Protein engineering for molecular electronics Stephen G. Sligar and F. Raymond Salemme University of Illinois at Urbana-Champaign, Urbana, Illinois and Sterling Winthrop Pharmaceutical Research Division, Malvern, Pennsylvania, USA
Recombinant DNA technology allows the manipulation of the physical properties of proteins that perform electron transport and photochemical processes. Recent work is reviewed that has a potential impact on the development of molecular electronic devices, within a general framework outlining strategies for device fabrication. This review is also published in Current Opinion in Structural Biology 1992, 2:587-592.
Current Opinion in Biotechnology 1992, 3:388-393
Introduction This review is p r o m p t e d b y the recent enthusiasm for the use of biological macromolecules as building blocks for the construction of molecular electronic devices (MEDs). Molecular electronics is a term that has a dual meaning. One meaning refers to electronic devices that use materials w h o s e unique properties result from their molecular structure. In this regard, proteins can be used in MEDs to sense, r e s p o n d to, or record chemical, electrical, or physical stimuli. Examples include the use of the bacterial photopigment bacteriorhodopsin as a three-dimensional medium for the storage and readout of optically e n c o d e d information [1",2",3]. The second definition of molecular electronics embodies the concepts that individual molecules are functional units responding to stimuli, and that they have the potential to b e interconnected in order to replicate electronic circuits functionally. G o o d examples of such molecular electronic devices do not yet exist, but the principles necessary for their design and assembly are beginning to emerge from a variety of areas.
From the perspective of practical device design, there are several levels at which the ability to modify protein molecules using recombinant DNA technology could have an impact on the fabrication of MEDs. These include alteration of intrinsic optical and electronic properties, increasing protein stability for operation in non-biological environments, and modification of surface properties to facilitate d e n o v o design of mesoscale molecular assemblies or molecular circuits. Current activities in each of these categories are described below.
Modification of protein electronic and physical properties Naturally occurring proteins incorporate a wide range of physical properties that are of potential use in electronic devices. In most cases, the useful properties derive from a prosthetic group such as a metal center, organic r e d o x cofactor, or chromophore. Although naturally occurring redox and photoactive proteins exhibit great functional diversity, this is achieved through a limited n u m b e r of prosthetic groups w h o s e properties are modulated through interaction with amino acid side chains of the surrounding protein moiety. Attempts to engineer proteins in order to modify their properties in useful ways have followed precedents suggested b y the study of structure-function relationships in natural systems. In this respect, heme-containing proteins, which comprise a diverse set of molecules with electron transfer, ligand binding, or catalytic function, are of particular interest. Alterations in the electronics or chemistry of the h e m e iron center frequently produce large changes in optical spectroscopic or magnetic properties which are useful as indicators for state assignment or device readout. Following pioneering w o r k on cytochrome b5 [4], myoglobin [5-8], and cytochrome p450 [9], recent w o r k reporting site-directed modifications of residues that are heme iron ligands, or can otherwise influence h e m e spectral properties, include additional studies on cytochrome p450 [10] and myoglobin [11], together with studies on cytochrome c peroxidase [12,13] and iso-l-cytochrome c [14]. Although the study of the structure-property relationships that affect heme proteins remains an area of active interest [15"], few systematic efforts have b e e n
Abbreviation MED--molecular electronic device. 388
© Current Biology Ltd ISSN 0958-1669
Protein engineering for molecular electronics Sligar and Salemme 389 directed at producing unusually stable molecules that could increase the reliability of state assignment. H o w ever, a serendipitous enhancement of protein stability was reported for a site-directed mutant of iso-l-cytochrome c [14], in which an internal asparagine had b e e n changed to a hydrophobic isoleucine residue. This alteration resulted in the loss of an internal water molecule w h o s e location in the interior of the native protein was postulated to b e energetically unfavorable. Similar hydrophobic enhancements could direct further efforts towards the important goal of stabilizing redox proteins in non-aqueous environments. A property central to the function of many MEDs is the regulation of electron-transfer rate between proteins or b e t w e e n proteins and an external oxidoreductant. Physical aspects of the electron-transfer p r o cesses in proteins are n o w relatively well understood, and suggest that electron-transfer rates d e p e n d primarily u p o n prosthetic group separation, difference in oxidoreduction potential and molecular reorganization energy [16",17]. This is consistent with recent site-directed modifications of the invariant Phe82 residue in yeast iso-l-cytochrome c, suggesting that intermolecular electron transfer does not require the participation of specific aromatic amino acids as electron wires between protein prosthetic groups [18,19]. Nevertheless, translation of the physical requirements for efficient electron transfer into a specific modification strategy remains complicated, as s h o w n by the results obtained b y Barker et al. for yeast iso-l-cytochrome c [20]. In addition to heine-containing redox proteins the introduction or regeneration of copper-binding sites in proteins has b e e n investigated. In one study, both type I and type II c o p p e r site properties could be obtained b y addition of an appropriate external ligand to an azurin mutant in which one of the native histidine c o p p e r ligands had b e e n deleted b y site-directed mutagenesis [21]. In a second study, site-directed modifications of cytochrome c were performed in order to introduce a c o p p e r ligand site [22]. In both cases, the modified proteins a p p e a r e d to lack the intrinsic stability of the native molecule, but may point the w a y to successive generations of molecules with useful functions.
Interaction specificity Because rates of intermolecular transfer d e p e n d very strongly on prosthetic group separation distance [16-, 17], reactions between reversibly binding electrontransfer proteins d e p e n d on interactions that ensure the proper relative intermolecular orientation. Early w o r k that m o d e l e d the interactions of cytochromes c and b5 established the importance of solvent exclusion and complementary electrostatic interactions in radox protein interactions [23]. This precedent has since b e e n generalized to m a n y other biological electron-transfer interactions. With the advent of site-directed mutagenesis methods, investigation of intermolecular recognition interactions has intensified [24-29]. Most notably,
methods have recently b e e n developed that discriminate between the relative contributions of the principal components of intermolecular recognition: polar interactions that result from hydrogen b o n d s and salt bridges forming in the interaction domain, and non-polar van der Waals interactions that occur w h e n protein surfaces dehydrate upon formation of the c o m p l e m e n tary complex [30"]. This m e t h o d o l o g y also facilitates m a p p i n g the interaction domain between two associating macromolecules, a k e y step in the d e v e l o p m e n t of generalized strategies to both study natural systems and engineer alternative interactions a m o n g molecules suitable for use in MEDs.
Bacteriorhodopsin in MED applications Bacteriorhodopsin (molecular weight 26000 D) functions as a light-driven proton p u m p in the purple m e m brane of the salt-marsh micro-organism Halobacterium halobium [1",2"]. Bacteriorhodopsin incorporates a retinal chromophore, which is covalently b o u n d as a protonated Schiff base in an all-trans conformation in the resting state. U p o n absorbing a light photon, the retinal photoisomerizes and subsequently undergoes a multistep photocycle that spans much of the visible spectrum. The photocycle can b e stopped at specific intermediates by low temperature trapping and recycled b y thermalization or application of a second light pulse of the appropriate wavelength [1",2"]. Photoactivated retinal switching occurs with high quantum yields at disparate wavelengths, which facilitate state assignment, and produces b o t h changes in protein refractive index and a photoelectric potential in oriented assemblies. These properties, together with the excellent stability of the protein w h e n immobilized in polym e r films or gels, form the basis for a variety of prototypes for data storage media, holographic memory, a n d electro-optic applications [1",2",3,31,32"]. Three-dimensional photochromic memories have incorporated bacteriorhodopsin that has b e e n oriented b y an external electric field and then immobilized in a solid polyacrylamide matrix [2"]. This application exploits the two-photon absorption cross section of bacteriorhodopsin and requires that a small v o l u m e of the matrix be simultaneously illuminated by intersecting laser beams for data storage. Readout is achieved by reillumination, which produces an electrical signal on the cuvette surface that depends o n the state of the irradiated volume in the data storage matrix. In this context, it is interesting to note recent experiments demonstrating the maintenance of redox protein properties w h e n immobilized in silicate glasses '[33"], as well as preservation and control of proteolytic e n z y m e activity in a photochromic azobenzene copolym e r [34]. Although most readily applied in biosensor applications, both schemes could potentially be useful in three-dimensional optical m e m o r y applications that utilize photochromic properties of incorporated
390
Protein engineering proteins or that couple protein catalytic properties to the photochromic matrix. Current w o r k on bacteriorhodopsin with relevance to molecular electronics involves finding additives, alternative pigments, or site-directed mutants of the protein that will alter the relative stabilities of the photo-intermediates [35], enhance stabilization of the final M intermediate at r o o m temperature [36], or alter other aspects of photocycle-coupled proton translocation [37,38] that might facilitate readout from information storage applications. Consequently, investigations of the Asp96 Asn modification [39], which alters coupling of retinal photoisomerization to proton translocation, continue to have substantial practical interest [40]. The physical changes in the modified protein include an increase in the lifetime of the intermediate from 10 ms to 750 ms, together with improved diffraction efficiency and photochromic sensitivity relative to the native protein. Considerable latitude for property improvement exists using a combination of site-directed modifications at the c h r o m o p h o r e binding site, introduction of alternative chromophores, or the engineering of modified ion-binding sites, all of which can affect intermediate lifetimes and spectral properties [1",2"]. In this regard, the recent structure determination of bacteriorhodopsin by electron diffraction methods [41"], coupled with rapidly advancing methods of computational simulation [42"], may provide n e w insights into the engineering of bacteriorhodopsin for MED applications.
