Tandem conjugation of enzyme and antibody on silica nanoparticle for enzyme immunoassay
Post on 26-Jun-2016
Tandem conjugation of enzyme and antibofor enzyme immunoassay
much higher signal and increased sensitivity. When used in an ELISA detection of the hepatitis B surface
ssay (s, pathsy maer, onrelativnces. T
on their surface. Thus,when they are used for preparation of enzymeand antibody conjugates, the resulting conjugates could vary signif-icantly in the enzyme-to-antibody ratios, resulting in poorreproducibility.
Itwould be advantageous to develop a conjugation approach thatcan provide controlled enzyme-to-antibody ratios. In the currentstudy,wedevelopeda layer-by-layer labelingstrategybyusing silicananoparticles as the conjugation scaffold. We rst immobilized theenzyme horseradish peroxidase (HRP) to the surface of preformedaminated silica nanoparticles. We then modied the HRP-labeled
Fujian 361005, China. Fax: +86 592 2187363.E-mail address: email@example.com (Q. Li).
1 These authors contributed equally to this work.2 Present address: Department of Genetics and Pathology, Rudbeck Laboratory,
Uppsala University, SE-751 85 Uppsala, Sweden.3 Abbreviations used: ELISA, enzyme-linked immunosorbent assay; HRP, horserad-
ish peroxidase; HBsAg, hepatitis B surface antigen; NaBH3CN, sodium cyanoborohy-dride; TEOS, tetraethyl orthosilicate; APTMS, 3-aminopropyltrimethoxysilane;AEAPTMS, 3-(2-aminoethylamino)propyltrimethoxysilane; AEAEAPTMS, 3-[2-(2-aminoethylamino)ethylamino]propyltrimethoxysilane; BSA, bovine serum albumin;IgG, immunoglobulin G; TMB, 3,30 ,5,50-tetramethylbenzidine; H2O2, hydrogen per-oxide; TEM, transmission electron microscope; UVVis, ultravioletvisible; S/N ratio,
Analytical Biochemistry 406 (2010) 813
Contents lists availab
journal homepage: www.esignal-to-noise ratio; CV, coefcient of variance.sitivity, uorogenic or chemiluminescent substrates are used insteadof chromogenic substrates . Unfortunately, the readout for uo-rescence or chemiluminescence requires expensive instrumentationthat increases the overall cost of the assay. An alternative way to in-crease the sensitivity was achieved by amplication of the signalsgenerated from the enzymeantibody conjugate with a highenzyme-to-antibody ratio . Such conjugates were assembled
Thepolymers, suchasdextranandpolylysinechains, usually con-tainmultiple functional groups that couldbe covalently linked toen-zyme and antibody molecules to form clustered enzymeantibodyconjugates . It was reported that by using polymer-basedconjugates, sensitivity could be improved signicantly. Micro-spheres, such as liposome [14,15], polystyrene microparticles[13,16,17], and silica nanoparticles , have been used to pre-pare uorescent labels for immunoassays. Because they can be ex-ibly introduced with active functional groups onto their surface,thesemicrospheres are potential scaffolds for enzyme and antibodyconjugation [13,22,23]. The disadvantage of these microspheres,however, is that they often have the same type of functional groups
* Corresponding author at:. Molecular Diagnostics Laboratory, Department ofBiomedical Sciences and Key Laboratory of Ministry of Education for Cell Biologyand Tumor Cell Engineering, School of Life Sciences, Xiamen University, Xiamen,Silica nanoparticlesTandem conjugationHepatitis B surface antigen
Enzyme-linked immunosorbent atial role in the detection of biomarkerals due to its cost effectiveness, eareadout interpretation . Howevtional absorption-based ELISA is itslimits its use to only abundant substa0003-2697/$ - see front matter 2010 Elsevier Inc. Adoi:10.1016/j.ab.2010.06.039antigen (HBsAg), thedetection limitwas three times lower than that of the commercially available ELISAkit. 2010 Elsevier Inc. All rights reserved.
