, J. Am.Chem. Soc. 2006, 128, 5834 –5850; b) F. Arnesano, L. Banci, I. Ber-tini, A. R. Thompsett, Structure 2002, 10, 1337 – 1347; c) F. Hussain,E. Sedlak, P. Wittung-Stafshede, Arch. Biochem. Biophys. 2007,467, 58 –66.

[27] I. Alicea, J. S. Marvin, A. E. Miklos, A. D. Ellington, L. L. Looger,E. R. Schreiter, J. Mol. Biol. 2011, 414, 356 – 369.

[28] B. Vergauwen, J. Elegheert, A. Dansercoer, B. Devreese, S. N. Sav-vides, Proc. Natl. Acad. Sci. USA 2010, 107, 13270 – 13275.

[29] R. M. De Lorimier, J. J. Smith, M. A. Dwyer, L. L. Looger, K. M.Sali, C. D. Paavola, S. S. Rizk, S. Sadigov, D. W. Conrad, L. Loew,H. W. Hellinga, Protein Sci. 2002, 11, 2655 –2675.

[30] M. Fehr, W. B. Frommer, S. Lalonde, Proc. Natl. Acad. Sci. USA2002, 99, 9846 –9851.

[31] M. Fehr, S. Lalonde, I. Lager, M. W. Wolff, W. B. Frommer, J. Biol.Chem. 2003, 278, 19127 –19133.

[32] K. Teasley Hamorsky, C. M. Ensor, Y. Wei, S. Daunert, Angew.Chem. 2008, 120, 3778 –3781; Angew. Chem. Int. Ed. 2008, 47,3718 – 3721.

[33] a) F. van den Ent, J. Moller-Jensen, L. A. Amos, K. Gerdes, J.Lowe, EMBO J. 2002, 21, 6935 –6943; b) S. Kunzelmann, M. R.Webb, Biochem. J. 2011, 440, 43 –49.

[34] S. Kunzelmann, M. R. Webb, J. Biol. Chem. 2009, 284, 33130 –33138.

[35] a) T. Wan, R. L. Beavil, S. M. Fabiane, A. J. Beavil, M. K. Sohi, M.Keown, R. J. Young, A. J. Henry, R. J. Owens, H. J. Gould, B. J.Sutton, Nat. Immunol. 2002, 3, 681 – 686; b) M. D. Holdom, A. M.

Davies, J. E. Nettleship, S. C. Bagby, B. Dhaliwal, E. Girardi, J.Hunt, H. J. Gould, A. J. Beavil, J. M. McDonnell, R. J. Owens, B. J.Sutton, Nat. Struct. Mol. Biol. 2011, 18, 571 – 576.

[36] J. Hunt, A. H. Keeble, R. E. Dale, M. K. Corbett, R. L. Beavil, J.Levitt, M. J. Swann, K. Suhling, S. Ameer-Beg, B. J. Sutton, A. J.Beavil, J. Biol. Chem. 2012, in press.

[37] P. Sarkar, T. Saleh, S. R. Tzeng, R. B. Birge, C. G. Kalodimos, Nat.Chem. Biol. 2011, 7, 51– 57.

[38] A. Severin, R. E. Joseph, S. Boyken, D. B. Fulton, A. H. Andreotti,J. Mol. Biol. 2009, 387, 726 – 743.

[39] S. C. Lummis, D. L. Beene, L. W. Lee, H. A. Lester, R. W. Broad-hurst, D. A. Dougherty, Nature 2005, 438, 248 – 252.

[40] W. Y. Lee, S. M. Sine, Nature 2005, 438, 243 –247.[41] a) C. Schiene-Fischer, T. Aum�ller, G. Fischer, Top Curr. Chem.

2012, in press; b) K. P. Lu, G. Finn, T. H. Lee, L. K. Nicholson, Nat.Chem. Biol. 2007, 3, 619 – 629; c) Y. C. Liou, X. Z. Zhou, K. P. Lu,Trends Biochem. Sci. 2011, 36, 501 –514.

[42] T. L. Radley, A. I. Markowska, B. T. Bettinger, J.-H. Ha, S. N. Loh,J. Mol. Biol. 2003, 332, 529 – 536.

