present and future of surface plasmon resonance

8/2/2019 Present and Future of Surface Plasmon Resonance

1/12

Abstract Surface plasmon resonance (SPR) biosensorsare optical sensors exploiting special electromagnetic waves surface plasmon-polaritons to probe interactions be-tween an analyte in solution and a biomolecular recogni-tion element immobilized on the SPR sensor surface. Ma-jor application areas include detection of biological ana-lytes and analysis of biomolecular interactions where SPRbiosensors provide benefits of label-free real-time analyt-ical technology. This paper reviews fundamentals of SPRaffinity biosensors and discusses recent advances in de-velopment and applications of SPR biosensors.

Keywords Optical sensor Biosensor Affinitybiosensor Immunosensor Surface plasmon resonance

Introduction

The last two decades have witnessed remarkable progressin the development of affinity biosensors and their appli-cations in areas such as environmental protection, bio-technology, medical diagnostics, drug screening, food safety,and security. An affinity biosensor consists of a transducer(electrochemical [1], piezoelectric [2], or optical [3]) and abiological recognition element which is able to interactwith a selected analyte. Various optical methods havebeen exploited in biosensors including fluorescence spec-troscopy [4], interferometry (reflectometric white light in-terferometry[5] and modal interferometry in opticalwaveguide structures[6]), spectroscopy of guided modes

of optical waveguides (grating coupler [7] and resonantmirror [8]), and surface plasmon resonance (SPR) [9, 10].Fluorescence-based biosensors offer high sensitivity but,due to the use of labels, they require either multi-step de-

tection protocols or delicately balanced affinities of inter-acting biomolecules for displacement assays, causing sen-sor cross-sensitivity to non-target analytes [11]. Sensorssuch as optical interferometers, grating coupler, resonantmirror, and SPR rely on the measurement of binding-in-duced refractive index changes and thus are label-freetechnologies. This paper focuses on SPR biosensor tech-nology, reviews fundamentals of SPR sensing, and dis-cusses advances in development and applications of SPRbiosensors and emerging trends in SPR biosensing.

Fundamentals of surface plasmon resonance (SPR)biosensors

Surface plasmon-polariton

A surface plasma wave (SPW) or a surface plasmon-po-lariton is an electromagnetic wave which propagates alongthe boundary between a dielectric and a metal, which be-haves like quasi-free electron plasma [12, 13]. An SPW isa transverse-magnetic (TM) wave (magnetic vector is par-allel to the plane of interface) and is characterized by thepropagation constant and electromagnetic field distribu-tion. The propagation constant of an SPW, , can be ex-pressed as:

(1)

where is the angular frequency, c is the speed of light invacuum, and D and M are dielectric functions of the di-electric and metal, respectively [12, 13]. This equation de-scribes an SPW if the real part ofM is negative and its ab-solute value is smaller than D. At optical wavelengthsthis condition is fulfilled for several metals of which goldis most commonly employed in SPR biosensors. The realand imaginary parts of the propagation constant describespatial periodicity and attenuation of an SPW in the direc-tion of propagation, respectively[12]. The electromag-netic field of an SPW is confined at the metaldielectricboundary and decreases exponentially into both media,

0 '

0 '

F

Ji Homola

Present and future of surface plasmon resonance biosensors

Anal Bioanal Chem (2003) 377: 528539DOI 10.1007/s00216-003-2101-0

Received: 20 March 2003 / Revised: 4 June 2003 / Accepted: 5 June 2003 / Published online: 19 July 2003

REVIEW

J. Homola ()Institute of Radio Engineering and Electronics,Academy of Sciences of the Czech Republic,Chabersk 57, Prague, Czech Republice-mail: [email protected]

Springer-Verlag 2003


2/12

Fig. 1. For an SPW at the boundary between gold and a di-electric with a refractive index of 1.32 the penetration depth(the distance from the interface at which the amplitude ofthe field falls to 1/e of its value at the metal surface) intothe dielectric is typically 100500nm in the visible andnear infrared regions [10, 13].

Concept of surface plasmon resonance biosensing

Owing to the fact that the vast majority of the field of anSPW is concentrated in the dielectric, the propagationconstant of the SPW is extremely sensitive to changes in

the refractive index of the dielectric. This property ofSPW is the underlying physical principle of affinity SPRbiosensors biomolecular recognition elements on the sur-face of metal recognize and capture analyte present in aliquid sample producing a local increase in the refractiveindex at the metal surface. The refractive index increasegives rise to an increase in the propagation constant ofSPW propagating along the metal surface (Fig. 2) whichcan be accurately measured by optical means.

The magnitude of the change in the propagation con-stant of an SPW depends on the refractive index changeand its distribution with respect to the profile of the SPWfield. There are two limiting cases:

1. analyte capture occurs only within a short distancefrom the metal surface (Fig.3a), and

2. analyte capture occurs within the whole extent of the

SPW field (Fig. 3b).