ELISA)3 plays an essen-ogens, and drug residu-nipulation, and simplee drawback of conven-ely low sensitivity thato achieve improved sen-
through special scaffolds, such as polymer and microsphere, to forma cluster of enzyme and antibody molecules. Therefore, when oneantibody molecule of this type of conjugate binds to one antigen,tens or hundreds of enzyme molecules can bind to a single antigen.Subsequently, all of the enzymes will contribute to the substratecatalysis reaction, consequently leading to signal amplication.Keywords:the conventional antibodyenzyme conjugate used in ELISA,which often has one or two enzymemoleculesper antibody, the new type of conjugate contained more enzyme molecules per antibody and provided aRongqin Ke a,1,2, Wei Yang a,1, Xiaohu Xia a, Ye Xu a, QaMolecular Diagnostics Laboratory, Department of Biomedical Sciences and Key LaboratSchool of Life Sciences, Xiamen University, Xiamen, Fujian 361005, ChinabKey Laboratory of Chemical Biology of Fujian, Xiamen, Fujian 361005, China
a r t i c l e i n f o
Article history:Received 21 February 2010Received in revised form 17 June 2010Accepted 28 June 2010Available online 1 July 2010
a b s t r a c t
Wepresent a new type of ethe sensitivity of enzyme-layer-by-layer immobilizatof enzyme was immobilizeconjugate could be easily pll rights reserved.dy on silica nanoparticle
gge Li a,b,*
of Ministry of Education for Cell Biology and Tumor Cell Engineering,
meantibody conjugate that simplies the labeling procedure and increasesed immunosorbent assay (ELISA). The conjugates were prepared throughof enzyme and antibody on a silica nanoparticle scaffold. Amaximal amountthe nanoparticle, followed by antibody linkage through Dextran 500. Theed from unreacted reagents by simple centrifugations. In comparison with
le at ScienceDirect
lsevier .com/locate /yabio
Preparation of aminated silica nanoparticles
y / RSilicananoparticleswere synthesizedusinga reversemicroemul-sionmethod [24,25]. Briey, 10 mlof cyclohexane, TritonX-100, andn-hexanol in a ratio of 3:1:1 (v/v) was stirred at room temperature,followed by the addition of 450 ll of water, 100 ll of TEOS, and70 ll of NH4OH (2830%), respectively. After reaction at room tem-perature for 24 h, the silica nanoparticles formed were isolated bycentrifugation (5900 RCF for 20 min) after adding 10 ml of acetone.The nanoparticles were rinsed with ethanol and water three times,silica nanoparticleswith ahydrophilic layer of oxidizeddextran, andthe antibody was conjugated to the oxidized dextran through Schiffreaction. The excess amount of reagents could be easily removed bybrief centrifugation. When used in an ELISA for the detection of thehepatitisB surfaceantigen (HBsAg), our conjugatesobtainedadetec-tion limit threefold lower than that of the commercial ELISA kit,whereas our results from 280 clinical serum samples agreed com-pletely with ELISA. In this article, we term our new type of en-zymeantibody conjugate as bifunctional silica nanoparticle.
Materials and methods
Triton X-100, sodium cyanoborohydride (NaBH3CN), tetraethylorthosilicate (TEOS), 3-aminopropyltrimethoxysilane (APTMS),3-(2-aminoethylamino)propyltrimethoxysilane (AEAPTMS), and 3-[2-(2-aminoethylamino)ethylamino]propyltrimethoxysilane (AEAEAPTMS), bovine serum albumin (BSA), and ProClinwere purchasedfrom SigmaAldrich (St. Louis, MO, USA). Dextran 500 (MW =500,000) was purchased from Amersham Biosciences (Uppsala,Sweden). HRPwas purchased fromBienzyme Laboratories (San Die-go, CA, USA). Monoclonal anti-HBsAg immunoglobulin G (IgG) anti-body (S04, B20), ELISAkits forHBsAg, colorimetric substrate forHRP,3,30,5,50-tetramethylbenzidine (TMB), hydrogen peroxide (H2O2),and human serum samples all were kindly provided by InTec Prod-ucts (Xiamen, China). All of the reagents were used without furtherpurication. Distilled water was used throughout the study.
Dilution buffer consisted of 0.01 M TrisHCl containing 2 g/LBSA, 0.1% ProClin, and 9 g/L NaCl (pH 7.8). Washing buffer con-sisted of 0.01 M phosphate-buffered saline containing 0.5 g/LTween 20 (pH 7.4). Blocking buffer consisted of 0.01 M TrisHClcontaining 2% BSA, 4% sucrose, and 1% glycine (pH 7.8). Carbonatebuffer 1 consisted of 0.05 M Na2CO3NaHCO3 (pH 9.5), and carbon-ate buffer 2 consisted of 0.025 M Na2CO3NaHCO3 (pH 9.5).