[43] a) G. Guntas, T. J. Mansell, J. R. Kim, M. Ostermeier, Proc. Natl.Acad. Sci. USA 2005, 102, 11224 – 11229; b) G. Guntas, S. F. Mitch-ell, M. Ostermeier, Chem. Biol. 2004, 11, 1483 –1487; c) G. Guntas,M. Ostermeier, J. Mol. Biol. 2004, 336, 263 – 273.

[44] J. Liang, J. R. Kim, J. T. Boock, T. J. Mansell, M. Ostermeier, Pro-tein Sci. 2007, 16, 929 –937.

[45] C. M. Wright, A. Majumdar, J. R. Tolman, M. Ostermeier, ProteinsStruct. Funct. Bioinf. 2010, 78, 1423 –1430.

[46] B. K. Shoichet, Nat. Biotechnol. 2007, 25, 1109 –1110.[47] A. Mçglich, K. Moffat, Photochem. Photobiol. Sci. 2010, 9, 1286 –

1300.[48] a) N. C. Rockwell, Y. S. Su, J. C. Lagarias, Annu. Rev. Plant Biol.

2006, 57, 837 –858; b) A. Nagatani, Curr. Opin. Plant Biol. 2010,13, 565 –570.

[49] a) A. T. Ulijasz, G. Cornilescu, C. C. Cornilescu, J. Zhang, M.Rivera, J. L. Markley, R. D. Vierstra, Nature 2010, 463, 250 –254;b) X. Yang, J. Kuk, K. Moffat, Proc. Natl. Acad. Sci. USA 2009,106, 15639 – 15644.

[50] J. E. Toettcher, D. Gong, W. A. Lim, O. D. Weiner, Methods Enzy-mol. 2011, 497, 409 – 423.

[51] S. Shimizu-Sato, E. Huq, J. M. Tepperman, P. H. Quail, Nat. Bio-technol. 2002, 20, 1041 –1044.

[52] D. W. Leung, C. Otomo, J. Chory, M. K. Rosen, Proc. Natl. Acad.Sci. USA 2008, 105, 12797 –12802.

[53] A. B. Tyszkiewicz, T. W. Muir, Nat. Methods 2008, 5, 303 – 305.[54] J. E. Toettcher, D. Gong, W. A. Lim, O. D. Weiner, Nat. Methods

2011, 8, 837 –839.[55] S. M. Harper, L. C. Neil, K. H. Gardner, Science 2003, 301, 1541 –

1544.[56] J. P. Zayner, C. Antoniou, T. R. Sosnick, J. Mol. Biol. 2012, in

press.[57] D. Strickland, K. Moffat, T. R. Sosnick, Proc. Natl. Acad. Sci. USA

2008, 105, 10709 –10714.[58] D. Strickland, X. Yao, G. Gawlak, M. K. Rosen, K. H. Gardner,

T. R. Sosnick, Nat. Methods 2010, 7, 623 –626.[59] a) J. Vreede, J. Juraszek, P. G. Bolhuis, Proc. Natl. Acad. Sci. USA

2010, 107, 2397 – 2402; b) A. Xie, L. Kelemen, J. Hendriks, B. J.White, K. J. Hellingwerf, W. D. Hoff, Biochemistry 2001, 40, 1510 –1517; c) C. Bernard, K. Houben, N. M. Derix, D. Marks, M. A. vander Horst, K. J. Hellingwerf, R. Boelens, R. Kaptein, N. A. van Nu-land, Structure 2005, 13, 953 –962; d) G. Cheng, M. A. Cusanovich,V. H. Wysocki, Biochemistry 2006, 45, 11744 –11751.

[60] a) S. A. Morgan, S. Al-Abdul-Wahid, G. A. Woolley, J. Mol. Biol.2010, 399, 94– 112; b) S. A. Morgan, G. A. Woolley, Photochem.Photobiol. Sci. 2010, 9, 1320 – 1326.

[61] H. Y. Fan, S. A. Morgan, K. E. Brechun, Y. Y. Chen, A. S. Jaikaran,G. A. Woolley, Biochemistry 2011, 50, 1226 –1237.

[62] S. Crosson, S. Rajagopal, K. Moffat, Biochemistry 2003, 42, 2– 10.[63] A. Mçglich, R. A. Ayers, K. Moffat, J. Mol. Biol. 2009, 385, 1433 –

1444.

Chem. Eur. J. 2012, 18, 7984 – 7999 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 7997

REVIEWProtein Conformational Switches

[64] A. Mçglich, R. A. Ayers, K. Moffat, J. Mol. Biol. 2010, 400, 477 –486.