Perturbation theory[14] suggests that if the binding oc-curs within the whole depth of the SPW field (Fig.3b), thebinding-induced refractive index change, n, produces achange in the real part of the propagation constant, , whichis directly proportional to the refractive index change:

(2)

where k denotes the free-space wavenumber [15]. If therefractive index change is caused by a binding event oc-curring within a distance from the surface d, much smallerthan the penetration depth of the SPW, the corresponding

change in the real part of the propagation constant can beexpressed as follows:

(3)

where nfand ns denote the refractive index of the biolayerand the refractive index of the background dielectric me-dium (sample), respectively. The binding-induced changein the propagation constant of the SPW is proportional tothe refractive index change and the depth of the areawithin which the change occurs. The factorF (F


3/12

530

ponent of the lights wave vector that is parallel to the

metal surface matches that of the SPW. This can be achievedby means of prism coupling, waveguide coupling, and grat-ing coupling.

In configurations based on prism coupling a light wavepasses through a high refractive index prism and is totallyreflected at the prismmetal layer interface generating anevanescent wave penetrating the metal layer (Fig. 4a). Thisevanescent wave propagates along the interface with a prop-agation constant which can be adjusted to match that ofthe SPW by controlling the angle of incidence. This methodis referred to as the attenuated total reflection (ATR) method[13]. The process of exciting an SPW in an optical wave-guide-based SPR structure (Fig. 4b) is similar to that in

the ATR coupler. The light wave is guided by an opticalwaveguide and, when entering the region with a thinmetal layer, it evanescently penetrates through the metallayer exciting an SPW at its outer boundary. Alternatively,an SPW can be excited by diffraction on a grating, Fig.4c.The component of the wave vector of the diffracted wavesparallel to the interface is diffraction-increased by an amountwhich is inversely proportional to the period of the gratingand can be matched to that of an SPW[16]. The interac-tion of a light wave with an SPW can alter lights charac-teristics such as amplitude, phase, polarization and spec-tral distribution. Changes in these characteristics can becorrelated with changes in the propagation constant of the

SPW. Therefore, binding-induced changes in the refrac-tive index at the sensor surface and, consequently, thepropagation constant of the SPW can be determined bymeasuring changes in one of these characteristics. Basedon which characteristic is measured, SPR biosensors canbe classified as angle, wavelength, intensity, phase, or po-larization modulation-based sensors. In SPR sensors withangular modulation the component of the light waveswavevector parallel to the metal surface matching that ofthe SPW is determined by measuring the coupling strengthat a fixed wavelength and multiple angles of incidence ofthe light wave and determining the angle of incidence

Fig. 4 Excitation of surfaceplasmon-polaritons: (a) by alight beam via prism coupling,(b) by a guided mode of opti-cal waveguide, and (c) by lightdiffraction on a diffractiongrating

Fig.5 Reflectivity and phase for light wave exciting an SPW inthe Kretschmann geometry (SF14 glass prism 50nm thick goldlayer dielectric) versus (a) the angle of incidence for two differ-ent refractive indices of the dielectric (wavelength 682 nm), and(b) wavelength for two different refractive indices of the dielectric(angle of incidence 54)


4/12

yielding the strongest coupling (Fig.5a, upper plot). InSPR sensors with wavelength modulation the componentof the light waves wavevector parallel to the metal surfacematching that of the SPW is determined by measuring thecoupling strength at a fixed angle of incidence and multiplewavelengths and determining the wavelength yielding thestrongest coupling (Fig. 5b). In SPR sensors with intensitymodulation, the change in the intensity of the light wave in-

teracting with the SPW is measured at a fixed wavelengthand angle of incidence (Fig.5b). In SPR sensors with phasemodulation, shift in phase of the light wave interactingwith the SPW is measured at a fixed wavelength and angleof incidence (Fig.5a, lower plot). In SPR sensors with po-larization modulation, changes in the polarization are mea-sured at a fixed wavelength and angle of incidence.

Performance characteristics

The main performance characteristics relevant for SPRbiosensors include sensitivity, accuracy, precision, repeata-

bility, and the lowest detection limit. Sensorsensitivity S,is the ratio of the change in sensor output,P (e.g. angle ofincidence, wavelength, intensity, phase, and polarizationof light wave interacting with an SPW) to the change inmeasurand (e.g. analyte concentration, c). SPR biosensorsensitivity can be decomposed into two components sensitivity to refractive index changes produced by thebinding of analyte to biomolecular recognition elementson the sensor surface SRI, and the efficiencyE, with whichthe presence of analyte at a concentration c is convertedinto the change in the refractive index n:

(4)

The efficiencyE depends on the properties of the biomo-lecular recognition elements and the target analyte. Therefractive index sensitivity SRI can be decomposed intotwo terms:

(5)

The first term S1 depends on the modulation method andthe method of excitation of an SPW[17, 18, 19, 20]. TheS2 term is independent of the modulation method and themethod of excitation of the SPW and describes the sensi-tivity of SPWs propagation constant to the refractive in-dex change, Eqs. (2) and (3).

