Affinity Biosensors, Leech 1994

8/13/2019 Affinity Biosensors, Leech 1994

1/9

Af fi ni ty BiosensorsDona1 LeechDepartement de Chimie Univers i te de Montreal C.P 6128 Succursale A Montreal Quebec Canada H3C J7

1 IntroductionThis review is not intended to be a comprehensive treatment ofaffinity biosensors. Its aim is to in trod uce the principle are as ofaffinity biosensor research a nd illustrate cur rent technology w ithselected examples. Mo re detailed info rmation o n biosensor andassay research a nd technology can be o btained by consulting theliterature cited and the references therein.A biosensor is normally defined as a device incorporating abiological molecular recognition component connected to atransducer which can output an electronic signal that is pro-portional to the concentration of the molecule being sensed.Research into biosensor technology has grow n rapidly in thepast decade because of its potential application in areas such asdiagnosis of disease, monitoring of clinical or environmentalsamples, fermentation and bioprocess monitoring, testing ofpharmaceutical or foo d prod ucts, and early warning chemicaland biological warfare detection systems. A range of combi-nations of receptor/transducers have been proposed and stu-died. Biosensors have been described that utilize enzymes,antibodies, receptors, cells an d tissue material for recogn ition ofthe analyte of interest. Tr ansd uctio n of the biochemical recogni-tion signal to an electronic signal has been accomplished usingoptical, electrochemical, calorimetric, an d piezoelectric detec-tion schemes. The term affinity biosensor in this article refers to adevice inco rpo ratin g immobilized biological receptor moleculesthat can reversibly detect receptor-ligand interactions with ahigh differential selectivity and in a non-destructive fashion.Th us, enzyme-based biocatalytic sensors are not covered by thisdescription.Excel1ent publications that describe biosensors areavailable for the interested reader. I have included in this reviewadvances in assay development and a discussion of novelcompetitive binding affinity systems for detection of ligandsalthough these are n ot generally tru e biosensors. These topicsare covered to d emo nstrate the advances in non-isotopic detec-tion metho ds and their potential application to biosensor tech-nology. I have purposely ignored the large body of literature onthe use of radioactive labels for the detection of ligand-receptorbinding in order to limit the length and scope of the article.

Dbnal Leech was born in Dublin Ireland. He obtained his B.S c. in1988 and his Ph.D. in electroanalytical chemistry under thedirection of M . R. Sm yth and J . G . Vos n 1991 ro m Dublin CityUniversity. During his Ph.D.studies a six month visit to J .Wang at New Mexico StateUniversity was undertaken.Post-doctoral research at theUniversity of Haw aii in thegroup of G . A. Rechnitzfollowed. In 1993 Dr. Leechaccepted a position as Assis-tant Professor of AnalyticalChem istry at the UniversitP deMontrkal. His research inter-ests currently f oc us on elec-troanalytical chemistry and itsapplication to chemical sen-sors, biosensors and assays.

2 Receptor Ligand Binding TheoryTh ere are many possible configurations for the design of biosen-sors. I will focus here o n the tw o schemes most likely to lead tooperational biosensors for reporting receptor/ligand bindingevents. These schemes, depicted in Figure 1 are the directaffinity sensor (a) and the competitive-binding affinity sensor(b). The competitive binding approach is considered herebecause of its applicability to reagentless biosensing.

a 0**. 0O O9 9 989Receptor

0 = Ligand0 = Labelled Ligand

Figure 1 The direct affinity sensor a) and the competitive-bindingaffinity sensor (b).Th e chemical reaction fo r an affinity sensor is

where R is the receptor, L the ligand, RL the receptor-ligandcomplex, k the association rate co nstant, k the dissociationrate constant , and K , the equilibrium constant (equilibriumaffinity or association con stant). The fraction of receptor m ole-cules in th e complexed form can be expressed as

where the total receptor concentration [R,] = [R] + [RL]. Adimensionless binding curve such as that depicted in Figure 2results, with obviously 1/K, = [L] when half o f the receptorbinding sites are occupied. This value can be used to approxi-mate the applicable dynamic concentration range for the sen-sor., A transformation of the binding curve is normally per-

100

*OE0 1 2 3 4 5 6 7 8L/KaFigure 2 Dimensionless binding curve for an affinity biosensor, withconcentration expressed as [L]/K,, and 100 x [RLJ/[R,] the YOLbound to R.

205

Downloaded

byFACDEQUIMICAon10October2012

Publishedon01Ja

nuary1994onhttp://pubs.rsc.org|doi:10.10

39/CS9942300205

View Online / Journal Homepage / Table of Contents for this issue
http://pubs.rsc.org/en/journals/journal/CS?issueid=CS1994_23_3http://pubs.rsc.org/en/journals/journal/CShttp://dx.doi.org/10.1039/cs9942300205


2/9

206 CHEMIC AL SOCIETY REVIEWS, 1994formed to determine the K , (or K d , dissociation constant = 1/K,) more accurately. Equation 3 represents the widely usedScatchard transformation. The K , K d ) an be determined fromthe slope of a linear plot of [RL]/[L] versus [RL] (or [bound]/[free] versus [bound] for the ligand).

[RLI= R,Ka - RLIK, 3)lJ4Direct affinity biosensing is limited to the instrumental meth-ods available for the direct detection of ligand binding. A moreuniversal approach involves competitive binding schemes suchas that in Figure 1b, where an analyte (ligand) or an analogue islabelled to allow the indirect detection of receptor-ligand bind-ing. The chemical reactions in this case are

4)

5 )