Oriented protein immobilization on surfaces
prosthetic groups in sensor or non-linear optical applications depends critically on the ability to control prosthetic g r o u p orientation precisely relative to the electrode or optical w a v e g u i d e surface. Site-specific introduction of anchoring points in a protein of k n o w n three-dimensional structure provides a straightforward solution to this problem. As an example, recent site-directed modifications of the heine protein cytochrome b5 carried out at the University of Illinois introduced cysteine residues at specified positions on the molecular surface. This allowed oriented coupling to a silane substrate and produced immobilized molecular arrays demonstrating a high level of heme prosthetic group orientation as determined b y linear dichroism spectroscopy. Methods to introduce two-dimensional periodicity on surface molecular arrays d e p e n d either on engineering both intermolecular interactions [30"] and anchoring sites for surface immobilization, or on schemes that erect molecular assemblies on a scaffold already possessing periodic order on the few tens of angstroms scale. Scaffold possibilities include natural two-dimensional lattices [43], DNA lattices [51"], or the very highly ordered synthetic lattices formed from streptavidin anchored to surfaces through biotinylated phospholipids [52"]. As streptavidin is a tetrameric protein with four biotin-binding sites arranged with approximate tetrahedral geometry, two biotin-binding sites per tetramer remain f r e e in the two dimensional arrays and provide a regular lattice for attaching additional molecules (Fig. 1). Molecular electronics
Streptavidin Bacteriorhodopsin is not unique in its ability to order spontaneously in two-dimensional films that resemble its native m e m b r a n e state and, indeed, other m e m brane protein systems have b e e n suggested for use or have potential in a variety of MED or related biosensor roles. Recent work includes studies of engineered bacterial transmembrane pores [43] and isolated bacterial reaction centers immobilized on electrode surfaces [44]. In these cases, surface orientation of the protein complexes results from preexisting interactions between the protein complexes or their m e m b r a n e environments. However, a n u m b e r of experiments have b e e n carried out that involve tethering proteins to electrode surfaces through polymeric 'wires' [45,46] or through surface electrostatic interactions [47] in order to enhance electron-transfer efficiences. Related studies of interest examine properties of soluble proteins conjugated to photo- or redox-active organic substituents that enhance enzyme catalytic function in the absence of natural cofactors or regeneration systems [48-50]. It is easy to envision hybrid systems that incorporate proteins with conjugated cofactors immobilized on electrode surfaces. Although the forgoing studies emphasize electron conduction b e t w e e n device components, the efficiency of m a n y devices incorporating
components
Biotinylated redox protein
Biotinylated DNA [~ giotinylated lipid ~ Lipid-anchored redox protein
Hypothetical molecular electronics architecture Fig. 1. Illustration of molecular electronics components and potential architectural features of self-assembling systems restricted to diffusion in two-dimensional lipid films.
Protein engineering for molecular electronics Sligar and Salemme
Strategies for protein patterning and circuit fabrication Although it a p p e a r s possible that quite complicated molecular assemblies could b e erected o n two-dimensional protein lattices using a combination of chemical conjugation, site-directed introduction o f a n c h o r points, and molecular fusions at the DNA level (rev i e w e d elsewhere in this issue), the materials described thus far all c o m e u n d e r our first definition o f molecular electronics as bulk materials. Although these substrates could u n d o u b t e d l y b e patterned using lithogr a p h y m e t h o d s similar to t h o s e used in fabricating integrated circuits, optical diffraction w o u l d limit detail to scales substantially larger than individual molecular assemblies. Such devices w o u l d not realize the ultimate in miniaturization of individually assignable devices inherent in molecular electronics of the s e c o n d definition, w h e r e a s individual molecular assemblies w o u l d lack the u n i q u e addressability and specific connectivity b e t w e e n c o m p o n e n t s c o m m o n l y associated with electronic circuits. Nevertheless, it is interesting to consider h o w it might be possible to construct true molecular circuits, as well as the dual p r o b l e m at the molecular level of h o w such devices could be addressed a n d read. Atomic p r o b e m i c r o s c o p y could potentially b e u s e d to manipulate and assemble molecules as a means o f constructing devices, as well as a m e a n s of setting molecular states or reading o u t information [53"]. The limitations of manipulating only o n e or a few c o m p o n e n t s at a time, however, w o u l d s e e m ultimately to defeat the objectives o f creating a molecular device. Alternatively, it m a y b e possible to 'condition' a preexisting molecular lattice of uniformly interconnected molecules so that it is able to store or process information as a neural n e t w o r k analog. Information processing using a distributed a p p r o a c h requires each m o l e c u l a r - a s s e m b l y to sense the state of its nearest neighbors a n d alter its configuration accordingly. With phasing to an external clock, this s c h e m e has the formal structure o f a cellular a u t o m a t o n operating with individual m a c r o m o l e c u l e s as building blocks. In such a highly cooperative device, it m a y n o t be necessary to manipulate individual molecules directly, but utilize intermolecular c o m m u n i c a t i o n s c o m m o n to m a n y cooperative multimeric proteins to p e r f o r m s o m e analog c o m p u t i n g function. Fabrication possibilities also exist using defect tessellation automata [54"], w h e r e very complicated structures can be generated b y defect introduction into a periodic lattice subject to a regular site r e p l a c e m e n t protocol. Although presently only the subject of c o m p u t e r simulation, it seems possible to e m b o d y the required properties in ligand-capture a n d displacement schemes using symmetric protein molecules with multiple b i n d i n g sites. Access and readout could potentially be achieved b y any of the methods outlined above. While these devices are presently w h o l l y conceptual, and m a y ultimately be limited b y molecular noise that will necessitate distributed or red u n d a n t processing to ensure reliability, it seems clear
that g e n u i n e a p p r o a c h e s exist to investigate the limits o f this technology.
Conclusion Practical application of e n g i n e e r e d proteins in molecular electronics and sensor materials applications are clearly o n the near horizon. Protein molecules represent the ultimate miniaturization possible in individual devices w h o s e state can potentially be controlled indep e n d e n t of its neighbors. This property, together with the e m e r g e n c e of a k n o w l e d g e base allowing protein properties and interaction specificity to be e n g i n e e r e d with relative ease, will u n d o u b t e d l y lead to progressively m o r e sophisticated assemblies, w h o s e functional limitations can only n o w be g u e s s e d at. Multidisciplinary p r o g r a m s such as those in the Soviet U n i o n that p i o n e e r e d d e v e l o p m e n t o f b a c t e r i o r h o d o p s i n based devices [1",3], or the Frontier Research P r o g r a m of the Riken Institute in J a p a n [55"], represent f o c u s e d efforts to realize the potential of this technology.
References and recommended reading Papers of particular interest, published within the annual period of review, have been highlighted as: of special interest •. of outstanding interest 1.
OESTERHELTD, BRAUCHLEC, HAMpp N: Bacteriorhodopsin: a Biological Material for Information Processing. Quart Rev Biophys !991, 24:425~78.
A recent review that summarizes structure-function relationships in bacteriorhodopsin and includes a discussion of site-specific mutants. BIRGERR: Photophysics and Molecular Electronic Applications o f the Rhodopsins. Annu Rev Phys Chem 1990, 41:683-733. This review emphasizes MED applications. 2.
3.
BELCHINSKYGY, BOGATOV PN, GRESKO AP, KONENKO AA, LUKASHEVEP, CHAMOROVSKYSK: Application o f Pigment-Protein Complexes as Carriers o f Optic Information. Biofizika 1991, 36:248-251.
4.
SLIGARS, EGEBERGK, SAGET, MORIKISD, CHAMPIONP: Alteration o f Heme Axial Ligands b y Site Directed Mutagenesis: a Cytochrome Becomes a Catalytic Demethylase. J Am Chem Soc 1987, 109:7896-7897.
5.
IAMBRIGHTDG, BALASUBRAMANIANS, BOXERSG: Ligand and Proton Exchange I)ylmmics i n Recombinant Human Myoglobin Mutants. J Mol Biol 1989, 207:289-299.
6.
VARADARAJAN R, ZEWERTWE, GRAy HB, BOXERSG: Effects of Buried Ionizable Amino Acids o n the Reduction Potential of Recombinant Myoglobin. Science 1989, 243:69-72.
7.
VARADARAJANR, LAMBRLGHTDG, BOXERSG: Electrostatic Interactions in Wild-type and Mutant Recombinant H u m a n Myoglobins. Biochemistry 1989, 28:3771-3781.
8.
EGEBERGKD, SPRINGERBA, MARTINISSA, SLIGARSG, MORIKIS D, CHAMPIONPM: Alteration of Sperm Whale Myoglobin Axial Ligation b y Site-directed Mutagenesis. Biochemistry 1990, 29:9783-9791.
391
392
Protein engineering 9.