Accuracy describes the degree to which a sensor outputrepresents the true value of the measurand (analyte con-centration). Accuracy is often confused with precision whichrefers to the way in which repeated measurements con-form to themselves without a reference to any true value.Repeatability refers to the capacity of a sensor to repro-duce output reading under the same measurement condi-tions over a short interval of time. The lowest detectionlimit describes the lowest concentration of analyte thatcan be measured by the sensor.

SPR biosensing formats

An interaction between a biomolecular recognition ele-ment on an SPR sensor surface and analyte in a liquidsample is governed by kinetic equations. In order to illus-trate fundamental properties of the interaction, we shalldiscuss the pseudo-first-order kinetic equation:

(6)

whereR is the relative amount of bound analyte, c is ana-lyte concentration, t is time, and ka and kd are the associa-tion and dissociation kinetic rate constants, respectively[21]. This interaction model assumes 1:1 binding, rapidmixing of the analyte from the bulk phase to the sensorsurface layer, and single-step binding. Observed binding,however, may deviate from this simple model due to morecomplex mechanisms of interaction and mass transportlimitations [22]. Equation (6) yields forR:

(7)

where R0 denotes the initial amount of analyte bound atthe time t=0 [21].

( ) ( )( )D GD

D G

N F N W N F

5 W 5 H 5N F N

+ = + +

( )D GG

G

5

N F 5 N 5W=

{ }

{ }5 ,

5H

5H

36 6 6

Q

= =

5 ,

3 Q6 6 (

Q F

= =

531

Fig.6 Direct detection

Fig.7 Direct detection. Binding between antibody and analyte cal-culated for four different concentrations of analyte, ka=310

5 mol1

L s1; kd=5104 s1


5/12

532

Various measurement formats have been adopted inSPR biosensing to ensure that the monitored binding eventproduces a measurable sensor response. The most fre-quently used measurement formats are direct detection,sandwich assay, and inhibition assay. In direct detectionformat, analyte in a sample interacts with a biomolecularrecognition element (antibody) immobilized on the sensorsurface, Fig.6. The resulting refractive index change is di-

rectly proportional to the concentration of analyte.Figure 7 which shows a kinetic model of the interac-tion between antibody and analyte suggests that the bind-ing between the target analyte and antibody is fast ini-tially. As the interaction progresses, the binding rate grad-ually decreases and eventually reaches a state in which theassociation and dissociation processes are in equilibrium.The time required for the interaction to reach the equilib-rium depends on the concentration of analyte and is longerfor lower concentrations of analyte.

Figure 8a shows dependence of the relative binding atequilibrium on the concentrations of analyte. At low ana-

lyte concentrations the equilibrium binding increases lin-early with analyte concentration. At higher analyte con-centrations, the binding sites provided by the biomolecu-lar recognition elements are saturated and a further in-crease in the analyte concentration produces a smaller in-crease in the amount of bound analyte. The initial bindingrate dR/dt (t=0) is directly proportional to the associationrate constant and analyte concentration (Fig.8b). Both the

amount of analyte bound at equilibrium and initial bind-ing rate can be used to determine analyte concentration.The measurement of the binding rate is faster and offers alarger dynamic range than measurement of the equilib-rium binding. In the sandwich assay format the measure-ment consists of two steps. In the first step, sample con-taining analyte is brought in contact with the sensor andthe analyte molecules bind to the antibodies on the sensorsurface. Then the sensor surface is incubated with a solu-tion containing secondary antibodies. The secondaryantibodies bind to the previously captured analyte furtherincreasing the number of bound biomolecules (Fig.9) andthus also the sensor response.

The inhibition assay is an example of a competitive as-say. In this detection format, a sample is initially mixedwith respective antibodies and then the mixture is broughtin contact with the sensor surface coated with analyte mol-ecules, so that the unoccupied antibodies could bind to theanalyte molecules (Fig.10).

The amount of bound analyte versus time may be esti-mated by calculating the equilibrium concentration of an-tibody which did not bind to the analyte in the sample andthen simulating the interaction between the unbound anti-body and the analyte-derivatized surface. Figure11 showsthe equilibrium binding and the initial binding rate as afunction of analyte concentration, assuming that antibodyat a concentration of 0.1nmolL1 was incubated with sam-ple and the mixture was provided with enough time toreach equilibrium. The amount of bound antibody and the

Fig. 8 Direct detection: (top) relative equilibrium binding as afunction of analyte concentration; (bottom) initial binding rate asa function of analyte concentration, ka=310

5 mol1 L s1; kd=5104 s1

Fig.9 Sandwich assay

Fig. 10 Inhibition assay


6/12

initial binding rate are inversely proportional to analyteconcentration, Fig. 11.

Features and challenges

SPR biosensor technology exhibits various advantageous

features. These include in particular:

1. Versatility generic SPR sensor platforms can be tai-lored for detection of any analyte, providing a biomo-lecular recognition element recognizing the analyte isavailable; analyte does not have to exhibit any specialproperties such as fluorescence or characteristic absorp-tion and scattering bands.