where L* represents the labelled ligand, k , and k, the respectiveassociation and dissociation rate constants, and K,* the equili-brium affinity constant for the reaction. The signal generatedcan thus arise from L* or RL*.For reagentless sensing of competitive binding processesseveral re~ ear che rs~ -~ave labelled an analyte-analogue with amacromolecule and entrapped the large receptors and labelledanalyte-analogue behind a semi-impermeable membrane on thetip of a sensor (a fibre optic cable in these cases) while allowingthe analyte to diffuse through the membrane, yielding a probe-type sensing system. Schultz2 has modelled the equilibriumresponse of such a sensor and concluded that the concentrationratio of the analyte-analogue to total receptor ([L*]/[Rt])controls the dynamic range available, with ratios of 0.01 to 0.1desirable for analytical applications depending on the dynamicrange required. This is in contrast to immunoassay requirementswhere ratios approaching 1 are desired. However the capabilityof the measuring instrument to detect the labelled analyte will bethe limiting factor governing selection of the concentration ratiofor high affinity receptor^.^ It is important to state that theultimate detection limit in any ligand-receptor interaction islimited by the intrinsic affinity of the receptor for the ligand,provided sufficient labelled species exist to be detected by theinstrument. The dynamic response of this sensor can also bemodified by modifying the binding constants.2 This can beachieved by selection of L* that has a different binding affinity orthat has a higher valency for the receptor than L.Miller and AndersonS have modelled the chemical kinetics ofa competitive binding immunosensor. Their simulations showthat the response time of such a sensor is controlled by thedissociation rate constant and is independent of the associationrate constant. Indeed Schultz2 has estimated that the associationrate constant is the same for all antibodies and is on the order of108M-ls-l, consistent with the view that the rate is diffusioncontrolled. The model predicts that equilibrium response timesin the order of minutes can be achieved for receptors withdissociation rate constants greater than approximately4 x 10- ,s- l. The use of receptors with higher dissociation rateconstants of course is limited by the reduction in the affinity ofthe sensor, leading to a trade-off between response time andsensitivity in the development of practical reversible biosensors.

3 Antibody-based BiosensorsImmunoassays are well established in clinical chemistry and areeven gaining acceptance for on-site environmental assay ofcontaminants. Below I will discuss the detection of antigen andantibody levels, both for assay and sensor development. Unfor-tunately the focus of most clinical diagnostic research is on thedevelopment of highly sensitive solid phase enzyme-immuno-

assays, using monoclonal antibodies with high affinities, thuslimiting the application to reversible sensing. Solid phase com-petitive-binding immunoassays have potential applications asimmunosensors, provided the reversal of the antibody-antigeninteraction can be achieved while maintaining the antibodybinding capacity, and the assay components (except targetedantigen) are available continuously. Indirect detection of anti-body-antigen binding is by far the most predominant schemeutilized in immunoassays. These schemes involve labelling ofeither the antibody or antigen with a species that is detectable orthat may generate a detectable species. Examples of labelsinclude enzymes, fluorescent, and electrochemical species.Enzyme-labelling of antibodies or antigens for amplificationof the signal and detection of antigen levels is the most widelyused non-isotopic assay procedure. The schemes involved in themost common enzyme immunoassays (EIAs), the hetero-geneous (sandwich and competitive) enzyme-linked immuno-sorbent assay (ELISA) and the homogeneous (competitive)enzyme-mediated immunoassay, are depicted in Figures 3 and 4respectively. Detection of antigen levels in ELISA is accom-plished by addressing the activity of the enzyme upon additionof substrate by monitoring the increase in products or decreasein substrate. Sandwich ELISA (Figure 3a) is the most sensitiveassay scheme and offers comparable detection limits to radio-immunoassay procedures.

S P S PX XYY

S P

YYy lry Antibodyo=ntigen

= Enzyme Labelled 2ry Antibody= Enzyme Labelled Antigen

Figure 3 Sandwic h two-site) a) and competitive-bind ingdb) enzyme -linked immunosorbent assays where S and P represent the enzymesubstrate and product, respectively.Sandwich ELISA techniques are limited to antigens thatpossess dual epitopes to enable simultaneous binding of twoantibodies. Detection of the smaller antigens (haptens) can beachieved using competitive binding ELISA (Figure 3b) orhomogeneous EIA schemes (Figure 4). Several enzyme-based

assay schemes for the detection of haptens have been deve-loped. These schemes use antibody-antigen binding to modu-late enzyme activity (Figure 4a), enzyme mediator (or modula-tor) activity (Figure 4b), or enzyme-substrate binding (Figure4c). Once again antigen levels are assayed by monitoring eitherthe enzyme substrate or product. Similar schemes to thoseoutlined in Figures 3 and 4may be performed using fluorescentor electroactive labels. Direct optical and mass detection ofantigen binding to immobilized antibody has also been demon-strated. The following sections will review the procedures deve-loped for immunoassays and sensors classified according to thetransducing element used for detection. Most of the assayschemes and detection principles outlined above and below havealso been developed for receptor-binding and DNA hybridiza-tion assays and will not be reiterated under those headings.3.1 Optical DetectionLarge-scale screening and assay of antigen or antibody levels isroutinely achieved using microtitre plate absorbance o r lumines-

Downloaded


Publishedon01Ja


39/CS9942300205

View Online
http://dx.doi.org/10.1039/cs9942300205


3/9

AFFINITY BIOSENSORS-D. LEECH 207

E-Ag Ab:Ag-EI n a c t i v eP

ACI Ab:Agxb:Ag-MA I n a c t i v eA ZS i ) P S i i ) PFigure 4 Homogeneous enzyme immunoassay schemes involvingmodulationofenzyme activity a), mediator or modulator activity b),or enzyme-substrate binding c) upon bindingof the labelled antigento the antibody. The symbols Ab, Ag, E, S, and P have their usualmeanings and Ag-M represents the antigen-mediator or enzymemodulator) complex.cence instruments in conjunction with ELISA schemes.Enzymes which have chromogenic substrates, such as horse-radish peroxidase (HRP), alkaline phosphatase (AP), and P Dgalactosidase 8-G), are normally used in ELISA procedures.More sensitive assays can be achieved using amplification orcascade schemes, where the enzyme product is a substrate for aseparate enzyme reaction. Increased sensitivities may also beobtained by using fluorescent labelled antibodies or antigens orby labelling with enzymes that generate luminescent products.Antibodies and antigens labelled with fluorescein and rhoda-mine derivatives have been successfully used in both competitiveand non-competitive fluoroimmunoassays. Substrates whichproduce fluorescent products are currently commercially avail-able for AP, HRP, and 8 G enzyme immunoassays. Conven-tional fluoresence-based immunoassays have limited sensitivi-ties, however, because of high background originating fromscattering from the instrument or the sample matrix, exacer-bated by the small Stokes shift of the fluorophores. Thepresence of background fluorescence signals in biologicalmatrices also limits the assay sensitivity. The use of time-resolved fluorescence systems have been introduced to overcomethis problem. Most systems employ the lanthanide chelateswhich have a long-lived fluorescence, large Stokes shift andnarrow emission band. The principle of time-resolved fluores-cence is straightforward. Short-lived fluorescence generated byirradiation of a sample with a burst of laser light will decay tozero in approximately loops. If an appropriate time lapse afterapplication of the pulse is allowed the long-lived lanthanidechelate fluorescence can be selectively measured. Immunoassaysusing lanthanide time-resolved fluorescence have beenreported.