DI PRIMO C, HUI BON HOA G, D o u z o u P, SL1GARS: Mut a g e n e s i s o f a Single H y d r o g e n B o n d i n C y t o c h r o m e P-450 Alters Cation B i n d i n g a n d H e i n e Solvation. J Biol Chem 1990, 265:5361-5363.
22.
TODD RJ, VANDAMME, CASIMIRO D, HAYMORE BL, ARNOLD FH: Cu(II)-binding P r o p e r t i e s of a C y t o c h r o m e - c w i t h a S y n t h e t i c Metal-binding Site - - His X3-His i n a n Alphahelix. Proteins 1991, 10:156-161.
10.
IWASAKIM, JUVONEN R, LINDBERG R, NEGISHI M: Alteration o f H i g h a n d Low S p i n E q u i l i b r i u m b y a Single Mutat i o n o f Amirto Acid-209 i n M o u s e C y t o c h r o m e - P 4 5 0 . J Biol Chem 1991, 266:3380-3382.
23.
SALEMMEFR: A H y p o t h e t i c a l Structure for a n Intermolecular Electron Transfer Complex of Cytochrome c a n d b5. J Mol Biol 1976, 102:563-568.
24.
11.
ADACHIS, NAGANOS, WATANABEY, ISHIMORI K, MORISHIMAI; A l t e r a t i o n o f H u m a n Myoglobin P r o x i m a l H i s t i d i n e to C y s t e i n e o r T y r o s i n e b y Site-directed M u t a g e n e s i s - C h a r a c t e r i z a t i o n a n d Their Catalytic Activities. Biochem Biophys Res C o m m u n 1991, 180:138-144.
RODGERSK, POCHAPSKY T, SLIGAR S: P r o b i n g t h e Mecha n i s m s o f M a c r o m o l e c u l a r Recognition: t h e Cyt o c h r o m e b 5 - c y t o c h r o m e c C o m p l e x . Science 1988, 240:1657-1659.
25.
STAY'YONPS, SLIGAR SG: T h e C y t o c h r o m e P - 4 5 0 c a m B i n d i n g Surface as D e f i n e d by Site-directed Mutagen e s i s a n d Electrostatic Modeling. Biochemistry 1990, 29:7381-7386.
26.
SHIMIZUT, TATEISHI T, HATANO M, FUJlIKURIyAMAY: Probi n g t h e Role o f Lysines a n d A r g i n i n e s i n t h e Catalytic F u n c t i o n o f C y t o c h r o m e - P 4 5 0 d by Site Directed Mutag e n e s i s - - I n t e r a c t i o n w i t h NADPH-cytochrome-P450 Reductase. J Biol Chem 1991, 266:3372-3375.
27.
COGHLANVM, VICKERYLE: Site-specific M u t a t i o n s i n Hum a n F e r r e d o x i n T h a t Affect B i n d i n g to F e r r o d o x i n R e d u c t a s e a n d Cytochrome-P450scc. J Biol Chem 1991, 266:18606-18612.
28.
SHIMIZU T, TATEISHI T, HATANO M, FUJlIKURIYAMA Y: P r o b i n g t h e Role o f Lysines a n d A r g l n i n e s i n t h e Catalytic F u n c t i o n o f Cytochrome-P450D b y Site-directed M u t a g e n e s i s - I n t e r a c t i o n w i t h NADPHC y t o c h r o m e - P 4 5 0 Reductase. J Biol Chem 1991, 266:3372-3375.
29.
SHIRABEK, YUBISUIT, NISHINO T, TAKESHITAM: Role o f Cyst e i n e R e s i d u e s i n H u m a n NADH-cytochrome-b5 Reductase Studies b y Site-directed M u t a g e n e s i s - - Cys-273 a n d Cys-283 are Located Close to t h e NADH-binding Site b u t a r e n o t Catalytically E s s e n t i a l J Biol Chem 1991, 266:7531-7536.
12.
13.
14.
GOODIN DB, DAVIDSON MG, ROE JA, MAUK AG, SMITH M: A m i n o Acid S u b s t i t u t i o n s at T r y p t o p h a n - 5 1 o f C y t o c h r o m e - c P e r o x i d a s e ~ Effects o n C o o r d i n a t i o n , Species P r e f e r e n c e f o r C y t o c h r o m e - c , a n d E l e c t r o n Transfer. Biochemistry 1991, 30:4953-4962. SMULEVICHG, MILLER /VIA, KRAUT J, SPIRO TG: Conform a t i o n a l C h a n g e a n d Histidine C o n t r o l o f Heine C h e m i s t r y i n C y t o c h r o m e - e P e r o x i d a s e - - Reson a n c e R a m a n E v i d e n c e f r o m Leu-52 a n d GIg-181 Mut a n t s o f C y t o c h r o m e - c Peroxidase. Biochemistry 1991, 30:9546-9558. HICKEYDR, BERGHUIS AM, LAFOND G, JAEGERJA, CARDILLO TS, MCLENDON D, DAS G, SHERMAN F, BRAYER GD: E n h a n c e d T h e r m o d y n a m i c Stabilities o f Yeast iso1 - c y t o c h r o m e - c w i t h A m i n o Acid R e p l a c e m e n t at Position-52 a n d Position-102. J Biol Chem 1991, 266:11686-11694.
15. •.
FRAUENFELDERH, SUGAR SG, WOLYNES PG: T h e E n e r g y L a n d s c a p e s a n d M o t i o n s o f P r o t e i n s . Science 1991, 254:1598-1603. Describes current concepts about the dynamic interconversion among conformational substates in proteins and the resulting physical manifestations. This is of fundamental importance in understanding how thermal phenomena will affect switching reliability in prorein molecules when excited by external stimuli. 16.
MOSER CC, KESKEJM, WARNCKEK, FARID RS, DLrI~ON PL: N a t u r e o f Biological Electron Transfer. Nature 1992, 355:796-802. This paper summarizes m u c h experimental data showing that electron transfer rates effectively depend on prosthetic group separation distance, differences in prosthetic group free energy and reorganization energies upon oxidoreduction. The treatment suggests that the protein presents a uniform electronic tunnelling barrier that derives from glass-like protein properties also revealed in dynamic studies. 17.
BERATANDN, ONUCHIJN, BETrS JN, BOWLER BE, GRAY HB: Electron Tunneling Pathways in Ruthenated Proteins. J A m Chem Soc 1990, 112:7915-7921.
18.
EVERESTAM, WALLIN SA, STEMP ED, NOCEKJM, MAUK AG, HOFFMAN B: A r o l n a t i c Hole S u p e r e x c h a n g e T h r o u g h Position-82 i n C y t o e h r o m e C is Not R e q u i r e d f o r I n t r a c o m p l e x E l e c t r o n T r a n s f e r to Z n - C y t o c h r o m e C P e r o x i d a s e . A m Chem Soc 1991, 113:4337--4338.
19.
INGLIS SC, GUILLEMETTEJG, JOHNSON JA, SMITH M: Analy s i s o f t h e I n v a r i a n t Phe82 R e s i d u e o f Yeast Iso1 - c y t o c h r o m e c b y Site-directed M u t a g e n e s i s U s i n g a P h a g e m i d Yeast Shuttle Vector. Protein Eng 1991, 4:569-574.
20.
BARKERPD, MAUKMR, MAUKAG: P r o t o n Titration Curve o f Yeast I s o - l - F e r r i c y t o c h r o m e - c - - Electrostatic a n d C o n f o r m a t i o n a l Effects o f P o i n t Mutations. Biochemistry 1991, 30:2377-2383.
21.
DENBLAAUWENT, VANDEKAMPM, CANTER GW: Type-I a n d Type-II C o p p e r Sites O b t a i n e d b y E x t e r n a l A d d i t i o n o f L i g a n d s to a HIS117GIy A z u r i n Mutant. J A m Chem Soc 1991, 113:5050-5052.
30.
RODGERSKK, SUGAR SG: M a p p i n g Electrostatic Interact i o n s i n M a c r o m o l e c u l a r Associations. Mol Biol 1991, 221:1453-1460. This paper describes how the relative contributions of complementaW electrostatic and hydrophobic interactions can be extracted from a variety of physical measurements on reversably binding electrontransfer proteins. 31.
MIYASAKAT, KOYAMAK, ITOH I: Q u a n t u m C o n v e r s i o n a n d I m a g e D e t e c t i o n b y a B a c t e r i o r h o d o p s i n - b a s e d Artificial P h o t o r e c e p t o r . Science 1992, 255:342-243.
TAKEI H, LEWIS A, CHEN Z, NEBENZAHLJ: I m p l e m e n t i n g Receptive Fields w i t h Excitatory a n d I n h i b i t o r y Optoelectrical R e s p o n s e s o f B a c t e r i o r h o d o p s i n Films. Applied Optics 1991, 30:500-509. This paper draws an interesting parallel between bacteriorhodopsin film response a n d visual receptor fields and illustrates the basic propel'ties of cooperative, distributed information-processing networks. 32.
33.
ELLERBYLM, NISHIDA CR, NISHIDA F, YAMANAKASA, DUNN B, VALENTINE JS, ZINK JI: E n c a p s u l a t i o n o f P r o t e i n s i n T r a n s p a r e n t P o r o u s Silicate Glasses P r e p a r e d b y t h e Sol-gel Method. Science 1992, 255:1113-1115. The redox proteins copper-zinc superoxide dismutase, cytochrome c, and myoglobin were immobilized in a glass matrix under gentle conditions that maintained the functional integrity of the protein. This matrix allows the diffusion of small molecules, making the composite material potentially useful for materials or biosensor applications. 34.