The colorimetric ELISAwasmeasuredbyamicrotiter plate reader(Thermo Labsystems Multiskan MK3 plate reader, Marietta, OH,USA). Transmission electron microscope (TEM) images of silicananoparticles were obtained from a JEOL JEM-2100 TEM (Tokyo, Ja-pan). Ultravioletvisible (UVVis) spectrawere recorded on a PGen-eral TU-1901 UVVis spectrophotometer (Beijing Purkinje GeneralInstrument,Beijing,China). All of thenanoparticleswere centrifugedon an Eppendorf 5415 D centrifuge (Hamburg, Germany) and soni-cated with a Scientz SB5200D ultrasonic cleaner (Ningbo, China).Microtitration plates were kindly provided by InTec Products.
Tandem conjugation of enzyme and antibodrespectively, to remove surfactant and unpolymerized materials.Silica nanoparticles (20 mg) were suspended in 1 ml of ethanol.
Then 2 ll of amino silane was added to the suspension to introduceamino groups onto the surface of silica nanoparticles (Fig. 1). Reac-tion was carried out at room temperature for 2 h with stirring, fol-lowed by incubation in a 65 C water bath for 5 min. Unreactedreagents were removed by washing with ethanol and water,respectively. The nanoparticles were resuspended in 1 ml of etha-nol. To conrm the existence of the amino groups on the surface ofsilica nanoparticles, one drop of salicylaldehyde was added to0.5 ml of nanoparticle suspension to observe color change .
Immobilization of HRP on surface of aminated silica nanoparticles
HRP was covalently linked to the aminated silica nanoparticlesusing the periodate oxidation method, as shown in Fig. 1. Here100 ll of HRP (4 mg/ml) was dialyzed against acetate buffer (pH5.2, 0.01 M) at 4 C and then mixed with 2 ll of 0.5 M NaIO4. Afterreaction for 25 min at room temperature in the dark with gentlestirring, the excess periodate was neutralized by adding 1 ll ofglycerol with stirring for 10 min. Unreacted reagents were re-moved by dialysis against acetate buffer (pH 5.2, 0.01 M). The solu-tion of oxidized HRP was adjusted to pH 9.5 with 0.1 M Na2CO3 andthen was added to 100 ll of aminated silica nanoparticle suspen-sion in carbonate buffer 1 (20 mg/ml). Aggregation occurred imme-diately on the addition of HRP, and the aggregated nanoparticleswere dispersed by ultrasonic treatment in an icewater bath. Afterreaction at 4 C for 6 h, NaBH3CN was added to a nal concentra-tion of 5 mM, followed by 12 h of incubation at 4 C. Finally, theunreacted reagents were removed by centrifugation and the pre-cipitated silica nanoparticles were washed with carbonate buffer1. After coupling, the color of nanoparticle suspension turned tolight brown from white.
Conjugation of antibody to surface of HRP-linked silica nanoparticles
Anti-HBsAg antibody (S04) was covalently linked to the HRP-immobilized silica nanoparticles through the oxidized dextran(Fig. 1), which was prepared by oxidation of Dextran 500 accordingto our previous work . For this purpose, 100 ll (33 mg/ml inwater solution) of the oxidized Dextran 500 was added to 100 ll(20 mg/ml) ofHRP-conjugated silica nanoparticle suspension in car-bonate buffer 1 and reacted at 4 Cwith stirring for 3 h. The obtainedDextran 500-bridged, HRP-immobilized silica nanoparticles werewashed three times with carbonate buffer 2 to remove the excessoxidized Dextran 500. The anti-HBsAg antibody (100 ll, 3 mg/ml)wasdialyzedagainst carbonatebuffer2 for6 handadded to theDex-tran 500-bridged, HRP-immobilized silica nanoparticles. The mix-ture was allowed to react at 4 C for 6 h, and then NaBH3CN wasadded to a nal concentration of 0.005 M, followed by incubationat 4 C overnight.