2. No labels required binding between the biomolecularrecognition element and analyte can be observed di-rectly without the use of radioactive or fluorescent la-bels.

3. Speed of analysis the binding event can be observedin real-time providing potentially rapid response.

4. Flexibility SPR sensors can perform continuous mon-itoring as well as one-time analyses.

SPR biosensors exhibit two inherent limitations:

1. Specificity of detection specificity is solely based onthe ability of biomolecular recognition elements to rec-

ognize and capture analyte. Biomolecular recognitionelements may exhibit cross-sensitivity to structurallysimilar but non-target molecules. If the non-target mol-ecules are present in a sample in a high concentration,sensor response due to the non-target analyte moleculesmay conceal specific response produced by low levelsof target analyte.

2. Sensitivity to interfering effects similar to other affin-ity biosensors relying on measurement of refractive in-dex changes, SPR biosensor measurements can becompromised by interfering effects which produce re-fractive index variations. These include non-specificinteraction between the sensor surface and sample (ad-

sorption of non-target molecules by the sensor surface),and background refractive index variations (due to sam-ple temperature and composition fluctuations).

Advances in SPR biosensor technology

This paper follows up an SPR sensor technology reviewpaper published in Sensors and Actuators B in 1999 [9],and therefore focuses primarily on recent advances in SPRbiosensor technology. In this section the following areasare discussed in particular: SPR sensor platforms, dataanalysis, and biomolecular recognition elements.

Instrumentation

In recent years, research into optical platforms for SPRbiosensors has been increasingly application-oriented, tar-geting specific application areas and providing solutionsmeeting unique requirements of specific applications. Twoimportant representatives of this trend are: developmentof laboratory SPR sensor platforms with a large numberof sensing channels for high-throughput screening appli-cations and development of mobile SPR sensor platformsfor analysis of complex samples in the field.

In traditional multichannel SPR sensors, SPWs wereexcited via a prism coupler in multiple areas which werearranged perpendicularly to the direction of propagationof SPWs; angular [23] or spectral [24] distribution of re-flected light was analyzed to yield information about themeasurand in each channel. While this spectroscopic ap-proach led to development of high-performance SPR sens-ing devices [23], the number of sensing channels whichcould be realized using this approach was rather limited(


7/12

in which a collimated light beam from a polychromaticlight source passes through a prism and is made incidenton an SPR-active metal layer; the reflected light is de-tected with a CCD camera after passing through a narrow-band interference filter [25]. This approach has beendemonstrated for monitoring of adsorption of a single-stranded DNA-binding protein on to single-stranded DNApatterned into an array of 500500-micron squares [29].

The choice of operating wavelength for imaging SPR sen-sors have been discussed by Johansen et al. [30]. An alter-native approach[26] is based on detection of spatialchanges in the phase of light exciting an SPW and inter-ferometry. Two interferometric schemes have been pro-posed. In the MachZehnder interferometer-based schememonochromatic light was split into reference and signalbeams; the signal beam passed through a prism and, afterreflection from an SPR-active metal layer, was recom-bined with the reference beam producing an interferencepattern on a CCD camera[28]. In the TETM polarizationinterferometer TE and TM polarized beams passed througha prism and, after reflection from an SPR-active surface,

were shifted with respect to each other and recombined bymeans of a polarizer producing an interference pattern ona CCD camera[26]. Another interesting approach is basedon SPR microscopy[31] and uses surface scanning andSPWs excited by means of an objective lens[27]. Most re-cently a new approach has been proposed which is basedon spectroscopy of SPW on an array of diffraction grat-ings [32].

A great deal of research has been focused on develop-ment of mobile SPR sensor platforms with referencing ca-pabilities enabling applications of SPR biosensors in out-of-laboratory environments and for analysis of complexsamples. The traditional approach consisted in simultane-

ous SPR measurements in two sensing channels containingdifferent biomolecular recognition element coatings, onewith (signal channel) and one without (reference channel)affinity for the analyte, and subtraction of the referencechannel response from that of the signal channel [24]. Re-cently, a new approach to multichannel SPR sensing hasbeen developed which is based on excitation of surfaceplasmons in different sensing channels at different wave-lengths and encoding information from different sensingchannels into different regions of the light spectrum. Thiscan be accomplished in a sensing element of special de-sign in which light is made incident on the SPR-activemetal at different angles of incidence (Fig.12a) [33], or byemploying a thin dielectric overlayer which shifts the res-onant wavelength for a part of the sensing surface tolonger wavelengths (Fig.12b) [34]. This approach has beendemonstrated to have capacity to discriminate effects oc-curring in the proximity of the sensor surface (specificbinding, non-specific adsorption) from those occurring inthe whole medium (interfering background refractive in-dex fluctuations) which is a prerequisite for advanced ref-erencing [35]. Referencing approaches have been studied[36, 37]. It was found that a residual error for compensa-tion of interfering background refractive index variationsis typically 13% of the total refractive index change;

compensation for temperature variations is less accuratewith an error of 510% of the total response due to thetemperature change [37].