WILLNERI, RUBIN S, ZOR T: P h o t o r e g u l a t i o n o f M p h a C h y m o t r y p s i n b y Its I m m o b i l i z a t i o n i n a Phot o c h r o n i c A z o b e n z e n e Copolymer. j A m Chem Soc 1991, 113:4013~i014.
35.
SUBRAMANIAMS, GREENHALGH DA, PATH 17, ROTHSCHILD KJ, KHORANAHG: R e p l a c e m e n t o f Leucine-93 b y A l a n i n e o r
Protein engineering for molecular electronics Sligar and Salemme 393 T h r e o n i n e Slows D o w n t h e Decay o f t h e N - I n t e r m e d i a t e a n d O - i n t e r m e d i a t e i n t h e P h o t o c y d e o f Bacter i o r h o d o p s i n - - I m p l i c a t i o n s for P r o t o n Uptake a n d 13-cis-retinal-~ All4rans-retinal Reisomerization. Proc Natl Acad Sci USA 1991, 88:6873-6877. 36.
THORGEIRSSONTE, MILDER SJ, MIERCKE LJW, BETLACH MC, SHAND RE, STROUD RM, KLIGER DS: Effects o f Asp96 -->Ash, Asp-85 -+ Ash, a n d Arg-82 -+ Gin Single-site S u b s t i t u t i o n s o n t h e Photocycle o f B a c t e r i o r h o d o p s i n . Biochemistry 1991, 30:9133-9142.
37.
SUBRAMANIAMS, MARTI T, ROSSELET SJ, ROTHSCHILD KJ, KHORANA HG: The Reaction o f H y d r o x y l a m i n e w i t h B a c t e r i o r h o d o p s i n Studied w i t h Mutants t h a t h a v e Altered Photocycles: Selective Reactivity o f Differe n t P h o t o i n t e r m e d i a t e s . Proc Natl Acad Sci USA 1991, 88:2583-2587.
38.
MARTIT, OTrO H, MOGI T, ROSSELETSJ, HEYN MP, KHORANA HG: B a c t e r i o r h o d o p s i n M u t a n t s C o n t a i n i n g Single Substitutions o f Serine o r T h r e o n i n e Residues a r e All Active i n P r o t o n Translocation, J Biol Chem 1991, 266:6919-6927.
39.
HOLZM, DRACHEVLA, MOG1 T, OTro H, KAULENAD, HEYN MP, SKULACHEVV'P, KHORANAHG: R e p l a c e m e n t o f Aspartic Acid-96 b y A s p a r a g i n e i n B a c t e r i o r h o d o p s i n Slows B o t h t h e Decay of t h e M I n t e r m e d i a t e a n d t h e Associated p r o t o n Movement. Proc Natl Acad Sci USA 1989, 86:2167-2171.
40.
HAMPPN, BRAUCHLE C, OESTERHELT D: B a c t e r i o r h o d o p s i n Wildtype a n d Variant Aspartate-96 to A s p a r a g i n e as Reversible H o l o g r a p h i c Media. Biophys J 1990, 58:83-93.
41. •-
HENDERSONR, BALDWINJM, CESKATA, ZEMLINF, BECKMANN E, DOWNING KH: Model f o r t h e Structure o f Bacterio r h o d o p s i n Based o n H i g h - r e s o l u t i o n E l e c t r o n Cryom i c r o s c o p y . J Mol Biol 1990, 213:899-929. A breakthrough paper describing the bacteriorhodopsin three-dimensional structure at the level of detail required for systematic protein engineering. 42.
NONELLAM, WINDEMUTHA, SHULTEN K: Structure o f Bact e r i o r h o d o p s i n a n d In Situ I s o m e r i z a t i o n o f Retinal: a Molecular D y n a m i c s Study. Photochem Photobiol 1991, 54:937-948. A state-of-the-art study of dynamical processes that are key to bacteriorhodopsin function. 43.
BAYLEYH: M o n o l a y e r s f r o m Genetically E n g i n e e r e d P r o t e i n Pores. Mat Res Soc Symp Proc 1991, 218:69-74.
44.
PICORELR, HOLT RE, HEALD R, COTTON TM, SEIBERTM: Stability o f Isolated Bacterial a n d P h o t o s y s t e m - I I React i o n C e n t e r C o m p l e x e s o n Ag Electrode Surfaces - - a S u r f a c e - e n h a n c e d R e s o n a n c e R a m a n Study. J A m Chem Soc 1991, 113:2839-2843.
45.
SCHUHMANNW, OHARATJ, SCHMIDTHL, HELLERA: E l e c t r o n T r a n s f e r B e t w e e n Glucose O x i d a s e a n d Electrodes Via R e d o x Mediators B o u n d w i t h Flexible C h a i n s to t h e E~e Surface. J A m Chem Society 1991, 113:1394-1397.
46.
FINKLEAHO, HANSHEWDD: Electron T r a n s f e r K i n e t i c s i n O r g a n i z e d Thiol M o n o l a y e r s w i t h Attached P e n t a a m m i n e ( p y r i d i n e ) r i t h e n i u m R e d o x Centers. A m Cbem Soc 1992, 114:3173-3182.
47.
TARLOVMJ, BOWDEN EF: E l e c t r o n - t r a n s f e r R e a c t i o n o f C y t o c h r o m e - c A b s o r b e d o n Carboxylic Acid Termi-
h a t e d A l k a n e t h i o l M o n o l a y e r Electrodes. J A m Chem Soc 1991, 113:1847-1849. 48.
WILLNERI, LAPIDOT N: Electrically Wired G l u a t h i o n e Red u c t a s e - - a Biocatalyst for t h e P h o t o c h e m i c a l Reduct i o n o f Glutathione. J A m Chem Soc 1991, 113:3625-3626.
49.
YOMO T, URABE I, OKADA H: P r e p a r a t i o n a n d Kinetic P r o p e r t i e s o f 5 - E t h y l p h e n a z i n e Polyethylene-glycol G l u t a m a t e - d e h y d r o g e n a s e Conjugate-A S e m i s y n t h e t i c NADH Oxidase. Eur J Biochem 1991, 196:343-348.
50.
PERSSONM, MAANSSONMO, Bi]LOW L, MOSBACH K: Cont i n u o u s R e g e n e r a t i o n o f N A D ( H ) Covalently B o u n d to a n E n g i n e e r e d C y s t e i n e R e s i d u e o f Glucose D e h y d r o g e n a s e . Biotechnolog3~ 1991, 9:280-285.
51.
SEEMANNC: DNA Structural E n g i n e e r i n g Using I m m o bile J u n c t i o n s . Curt Opin Struct Biol 1991, 1:653-661. DNA lattices offer extraordinary potential for two- and three-dimensional scaffolding for MEDs, particularly when they additionally incorporate internal or terminal biotinylation sites. These sites can interconnect DNA lattices with avidin or streptavidin molecules to produce addffional symmetric, multivalertt biotin attachment sites. 52.
DARSTSA, AHLERSM, MELLERPH, KUBALEKEW, BLAKENBURG R, RIBI HO, RINGSDORF H, KORNBERG RE): Two-dimeaxsional Crystals of Streptavidin on Biotinylated Lipid Layers a n d T h e i r I n t e r a c t i o n s w i t h Biotinylated Macrom o l e c u l e s . Biophys J 1991, 59:387-396. Streptavidin monolayers are demonstrated to have extraordinary surface order which, together with their ability to irreversibly bind virtually anything that can be conjugated to biotin, suggest a central role for the system in MED fabrication. 53.
EIGLER DM, SCHWEIZER EK: P o s i t i o n i n g Single A t o m s w i t h a S c a n n i n g T u n n e l l i n g Microscope. Nature 1990, 344:524-526. An initial demonstration of the possibility for patterning surfaces by manipulation of single atoms, 54.
PICKOVERC: M a t h e m a t i c s a n d Beauty VIII: Tessellation A u t o m a t a Derived f r o m a Single Defect. Comp Math with Appl 1989, 17:321-336. A remarkable paper illustrating the temporal evolution of complex interconnected lattices bearing a striking resemblance to integrated circuits. 55.
SaSABEH: P r o c e e d i n g s o f t h e F r o n t i e r R e s e a r c h F o r u m o n Bioelectronic Materials. 1991, Nishina Memorial Hall, RIKEN, Wako-Shi, Japan. Creative efforts to capitalize on the use of proteins as biomolecular devices has come from extensive work by the Frontier Research Program at the RIKEN Institute in Japan under the leadership of Dr Hiroyuki Sasabe. This has included ordered assembly of such photoactive systems as nitrile hydratase, photosynthetic reaction centers and bacteriorhodopsin, as well as nonqinear optical properties of polymer protein systems.
SG Sligar, School of Chemical Sciences, University of Illinois at Urbana-Champaign, 303 Roger Adams Laboratory, 1209 West California Street, Urbana, Illinois, IL 61801, USA. PR Salemme, Sterling Winthrop Pharmaceutical Research Division, 9 Great Valley Parkway, Malvern, Pennsylvania, I1 61801, PA 19355, USA.