Chemiluminescent and bioluminescent immunoassays havebecome popular because of the analytical sensitivity obtainable.Assays that use chemiluminogenic substrates of the enzymeHRP have been improved by the development of variouscompounds which enhance and prolong the luminescent signal.

Similar improvements in AP-based chemiluminescent assayshave been achieved by the synthesis of novel substrates whichcan generate longer-lived chemiluminescencefollowing dephos-phorylation.* The use of bioluminescent systems has the poten-tial advantage of offering higher quantum yields (up to goyo)compared to chemiluminescent systems (usually below 5YO).However, routine use of assays utilizing the firefly luciferasebioluminescence system are at present hindered by the com-plexity of the enzyme system and its low turnover number.Although these assays are simple and versatile they are timeconsuming and are not configured as probe-type reversiblebiosensors. Several research groups have attempted to developoptical immunosensing strategies using fibre optic cables aswaveguides. A bioaffinity fibre optic immunosensor has beendeveloped for the direct detection of the fluorescent antigenbenzo[a]pyrene tetraol (BPT) using a microscale regenerablesystemg o aspirate reagents and washing solution over the distaltip of the cable. Barnard and Walt1 have devised an alternativestrategy for the delivery of reagents to a fibre opticfluoroimmunosensor. A competitive-binding fluoroimmuno-sensor based on fluorescence energy transfer and the controlledrelease of a fluorescein-labelled antibody and Texas Red-labelled antigen from ethylene-vinyl acetate polymer reservoirsat the tip of the fibre optic cable was described. Binding of theunlabelled antigen IgG was detected by monitoring the ratio offluorescent intensity following fluorescence energy transfer fromthe fluorescein molecules to Texas Red upon formation of thedual labelled immune complex. A reversible immunosensorbased on the fluoresence energy transfer mechanism and anti-bodies with high dissociation rate constants was also reported. sThis sensor used a competitive binding scheme of Texas Redlabelled antiphenytoin antibody and b-phycoerythrin-labelledphenytoin for the detection of phenytoin. A similar reversiblesensing scheme was first described by Schultz for an opticalaffinity sensor for glucose based on fluorescein labelled dextran(an analyte-analogue) and the receptor Concanavalin-A (Con-A) confined at a fibre optic tip behind a semi-permeable dialysismembrane2*3as shown in Figure 5 . Operation of the sensorinvolved allowing the small unlabelled glucose to diffuse into thereaction chamber to displace the fluorescein isothiocyanate-labelled dextran (FITC-Dextran) from the Con-A. The dissocia-tion kinetics of these competitive-binding affinity biosensors isthe determining factor in reversibility and regeneration of thereceptor, as discussed above.

ImmobilizedConca nava in-A\ . 1Exc i ta t i onEmiss ion I

-Labe l ledDextran/ \ lucoseHollowDia l ys i sFibre

Figure 5 Schematic diagram of a fibre optic equilibrium displacementAdapted with permission from reference2.)affinity biosensor.

Immunosensors based on the immobilization of the antibodyalong the core of the fibre optic cable and the detection offluorescent labels in the evanescent field of the fibre have beendescribed. This electromagnetic wave can be used to confersurface selectivity to bioaffinity sensors as the typical pene-tration depth of the field is 40-100nm. Changes in the fluores-cence excited by the evanescent wave in this field upon binding of

Downloaded


Publishedon01Ja


39/CS9942300205

View Online
http://dx.doi.org/10.1039/cs9942300205


4/9

208 CHEMICAL SOCIETY REVIEWS, 1994the antigen can thus be used for biosensing purposes Improve-ments in the sensitivity and detection limit of evanescent waveimmunosensors are still required for them to compete with solid-phase enzyme immunoassays Progress in this area could beachieved by the use of chemiluminescent or long-lived fluor-escent labels to increase the signal-to-noise ratio and eliminatebackground problems Other surface-sensitive optical methodshave the capability to be used in bioaffinity sensor design asdemonstrated by the Surface Enhanced Raman Scattering(SERS) sandwich assay for Thyroid Stimulation Hormone(TSH) l 2While these indirect sensors possess some of the desirablecharacteristics for reagentless, reversible affinity biosensing, thesimpler direct approach to the detection of ligand bindingoffered by some Internal Reflectance Spectroscopy (IRS) tech-niques seems more attractive One such technique receivingmuch attention recently is Surface Plasmon Resonance (SPR)Surface plasmons, which represent quanta of oscillations, maybe produced at metal-glass interfaces by the resonant excitationof electrons in the metal by electron beams or light Plasmons forsensing applicationsare produced by the evanescent waves in theAttenuated Total Reflectance (ATR) configuration as depictedin Figure 6 When the light energy is transferred to the surfaceelectrons a corresponding decrease in the intensity of the ref-lected light occurs giving a minimum in the reflectivity versuslight angle plot The system is sensitive to changes in therefractive index n)of the media at the surface and thus changesin n due to binding reactions can be detected by a change in theresonance angle of the incident light SPR has been used todetect the Human Chorionic Gonadotrophin (HCG) hormoneto a detection limit of 1 x 10-*M using a multilayer biotin-avidin-antibody system While this technology offers excitingprospects for the direct detection of receptor-ligand bindinginteractions, SPR still suffers from problems associated withnon-specific binding of proteins, lack of comparable sensitivityto ELISA technology and application to reversible-bindingsensing strategies

lane Polanzedlight

Glass SupportMetal (gold) LayerImmobilized Layer1 ample

Figure 6 Schematic diagram of the experimental arrangement forsurface plasmon resonance detection of ligand binding The angle ofminimim reflection of the totally internally reflected laser light, 8,shifts with changes in the refractive index of the sample medium closeto the gold layer allowing monitonng of ligand-binding processes(Adapted with permission from reference 12