There is continuing effort to develop miniature SPRplatforms based on miniaturized prism couplers [38] andoptical fibers. The first fiber optic SPR sensors were basedon wavelength-modulation in multimode optical fibers withpartly removed cladding and a metal film deposited sym-metrically around the exposed section of fiber core [39] oron intensity-modulation in single-mode optical fibers whichwere side-polished and coated with a thin metal film[40].These SPR sensors suffered from modal noise (multimodefiber-based sensors) and polarization instability (single-

mode fiber-based sensors). Two methods of overcomingthe limitation originating from the need to precisely con-trol polarization of light in the single-mode optical fiber-based SPR sensors have been developed. In the first method,light from a polychromatic light source passes through aLyott fiber optic depolarizer which produces unpolarizedlight which is then coupled into a fiber optic SPR sensingelement. The transmitted light is analyzed with a spectro-graph [41]. The second method uses a polarization-main-taining fiber to control polarization of light in the fiber-optic SPR sensing element, Fig. 13. This approach pro-vides best suppression of polarization effects due to fiber

534

Fig. 12 Dual-surface-plasmon spectroscopy. (a) Geometry withtwo different angles of incidence [33]. (b) Geometry with a highrefractive index overlayer [34]


8/12

deformations resulting in enhanced-stability fiber optic SPRsensors with performance comparable to bulk-optic SPRsensors [42].

Research is also carried out to improve detection capa-bilities of SPR biosensors by exploiting a special type ofan SPW, the so-called long-range surface plasmon. Theuse of long-range surface plasmons provides two benefits increased sensitivity and very narrow angular or spectraldips, which makes it possible to determine the spectral po-sition of the SPR dip with a high accuracy[43]. As long-range surface plasma waves penetrate deeper into the

probed medium, their use benefits especially biosensorswith extended matrices of biomolecular recognition ele-ments [44].

Sensor data analysis

Detection limits of SPR biosensors are ultimately con-strained by the noise-based precision of the SPR instru-ment itself. The precision depends on noise contributionsmade by individual components of an SPR sensor anddata processing method. Therefore, study of noise in SPRsensors and development of optimized algorithms for pro-cessing data from SPR sensors have received much atten-tion lately. Earlier approaches to understanding the per-formance of SPR sensor data analysis have included stud-ies of the effects of noise and comparisons between algo-rithms. Locally-weighted parametric regression and othermethods were compared for low- and high-noise detec-tors, demonstrating that a small number of low-noise de-tector pixels may outperform a larger but noisier detectorarray [45]. Linearization of data processing algorithmsand an optimal linear data analysis method were proposedas a means of optimizing algorithm parameters [46]. Con-tribution of analog-to-digital converter resolution and num-berof pixels in the detector array to the performance of anumber of SPR data-analysis algorithms was studied [47].Also, effects of sensitivity deviations on concentrationanalyses and kinetic studies have been investigated[48].Most recently, sources of noise were investigated for awavelength-modulated SPR sensor. Shot noise of the de-tector was found to be the dominant source of noise andan analytical formula was derived which allows predic-tion of the noise of the sensor output based on the detec-tor noise [49].

Biomolecular recognition elementsand their immobilization

The main types of biomolecular recognition elements usedin SPR biosensors include antibodies, nucleic acids andbiomimetic materials. Antibodies are used most frequentlybecause of their high affinity, versatility, and commercialavailability. Various immobilization chemistries have beendeveloped to attach antibodies to SPR-active gold films.Traditional approaches include formation of streptavidinlayer on the gold surface followed by attachment of bi-

otinylated antibodies [50], use of self-assembled alka-nethiol films with suitable reactive groups [51], use of ahydrogel matrix composed of carboxyl-methylated dex-tran chains which can be modified allowing antibodies tobe attached via surface-exposed amine, carboxyl, sulfhydryl,and aldehyde groups [52]. Alternatively, SPR sensing sur-faces may be functionalized by thin polymer films towhich antibodies are coupled via amino groups[53]. Pro-tein contact printing has been examined for spatially-con-trolled attachment of bovine serum albumin (BSA) anddinitrophenylated BSA onto adjacent reference and signalchannels of a dual-channel SPR sensor [54]. DNA can beimmobilized on gold SPR sensor surfaces by formation ofa streptavidin layer on a gold surface followed by attach-ment of biotinylated DNA[55]. A multistep surface mod-ification based on alkanethiol self-assembled monolayershas also been used to attach DNA to gold surfaces[56].Recently, biomimetic materials consisting in molecularlyimprinted polymers (MIPs) have been exploited in SPRbiosensors[57].

Applications of SPR biosensors

Two major application areas for SPR biosensing are in de-tection and identification of biological analytes and bio-physical analysis of biomolecular interactions. This re-view focuses on applications for detection and identifica-tion of biological analytes; recent advances in SPR-basedbiomolecular interaction analysis can be found in Ref.[58].