An equal volume of blocking buffer was added to the mixtureand incubated overnight at 4 C. Then the nanoparticles were cen-trifuged and rinsed with 0.01 M TrisHCl buffer (pH 7.8) threetimes and nally suspended in the dilution buffer. The preparedimmunoconjugates were stored at 4 C before use.
Detection of HBsAg by ELISA
Bifunctional silica nanoparticles were diluted 1000-fold in thedilution buffer before use. Detection of HBsAg followed a standardELISA procedure. Briey, the monoclonal anti-HBsAg antibody B20(5 lg/ml in 0.02 M TrisHCl buffer, pH 7.4) was physically coatedonto the microtitration wells (100 ll/well). After incubation at 4 Covernight, themicrotitrationwells werewashed oncewithwashingbuffer andblockedwithblockingbuffer (200 ll/well) for12 hat4 C.
. Ke et al. / Anal. Biochem. 406 (2010) 813 9Then the blocking bufferwas thrown away, and 50 ll of HBsAg stan-dard solutionsor serumsampleswasadded toeachwell, followedbythe addition of 50 ll of diluted bifunctional silica nanoparticles or
y / R10 Tandem conjugation of enzyme and antibodHRP-labeled anti-HBsAg antibody. The plate was incubated at 37 Cfor 1 h and then washed ve times with washing buffer. SubstrateTMB + H2O2 (100 ll/well) was added, and the plate was incubatedat 37 C for 15 min. The reaction was then terminated by adding50 ll of 2 M H2SO4, and the absorbance at 450 nm with 630 nm asreference was recorded on a ThermoMK3 plate reader.
Results and discussion
Preparation of bifunctional silica nanoparticles
In the current work, silica nanoparticles were employed as thecarrier for enzyme and antibody immobilization. Thus, monodi-spersed silica nanoparticles with uniform morphology were vitalfor loading biomolecules on eachnanosphere, and this in turnwouldinuence the sensitivity, reproducibility, and analytical perfor-mance of the resulting immunoassay. Monodispersed spherical sil-ica nanoparticles prepared by a reverse microemulsion methodwere uniform (50 5 nm in diameter [mean standard deviation]),as observed with TEM (Fig. 2).
To introduce amino groups onto the surface, the nanoparticleswere modied with the amino silane under mild conditions. Theexistence of amino groups on the surface was conrmed by salicyl-aldehyde-mediated yellow color change.
Instead of direct encapsulation of HRP into the core of silicananoparticles, we immobilized HRP onto the surface of silica nano-particles. This method obviated the direct contact of HRP with or-ganic reagents that may cause a decrease or loss of the enzyme
Fig. 1. Schematic illustration of preparation procedure of the bifunctional silica nanoparwith HRP and antibody. Bare silica nanoparticles were rst aminated using amino silanperiodate oxidation method. The antibody was covalently linked to the HRP-conjugananoparticle-based ELISA. The immobilized capture antibody and the detection antibodsandwich with the antigen to be detected.. Ke et al. / Anal. Biochem. 406 (2010) 813activity. The large surface area of silica nanoparticle carriers al-lowed us to immobilize the maximal amount of HRP onto eachnanoparticle. Before antibody conjugation, we added a layer of
ticle and its use in immunoassay. (A) Preparation of silica nanoparticles conjugatedes. HRP was then covalently linked to the aminated silica nanoparticles using theted silica nanoparticles through oxidized Dextran 500. (B) Working principle ofy, which was conjugated onto the silica nanoparticle-harboring enzyme, formed a
Fig. 2. TEM image of the prepared silica nanoparticles.
Dextran 500 to the surface of the nanoparticles. One use of thislayer is to offer active groups for antibody conjugation, and anotheruse is to add hydrophilicity of the silica nanoparticle, rendering itmore easily dispersed in aqueous solution. Compared with a shortcross-linker such as glutaraldehyde, the use of Dextran 500 polyal-dehyde for enzyme conjugation resulted in minimal loss of enzymeactivity [10,11,27]. The enzyme conjugate could be kept at 4 C forat least 1 year without obvious loss of activity.