[65] Y. I. Wu, D. Frey, O. I. Lungu, A. Jaehrig, I. Schlichting, B. Kuhl-man, K. M. Hahn, Nature 2009, 461, 104 –108.

[66] a) S. K. Yoo, Q. Deng, P. J. Cavnar, Y. I. Wu, K. M. Hahn, A. Hut-tenlocher, Dev. Cell 2010, 18, 226 – 236; b) X. Wang, L. He, Y. I.Wu, K. M. Hahn, D. J. Montell, Nat. Cell Biol. 2010, 12, 591 –597.

[67] D. Wildemann, C. Schiene-Fischer, T. Aumuller, A. Bachmann, T.Kiefhaber, C. Lucke, G. Fischer, J. Am. Chem. Soc. 2007, 129,4910 – 4918.

[68] Y. Huang, Z. Cong, L. Yang, S. Dong, J. Pept. Sci. 2008, 14, 1062 –1068.

[69] Y. Huang, G. Jahreis, C. Lucke, D. Wildemann, G. Fischer, J. Am.Chem. Soc. 2010, 132, 7578 –7579.

[70] A. Reiner, D. Wildemann, G. Fischer, T. Kiefhaber, J. Am. Chem.Soc. 2008, 130, 8079 –8084.

[71] X. Shen, C. A. Valencia, J. W. Szostak, B. Dong, R. Liu, Proc. Natl.Acad. Sci. USA 2005, 102, 5969 – 5974.

[72] A. Miyawaki, J. Llopis, R. Heim, J. M. McCaffery, J. A. Adams, M.Ikura, R. Y. Tsien, Nature 1997, 388, 882 –887.

[73] T. Nagai, A. Sawano, E. S. Park, A. Miyawaki, Proc. Natl. Acad.Sci. USA 2001, 98, 3197 –3202.

[74] G. S. Baird, D. A. Zacharias, R. Y. Tsien, Proc. Natl. Acad. Sci.USA 1999, 96, 11241 – 11246.

[75] a) B. N. G. Giepmans, S. R. Adams, M. H. Ellisman, R. Y. Tsien,Science 2006, 312, 217 – 224; b) N. Demaurex, M. Frieden, Cell Cal-cium 2003, 34, 109 – 119.

[76] I. V. Korendovych, D. W. Kulp, Y. Wu, H. Cheng, H. Roder, W. F.DeGrado, Proc. Natl. Acad. Sci. USA 2011, 108, 6823 – 6827.

[77] a) D. Rçthlisberger, O. Khersonsky, A. M. Wollacott, L. Jiang, J.DeChancie, J. Betker, J. L. Gallaher, E. A. Althoff, A. Zanghellini,O. Dym, S. Albeck, K. N. Houk, D. S. Tawfik, D. Baker, Nature2008, 453, 190 –195; b) L. Jiang, E. A. Althoff, F. R. Clemente, L.Doyle, D. Rothlisberger, A. Zanghellini, J. L. Gallaher, J. L.Betker, F. Tanaka, C. F. Barbas III, D. Hilvert, K. N. Houk, B. L.Stoddard, D. Baker, Science 2008, 319, 1387 – 1391.

[78] a) R. C. Tyler, N. J. Murray, F. C. Peterson, B. F. Volkman, Bio-chemistry 2011, 50, 7077 –7079; b) A. G. Murzin, Science 2008, 320,1725 – 1726.

[79] R. L. Tuinstra, F. C. Peterson, S. Kutlesa, E. S. Elgin, M. A. Kron,B. F. Volkman, Proc. Natl. Acad. Sci. USA 2008, 105, 5057 – 5062.

[80] a) X. Luo, Z. Tang, G. Xia, K. Wassmann, T. Matsumoto, J. Rizo,H. Yu, Nat. Struct. Mol. Biol. 2004, 11, 338 –345; b) X. Luo, H. Yu,Structure 2008, 16, 1616 –1625.

[81] C. G. Roessler, B. M. Hall, W. J. Anderson, W. M. Ingram, S. A.Roberts, W. R. Montfort, M. H. Cordes, Proc. Natl. Acad. Sci. USA2008, 105, 2343 –2348.

[82] I. Yadid, N. Kirshenbaum, M. Sharon, O. Dym, D. S. Tawfik, Proc.Natl. Acad. Sci. USA 2010, 107, 7287 –7292.