Numerous SPR biosensors have been developed fordetection and identification of specific analytes. Thesebiosensors use a number of platform designs, biomolecu-lar recognition elements and detection formats. The choiceof detection format for a particular application depends onsize of target analyte molecules, binding characteristics ofbiomolecular recognition element, and the range of ana-

535

Fig. 13 SPR sensor based on apolarization-maintaining sin-gle-mode optical fiber


9/12

lyte concentrations to be measured. Direct detection isusually preferred in applications where direct binding ofanalyte of concentrations of interest produces a sufficientresponse. If necessary, the lowest detection limits of thedirect SPR biosensors can be improved by using sandwichassay. The secondary antibodies may also be coupled tolarge particles such as latex particles [59] and gold beads[60] to further enhance the SPR sensor response. Smaller

analytes (molecular weight


10/12

enterotoxin B was performed using a commercial angularsensor (Biacore 3000)[73] and laboratory wavelengthmodulation-based SPR sensors [74]. The lowest detectionlimits were 110 ng mL1 (in milk and meat) [73], 5 ng mL1

(direct detection in buffer), and 0.5ng mL1 (sandwich as-say in buffer and milk) [74]. A typical sensorgram for de-tection of Staphylococcal enterotoxins B (SEB) is shownin Fig.16 [74]. Figure17a shows the equilibrium sensor

response for both the direct capture of SEB and amplifi-

cation by secondary antibodies as a function of SEB con-

centration. The use of secondary antibodies provides 10-foldincrease in the sensor response. In Fig.17b, initial bindingrates for SEB and secondary antibody concentration areshown for several SEB concentrations[74].

Detection of large analytes

Representatives of large analytes targeted by SPR biosen-sor technology include bacterial pathogens such asEsche-richia coli, Salmonella enteritidis, andListeria monocyto-genes. Detection ofEscherichia coli O157:H7 was per-formed by Fratamico et al. who used an angular-modula-tion commercial SPR sensor and sandwich assay [75].They used monoclonal antibodies immobilized on thesensor surface for capturingE. coli and polyclonal second-ary antibodies for enhancing the specific sensor response.The lowest detection limit forE. coli was established at5107 cfu mL1. Direct detection ofSalmonella enteritidisand Listeria monocytogenes at concentrations down to106 cfu mL1 was demonstrated by Koubov et al. using alaboratory wavelength-modulated SPR sensor and mono-clonal antibodies [76]. A sensorgram illustrating direct de-tection ofSalmonella enteritidis is shown in Fig. 18.

The main challenges in detecting bacterial analytes atlower levels include the low concentration of a particularantigen relative to total cellular material and slow diffu-sion of bacterial cells to sensor surface[77].

Commercialization of SPR biosensors

The first commercial SPR biosensor was launched by Bia-core International AB in 1990. In the following decade,Biacore has developed a range of laboratory SPR instru-ments (Biacore 1000, Biacore 2000, Biacore 3000, Bia-core C, Biacore X, Biacore J, Biacore Q)[78]. Most re-cently the Biacore S51 has been developed which offers

537

Fig. 16 Detection of Staphylococcal enterotoxin B using sand-wich assay. Kinetic response [74]

Fig. 17 Detection of Staphylococcal enterotoxin B. (a) Equilib-rium response for different concentrations of SEB. (b) Initial bind-ing rate for different concentrations of SEB[74]

Fig. 18 Direct detection of Salmonella enteritidis. Kinetic re-sponse to Salmonella at a concentration of 106 cfu mL1


11/12

538

higher sensitivity and throughput. Other SPR sensors havebeen developed by British Windsor Scientific (IBIS) [79],Nippon Laser and Electronics Laboratory (SPR-670 andSPR-CELLIA) [80], Texas Instruments (Spreeta) [81], andAnalytical -Systems (BIO-SUPLAR 2) [82].

Outlook

Over the past ten years, surface plasmon resonance (SPR)biosensor technology has made great strides, and a largenumber of SPR sensor platforms, biomolecular recogni-tion elements, and measurement formats have been devel-oped. There has been growing interest in commercializa-tion of SPR biosensor technology leading to a number ofsystems available on todays market. SPR biosensors haveplayed a significant role in research into biomolecules andtheir interactions and have been increasingly used for de-tection and identification of chemical and biological sub-stances. SPR biosensors have been particularly successfulfor detection of small and medium size analytes, a number

of which have been detected at practically relevant levels.Detection limits for large analytes such as bacteria andviruses still need to be improved to meet todays needs.