3.2 Electrochemical DetectionElectrochemical immunoassays and sensors have some advan-tages over optical-based systems in that they can operate inturbid media, offer comparable instrumental sensitivity, and aremore amenable to miniaturizationPotentiometric sensors measure the potential differencebetween a working and a reference electrode Potentiometricimmunoassays for the direct detection of antigen binding havebeen proposed where the protein antigen or antibody is viewedas a charged polyelectrolyte which, upon binding to the surfaceof an indicator electrode, would change the potential differenceUnfortunately these assays suffer from a lack of sensitivity andfrom interference problems from other ions in the matrixAttempts to improve the sensitivityof potentiometric immuno-assays using enzyme-labelledantibodies that generate detectableions or gases have been relatively unsuccessful as the problems

of interferences and non-specific binding events still exist Theadvent of the Ion Selective Field Effect Transistors (ISFETs)and Chemically Selective FETs (CHEMFETs) has resulted inthe capability of microfabrication of ion and pH sensors andshowed great early promise in application to immunosensingThe FET devices are insulated semiconductor devices thatrespond by detecting the effect surface potentials at the elec-trode-solution interface have on the electric fields in the semi-conductor Several researchers have attempted to utilize affinityspecies attached to the channel between the source and drainelectrodes to measure changes upon binding to the antigenSome controversy exists, however, about the origin of thepotential shifts14 and interference and non-specific bindingwould also seem to limit these devicesThe recently developed Light Addressable PotentiometricSensor (LAPS) based on FET technology has, in contrast,proved to be highly successful for immunoassay and receptorbinding studies The basis of the technique is shown in Figure 7and involves the application of a bias potential to a silicon platecontaining a pH-sensitive surface insulator of silicon oxynitrideMolecular

SiliconOxyni t r ideS i l i c onP l a t e

L i g h t E m i t t i n gDiodesFigure 7 The light-addressable potentiometric sensor The silicon platewith pH-sensitive surface layer is photoresponsive to the light-emitting diodes The photocurrent in the external circuit is dependenton the applied potential and responds to changes in the surface

potential caused by localized chemistries PH changes) induced bymolecular recognition (see text for further details)(Adapted with permission from reference 14An intensity-modulated light source from Light-EmittingDiodes (LEDs) gives rise to an alternating photocurrent whichhas an amplitude proportional to the solution/surface pH at agiven bias potential The immunoassay instrument developedmeasures pH changes in a sandwich-type assay using urease-labelled antibodies to produce local pH changes upon additionof urea substrate A novel potentiometric affinity biosensoremploying catalytic antibodies as the molecular recognitionelement has recently been described The model catalyticantibody (20G9) raised against phenyl phosphonate wasdesigned to catalyse the hydrolysis of phenyl acetate to acidicproducts that change the pH at the underlying potentiometricpH electrode Antibodies raised against transition-state analo-gues are capable of catalysing chemical reactions by lowering thepotential energy for the transition state of the reaction uponbinding Although a detection limit of only 5mM and a slowresponse time was obtained for phenyl acetate, the design ofcatalytic antibodies for specific affinity biosensor applicationshas immense potential, because of the rapid dissociation of theantigen and the regeneration of the molecular recognition siteAmperometnc sensors measure the current produced uponoxidation or reduction of an electroactive species at a polarizedworking electrode Research on amperometric immunoassayshave in recent years focused on the use of heterogeneous ELISAschemes using AP-labelled antibodies because of the availabilityof the non-electroactive substrate 4-aminophenyl phosphateThe product of the enzyme reaction, 4-aminophenol, can beeasily oxidized at a low potential thus minimizing interferencefrom other electroactive species Other researchers have devisedelectrochemical schemes for substrate delivery in electrochemi-

Downloaded


Publishedon01Ja


39/CS9942300205

View Online
http://dx.doi.org/10.1039/cs9942300205


5/9

AFFINITY BIOSENSORS-D. LEECH 209cal immunoassays. This entails either the use of an enzymesubstrate that can be produced electrochemically, such as H 2 0 2for HRP enzyme immunoassays, or an electrochemical media-tor of enzyme oxidation-reduction that can be regeneratedelectrochemically, such as the benzoquinone/hydroquinoneredox couple used for glucose oxidase (GOD) labelled immuno-assays.19 An automatic flow injection apparatus for hetero-geneous immunoassays, based on the GOD-benzoquinone elec-trocatalytic scheme, allowed the electrochemical regeneration ofthe electrode, for subsequent primary antibody adsorption, byoxidizing it in the presence of nitric acid.19The use of enzyme mediation schemes for the development ofhomogeneous amperometric immunoassays has also beenattempted. The initial studyzoawas based on the ability of aferrocene-lidocaine conjugate to mediate the oxidation of glu-cose by GOD in the scheme depicted in Figure 8a. Competitivebinding of unlabelled lidocaine yielded an increase in the cata-lytic current. Coupling of homogeneous amperometric im-munoassay schemes to an enzyme cascade could lead toimproved sensitivity, such as demonstrated for the detection ofbiotin (Figure 8b).206An extremely sensitive competitive im-munoassay procedure using an electrochemiluminescent label iscommercially available.2 The assay utilizes a Ru(bpy): label,where bpy is the ligand 2,2-bipyridine, which when oxidized inconjuction with tripropylamine (TPA) forms the excited-stateRu(bpy)g+* that emits a photon of 620nm wavelength upondecay to the ground state. Extremely low detection limits of 200fh for the label was reported.Although advances in the design of sensitive electrochemicalassays have been significant there has as yet been no report of atrue amperometric immunosensor. The approach taken for thedesign of optical immunosensors described above2-* would

a)Ag

Fc-Ag Fc-Ag:Ab inactive)

e - C+-Ag G I u c o s e

Electrode Gluconolactone

BwB ATrypsin-B:A inactive)

nChymotrypsinogen ChymotrypsinGlucosee- HO-Tyr -CO-Fc+

GluconolactoneE lec t r ode

Figure 8 Homogeneous competitive amperometric immunoassaysbased on enzyme-mediated amplification of the binding. In (a),binding of the antibody to the ferrocene-labelled antigen (Fc-Ag)inactivates the mediated electron transfer from the enzyme GOD tothe electrode. In (b), avidin (A) binding to biotinylated trypsin(trypsin-B) results in inactivation of the enzyme cascade that wouldresult in the release of a small ferrocene complex capable of mediationof electron transfer from GOD to the electrode surface.(Adapted with permission from references 19 and 20.)