Recombinant DNA technology allows the manipulation of the physical properties of proteins that perform electron transport and photochemical processes. Recent work is reviewed that has a potential impact on the development of molecular electronic devices, within a general framework outlining strategies for device fabrication. This review is also published in Current Opinion in Structural Biology 1992, 2:587-592.
Current Opinion in Biotechnology 1992, 3:388-393
Introduction This review is p r o m p t e d b y the recent enthusiasm for the use of biological macromolecules as building blocks for the construction of molecular electronic devices (MEDs). Molecular electronics is a term that has a dual meaning. One meaning refers to electronic devices that use materials w h o s e unique properties result from their molecular structure. In this regard, proteins can be used in MEDs to sense, r e s p o n d to, or record chemical, electrical, or physical stimuli. Examples include the use of the bacterial photopigment bacteriorhodopsin as a three-dimensional medium for the storage and readout of optically e n c o d e d information [1",2",3]. The second definition of molecular electronics embodies the concepts that individual molecules are functional units responding to stimuli, and that they have the potential to b e interconnected in order to replicate electronic circuits functionally. G o o d examples of such molecular electronic devices do not yet exist, but the principles necessary for their design and assembly are beginning to emerge from a variety of areas.
From the perspective of practical device design, there are several levels at which the ability to modify protein molecules using recombinant DNA technology could have an impact on the fabrication of MEDs. These include alteration of intrinsic optical and electronic properties, increasing protein stability for operation in non-biological environments, and modification of surface properties to facilitate d e n o v o design of mesoscale molecular assemblies or molecular circuits. Current activities in each of these categories are described below.
Modification of protein electronic and physical properties Naturally occurring proteins incorporate a wide range of physical properties that are of potential use in electronic devices. In most cases, the useful properties derive from a prosthetic group such as a metal center, organic r e d o x cofactor, or chromophore. Although naturally occurring redox and photoactive proteins exhibit great functional diversity, this is achieved through a limited n u m b e r of prosthetic groups w h o s e properties are modulated through interaction with amino acid side chains of the surrounding protein moiety. Attempts to engineer proteins in order to modify their properties in useful ways have followed precedents suggested b y the study of structure-function relationships in natural systems. In this respect, heme-containing proteins, which comprise a diverse set of molecules with electron transfer, ligand binding, or catalytic function, are of particular interest. Alterations in the electronics or chemistry of the h e m e iron center frequently produce large changes in optical spectroscopic or magnetic properties which are useful as indicators for state assignment or device readout. Following pioneering w o r k on cytochrome b5 [4], myoglobin [5-8], and cytochrome p450 [9], recent w o r k reporting site-directed modifications of residues that are heme iron ligands, or can otherwise influence h e m e spectral properties, include additional studies on cytochrome p450 [10] and myoglobin [11], together with studies on cytochrome c peroxidase [12,13] and iso-l-cytochrome c [14]. Although the study of the structure-property relationships that affect heme proteins remains an area of active interest [15"], few systematic efforts have b e e n
Abbreviation MED--molecular electronic device. 388
© Current Biology Ltd ISSN 0958-1669
Protein engineering for molecular electronics Sligar and Salemme 389 directed at producing unusually stable molecules that could increase the reliability of state assignment. H o w ever, a serendipitous enhancement of protein stability was reported for a site-directed mutant of iso-l-cytochrome c [14], in which an internal asparagine had b e e n changed to a hydrophobic isoleucine residue. This alteration resulted in the loss of an internal water molecule w h o s e location in the interior of the native protein was postulated to b e energetically unfavorable. Similar hydrophobic enhancements could direct further efforts towards the important goal of stabilizing redox proteins in non-aqueous environments. A property central to the function of many MEDs is the regulation of electron-transfer rate between proteins or b e t w e e n proteins and an external oxidoreductant. Physical aspects of the electron-transfer p r o cesses in proteins are n o w relatively well understood, and suggest that electron-transfer rates d e p e n d primarily u p o n prosthetic group separation, difference in oxidoreduction potential and molecular reorganization energy [16",17]. This is consistent with recent site-directed modifications of the invariant Phe82 residue in yeast iso-l-cytochrome c, suggesting that intermolecular electron transfer does not require the participation of specific aromatic amino acids as electron wires between protein prosthetic groups [18,19]. Nevertheless, translation of the physical requirements for efficient electron transfer into a specific modification strategy remains complicated, as s h o w n by the results obtained b y Barker et al. for yeast iso-l-cytochrome c [20]. In addition to heine-containing redox proteins the introduction or regeneration of copper-binding sites in proteins has b e e n investigated. In one study, both type I and type II c o p p e r site properties could be obtained b y addition of an appropriate external ligand to an azurin mutant in which one of the native histidine c o p p e r ligands had b e e n deleted b y site-directed mutagenesis [21]. In a second study, site-directed modifications of cytochrome c were performed in order to introduce a c o p p e r ligand site [22]. In both cases, the modified proteins a p p e a r e d to lack the intrinsic stability of the native molecule, but may point the w a y to successive generations of molecules with useful functions.
Interaction specificity Because rates of intermolecular transfer d e p e n d very strongly on prosthetic group separation distance [16-, 17], reactions between reversibly binding electrontransfer proteins d e p e n d on interactions that ensure the proper relative intermolecular orientation. Early w o r k that m o d e l e d the interactions of cytochromes c and b5 established the importance of solvent exclusion and complementary electrostatic interactions in radox protein interactions [23]. This precedent has since b e e n generalized to m a n y other biological electron-transfer interactions. With the advent of site-directed mutagenesis methods, investigation of intermolecular recognition interactions has intensified [24-29]. Most notably,
methods have recently b e e n developed that discriminate between the relative contributions of the principal components of intermolecular recognition: polar interactions that result from hydrogen b o n d s and salt bridges forming in the interaction domain, and non-polar van der Waals interactions that occur w h e n protein surfaces dehydrate upon formation of the c o m p l e m e n tary complex [30"]. This m e t h o d o l o g y also facilitates m a p p i n g the interaction domain between two associating macromolecules, a k e y step in the d e v e l o p m e n t of generalized strategies to both study natural systems and engineer alternative interactions a m o n g molecules suitable for use in MEDs.
Bacteriorhodopsin in MED applications Bacteriorhodopsin (molecular weight 26000 D) functions as a light-driven proton p u m p in the purple m e m brane of the salt-marsh micro-organism Halobacterium halobium [1",2"]. Bacteriorhodopsin incorporates a retinal chromophore, which is covalently b o u n d as a protonated Schiff base in an all-trans conformation in the resting state. U p o n absorbing a light photon, the retinal photoisomerizes and subsequently undergoes a multistep photocycle that spans much of the visible spectrum. The photocycle can b e stopped at specific intermediates by low temperature trapping and recycled b y thermalization or application of a second light pulse of the appropriate wavelength [1",2"]. Photoactivated retinal switching occurs with high quantum yields at disparate wavelengths, which facilitate state assignment, and produces b o t h changes in protein refractive index and a photoelectric potential in oriented assemblies. These properties, together with the excellent stability of the protein w h e n immobilized in polym e r films or gels, form the basis for a variety of prototypes for data storage media, holographic memory, a n d electro-optic applications [1",2",3,31,32"]. Three-dimensional photochromic memories have incorporated bacteriorhodopsin that has b e e n oriented b y an external electric field and then immobilized in a solid polyacrylamide matrix [2"]. This application exploits the two-photon absorption cross section of bacteriorhodopsin and requires that a small v o l u m e of the matrix be simultaneously illuminated by intersecting laser beams for data storage. Readout is achieved by reillumination, which produces an electrical signal on the cuvette surface that depends o n the state of the irradiated volume in the data storage matrix. In this context, it is interesting to note recent experiments demonstrating the maintenance of redox protein properties w h e n immobilized in silicate glasses '[33"], as well as preservation and control of proteolytic e n z y m e activity in a photochromic azobenzene copolym e r [34]. Although most readily applied in biosensor applications, both schemes could potentially be useful in three-dimensional optical m e m o r y applications that utilize photochromic properties of incorporated
390
Protein engineering proteins or that couple protein catalytic properties to the photochromic matrix. Current w o r k on bacteriorhodopsin with relevance to molecular electronics involves finding additives, alternative pigments, or site-directed mutants of the protein that will alter the relative stabilities of the photo-intermediates [35], enhance stabilization of the final M intermediate at r o o m temperature [36], or alter other aspects of photocycle-coupled proton translocation [37,38] that might facilitate readout from information storage applications. Consequently, investigations of the Asp96 Asn modification [39], which alters coupling of retinal photoisomerization to proton translocation, continue to have substantial practical interest [40]. The physical changes in the modified protein include an increase in the lifetime of the intermediate from 10 ms to 750 ms, together with improved diffraction efficiency and photochromic sensitivity relative to the native protein. Considerable latitude for property improvement exists using a combination of site-directed modifications at the c h r o m o p h o r e binding site, introduction of alternative chromophores, or the engineering of modified ion-binding sites, all of which can affect intermediate lifetimes and spectral properties [1",2"]. In this regard, the recent structure determination of bacteriorhodopsin by electron diffraction methods [41"], coupled with rapidly advancing methods of computational simulation [42"], may provide n e w insights into the engineering of bacteriorhodopsin for MED applications.