The simple puricationof ourHRPand antibody conjugates is an-other advantage of our labeling strategy. Purication of the conven-tional antibodyenzyme conjugates often demands gel ltrationchromatography, which is time-consuming and laborious and oftenresults in loss of overall yield of the conjugate. In contrast, ourbifunctional silica nanoparticles could be puried by simple low-speed centrifugation. Moreover, the centrifugation procedure couldrecoverall of theenzymeconjugates,whichcouldbeconcentrated toany desired volume for later use. Therefore, ourmethod ismore con-venient and effective than the conventional preparations.
immunoassay fordetectingHBsAgstandardsof 1 ng/ml. Theaverage
signal-to-noise (S/N) ratiowas calculated from four replicas. The re-sult turnedout tobe that the longer the spacer arm, the greater the S/N value (Fig. 3B). Although the silica nanoparticles modiedwith 7-and 10-atom spacer arms could immobilize nearly the same amountof HRP molecules, the latter tended to generate higher S/N ratios inthe enzyme immunoassay, probably due to the increased antibodymolecules immobilized, whichmight also increase the overall afn-ity of the antibody. However, we observed that although the nano-particle conjugates with the 10-atom spacer arm provided thehighest S/N ratio, they turned to aggregateduring long-termstorage.We attributed this phenomenon to the increased hydrophobicity ofthe nanoparticles with the long-chain alkanes. For this reason, thenanoparticles with the 7-atom spacer arm were used in the follow-ing experiments.
Effect of amount of HRP and antibody used for conjugation
The effect of the amount of HRP used for conjugation was also
Tandem conjugation of enzyme and antibody / R. Ke et al. / Anal. Biochem. 406 (2010) 813 11Effect of different lengths of spacer arms
Three different amino silanes were used to generate aminogroups on the surface of silica nanoparticles. These amino silanesare APTMS, AEAPTMS, and AEAEAPTMS, and they will generateamino groups that are 4, 7, and 10 atoms away from the surfaceof nanoparticles. The effect of different lengths of the spacer armson the conjugation of HRP with aminated silica nanoparticles wasinvestigated. Experimental results indicated that the amount ofunconjugated HRP that remained in the reaction mixture withthe 10-atom spacer arm was the smallest, whereas that with the4-atom spacer arm was the largest (Fig. 3A). We calculated thenumber of HRP molecules per silica particle based on the densityof SiO2 (1.96 g/cm3) , and the data are summarized in Table 1.This result showed that with an increase in the length of the spacerarm, more HRP molecules could be immobilized on the surface ofthe silica nanoparticles. The increased conjugation efciency withlonger spacer arms could be attributed to the reduced steric hin-drance of HRP in access to the active sites on the surface of thenanoparticles .
The above results were also conrmed indirectly by using thesebifunctional nanoparticles in enzyme immunoassay. These bifunc-tional silica nanoparticles were diluted and used in the enzymeFig. 3. Effects of different lengths of spacer arms. (A) Absorption spectrum of unconjugatspacer arms on the enzyme immunoassay. The bifunctional silica nanoparticles had thdilution buffer. The concentration of HBsAg standard measured here was 1 ng/ml.examined.When0.2 mgofHRPwasadded to2 mgof aminated silicananoparticles, nearly all HRP molecules were immobilized on thesurface of the particles. This result indicated that 0.2 mg of HRPwas nonsaturated for conjugation of 2 mg of nanoparticles. In con-trast, when 0.4 mg of HRP was added, excess HRP remained uncon-jugated, indicating that 0.4 mg of HRP had already been saturatedfor conjugation of such an amount of nanoparticles. A larger amountof HRP was not examined after that.
We then compared these two conjugates in their performance forimmunoassay. To each 2.0 mg of HRP-modied silica nanoparticleswith different dilution ratios, 0.3 mg of anti-HBsAg antibodies wasadded. The resulting conjugates were used to detect 1 ng/ml HBsAgstandard solution. When compared with the HRP-nonsaturatedbifunctional silica nanoparticles, at all dilution ratios, the HRP-satu-rated nanoparticles always generated higher S/N ratios in the detec-tion of 1 ng/ml HBsAg standards (Fig. 4A). The result indicated that,to generate maximal S/N ratios in silica nanoparticle-based enzymeimmunoassay, the HRP used for conjugation should be saturated.Fortunately, saturated HRP conjugation could be easily achievedby using excess HRP for immobilization, and a simple centrifugationcould eliminate all unconjugated HRP.