seem feasible also for the development of amperometric im-munosensors but has as yet not been demonstrated. Ampero-metric immunoassays employing flow-injection schemes seemmore practical than microtitre-plate technology for the rapidassay of antigen levels. Improvements in design of these assayscould include methods for continuous delivery of substratesand regeneration of antibody binding sites. An attractive schemefor the reagentless detection of enzyme activity has been des-cribed by Smit and Rechnitz,22where the enzyme tyrosinase wasdemonstrated to oxidize an artificial substrate, Fe(CN) ,which could be re-reduced at an electrode, while consuming(reducing) only oxygen.3.3 Piezoelectric DetectionThe converse piezoelectric effect, where an electric field appliedto a piezoelectric material induces stress in the material, has beenexploited for the detection of minute mass changes. The range ofpiezoelectric acoustic wave devices available and the applicationof the effect to chemical and biosensing has been recentlyr e v i e ~ e d ~ ~ - ~nd will not be dealt with in detail here. The twomost widely studied piezoelectric biosensor configurations arethe Quartz Crystal Microbalance (QCM) and the SurfaceAcoustic Wave (SAW) modes. The QCM devices consist of anoscillator circuit that drives an AT-cut quartz crystal, sand-wiched between two electrodes, at its fundamental resonantfrequency that is tracked with a frequency counter. Changes inthe resonant frequency are related to changes in the mass of thecrystal, thus allowing the QCM to operate as a sensitive massdetector. SAW devices operating on the Raleigh wave principlecan also be used for mass detection by monitoring changes inacoustic wave propagation between the generator and receiversets of electrodes separated by a piezoelectric material.Although the ability of these devices to measure gaseous sampleshas received considerable attention, their most interestingfeature is their ability to operate in liquids. Some controversyexists, however, concerning the SAW devices acoustic mode inliquids and the origination of the frequency changes for theQCM.23Piezoelectric immunosensors have been fabricated by modify-ing the piezoelectric surface of the QCM with antibodies orantigens, and several researchers report the direct detection ofantibody-antigen binding by monitoring of frequency changesin l i q ~ i d s . ~ ~ , ~owever, the variability, lack of sensitivity, andinadequate signal-to-noise ratios of these devices in liquidswould seem to limit their application to direct liquid sensing.Indeed it is questionable whether the total frequency changeupon binding is due to mass changes alone since changing liquidproperties such as density and viscosity and entrapped surfaceliquid changes may contribute to the frequency response of thesystem, although the use of a network analyser to operate theQCM devices can help distinguish the mechanisms involved inthe sensor response.23b Binding of antibody or antigen toimmobilized species on the QCM in the liquid phase followed bydrying of the crystal and measurement of the subsequent fre-quency shift has been demonstrated to be a useful assay for largeantigens (such as microbial ~pecies).~~Jontinuous liquidphase operation is, however, more preferable for biosensordesign. A problem also encountered in the direct detection ofmass changes is the effect of non-specific adsorption on thesensor response. This can be alleviated to some extent by the useof matching reference crystals and the proper use of blockingagents.The use of amplification schemes such as the proposedAmplified Mass Immunosorbent Assay (AMISA) can lead toimproved detection limits over direct detection schemes.26 Inthis scheme, shown in Figure 9a, AP-labelled secondary anti-body to adenosine 5-sulfonate reductase (APS) was utilized todeposit the insoluble dimer of the enzyme substrate 5-bromo-4-chloro-3-indolyl phosphate (BCIP) on the piezoelectric surfaceupon oxidation of the substrate by the bound enzyme, thusamplifying the response. A detection limit of 10 I4Mwas

Downloaded


Publishedon01Ja


39/CS9942300205

View Online
http://dx.doi.org/10.1039/cs9942300205


6/9

210 CHEMICAL SOCIETY REVIEWS, 1994

S s a)

y- ry Antibody o=ntigeny=nzyme Labelled 2ry Ant ibodyFigure9 Schematic diagram of a QCM sandwich-based amplified massimmunosorbent assays. In (a), the alkaline phosphatase-labelledsecondary antibody converts the BCIP substrate (S) into the insolubledimer (P) which deposits on the gold QCM electrode decreasing theresonant frequency. In (b), the horseradish peroxidase-labelledsecondary antibody on the nylon membrane converts the H 2 0 2 / I -substrate (S) into I; (P) which can diffuse to, insert into, and oxidizethe nearby PVF film on the QCM decreasing the resonant frequency.

Electrochemical reduction of the PVF film results in expulsion of theI; and regeneration of the neutral film for subsequent assays.(Adapted with permission from reference 26.)

established for the APS antigen, whereas direct detection of APSwas not possible because of the small frequency changesinvolved. In the same report the authors described an approachfor the development of a two-component sensor consistingof areusable crystal and a disposable antibody membrane asdepicted in Figure 9b. The gold-covered quartz crystal electrodewas modified with a redox-active PolyVinylFerrocene (PVF)layer while a nylon membrane containing immobilized anti-HCG antibodies was placed opposite the electrode. HRP-labelled secondary antibodies enzymatically reduced the I toI;/12 and the 1 could in turn oxidize and insert into the PVF filmresulting in large frequency changes. Reversion of the film to itsneutral state could be achieved by electrochemical reduction ofthe PVF and rapid expulsion of I;, resulting in a reusable sensingsurface/disposable recognition surface biosensor.It should be noted that rapid, non-destructive reversal of thebinding event is still required for the development of a truepiezoelectric affinity biosensor, even if the devices are proven tobe capable of measuring direct binding of antigens in solution.Chemical reversal of antibody binding for reusable assays hasbeen achieved using low pH buffers and/or chaotropic agents.However the design of antibodies possessing kinetic bindingcharacteristics specially tailored for biosensing purposes wouldresult in the development of continuous-use reversible biosen-sors that would not require a heteregeneous washing or regene-ration step.

3.4 Antibody Immobilization and LabellingAn important factor involved in the design of solid-phaseimmunosensors and assays not dealt with above is the immobili-zation procedure of the antibody. Adsorption of the antibodyonto polystyrene wells is normally performed for ELISA, while

methods for the chemical attachment of the antibody to solidphases have been examined. An apparently simple procedurefor antibody immobilization involves encapsulation within asol-gel matrix, as reported recently.28 This procedure hasapplication to both optical and piezoelectric biosensor develop-ment provided antigen-binding activity is maintained within thematrix. New genetic engineering technology can also play a rolein immunosensor research. The advent of molecular biologicaltechniques has resulted in methods for the production of fusionproteins.28 Fusion proteins of antibody-nzyme or antibody-peptide conjugates are produced by transforming bacterial cellswith the enzyme-peptide vector and subsequent culturing of thecells to express the conjugate. Genetic engineering of antibody-enzyme conjugates is an exciting prospect and could lead to site-directed labelling methods for optimum antibody and enzymeactivities and will also minimize the variability associated withthe current chemical labelling schemes.4 Receptor based Assays and SensorsNeurotransmitter and hormonal receptors are the body s ownbiosensors. Binding of a ligand (an agonist) to the receptortriggers an amplified physiological response such as ion channelopening, second messenger systems and activation of enzymes.Biological receptors are normally capable of binding to a seriesof structurally related compounds rather than to one specificanalyte, which make their potential use as biosensors for classesof pharmacologically active compounds attractive.RadioReceptor Assays (RRAs) have been routinely used forinvestigation of receptor-ligand interactions in neurobiology,pharmacology, and other related areas. In recent years the use ofnon-isotopic receptor-binding assays has received growingattention in an attempt to address the limitations of RRAs. Thenon-isotopic receptor-based assays and sensors can be dividedinto those that utilize intact receptors and those that utilizeisolated, purified receptor preparations.