Oriented protein immobilization on surfaces
prosthetic groups in sensor or non-linear optical applications depends critically on the ability to control prosthetic g r o u p orientation precisely relative to the electrode or optical w a v e g u i d e surface. Site-specific introduction of anchoring points in a protein of k n o w n three-dimensional structure provides a straightforward solution to this problem. As an example, recent site-directed modifications of the heine protein cytochrome b5 carried out at the University of Illinois introduced cysteine residues at specified positions on the molecular surface. This allowed oriented coupling to a silane substrate and produced immobilized molecular arrays demonstrating a high level of heme prosthetic group orientation as determined b y linear dichroism spectroscopy. Methods to introduce two-dimensional periodicity on surface molecular arrays d e p e n d either on engineering both intermolecular interactions [30"] and anchoring sites for surface immobilization, or on schemes that erect molecular assemblies on a scaffold already possessing periodic order on the few tens of angstroms scale. Scaffold possibilities include natural two-dimensional lattices [43], DNA lattices [51"], or the very highly ordered synthetic lattices formed from streptavidin anchored to surfaces through biotinylated phospholipids [52"]. As streptavidin is a tetrameric protein with four biotin-binding sites arranged with approximate tetrahedral geometry, two biotin-binding sites per tetramer remain f r e e in the two dimensional arrays and provide a regular lattice for attaching additional molecules (Fig. 1). Molecular electronics
Streptavidin Bacteriorhodopsin is not unique in its ability to order spontaneously in two-dimensional films that resemble its native m e m b r a n e state and, indeed, other m e m brane protein systems have b e e n suggested for use or have potential in a variety of MED or related biosensor roles. Recent work includes studies of engineered bacterial transmembrane pores [43] and isolated bacterial reaction centers immobilized on electrode surfaces [44]. In these cases, surface orientation of the protein complexes results from preexisting interactions between the protein complexes or their m e m b r a n e environments. However, a n u m b e r of experiments have b e e n carried out that involve tethering proteins to electrode surfaces through polymeric 'wires' [45,46] or through surface electrostatic interactions [47] in order to enhance electron-transfer efficiences. Related studies of interest examine properties of soluble proteins conjugated to photo- or redox-active organic substituents that enhance enzyme catalytic function in the absence of natural cofactors or regeneration systems [48-50]. It is easy to envision hybrid systems that incorporate proteins with conjugated cofactors immobilized on electrode surfaces. Although the forgoing studies emphasize electron conduction b e t w e e n device components, the efficiency of m a n y devices incorporating
components
Biotinylated redox protein
Biotinylated DNA [~ giotinylated lipid ~ Lipid-anchored redox protein
Hypothetical molecular electronics architecture Fig. 1. Illustration of molecular electronics components and potential architectural features of self-assembling systems restricted to diffusion in two-dimensional lipid films.
Protein engineering for molecular electronics Sligar and Salemme
Strategies for protein patterning and circuit fabrication Although it a p p e a r s possible that quite complicated molecular assemblies could b e erected o n two-dimensional protein lattices using a combination of chemical conjugation, site-directed introduction o f a n c h o r points, and molecular fusions at the DNA level (rev i e w e d elsewhere in this issue), the materials described thus far all c o m e u n d e r our first definition o f molecular electronics as bulk materials. Although these substrates could u n d o u b t e d l y b e patterned using lithogr a p h y m e t h o d s similar to t h o s e used in fabricating integrated circuits, optical diffraction w o u l d limit detail to scales substantially larger than individual molecular assemblies. Such devices w o u l d not realize the ultimate in miniaturization of individually assignable devices inherent in molecular electronics of the s e c o n d definition, w h e r e a s individual molecular assemblies w o u l d lack the u n i q u e addressability and specific connectivity b e t w e e n c o m p o n e n t s c o m m o n l y associated with electronic circuits. Nevertheless, it is interesting to consider h o w it might be possible to construct true molecular circuits, as well as the dual p r o b l e m at the molecular level of h o w such devices could be addressed a n d read. Atomic p r o b e m i c r o s c o p y could potentially b e u s e d to manipulate and assemble molecules as a means o f constructing devices, as well as a m e a n s of setting molecular states or reading o u t information [53"]. The limitations of manipulating only o n e or a few c o m p o n e n t s at a time, however, w o u l d s e e m ultimately to defeat the objectives o f creating a molecular device. Alternatively, it m a y b e possible to 'condition' a preexisting molecular lattice of uniformly interconnected molecules so that it is able to store or process information as a neural n e t w o r k analog. Information processing using a distributed a p p r o a c h requires each m o l e c u l a r - a s s e m b l y to sense the state of its nearest neighbors a n d alter its configuration accordingly. With phasing to an external clock, this s c h e m e has the formal structure o f a cellular a u t o m a t o n operating with individual m a c r o m o l e c u l e s as building blocks. In such a highly cooperative device, it m a y n o t be necessary to manipulate individual molecules directly, but utilize intermolecular c o m m u n i c a t i o n s c o m m o n to m a n y cooperative multimeric proteins to p e r f o r m s o m e analog c o m p u t i n g function. Fabrication possibilities also exist using defect tessellation automata [54"], w h e r e very complicated structures can be generated b y defect introduction into a periodic lattice subject to a regular site r e p l a c e m e n t protocol. Although presently only the subject of c o m p u t e r simulation, it seems possible to e m b o d y the required properties in ligand-capture a n d displacement schemes using symmetric protein molecules with multiple b i n d i n g sites. Access and readout could potentially be achieved b y any of the methods outlined above. While these devices are presently w h o l l y conceptual, and m a y ultimately be limited b y molecular noise that will necessitate distributed or red u n d a n t processing to ensure reliability, it seems clear
that g e n u i n e a p p r o a c h e s exist to investigate the limits o f this technology.
Conclusion Practical application of e n g i n e e r e d proteins in molecular electronics and sensor materials applications are clearly o n the near horizon. Protein molecules represent the ultimate miniaturization possible in individual devices w h o s e state can potentially be controlled indep e n d e n t of its neighbors. This property, together with the e m e r g e n c e of a k n o w l e d g e base allowing protein properties and interaction specificity to be e n g i n e e r e d with relative ease, will u n d o u b t e d l y lead to progressively m o r e sophisticated assemblies, w h o s e functional limitations can only n o w be g u e s s e d at. Multidisciplinary p r o g r a m s such as those in the Soviet U n i o n that p i o n e e r e d d e v e l o p m e n t o f b a c t e r i o r h o d o p s i n based devices [1",3], or the Frontier Research P r o g r a m of the Riken Institute in J a p a n [55"], represent f o c u s e d efforts to realize the potential of this technology.
References and recommended reading Papers of particular interest, published within the annual period of review, have been highlighted as: of special interest •. of outstanding interest 1.
OESTERHELTD, BRAUCHLEC, HAMpp N: Bacteriorhodopsin: a Biological Material for Information Processing. Quart Rev Biophys !991, 24:425~78.
A recent review that summarizes structure-function relationships in bacteriorhodopsin and includes a discussion of site-specific mutants. BIRGERR: Photophysics and Molecular Electronic Applications o f the Rhodopsins. Annu Rev Phys Chem 1990, 41:683-733. This review emphasizes MED applications. 2.
3.
BELCHINSKYGY, BOGATOV PN, GRESKO AP, KONENKO AA, LUKASHEVEP, CHAMOROVSKYSK: Application o f Pigment-Protein Complexes as Carriers o f Optic Information. Biofizika 1991, 36:248-251.
4.
SLIGARS, EGEBERGK, SAGET, MORIKISD, CHAMPIONP: Alteration o f Heme Axial Ligands b y Site Directed Mutagenesis: a Cytochrome Becomes a Catalytic Demethylase. J Am Chem Soc 1987, 109:7896-7897.
5.
IAMBRIGHTDG, BALASUBRAMANIANS, BOXERSG: Ligand and Proton Exchange I)ylmmics i n Recombinant Human Myoglobin Mutants. J Mol Biol 1989, 207:289-299.
6.
VARADARAJAN R, ZEWERTWE, GRAy HB, BOXERSG: Effects of Buried Ionizable Amino Acids o n the Reduction Potential of Recombinant Myoglobin. Science 1989, 243:69-72.
7.
VARADARAJANR, LAMBRLGHTDG, BOXERSG: Electrostatic Interactions in Wild-type and Mutant Recombinant H u m a n Myoglobins. Biochemistry 1989, 28:3771-3781.
8.
EGEBERGKD, SPRINGERBA, MARTINISSA, SLIGARSG, MORIKIS D, CHAMPIONPM: Alteration of Sperm Whale Myoglobin Axial Ligation b y Site-directed Mutagenesis. Biochemistry 1990, 29:9783-9791.
391
392
Protein engineering 9.
DI PRIMO C, HUI BON HOA G, D o u z o u P, SL1GARS: Mut a g e n e s i s o f a Single H y d r o g e n B o n d i n C y t o c h r o m e P-450 Alters Cation B i n d i n g a n d H e i n e Solvation. J Biol Chem 1990, 265:5361-5363.