To study the effect of the amount of antibodies used for conjuga-tion, the HRP-saturated silica nanoparticles were used to conjugatewith a varied amount of anti-HBsAg antibody. The resulting conju-gates were evaluated under different dilution ratios by detectinged HRP in the supernatant of the reaction mixture. (B) Effects of different lengths ofe same original concentration and were diluted to different concentrations by the
1 ng/ml HBsAg standards. Interestingly, the results showed that theS/N ratio was not in direct proportion to the amount of antibodiesconjugated onto the nanoparticles, although nanoparticles harbor-ing more antibody molecules largely gave greater S/N ratios. Thenanoparticles carrying the largest number of antibodies behavedpoorly except at the lowest dilution (Fig. 4B). These data revealedthat the number of antibody molecules on the surface needs to bebalanced to achieve high afnity for best sensitivity.
Taken together, the above results demonstrated that using silica
sults for all of the samples. To study the correlationbetween conven-tional ELISA and our method, 30 positive serum samples weresubjected to quantitative detection by both methods. Because theconcentration ofHBsAg in serumsamplesmaybehigh enough to ex-ceed the upper limit of the linear range of both methods, theseHBsAg-positive serum samples were diluted with dilution buffertomake sure that the concentration of HBsAgwas located in the lin-ear range. The original concentrations of HBsAg in serum sampleswere subsequently calculated according to the dilution ratios. Theobtained correlation coefcient, r = 0.992 (P < 0.0001, n = 30), dem-onstrated excellent correlation between the two methods (Fig. 5).
Table 1Effects of different lengths of spacer arms on conjugation of HRP to silicananoparticles.
Amino silane Spacer arm length (atoms) Number of HRP moleculesper SiO2 nanoparticle
APTMS 4 77AEAPTMS 7 310AEAEAPTMS 10 329
Table 2Intra- and interassay precision of SiO2 nanoparticle-based HBsAg immunoassay.
Concentration (ng/ml) 5.0 2.0 1.0
Intraassay CV (%) (n = 8) 6.10 5.40 5.50Interassay CV (%) (n = 6) 8.93 9.96 11.08
12 Tandem conjugation of enzyme and antibody / R. Ke et al. / Anal. Biochem. 406 (2010) 813nanoparticles as a carrier for conjugation between antibody andenzyme allowedmore controllable and adjustable labeling. In com-parison, it is hard for current homobifunctional or heterobifunc-tional cross-linkers to control the amounts of the enzyme andantibody molecules in the conjugates.
Determination of HBsAg by sandwich ELISA
The linear range of the silica nanoparticle-based ELISAwas 0.1512.5 ng/mlwith a linear regression value r2 = 0.998. The linear rangewas identical with conventional ELISA. The coefcients of variance(CVs) along the entire concentration range were less than 10%(n = 8). Thedetection limit, calculatedas the responseat the zero cal-ibrator plus 3 standard deviations, was 0.06 ng/ml, which was threetimes lower than a conventional ELISA described previously .The precision study of the silica nanoparticle-based ELISA usingthree different concentrations of HBsAg standards showed that theCV values of the intra- and interassays were between 5.4% and6.1% and between 8.93% and 11.08%, respectively (Table 2).
A total of 280serumsamples (220HBsAg-negativeand60HBsAg-positive)weremeasured by both conventional ELISA and nanoparti-cle-based ELISA. When the cutoff value was dened as two timesbackground signal, the twomethods gave completely consistent re-Fig. 4. Effects of the amount of HRP (A) and antibody (B) used for preparation of bifuncmodied 2-mg aminated silica nanoparticles. (B) Here 0.15 mg (1), 0.3 mg (2), and 0.45 mfor conjugation. The bifunctional silica nanoparticles used had the same original concentHBsAg standard measured here is 1.0 ng/ml, and the immunoassay followed a standardFig. 5. Correlation analysis between ELISA and bifunctional silica nanoparticle-based ELISA for quantication of HBsAg in 30 positive serum samples (r = 0.992,P < 0.0001, n = 30).tional silica nanoparticles. (A) Here 0.4 mg (a) and 0.2 mg (b) of HRP were used forg (3) of antibody were used for coupling 2 mg of HRP-modied silica nanoparticles
ration, and all were diluted to different folds by dilution buffer. The concentration ofELISA.
