Single Particle Methods For Homogeneous Immunoassays Using Noble Metal Nanoparticles As Labeling Probes

Currently, immunoassay is one of the most important analytical methods forproteins, drugs, hormones, toxins, microbes and other substances based on theantibody-antigen specific binding reaction. Due to its high selectivity and affinityimmunoassay has been widely used in clinical medicine, chemistry, pharmaceuticaland other fields. The heterogeneous assay mode was widely applied in traditionalimmunoassays such as enzyme linked immunosorbent assay (ELISA). However, theheterogeneous immunoassays involve antibody immobilization, immune reaction andwashing cycles, and thus this assay is labor-intensive and time-consuming. Generally,homogeneous immunoassay is an attractive detection format because it is amenable toautomation, reduced the risk of contamination and shorten analysis time. The key inhomogeneous immunoassay is to develop a new detection method for quantitativelyand sensitively distinguishing properties of antibodies (or antigen) andantigen-antibody complexes in the reaction solution. In this thesis, combined singleparticle detection methods with noble metal nanoparticles (such as gold or silvernanoparticles) labelling techniques, we devepled homogeneous immunoassays withhigh sensitivity, and main work is included the following parts.As one of the optical probes, gold nanoparticles (GNPs) have many advantages,such as facile synthesis and modification, strong light scattering properties, goodbiocompatibilty and so on. We established a setup of resonance light scatteringcorrelation spectroscopy (RLSCS). The principle of RLSCS is similar to fluorescencecorrelation spectroscopy (FCS). When GNPs pass through the detection volume ofRLSCS system due to Brownian motion, the fluctuation of the scattering lightintensity can be tracked. Certain information such as the concentration or diffusioncoefficient of the detected nanoparticles can be obtained through the correlation analysis of the light fluctuation.1. We conjugated two different antibodies (Ab) with GNPs respectively. Inimmunoassays, when two different GNPs labeld with antibodies were mixed in asample containing antigen (Ag) targets, the binding of targets will cause GNPs toform dimers (or oligomers), which leads to the significant increase in thecharacteristic diffusion time of GNPs in the detection volume. The RLSCS methodcan sensitively detect the changes in the characteristic diffusion time of GNPs beforeand after immune reactions. We used this technology in homogeneous immunoassaysfor the liver cancer biomarker alpha-fetoprotein (AFP). The linear range of this assayis from1pmol/L to1nmol/L and the detection limit is1pmol/L for AFP. This newmethod was successfully applied for the direct determination of AFP levels in serafrom healthy subjects and cancer patients. Our results were in good agreement withELISA assays.2. Competitive immunoassay mode can be applied to both large and smallmolecules. Silver nanoparticles (SNPs) as a probe carrier have stronger plasmonicscattering property than GNPs. The antibody, antigen and hapten were attached to thesurface of SNPs. The antibody-labeled SNPs were firstly mixed with a samplecontaining antigens, and the part of antibody-labeled SNPs was bound to antigens ofinterest in the sample. And then, the antigen-labeled SNPs were added into the mixedsolution above, and they were bound to free antibody-labeled SNPs (excess) to formdimers (or oligomers), which led to the significant increase in the characteristicdiffusion time of SNPs in a tinny detection volume. In the competitive model, thecharacteristic diffusion time of SNPs decreased with an increase of antigenconcentration. The RLSCS can sensitively detects the changes in the characteristicdiffusion time of SNPs before and after the immunoreactions. In order to demonstratethe universality of this new method, small biomolecules,17-β estradiol (E2), andbiomacromolecules AFP, were used as assay models. The linear ranges of this methodwere from10pmol/L to10nmol/L for E2and100pmol/L to10nmol/L for AFP,respectively, and the detection limits were10pmol/L for E2and100pmol/L for AFP,respectively. The presented method was successfully used to the determination of E2levels in human urine and AFP levels in human sera. Our results were in goodagreement with ELISA assays.3. The photon bursting of GNPs will be generated in a highly focused laser beam (less than1fL) due to the plasmon resonance scattering and Brownian motion ofGNPs. The number of photon burst and the number of particles in solution GNPsshowed a good linear relationship. We observed that the noise shows Gaussiandistribution, and photon burst signal of GNPs meets Poisson distribution. Based onthe behaviors of noise and photon burst signal of GNPs, we established the photonburst counting method. On the plot between the intensity and the number of thephoton burst, the relationship between the intensity and the number of the photonburst can be obtained. Through the curve fitting, the noise distribution range ofphoton burst and its mathematical expectation μ and variance σ were determined, wedefined the threshold equals to μ+8×σ. Through software, the number of photonburst can be obtained. This method has good reproducibility and high sensitivity.4. Based on the photon burst counting method, we developed the homogeneoussandwich and competitive immunoassay. In assay, sandwich or competitive immunoreactions will cause GNPs to form dimers (or oligomers), which will cause a decreasein the overall number of GNPs in solution and the number of photon burst, thischanges can be detected by photon burst counting technique sensitively. We appliedprostate-specific antigen (PSA) and alpha-fetoprotein (AFP) as modes to evaluate thesystem. The linear range of PSA sandwich immunoassay was1pmol/L-10nmol/L,and the detection limit was1pmol/L; The linear range of PSA competitiveimmunoassay was10pmol/L-10nmol/L, and the detection limit was10pmol/L; Thelinear range of AFP sandwich immunoassay was1pmol/L-10nmol/L, and thedetection limit was1pmol/L; The linear range of AFP competitive immunoassay was1pmol/L-1nmol/L, and the detection limit was1pmol/L. Our results were in goodagreement with ELISA assays.