4.1 Intact Receptor-basedBiosensorsPerhaps the most important biosensing device developed inrecent years is the patch clamp apparatus.29 The patch clamp isused by electrophysiologsts to investigate the mechanisms ofchannel gating, membrane transport, and the molecular phar-macology of membrane-bound receptors by examining effectson the ion currents in a small patch of membrane located on thetip of a glass microelectrode. The application of the patch clampto practical biosensing is limited however by the instability ofthese patches, the small currents measured (requiring elaborateequipment and attention to grounding) and the fragility of theglass microelectrodes.Improved lifetimes of neuronal biosensors can be achievedby using less invasive electrophysiologicaldetection proceduresto measure the ionic flux associated with ion-channel opening inmembrane-bound receptors. Extracellular monitoring of nervefiring, associated with the transmission of signals in nervoustissue, for application to biosensing, has been studied in recentyears following the report on the prototype chernoreceptor-based This sensor utilized excised crustacean olfac-tory structures (antennules) for neuronal sensing of chemore-ceptor ligands. The antennules contain receptors responsible forolfactory sensing of compounds such as amino acids, sexhormones, pyridine-based chemicals, and other food markers.The binding of the ligand to the corresponding chemoreceptortriggers the firing of an action potential (AP) in the nerve thatcan be detected with extracellular glass microelectrodes. Thefiring frequency of the AP event is related in a dose-dependantmanner to the concentration of the ligand (stimulant). Reviewson chemoreceptor-based biosensors have been published inrecent years outlining the sensitivity and response ranges achiev-able and describing the principles and applications of theseMore recently our and that of hasbeen directed to using intact neurond receptors for the detection

Downloaded


Publishedon01Ja


39/CS9942300205

View Online
http://dx.doi.org/10.1039/cs9942300205


7/9

AFFINITY BIOSENSORS-D LEECH 21of neuromodulators While the Van Wie group33 has concen-trated on voltage-clamp experiments on the isolated ganglion ofthe pond snail Lzmnea stagnalzs for the detection of serotoninand its antagonists, we have focused our attention on lessinvasive extracellular detection for sensing agents that modulateor block the propagation of the AP in the walking leg nerves andgiant axons of the crayfish Procambarus clarkiz 3 2 a The stimu-lation of AP firing can be achieved using either orelectrical32* timulus pulses The experimental system, shown inFigure 10 allows the detection of compounds which modulatethe propagation of the electrically stimulated AP event Thesystem was used to construct dose-response curves for detectionof the local anaesthetics (LAs), which can bind in the ion channelof the voltage-gated sodium channel receptor responsible for APpropagation in nervous tissue, thus blocking AP conduction

SwitchingValveLSalineplaced on micromanipulators

GI Mounted on

Current Pulse Generator

Micros cope Stage

ct ion Microelect rodes

, 1

Cassette Record er

AudioSpeaker

PhysiologicalPreamplif ier

o\cllloscopcComputer

Figure 10 Block diagram o f the experimental set-up for microelectrodestimulationand monitoringof AP firing in crayfish walking leg nervesfor the detection of neu romodulators such as the local anaestheticsThe applicability of the system to screening natural productextracts for neuromodulatory drugs or toxins was proposed, asseveral important classes of compounds such as anti-depres-sants, narcotics, alcohols, and venoms and toxins are known toaffect AP conduction Although this system shows promise as aneuronal biosensor the limited lifetime of the sensor (approxi-mately 4-8h) is a severe drawback Recently we have demon-strated that channel blockers such as the LAs can be quantifiedby monitoring the biomagnetic current induced in a toroidalprobe surrounding the nerve 32c This system allowsnon-znvaszvedetection of the compound action current (CAC) associatedwith the AP event in crayfish giant axons, as depicted in Figure11 A dose-response curve for the LA lidocaine can be con-structed, as shown in Figure 12, by monitoring the duration ofthe conduction block in the crayfish giant axon upon applicationof standard solutions of the LA to the specially designed flowcell Current research is aimed at studying the effect the non-invasive detection of the AP has on the lifetime of the prep-aration Attempts to replace the invasive extracellular micro-electrodes used to apply stimulus pulses to the preparation withnon-invasive stimulus procedures are also under investigationOther researchers have reported on the monitoring of thecellular response of microorganisms for drug-screeningpurposes 33 34 The LAPS system previously described has beenused to detect changes in the pH of the environment surroundingimmobilized cells, reflecting changing cellular metabolism, in

O 4

0 60 1 2 3 4 5 6 7 8Time ms)

O O60 1 2 3 4 5 6 7 8T i me ms )

Figure 11 Non-invasive recordings of the compound action currentCA C) in crayfish giant ax ons with a b ioma gnetic current probesurrounding the nerves before abov e) and after below ) app licationof 50 nM lidoca ine to the cell The rectangular pulse evident ms aftertriggering of data a cquisition is a 1 ms, 0 2 pA calibration pulse

12001000 18 800 -i

600 -

2C 400 -

0 200 -

0 5 10 15 20 25 30 35[L idocane] mM)Figure 12 Dose-response curve for the local anaesthetic lidocaineconstructed by monitoring with a biomagnetic current probe theduration of the conduction block induced in the crayfish giant axonfollowing a 5 minute flow of standard solutions of lidocaine throughthe cellresponse to various drugs and cell-affecting agents 3 3 McClint-lock el a1 34 have described a useful bioassay for investigatingthe G-protein-coupled receptors by expressing them in Xenopuslaevis melanophores Receptor-mediated stimulation/inhibitionof the second messenger systems caused dispersion/aggregationof the melanosomes Translocation of the melanosomes couldthen be tracked using video microscopy leading to an assay fordetection of receptor expression The combined use of receptorexpression in melanophores and video imaging technologycould clarify the mechanisms of ligand binding and secondmessenger activation pathways

4.2 Isolated Receptor-basedBiosensorsThe predominant scheme in use for receptor- based assays usingisolated receptors is the RRA, although recent research hasdemonstrated the use of fluorescent-labelled, biotinylated, andenzyme-labelled ligands for the assay of receptor agonist andantagonists using schemes similar to those described previously