In this work, we have developed a model system for preparationof antibodyenzymeconjugates on silica nanoparticle scaffold usingHRP and anti-HBsAg. The generated conjugates were used in ELISAfor thedetectionofHBsAg. Their easypreparation, lowcost, andex-ibilitymake silica nanoparticles an ideal scaffold for the preparationof antibodyenzyme conjugates. It can be conceived that a similarstrategy should be applicable to other enzyme labels as well as tosmall molecule labels used in immunoassays.
The benecial nature of using silica nanoparticles as a scaffold isthat it converts a liquidphase-based, single-step reaction into a solidliquid-based, layer-by-layer reaction for enzyme conjugation. Thus,
detection of severe acute respiratory syndrome viral infection, J. Clin.Microbiol. 46 (2008) 302310.
 S. Dhawan, Signal amplication systems in immunoassays: implications forclinical diagnostics, Expert Rev. Mol. Diagn. 6 (2006) 749760.
 C. Mateo, J.M. Palomo, L.M. van Langen, F. van Rantwijk, R.A. Sheldon, A new,mild cross-linking methodology to prepare cross-linked enzyme aggregates,Biotechnol. Bioeng. 86 (2004) 273276.
 M. Fuentes, R.L. Segura, O. Abian, L. Betancor, A. Hidalgo, C.Mateo, R. Fernandez-Lafuente, J.M. Guisan, Determination of proteinprotein interactions throughaldehydedextran intermolecular cross-linking, Proteomics 4 (2004) 26022607.
 S. Dhawan, Design construction of novel molecular conjugates for signalamplication: I. Conjugation of multiple horseradish peroxidase molecules toimmunoglobulin via primary amines on lysine peptide chains, Peptides 23(2002) 20912098.
 S. Dhawan, Design construction of novel molecular conjugates for signalamplication: II. Use of multivalent polystyrene microparticles lysine peptide
Tandem conjugation of enzyme and antibody / R. Ke et al. / Anal. Biochem. 406 (2010) 813 13the one-step conjugation is divided into several controllable steps,and in each step the reactant ratio can be adjustable while the prod-ucts can be simply isolated through centrifugations. Because the ex-act labeling ratio is exibly adjustable rather than strictly restrictedwith the number of active groups of the conjugated molecules, thewhole conjugationprocess can be optimized to achieve the best per-formancewith high reproducibility. Therefore, not only is the puri-cation of the conjugate greatly simplied, but also the sensitivity isimproved. We expect that this novel labeling strategy could ndwide applications in various immunoassay formats.
We thank InTec Products (Xiamen, China) for providing humanserum samples and materials used in conventional ELISA. We alsothank Jieli Zhang and Ping Chen for technical assistance. The studywas nancially supported by the National Natural Science Founda-tion of China (30500454) and the Natural Science Foundation ofFujian Province of China (2007J0112).
 J.R. Crowther, The ELISA Guidebook, second ed., Humana, Totowa, NJ, 2009. R. Rapley, J.M. Walker, Molecular Biomethods Handbook, second ed., Humana,
Totowa, NJ, 2008. pp. 657682. A.H.B. Wu, A selected history future of immunoassay development and
applications in clinical chemistry, Clin. Chim. Acta 369 (2006) 119124. R.M. Lequin, Enzyme immunoassay (EIA)/enzyme-linked immunosorbent
assay (ELISA), Clin. Chem. 51 (2005) 24152418. M.L. Sanchez-Martinez, M.P. Aguilar-Caballos, A. Gomez-Hens, Long-
wavelength homogeneous enzyme immunoassay for the determination ofamikacin in water samples, Talanta 78 (2009) 305309.
 Y.Meng, K. High, J. Antonello,M.W.Washabaugh,Q.J. Zhao, Enhanced sensitivityprecision in an enzyme-linked immunosorbent assay with uorogenicsubstrates compared with commonly used chromogenic substrates, Anal.Biochem. 345 (2005) 227236.
 M.M. Vdovenko, L. Della Ciana, I.Y. Sakharov, 3-(100-phenothiazinyl)propane-1-sulfonate is apotentenhancer of soybeanperoxidase-induced chemiluminescence,Anal. Biochem. 392 (2009) 5458.
 K. Fujimoto, K.H. Chan, K. Takeda, K.F. Lo, R.H.K. Leung, T. Okamoto, Sensitivespecic enzyme-linked immunosorbent assay using chemiluminescence forchains to generate immunoglobulinhorseradish peroxidase conjugates,Peptides 23 (2002) 20992110.