[83] W. J. Anderson, L. O. Van Dorn, W. M. Ingram, M. H. Cordes, Pro-tein Eng. Des. Sel. 2011, 24, 765 – 771.

[84] a) T. A. Anderson, M. H. Cordes, R. T. Sauer, Proc. Natl. Acad.Sci. USA 2005, 102, 18344 –18349; b) M. C. Mossing, R. T. Sauer,Science 1990, 250, 1712 – 1715.

[85] a) X. I. Ambroggio, B. Kuhlman, J. Am. Chem. Soc. 2006, 128,1154 – 1161; b) Y. Hori, Y. Sugiura, J. Am. Chem. Soc. 2002, 124,9362 – 9363; c) E. Cerasoli, B. K. Sharpe, D. N. Woolfson, J. Am.Chem. Soc. 2005, 127, 15008 –15009.

[86] S. Dalal, L. Regan, Protein Sci. 2000, 9, 1651 – 1659.[87] a) P. A. Alexander, Y. He, Y. Chen, J. Orban, P. N. Bryan, Proc.

Natl. Acad. Sci. USA 2009, 106, 21149 –21154; b) P. A. Alexander,Y. He, Y. Chen, J. Orban, P. N. Bryan, Proc. Natl. Acad. Sci. USA2007, 104, 11963 –11968; c) P. A. Alexander, D. A. Rozak, J.Orban, P. N. Bryan, Biochemistry 2005, 44, 14045 –14054; d) Y. He,Y. Chen, P. Alexander, P. N. Bryan, J. Orban, Proc. Natl. Acad. Sci.USA 2008, 105, 14412 – 14417.

[88] Y. He, Y. Chen, P. A. Alexander, P. N. Bryan, J. Orban, Structure2012, 20, 283 –291.

[89] V. N. Uversky, A. K. Dunker, Biochim. Biophys. Acta Proteins Pro-teomics 2010, 1804, 1231 – 1264.

[90] a) P. J. O�Donovan, D. M. Livingston, Carcinogenesis 2010, 31,961 – 967; b) M. S. Huen, S. M. Sy, J. Chen, Nat. Rev. Mol. Cell Biol.2010, 11, 138 –148.

[91] W. Y. Mark, J. C. Liao, Y. Lu, A. Ayed, R. Laister, B. Szymczyna,A. Chakrabartty, C. H. Arrowsmith, J. Mol. Biol. 2005, 345, 275 –287.

[92] H. Zhang, K. Somasundaram, Y. Peng, H. Tian, D. Bi, B. L. Weber,W. S. El-Deiry, Oncogene 1998, 16, 1713 –1721.

[93] K. A. Cissell, S. Shrestha, J. Purdie, D. Kroodsma, S. K. Deo, Anal.Bioanal. Chem. 2008, 391, 1721 – 1729.

[94] J. M. Flanagan, M. Kataoka, D. Shortle, D. M. Engelman, Proc.Natl. Acad. Sci. USA 1992, 89, 748 –752.

[95] J. E. Kohn, K. W. Plaxco, Proc. Natl. Acad. Sci. USA 2005, 102,10841 – 10845.

[96] M. M. Stratton, D. M. Mitrea, S. N. Loh, ACS Chem. Biol. 2008, 3,723 – 732.

[97] D. M. Mitrea, L. Parsons, S. N. Loh, Proc. Natl. Acad. Sci. USA2010, 107, 2824 –2829.

[98] a) M. M. Stratton, S. N. Loh, Proteins Struct. Funct. Bioinf. 2010,78, 3260 – 3269; b) M. M. Stratton, S. McClendon, D. Eliezer, S. N.Loh, Biochemistry 2011, 50, 5583 –5589.

[99] N. J. Skelton, J. Kordel, W. J. Chazin, J. Mol. Biol. 1995, 249, 441 –462.

[100] M. Gautel, Pfluegers Arch. 2011, 462, 119 –134.[101] O. Mayans, P. F. van der Ven, M. Wilm, A. Mues, P. Young, D. O.

Furst, M. Wilmanns, M. Gautel, Nature 1998, 395, 863 –869.[102] a) E. M. Puchner, A. Alexandrovich, A. L. Kho, U. Hensen, L. V.