22.
TODD RJ, VANDAMME, CASIMIRO D, HAYMORE BL, ARNOLD FH: Cu(II)-binding P r o p e r t i e s of a C y t o c h r o m e - c w i t h a S y n t h e t i c Metal-binding Site - - His X3-His i n a n Alphahelix. Proteins 1991, 10:156-161.
10.
IWASAKIM, JUVONEN R, LINDBERG R, NEGISHI M: Alteration o f H i g h a n d Low S p i n E q u i l i b r i u m b y a Single Mutat i o n o f Amirto Acid-209 i n M o u s e C y t o c h r o m e - P 4 5 0 . J Biol Chem 1991, 266:3380-3382.
23.
SALEMMEFR: A H y p o t h e t i c a l Structure for a n Intermolecular Electron Transfer Complex of Cytochrome c a n d b5. J Mol Biol 1976, 102:563-568.
24.
11.
ADACHIS, NAGANOS, WATANABEY, ISHIMORI K, MORISHIMAI; A l t e r a t i o n o f H u m a n Myoglobin P r o x i m a l H i s t i d i n e to C y s t e i n e o r T y r o s i n e b y Site-directed M u t a g e n e s i s - C h a r a c t e r i z a t i o n a n d Their Catalytic Activities. Biochem Biophys Res C o m m u n 1991, 180:138-144.
RODGERSK, POCHAPSKY T, SLIGAR S: P r o b i n g t h e Mecha n i s m s o f M a c r o m o l e c u l a r Recognition: t h e Cyt o c h r o m e b 5 - c y t o c h r o m e c C o m p l e x . Science 1988, 240:1657-1659.
25.
STAY'YONPS, SLIGAR SG: T h e C y t o c h r o m e P - 4 5 0 c a m B i n d i n g Surface as D e f i n e d by Site-directed Mutagen e s i s a n d Electrostatic Modeling. Biochemistry 1990, 29:7381-7386.
26.
SHIMIZUT, TATEISHI T, HATANO M, FUJlIKURIyAMAY: Probi n g t h e Role o f Lysines a n d A r g i n i n e s i n t h e Catalytic F u n c t i o n o f C y t o c h r o m e - P 4 5 0 d by Site Directed Mutag e n e s i s - - I n t e r a c t i o n w i t h NADPH-cytochrome-P450 Reductase. J Biol Chem 1991, 266:3372-3375.
27.
COGHLANVM, VICKERYLE: Site-specific M u t a t i o n s i n Hum a n F e r r e d o x i n T h a t Affect B i n d i n g to F e r r o d o x i n R e d u c t a s e a n d Cytochrome-P450scc. J Biol Chem 1991, 266:18606-18612.
28.
SHIMIZU T, TATEISHI T, HATANO M, FUJlIKURIYAMA Y: P r o b i n g t h e Role o f Lysines a n d A r g l n i n e s i n t h e Catalytic F u n c t i o n o f Cytochrome-P450D b y Site-directed M u t a g e n e s i s - I n t e r a c t i o n w i t h NADPHC y t o c h r o m e - P 4 5 0 Reductase. J Biol Chem 1991, 266:3372-3375.
29.
SHIRABEK, YUBISUIT, NISHINO T, TAKESHITAM: Role o f Cyst e i n e R e s i d u e s i n H u m a n NADH-cytochrome-b5 Reductase Studies b y Site-directed M u t a g e n e s i s - - Cys-273 a n d Cys-283 are Located Close to t h e NADH-binding Site b u t a r e n o t Catalytically E s s e n t i a l J Biol Chem 1991, 266:7531-7536.
12.
13.
14.
GOODIN DB, DAVIDSON MG, ROE JA, MAUK AG, SMITH M: A m i n o Acid S u b s t i t u t i o n s at T r y p t o p h a n - 5 1 o f C y t o c h r o m e - c P e r o x i d a s e ~ Effects o n C o o r d i n a t i o n , Species P r e f e r e n c e f o r C y t o c h r o m e - c , a n d E l e c t r o n Transfer. Biochemistry 1991, 30:4953-4962. SMULEVICHG, MILLER /VIA, KRAUT J, SPIRO TG: Conform a t i o n a l C h a n g e a n d Histidine C o n t r o l o f Heine C h e m i s t r y i n C y t o c h r o m e - e P e r o x i d a s e - - Reson a n c e R a m a n E v i d e n c e f r o m Leu-52 a n d GIg-181 Mut a n t s o f C y t o c h r o m e - c Peroxidase. Biochemistry 1991, 30:9546-9558. HICKEYDR, BERGHUIS AM, LAFOND G, JAEGERJA, CARDILLO TS, MCLENDON D, DAS G, SHERMAN F, BRAYER GD: E n h a n c e d T h e r m o d y n a m i c Stabilities o f Yeast iso1 - c y t o c h r o m e - c w i t h A m i n o Acid R e p l a c e m e n t at Position-52 a n d Position-102. J Biol Chem 1991, 266:11686-11694.
15. •.
FRAUENFELDERH, SUGAR SG, WOLYNES PG: T h e E n e r g y L a n d s c a p e s a n d M o t i o n s o f P r o t e i n s . Science 1991, 254:1598-1603. Describes current concepts about the dynamic interconversion among conformational substates in proteins and the resulting physical manifestations. This is of fundamental importance in understanding how thermal phenomena will affect switching reliability in prorein molecules when excited by external stimuli. 16.
MOSER CC, KESKEJM, WARNCKEK, FARID RS, DLrI~ON PL: N a t u r e o f Biological Electron Transfer. Nature 1992, 355:796-802. This paper summarizes m u c h experimental data showing that electron transfer rates effectively depend on prosthetic group separation distance, differences in prosthetic group free energy and reorganization energies upon oxidoreduction. The treatment suggests that the protein presents a uniform electronic tunnelling barrier that derives from glass-like protein properties also revealed in dynamic studies. 17.
BERATANDN, ONUCHIJN, BETrS JN, BOWLER BE, GRAY HB: Electron Tunneling Pathways in Ruthenated Proteins. J A m Chem Soc 1990, 112:7915-7921.
18.
EVERESTAM, WALLIN SA, STEMP ED, NOCEKJM, MAUK AG, HOFFMAN B: A r o l n a t i c Hole S u p e r e x c h a n g e T h r o u g h Position-82 i n C y t o e h r o m e C is Not R e q u i r e d f o r I n t r a c o m p l e x E l e c t r o n T r a n s f e r to Z n - C y t o c h r o m e C P e r o x i d a s e . A m Chem Soc 1991, 113:4337--4338.
19.
INGLIS SC, GUILLEMETTEJG, JOHNSON JA, SMITH M: Analy s i s o f t h e I n v a r i a n t Phe82 R e s i d u e o f Yeast Iso1 - c y t o c h r o m e c b y Site-directed M u t a g e n e s i s U s i n g a P h a g e m i d Yeast Shuttle Vector. Protein Eng 1991, 4:569-574.
20.
BARKERPD, MAUKMR, MAUKAG: P r o t o n Titration Curve o f Yeast I s o - l - F e r r i c y t o c h r o m e - c - - Electrostatic a n d C o n f o r m a t i o n a l Effects o f P o i n t Mutations. Biochemistry 1991, 30:2377-2383.
21.
DENBLAAUWENT, VANDEKAMPM, CANTER GW: Type-I a n d Type-II C o p p e r Sites O b t a i n e d b y E x t e r n a l A d d i t i o n o f L i g a n d s to a HIS117GIy A z u r i n Mutant. J A m Chem Soc 1991, 113:5050-5052.
30.
RODGERSKK, SUGAR SG: M a p p i n g Electrostatic Interact i o n s i n M a c r o m o l e c u l a r Associations. Mol Biol 1991, 221:1453-1460. This paper describes how the relative contributions of complementaW electrostatic and hydrophobic interactions can be extracted from a variety of physical measurements on reversably binding electrontransfer proteins. 31.
MIYASAKAT, KOYAMAK, ITOH I: Q u a n t u m C o n v e r s i o n a n d I m a g e D e t e c t i o n b y a B a c t e r i o r h o d o p s i n - b a s e d Artificial P h o t o r e c e p t o r . Science 1992, 255:342-243.
TAKEI H, LEWIS A, CHEN Z, NEBENZAHLJ: I m p l e m e n t i n g Receptive Fields w i t h Excitatory a n d I n h i b i t o r y Optoelectrical R e s p o n s e s o f B a c t e r i o r h o d o p s i n Films. Applied Optics 1991, 30:500-509. This paper draws an interesting parallel between bacteriorhodopsin film response a n d visual receptor fields and illustrates the basic propel'ties of cooperative, distributed information-processing networks. 32.
33.
ELLERBYLM, NISHIDA CR, NISHIDA F, YAMANAKASA, DUNN B, VALENTINE JS, ZINK JI: E n c a p s u l a t i o n o f P r o t e i n s i n T r a n s p a r e n t P o r o u s Silicate Glasses P r e p a r e d b y t h e Sol-gel Method. Science 1992, 255:1113-1115. The redox proteins copper-zinc superoxide dismutase, cytochrome c, and myoglobin were immobilized in a glass matrix under gentle conditions that maintained the functional integrity of the protein. This matrix allows the diffusion of small molecules, making the composite material potentially useful for materials or biosensor applications. 34.