Downloaded


Publishedon01Ja


39/CS9942300205

View Online
http://dx.doi.org/10.1039/cs9942300205


8/9

212 CHEMICAL SOCIETY REVIEWS, 1994for immunoassays. Isolated receptors have also been success-fully reconstituted into bilayer lipid membranes on patchpipettes and at interfaces of troughs for the electrophysiological/electrochemical investigation of the ion channel and receptor-ligand binding properties.Receptor-based biosensors have been constructed using thewell studied nicotinic Acetylcholine Receptor (nAChR) invarious formats. An ISFET-based biosensor relied on thepotential shifts in the gate voltage associated with ligand bind-ing35but is unlikely to lead to a practical biosensor because ofpreviously mentioned problems with ISFET biosensors. Themeasurement of capacitance changes in receptor liposomesimmobilized on planar interdigitated electrodes upon ligandbinding has also been proposed for biosensor application^^^although it is doubtful that ligand binding is the only contribu-tion to the reported responses. Eldefrawi et al.37have describedthe development of an nAChR-based fibre-optic biosensordepending on evanescent wave excitation of fluorescence inlabelled ligands bound to the immobilized receptor. An irrever-sible system using fluorescein isothiocyanate (F1TC)-labelledabungarotoxin was subsequently improved upon by using thereversible ligand FITC-a-conotoxin. The extension of thissystem to other receptors is envisaged for the non-isotopic assayand detection of therapeutic drugs.None of the biosensing systems above have taken advantageof the inherent amplification associated with ion-channel open-ing of the ligand- and voltage-gated receptors upon agonistbinding, as described for intact-based biosensors. Minami etaL3* have described an ion-channel sensor that utilizes theglutamate receptor reconstituted into a bilayer lipid membranefor the coulometric detection of glutamate. Problems with thestability of the sensor, however, would seem to hinder furtherapplication.Although the use of isolated, immobilized receptors forbiosensing has been demonstrated, practical application of thesesensors requires improvements in receptor purification andreconstitution. Improvements in non-invasive technology forstimulation and/or detection of AP events in intact receptors,and improvements in tissue culture and nerve growth media forthe maintenance of nervous activity in excised nerves showspromise for the development of practical neuronal biosensors.

5 Other A ffi ni ty Biosensors5.1 DNA DetectionNucleic acid hybridization assays can be monitored non-isotopi-cally by attachment of an enzyme, fluorescent, or chemilumines-cent label to a complementary portion of DNA or RNA, or byselection of antibodies specific for double-stranded DNA or alabel. The application of non-isotopic techniques to DNAdetection has been reviewed recently39and will not be presentedin detail here. In general schemes such as those presented underimmunoassays may be used for DNA-based assays. A new andpromising approach to DNA biosensing involves the fluor-escent40 or e lec tro~ hem ica l~~etection of inorganic DNA inter-calators. These compounds, most notably the transition metalcomplexes of Pt, Ru, Co, and Fe with bi- and tridentate ligands,intercalate into double-stranded DNA or bind electrostatic-ally to the DNA phosphate ba~kbone.~O*~lhe electrochem-ical DNA was capable of reversibly detectingoligo(dA), oligonucleotides by immobilization ofolig(dT),,(dG),, on glassy carbon electrode surfaces. Immobili-zation of probe strands of DNA onto electrodes or fibre opticcables for the detection of DNA hybridization using theseintercalators is a promising avenue of research for the develop-ment of biosensors for diagnosis of disease. The direct detectionof DNA hybridi~ation~~r of i n t e r~a la to r s~~as also beendemonstrated using oligonucleotides bound to the QCM,although problems of non-specific binding events need to beaddressed for practical application of these sensors,

5.2 Binding ProteinsThe affinity biosensor concept was first introduced by Schultzetal.3 as described previously, who demonstrated the opticaldetection of glucose, based on competitive affinity binding ofglucose and fluorescent-labelled dextran to concanavalin-A.Other binding proteins, such as protein-A, avidin, and enzymeshave potential uses in affinity biosensing schemes although theiruse has mainly been confined to affinity purification of anti-bodies, amplification of binding events and affinity immobiliza-tion, and biocatalytic sensors, respectively. Ikariyama etdeveloped an affinity biosensor for biotin based on the differen-tial affinity of catalase-labelled avidin for biotin and membrane-bound immobilized 2-[ 4-hydroxyphenyl)azo]benzoic acid(HABA), a biotin analogue. Electrochemical detection of bind-ing was achieved by monitoring production of oxygen by theenzyme reaction following binding to HABA. Wang et al .4simmobilized the enzyme tyrosine hydroxylase (TH) in a Lang-muir-Blodgett film on a gold electrode for the affinity concent-ration of the TH ligands such as the neuroleptic drugs. Appli-cation of a voltammetric pulse scheme resulted in quantitationof the electroactive drugs at the micromolar level and dissocia-tion of the drug from the receptor. The extension of this schemeto other receptors and drugs shows promise, provided a stablefilm can beestablished and that the ligands are electrochemicallyor optically active.

6 ConclusionsResearch into the development of affinity-based biosensors hasincreased considerably in recent years following the first descrip-tion of the affinity biosensor by Schultzet al.3and the report onthe prototype intact receptor-based While severalviable procedures have been demonstrated, problems associatedwith non-specific binding effects, receptor stability, interfer-ences, detection limits, continuous operation, and reversibilityof the binding still exist. The selection of monoclonal antibodieswith the kinetic properties required for reversible binding,coupled with advances in direct detection schemes, such as fibreoptics, SPR and QCM detection, requires close collaborationbetween immunologists and biosensor researchers for thedevelopment of continuous use immunosensors. Competitive-binding optical affinity biosensors such as those describedr e ~ e n t l y ~ - ~how great promise if this problem can be solved.Advances in genetic engineering of fusion proteins, antibodyimmobilization procedures, and reagentless assay of enzymeactivity will improve immunoassay technology immensely.Interdisciplinary approaches to receptor cloning, purification,and immobilization are also required to develop operationalreceptor-based biosensors for drug screening applications. Inconclusion, although several promising affinity biosensorformats have been reported, a practical commercial affinitybiosensor device has yet to be constructed. Future researchrequires collaboration between chemists, biochemists, immuno-logists, pharmacologists, neuroscientists, and engineers in orderto obtain stable, isolated, immobilized receptors (antibodies,receptors, binding proteins) that possess the required character-istics for biosensing (affinity, reversibility, sensitivity, selecti-vity), while achieving advances in instrumentation and detectionprocedures to allow direct detection of the binding event.