Undoubtedly, future development of SPR biosensors willbe driven by the needs of the consumer. SPR biosensortechnology has the potential to benefit numerous impor-tant fields including pharmaceutical research, medical di-agnostics, environmental monitoring, food safety, and se-curity. Applications in these areas present unique chal-lenges and impose special requirement on analytical tech-nologies. Pharmaceutical research, which was fast to adoptoptical biosensors, needs laboratory-based high-through-put systems with high sensitivity to facilitate parallel

screening. Medicine would benefit from high-throughputdiagnostic tools for centralized laboratories and from com-pact diagnostic systems dedicated to selected diagnosticapplications which could be used at clinics. In addition,there is a growing interest in tools for home medical (self-)diagnostics. Mobile analytical systems enabling rapid de-tection of food-borne pathogens in food would be impor-tant for food producers, processors, distributors and regu-latory agencies and thus benefit the food safety. Environ-mental monitoring would benefit from detection systemswhich could be deployed in the field for continuous mon-itoring and from mobile systems enabling fast identifica-tion of environmental threat. SPR biosensors could alsoplay an important role in defense, where fast, portable andrugged units are needed for early detection and identifica-tion of biological warfare agents in the field. Develop-ment of these systems will require significant advances inminiaturization of SPR biosensing platforms, develop-ment of robust biomolecular recognition elements withhigh specificity and long storage life, integration of SPRsensor platforms with microfluidic devices, and applica-tion-specific sampling systems.

Acknowledgement This work was supported by the Grant Agencyof the Czech Republic under contracts 102/03/0633, 303/03/0249,and 203/02/1326.

References

1. Ghindilis AL, Atanasov P, Wilkins M, Wilkins E (1998)Biosens Bioelectron 13:113131

2. Chu X, Lin ZH, Shen GL, Yu RQM (1995) Analyst 120:28292832

3. Gauglitz G (1996) Opto-chemical and opto-immuno sensors,sensor update, vol 1. VCH, Weinheim

4. Rowe-Taitt CA, Hazzard JW, Hoffman KE, Cras JJ, Golden

JP, Ligler FS (2000) Biosens Bioelectron 15:5795895. Piehler J, Brecht A, Gauglitz G (1996) Anal Chem 68:1391436. Heideman RG, Kooyman RPH, Greve J (1993) Sens Actuators

B 10:2092177. Clerc D, Lukosz W (1994) Sens Actuators B 19:5815868. Cush R, Cronin JM, Stewart WJ, Maule CH, Molloy J, God-

dard NJ (1993) Biosens Bioelectron 8:3473539. Homola J, Yee S, Gauglitz G (1999) Sens Actuators B 54:315

10. Homola J, Yee S, Myszka D (2002) Surface plasmon biosen-sors. In: Ligler FS, Taitt CR (eds) Optical biosensors: presentand future. Elsevier

11. Rabbany SY, Lane WJ, Marganski WA, Kusterbeck AW,Ligler FS (2000) J Immunol Methods 246:6977

12. Boardman AD (1982) (ed) Electromagnetic surface modes.Wiley, Chichester

13. Reather H (1983) Surface plasmons on smooth and rough sur-

faces and on gratings, Springer tracts in modern physics, vol111. Springer, Berlin Heidelberg New York

14. Snyder AW, Love JD (1983) Optical waveguide theory. Chap-man and Hall, London

15. Parriaux O, Voirin G (1990) Sens Actuators A 2123:113716. Hutley MC (1982) Diffration gratings. Academic Press, Lon-

don17. Kooyman RPH, Kolkman H, van Gent J, Greve J (1988) Anal

Chim Acta 213:354518. Yeatman EM (1996) Biosens Bioelectron 11:63564919. Homola J (1997) Sens Actuators B 41:20721120. Homola J, Koudela I, Yee S (1999) Sens Actuators B 54:162421. Edwards PR, Leatherbarrow RJ (1997) Anal Biochem 246:1622. Vijayendran RA, Ligler FS, Leckband DE (1999) Anal Chem

71:5405541223. Lfs S, Malmqvist M, Rnnberg I, Stenberg E, Liedberg B,

Lundstrm I (1991) Sens Actuators B 5:7924. Nenninger GG, Clendenning JB, Furlong CE, Yee S (1998)Sens Actuators B51:38

25. Jordan CE, Frutos AG, Thiel AJ, Corn RM (1997) Anal Chem69:4939

26. Nikitin PI, Beloglazov AA, Kabashin AV, Valeiko MV,Kochergin V E (1999) Sens Actuators B54:43

27. Kano H, Knoll W (2000) Optics Communications 182:111528. Kabashin A V, Nikitin P I (1998) Opt Commun 150:5829. Nelson BP, Frutos AG, Brockman JM, Corn RM (1999) Anal

Chem 71:3928393430. Johnasen K, Arwin H, Lundstrm I, Liedberg B (2000) Rev Sci

Instrum 71:3530353831. Rothenhusler B, Knoll W (1988) Nature 332:68832. Dostlek J, Homola J, Miler M (2002) EUROPT(R)ODE VI,

Manchester, UK, Book of Abstracts, 265

33. Homola J, Dostlek J, tyrok J (2001) Proc SPIE 4416:8634. Homola J, Lu HB, Yee S (1999) Electr Lett 35:110535. Homola J, Lu HB, Nenninger GG, Dostlek J, Yee S (2001)

Sens Actuators B76:40341036. Ober RJ, Ward E S (1999) Anal Biochem 271:708037. Homola J, Dostlek J, Piliarik M, Yee S (2002) EUROP-