WILLNERI, RUBIN S, ZOR T: P h o t o r e g u l a t i o n o f M p h a C h y m o t r y p s i n b y Its I m m o b i l i z a t i o n i n a Phot o c h r o n i c A z o b e n z e n e Copolymer. j A m Chem Soc 1991, 113:4013~i014.
35.
SUBRAMANIAMS, GREENHALGH DA, PATH 17, ROTHSCHILD KJ, KHORANAHG: R e p l a c e m e n t o f Leucine-93 b y A l a n i n e o r
Protein engineering for molecular electronics Sligar and Salemme 393 T h r e o n i n e Slows D o w n t h e Decay o f t h e N - I n t e r m e d i a t e a n d O - i n t e r m e d i a t e i n t h e P h o t o c y d e o f Bacter i o r h o d o p s i n - - I m p l i c a t i o n s for P r o t o n Uptake a n d 13-cis-retinal-~ All4rans-retinal Reisomerization. Proc Natl Acad Sci USA 1991, 88:6873-6877. 36.
THORGEIRSSONTE, MILDER SJ, MIERCKE LJW, BETLACH MC, SHAND RE, STROUD RM, KLIGER DS: Effects o f Asp96 -->Ash, Asp-85 -+ Ash, a n d Arg-82 -+ Gin Single-site S u b s t i t u t i o n s o n t h e Photocycle o f B a c t e r i o r h o d o p s i n . Biochemistry 1991, 30:9133-9142.
37.
SUBRAMANIAMS, MARTI T, ROSSELET SJ, ROTHSCHILD KJ, KHORANA HG: The Reaction o f H y d r o x y l a m i n e w i t h B a c t e r i o r h o d o p s i n Studied w i t h Mutants t h a t h a v e Altered Photocycles: Selective Reactivity o f Differe n t P h o t o i n t e r m e d i a t e s . Proc Natl Acad Sci USA 1991, 88:2583-2587.
38.
MARTIT, OTrO H, MOGI T, ROSSELETSJ, HEYN MP, KHORANA HG: B a c t e r i o r h o d o p s i n M u t a n t s C o n t a i n i n g Single Substitutions o f Serine o r T h r e o n i n e Residues a r e All Active i n P r o t o n Translocation, J Biol Chem 1991, 266:6919-6927.
39.
HOLZM, DRACHEVLA, MOG1 T, OTro H, KAULENAD, HEYN MP, SKULACHEVV'P, KHORANAHG: R e p l a c e m e n t o f Aspartic Acid-96 b y A s p a r a g i n e i n B a c t e r i o r h o d o p s i n Slows B o t h t h e Decay of t h e M I n t e r m e d i a t e a n d t h e Associated p r o t o n Movement. Proc Natl Acad Sci USA 1989, 86:2167-2171.
40.
HAMPPN, BRAUCHLE C, OESTERHELT D: B a c t e r i o r h o d o p s i n Wildtype a n d Variant Aspartate-96 to A s p a r a g i n e as Reversible H o l o g r a p h i c Media. Biophys J 1990, 58:83-93.
41. •-
HENDERSONR, BALDWINJM, CESKATA, ZEMLINF, BECKMANN E, DOWNING KH: Model f o r t h e Structure o f Bacterio r h o d o p s i n Based o n H i g h - r e s o l u t i o n E l e c t r o n Cryom i c r o s c o p y . J Mol Biol 1990, 213:899-929. A breakthrough paper describing the bacteriorhodopsin three-dimensional structure at the level of detail required for systematic protein engineering. 42.
NONELLAM, WINDEMUTHA, SHULTEN K: Structure o f Bact e r i o r h o d o p s i n a n d In Situ I s o m e r i z a t i o n o f Retinal: a Molecular D y n a m i c s Study. Photochem Photobiol 1991, 54:937-948. A state-of-the-art study of dynamical processes that are key to bacteriorhodopsin function. 43.
BAYLEYH: M o n o l a y e r s f r o m Genetically E n g i n e e r e d P r o t e i n Pores. Mat Res Soc Symp Proc 1991, 218:69-74.
44.
PICORELR, HOLT RE, HEALD R, COTTON TM, SEIBERTM: Stability o f Isolated Bacterial a n d P h o t o s y s t e m - I I React i o n C e n t e r C o m p l e x e s o n Ag Electrode Surfaces - - a S u r f a c e - e n h a n c e d R e s o n a n c e R a m a n Study. J A m Chem Soc 1991, 113:2839-2843.
45.
SCHUHMANNW, OHARATJ, SCHMIDTHL, HELLERA: E l e c t r o n T r a n s f e r B e t w e e n Glucose O x i d a s e a n d Electrodes Via R e d o x Mediators B o u n d w i t h Flexible C h a i n s to t h e E~e Surface. J A m Chem Society 1991, 113:1394-1397.
46.
FINKLEAHO, HANSHEWDD: Electron T r a n s f e r K i n e t i c s i n O r g a n i z e d Thiol M o n o l a y e r s w i t h Attached P e n t a a m m i n e ( p y r i d i n e ) r i t h e n i u m R e d o x Centers. A m Cbem Soc 1992, 114:3173-3182.
47.
TARLOVMJ, BOWDEN EF: E l e c t r o n - t r a n s f e r R e a c t i o n o f C y t o c h r o m e - c A b s o r b e d o n Carboxylic Acid Termi-
h a t e d A l k a n e t h i o l M o n o l a y e r Electrodes. J A m Chem Soc 1991, 113:1847-1849. 48.
WILLNERI, LAPIDOT N: Electrically Wired G l u a t h i o n e Red u c t a s e - - a Biocatalyst for t h e P h o t o c h e m i c a l Reduct i o n o f Glutathione. J A m Chem Soc 1991, 113:3625-3626.
49.
YOMO T, URABE I, OKADA H: P r e p a r a t i o n a n d Kinetic P r o p e r t i e s o f 5 - E t h y l p h e n a z i n e Polyethylene-glycol G l u t a m a t e - d e h y d r o g e n a s e Conjugate-A S e m i s y n t h e t i c NADH Oxidase. Eur J Biochem 1991, 196:343-348.
50.
PERSSONM, MAANSSONMO, Bi]LOW L, MOSBACH K: Cont i n u o u s R e g e n e r a t i o n o f N A D ( H ) Covalently B o u n d to a n E n g i n e e r e d C y s t e i n e R e s i d u e o f Glucose D e h y d r o g e n a s e . Biotechnolog3~ 1991, 9:280-285.
51.
SEEMANNC: DNA Structural E n g i n e e r i n g Using I m m o bile J u n c t i o n s . Curt Opin Struct Biol 1991, 1:653-661. DNA lattices offer extraordinary potential for two- and three-dimensional scaffolding for MEDs, particularly when they additionally incorporate internal or terminal biotinylation sites. These sites can interconnect DNA lattices with avidin or streptavidin molecules to produce addffional symmetric, multivalertt biotin attachment sites. 52.
DARSTSA, AHLERSM, MELLERPH, KUBALEKEW, BLAKENBURG R, RIBI HO, RINGSDORF H, KORNBERG RE): Two-dimeaxsional Crystals of Streptavidin on Biotinylated Lipid Layers a n d T h e i r I n t e r a c t i o n s w i t h Biotinylated Macrom o l e c u l e s . Biophys J 1991, 59:387-396. Streptavidin monolayers are demonstrated to have extraordinary surface order which, together with their ability to irreversibly bind virtually anything that can be conjugated to biotin, suggest a central role for the system in MED fabrication. 53.
EIGLER DM, SCHWEIZER EK: P o s i t i o n i n g Single A t o m s w i t h a S c a n n i n g T u n n e l l i n g Microscope. Nature 1990, 344:524-526. An initial demonstration of the possibility for patterning surfaces by manipulation of single atoms, 54.
PICKOVERC: M a t h e m a t i c s a n d Beauty VIII: Tessellation A u t o m a t a Derived f r o m a Single Defect. Comp Math with Appl 1989, 17:321-336. A remarkable paper illustrating the temporal evolution of complex interconnected lattices bearing a striking resemblance to integrated circuits. 55.
SaSABEH: P r o c e e d i n g s o f t h e F r o n t i e r R e s e a r c h F o r u m o n Bioelectronic Materials. 1991, Nishina Memorial Hall, RIKEN, Wako-Shi, Japan. Creative efforts to capitalize on the use of proteins as biomolecular devices has come from extensive work by the Frontier Research Program at the RIKEN Institute in Japan under the leadership of Dr Hiroyuki Sasabe. This has included ordered assembly of such photoactive systems as nitrile hydratase, photosynthetic reaction centers and bacteriorhodopsin, as well as nonqinear optical properties of polymer protein systems.
SG Sligar, School of Chemical Sciences, University of Illinois at Urbana-Champaign, 303 Roger Adams Laboratory, 1209 West California Street, Urbana, Illinois, IL 61801, USA. PR Salemme, Sterling Winthrop Pharmaceutical Research Division, 9 Great Valley Parkway, Malvern, Pennsylvania, I1 61801, PA 19355, USA.