7 ReferencesI a ) Biosensors: Fundamentals and Applications, ed. A. P. F .Turner, I. Karube, and G. S. Wilson, Oxford University Press,Oxford, 1987. b) Methods in Enzymology: Immobilized Enzymesand Cells, ed. K . Mosbach, Academic Press, San Dieg o, Vol. 137,Part D, 1988. c) E. Hall, Biosensors, Horwood, Chichester, 1991.2 J. S. Schultz, Ann. N . Y . Acad. Sci. 987,506,406.3 S. Mansouri and J . S. Schultz, Biotechnology, 1984,2, 385.4 S. G. Weber and A . Weber, Anal. Chem., 1993,65,223.5 W. G. Miller and F. P. Anderson, Anal. Chim. Acta , 1989,227, 135.6 Enzyme-Mediated Immunoassay, ed. T . Ngo and H . M. Len-hoff, Plenum Press, N . Y ., 1985.

Downloaded


Publishedon01Ja


39/CS9942300205

View Online
http://dx.doi.org/10.1039/cs9942300205


9/9

AFFINITY BIOSENSORS-D LEECH 2137 E P Diamandis and T K Christopoulos, Anal Chem , 1990, 62,8 T T Ngo, Curr Opinion Biotech , 1991,2, 1029 J P Alane, J R Bowyer, M J Sepaniak,A M Hoyt, and T Vo-

1149A

Dinh, Anal Chim Acta, 1990,236,23710 S M Barnard and D R Walt, Science, 1991,251,92711 R B Thompson and F S Ligler in Biosensorswith Fiberoptics, edD L Wise and L B Wingard, Humana Press, N Y , 1991, p 11 112 R E Tohr , T Cotton, N Fan, and P J Tarcha, Anal Biuchem ,1989,182,38813 J Spinke, M Liley, H -J Gruder, L Angermaier, and W Knoll,Langmuir, 1993,9, 182114 J Janata, Anal Chem , 1992,64, 196R15 D G Hafeman, J W Parce, and H M McConnell, Science, 1988,240 118216 G F Blackburn, D B Talley, P M Booth, C N Durfor, M TMartin, A D Napper, and A R Rees, Anal Chem ,1990,62,221117 Y Xu, H B Halsall, and W R Heineman in ImmunochemicalAssays and Biosensors Technology for the 1990s, ed R M Naka-mura, Y Kasahara, and G A Rechnitz, American Society forMicrobiology, Washington D C , 1992,p 29118 W J Albery in Host-Guest Molecular Interactions FromChemistry to Biology, ed D J Chadwick and K Widdows, CibaFoundation, London, 1991, p 6319 D Huet and C Bourdillon, Anal Chzm Acta, 1993,272, 20520 a ) K DiGlena , H A Hi11,C J McNeil, an dM J Green, AnalC h e m , 1986,58, 1203 (b) K DiGlena, H A Hill, and J AChambers, J Electroanal Chem , 1989,267,8321 G F Blackburn,H P Shah,J H Kenten,J Leland,R A Kemin,JLink, J Peterman, M J Powell, A Shah, D B Talley, S K Tyagi,E Wilkins, T -G Wu, and R J Massey, Clin Chem ,1991,37,153422 M H Smit and G A Rechnitz, Anal Chem , 1993,65,38023 J W Grate, S J Martin, and R M White, Anal Chem ,1993,65, a )

24 A A Suleiman and G G Guilbault, Anal Lett , 1991,24, 128325 M D Ward and D A Buttry, Science, 1990,249, 100026 R C Ebersole and M D Ward, J Am Chem SOC, 1988,110,862327 R Wang,U Narang,P N Prasad,andF V Bright, Anal Chem,

940A, (b) 987A

1993,65,2671

28 C Lindbladh, K Mosbach, and L Bulow, Tr Biochem Sci , 1993,29 E Neher and B Sakman, Nature, 1976,260,79930 L Belli and G A Rechnitz, An a l L e t t , 1986,19,4033 1 a )R M Buch and G A Rechnitz, Anal Chem ,l989,61,533A b )D Leech and G A Rechnitz, Electroanalysis, 1993, 5, 10332 (a) D Leech and G A Rechnitz, Anal Chim Acta, 1993,274,25 b)D Leech and G A Rechnitz, Anal Let t , 1993, 26, 1259 (c) DLeech and G A Rechnitz, Anal Chem , 1993,65,326233 a )R S Skeen, W S Kisaalita, B J Van Wie, S J Fung, and C DBarnes, Biosen Bioelectrunics 5, 1990,491 b)R S Skeen, B J VanWie, S J Fung, and C D Barnes,Biosen Bioelectronics 7, 1992,9134 T S McClintlock, G F Graminski, M N Potenza, C K Jayawick-reme, A Roby-Shemkovitz, and M R Lerner, Anal Biochem ,1993,209,29835 M Gotoh, E Tamiya, M Momoi, Y Kagawa, and I Karube, AnalL e t t , 1987,20,85736 R F Taylor, I G Marenchic, and E J Cook, Anal Chim Acta,1988,213, 13137 M E Eldefrawi, A M Eldefrawi, K R Rogers, and J J Valdes, inImmunochemical Assays and Biosensors Technology for the 1990s,ed R M Nakamura, Y Kasahara, and G A Rechnitz, AmencanSociety for Microbiology, Washington D C , 1992, p 39138 H Minami, M Sugawara, K Odashima, Y Umezawa, M Uto, EK Michaelis, and T kuwana, Anal Chem , 1991,63,278739 a ) Nonisotopic DNA Probe Techniques, ed L J Kricka, Aca-demic Press, San Diego, 1992 (b) M E A Downs, S Kobayashi,and I Karube, Anal Let t , 1987,20, 189740 A E Friedman,C V Kumar,N J Turro,andJ K Barton, NucleicAci ds Re s , 1991,19,259541 K M Millan and S R Mikkelsen, Anal Chem , 993,65,231742 N C Fawcett, J A Evans, L C Chien, and N Flowers, Anal Lett ,1988,21, 109943 Y Okahata, K Ijiro, and Y Matsuzaki, Langmuir, 1993,9, 1944 Y Ikanyama, M Furuki, and M Aizawa, Anal Chem , 1985,57,45 J Wang, Y Lin, A V Eremenko, I N Kurochkin, and M F

18,279

496Mineyeva, Anal Chem , 1993,65, 513

Downloaded


Publishedon01Ja


39/CS9942300205

View Online
http://dx.doi.org/10.1039/cs9942300205

Affinity Biosensors, Leech 1994

Documents

Transcript of Affinity Biosensors, Leech 1994