 K.A. Edwards, A.J. Baeumner, Liposomes in analyses, Talanta 68 (2006) 14211431.
 W. Zhan, A.J. Bard, Electrogenerated chemiluminescence: 83. Immunoassay ofhuman C-reactive protein by using Ru(bpy)(3)(2+)-encapsulated liposomes aslabels, Anal. Chem. 79 (2007) 459463.
 J.L. Yan, M.C. Estevez, J.E. Smith, K.M. Wang, X.X. He, L. Wang, W.H. Tan, Dye-doped nanoparticles for bioanalysis, Nano Today 2 (2007) 4450.
 A. Valanne, J. Suojanen, J. Peltonen, T. Soukka, P. Hanninen, H. Harma, Multiplesized europium(III) chelate-dyed polystyrene particles as donors in FRET: anapplication for sensitive protein quantication utilizing competitiveadsorption, Analyst 134 (2009) 980986.
 Y.F. Wu, C.L. Chen, S.Q. Liu, Enzyme-functionalized silica nanoparticles assensitive labels in biosensing, Anal. Chem. 81 (2009) 16001607.
 Y. Piao, A. Burns, J. Kim, U. Wiesner, T. Hyeon, Designed fabrication of silica-based nanostructured particle systems for nanomedicine applications, Adv.Funct. Mater. 18 (2008) 37453758.
 L. Chen, C. Chen, R. Li, Y. Li, S. Liu, CdTe quantum dot functionalized silicananosphere labels for ultrasensitive detection of biomarker, Chem. Commun.19 (2009) 26702672.
 J. Wang, G. Liu, M.H. Engelhard, Y. Lin, Sensitive immunoassay of a biomarkertumor necrosis factor-a based on poly(guanine)-functionalized silicananoparticle label, Anal. Chem. 78 (2006) 69746979.
 M.A. Jones, P.K. Kilpatrick, R.G. Carbonell, Preparation and characterization ofbifunctional unilamellar vesicles for enhanced immunosorbent assays,Biotechnol. Progr. 9 (1993) 242258.
 K.G. Nilsson, Preparation of nanoparticles conjugated with enzyme andantibody and their use in heterogeneous enzyme immunoassays, J. Immunol.Methods 122 (1989) 273277.
 X.J. Zhao, R.P. Bagwe, W.H. Tan, Development of organic-dye-doped silicananoparticles in a reverse microemulsion, Adv. Mater. 16 (2004) 173176.
 R.P. Bagwe, C.Y. Yang, L.R. Hilliard, W.H. Tan, Optimization of dye-doped silicananoparticles prepared using a reverse microemulsion method, Langmuir 20(2004) 83368342.
 Y. Xu, Q.G. Li, Multiple uorescent labeling of silica nanoparticles withlanthanide chelates for highly sensitive time-resolved immunouorometricassays, Clin. Chem. 53 (2007) 15031510.
 G. Penzol, P. Armisen, R. Fernandez-Lafuente, L. Rodes, J.M. Guisan, Use ofdextrans as long hydrophilic spacer arms to improve the performance ofimmobilized proteins acting on macromolecules, Biotechnol. Bioeng. 60(1998) 518523.
 S. Santra, P. Zhang, K. Wang, R. Tapec, W. Tan, Conjugation of biomoleculeswith luminophore-doped silica nanoparticles for photostable biomarkers,Anal. Chem. 73 (2001) 49884993.
 M.H.C.Bieniarz, G. Barnes, C.A. King, C.J.Welch, Extended lengthheterobifunctionalcoupling agents for protein conjugations, Bioconjug. Chem. 7 (1996) 8895.
Tandem conjugation of enzyme and antibody on silica nanoparticle for enzyme immunoassayMaterials and methodsChemicalsBuffersInstrumentsPreparation of aminated silica nanoparticlesImmobilization of HRP on surface of aminated silica nanoparticlesConjugation of antibody to surface of HRP-linked silica nanoparticlesDetection of HBsAg by ELISA
Results and discussionPreparation of bifunctional silica nanoparticlesEffect of different lengths of spacer armsEffect of amount of HRP and antibody used for conjugationDetermination of HBsAg by sandwich ELISA