T(R)ODE VI, Manchester, UK, Book of Abstracts, 7138. Stemmler I, Brecht A, Gauglitz G (1999) Sens Actuators B

54:9810539. Jorgenson RC, Yee S (1993) Sens Actuators B12:21340. Homola J (1995) Sens Actuators B 29:40140541. Homola J, Piliarik M, Slavk R, tyrok J (2001) SPIE Proc

4416:828542. Piliarik M, Homola J, Mankov Z, tyrok J (2003) Sens Ac-

tuators B 90:236242


12/12

539

43. Nenninger GG, Tobika P, Homola J, Yee S (2001) Sens Actu-ators B74:145

44. Liedberg B, Lundstrom I, Stenberg E (1993) Sens ActuatorsB11:63

45. Johnston KS, Booksh KS, Chinowsky TM, Yee S (1999) SensActuators B 54:8088

46. Chinowsky TM, Jung LS, Yee S (1999) Sens Actuators B 54:8997

47. Johansen K, Stlberg R, Lunstrm I, Liedberg B (2000) MeasSci Technol 11:16301638

48. Johansen K, Lundstrm I, Liedberg B (2000) Biosens Bioelec-tron 15:503509

49. Nenninger GG, Piliarik M, Homola J (2002) Meas Sci Technol13:20382046

50. Morgan H, Taylor DM, (1992) Biosens Bioelectron 7:40541051. Duschl C, Sevin-Landais A, Vogel H (1995) Biophys J 70:

198552. Lfs S, Malmqvist M, Rnnberg I, Stenberg E, Liedberg B,

Lundstrm I (1991) Sens Actuators B 5:7953. Nakamura R, Muguruma H, Ikebukuro K, Sasaki S, Nagata R,

Karube I, Pedersen H (1997) Anal Chem 69:464954. Lu HB, Homola J, Campbell CT, Nenninger GG, Yee S, Rat-

ner BD (2001) Sens Actuators B 74:919955. Watts HJ, Yeung D, Parkes H (1995) Anal Chem 67:4283

428956. Brockman JM, Frutos AG, Corn RM (1999) J Am Chem Soc

121:8044805157. Lai EPC, Fafara A, VanderNoot VA, Kono M, Polsky B (1998)Can J Chem 76:265273

58. Rich RL, Myszka DG (2002) J Mol Recognit 15:35237659. Severs AH, Schasfoort RBM (1993) Biosens Bioelectron 8:36560. Leung PT, Pollard-Knight D, Malan GP, Finlan MF (1994)

Sens Actuators B22:17561. Minunni M, Mascini M (1993) Anal Lett 26:144162. Mouvet C, Harris RD, Maciag C, Luff BJ, Wilkinson JS,

Piehler J, Brecht A, Gauglitz G, Abuknesha R, Ismail G (1997)Anal Chim Acta 338:109

63. Harris RD, Luff BJ, Wilkinson JS, Piehler J, Brecht A,Gauglitz G, Abuknesha RA (1999) Biosens Bioelectron 14:377

64. Miura N, Ogata K, Sakai G, Uda T, Yamazoe N (1997) ChemLett 8:713

65. Sakai G, Nakata S, Uda T, Miura N, Yamazoe N (1999) Electro-chim Acta 44:3849

66. Mullett W, Edward PC, Yeung MJ (1998) Anal Biochem 258:161167

67. Sternesjo A, Mellgren C, Bjorck L (1996) ACS Symp Ser621:463470

68. Gaudin V, Pavy M-L (1999) JAOAC Int 82:1316132069. Baxter GA, Ferguson JP, OConner MC, Elliott CT (2001)J Agric Food Chem 49:32043207

70. Elliot CT, Baxter GA, Crooks SRH, McCaughey WJ (1999)Food Agric Immunol 11:1928

71. Choi K, Seo W, Cha S, Choi J (1998) J Biochem Mol Biol31:101105

72. Spangler BD, Wilkinson EA, Murphy JT, Tyler BJ (2001) AnalChim Acta 444:149161

73. Rasooly A (2001) J Food Prot 64:374374. Homola J, Dostlek J, Chen S, Rasooly A, Jiang S, Yee S

(2002) Int J Food Microbiol 75:616975. Fratamico PM, Strobaugh TP, Medina MB, Gehring AG (1998)

Biotechnol Tech 12:57157676. Koubov V, Brynda E, Karasov L, kvor J, Homola J, Dos-

tlek J, Tobika P, Roick J (2001) Sens Actuators B 74:

10010577. Perkins EA, Squirrell DJ (2000) Biosens Bioelectron 14:85378. Biacore websitewww.biacore.com79. IBIS Technologies website www.ibis-spr.nl80. Nippon Laser and Electronics Laboratory website www.nle-lab.

co.jp/English/ZO-HOME.htm81. Texas Instruments website www.ti.com/sc/docs/products/msp/

control/spreeta82. Analytical -Systems website www.micro-systems.de

present and future of surface plasmon resonance

Documents