Schafer, B. Brandmeier, F. Grater, H. Grubmuller, H. E. Gaub, M.Gautel, Proc. Natl. Acad. Sci. USA 2008, 105, 13385 – 13390; b) F.Gr�ter, J. Shen, H. Jiang, M. Gautel, H. Grubmuller, Biophys. J.2005, 88, 790 –804.

[103] a) D. E. Leckband, Q. le Duc, N. Wang, J. de Rooij, Curr. Opin.Cell Biol. 2011, 23, 523 – 530; b) G. A. Gomez, R. W. McLachlan,A. S. Yap, Trends Cell Biol. 2011, 21, 499 –505.

[104] M. A. Schwartz, D. W. DeSimone, Curr. Opin. Cell Biol. 2008, 20,551 – 556.

[105] B. A. Kesner, F. Ding, B. R. Temple, N. V. Dokholyan, Proteins2010, 78, 12–24.

[106] a) J. R. Forman, Z. T. Yew, S. Qamar, R. N. Sandford, E. Paci, J.Clarke, Structure 2009, 17, 1582 – 1590; b) G. Dalagiorgou, E. K.Basdra, A. G. Papavassiliou, Int. J. Biochem. Cell Biol. 2010, 42,1610 – 1613.

[107] J. Chen, A. Zolkiewska, PLoS One 2011, 6, e22837.[108] C. A. Kumamoto, Nat. Rev. Microbiol. 2008, 6, 667 – 673.[109] M. Mederos y Schnitzler, U. Storch, T. Gudermann, Curr. Opin.

Pharmacol. 2011, 11, 112 –116.[110] a) T. Cutler, S. N. Loh, J. Mol. Biol. 2007, 371, 308 –316; b) T.

Cutler, B. M. Mills, D. J. Lubin, L. T. Chong, S. N. Loh, J. Mol.Biol. 2009, 386, 854 – 868.

[111] J.-H. Ha, J. S. Butler, D. M. Mitrea, S. N. Loh, J. Mol. Biol. 2006,357, 1058 – 1062.

[112] J.-H. Ha, J. M. Karchin, N. Walker-Kopp, L.-S. Huang, E. A. Berry,S. N. Loh, J. Mol. Biol. 2012, 416, 495 – 502.

[113] Q. Peng, H. Li, J. Am. Chem. Soc. 2009, 131, 13347 – 13354.[114] L. M. Luheshi, C. M. Dobson, FEBS Lett. 2009, 583, 2581 –2586.[115] C. Tang, C. D. Schwieters, G. M. Clore, Nature 2007, 449, 1078 –

1082.[116] D. M. Miller III, J. S. Olson, J. W. Pflugrath, F. A. Quiocho, J. Biol.

Chem. 1983, 258, 13665 –13672.[117] S. Kunzelmann, M. R. Webb, ACS Chem. Biol. 2010, 5, 415 – 425.[118] D. Pan, A. Philip, W. D. Hoff, R. A. Mathies, Biophys. J. 2004, 86,

2374 – 2382.[119] S. R. Martin, S. Linse, C. Johansson, P. M. Bayley, S. Forsen, Bio-

chemistry 1990, 29, 4188 –4193.[120] A. Vall�e-B�lisle, F. Ricci, K. W. Plaxco, J. Am. Chem. Soc. 2012,

134, 2876 – 2879.

www.chemeurj.org � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2012, 18, 7984 – 79997998

S. N. Loh and J.-H. Ha

[121] H. N. Motlach, V. J. Hilser, Proc. Natl. Acad. Sci. USA 2012, inpress.

[122] A. J. Bjçrkman, R. A. Binnie, H. Zhang, L. B. Cole, M. A. Her-modson, S. L. Mowbray, J. Biol. Chem. 1994, 269, 30206 –30211.

[123] A. S. Halavaty, K. Moffat, Biochemistry 2007, 46, 14001 –14009.[124] E. D. Getzoff, K. N. Gutwin, U. K. Genick, Nat. Struct. Biol. 2003,

10, 663 –668.

[125] M. Zhang, T. Tanaka, M. Ikura, Nat. Struct. Biol. 1995, 2, 758 – 767.[126] M. Ikura, G. M. Clore, A. M. Gronenborn, G. Zhu, C. B. Klee, A.

Bax, Science 1992, 256, 632 –638.

Published online: June 11, 2012

Chem. Eur. J. 2012, 18, 7984 – 7999 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 7999

REVIEWProtein Conformational Switches