Development of a Multiplexed Particle-based Immunoassay for Biomarker Screening

Annie Chen (2010). Development of a Multiplexed Particle-based Immunoassay for Biomarker Screening PhD Thesis, Aust Institute for Bioengineering & Nanotechnology, The University of Queensland.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s40291822_phd_abstract.pdf s40291822_phd_abstract.pdf application/pdf 65.22KB 13
s40291822_phd_totalthesis.pdf s40291822_phd_totalthesis.pdf application/pdf 4.41MB 26
Author Annie Chen
Thesis Title Development of a Multiplexed Particle-based Immunoassay for Biomarker Screening
School, Centre or Institute Aust Institute for Bioengineering & Nanotechnology
Institution The University of Queensland
Publication date 2010-05
Thesis type PhD Thesis
Supervisor Prof. Matt Trau
Dr Bronwyn Battersby
Dr Darby Kozak
Total pages 179
Total colour pages 19
Total black and white pages 160
Subjects 06 Biological Sciences
Abstract/Summary Early disease detection has been identified as one of the most effective means of preventing, treating or reducing the transmission and mortality associated with diseases. In particular, it has been shown that the early and accurate diagnosis of cancers can significantly improve treatment success rates and prolong patients’ life expectancy. However, cancer detection is often hindered by the difficulty of early stage diagnosis, as current symptom-based diagnosis is often vague and ambiguous. Amongst all cancers, this is especially true for ovarian cancer. Ovarian cancer is one of the most aggressive gynecological cancers, and is ranked as the 5th most lethal cancer in females in both the U.S. and Australia. As a result, diagnostics based on the detection of distinct biological signatures, or biomarkers, that are able to indicate the presence of ovarian cancer at an early stage, have generated considerable interest. Currently, over 445 proteins have been identified and proposed as serological biomarker candidates for ovarian cancer diagnosis. In this study, three of these proteomic biomarkers Mesothelin (MSLN), Human Epididymis Protein 4 (HE4), and Secretory Leukocyte Peptidase Inhibitor (SLPI) have been chosen as targets for the detection of ovarian cancer, as they are typically over-expressed in ovarian carcinomas. It is believed that the future of disease diagnostics, especially for cancer, will rely on the detection of multiple biomarkers for improved diagnosis accuracy. Simultaneous detection of multiple biomarkers from single sample can be potentially achieved using a mixed suspension of various optically encoded particles, where each optically encoded particle population corresponds to one target biomarker. This thesis presents the optimisation and clinical validation of an optically encoded organosilica microsphere assay platform for the early detection of three ovarian cancer biomarkers from 24 clinical ovarian cancer patient serum samples. In order to improve diagnosis accuracy, the quantification and improvement of assay signal-to-noise ratio (S/N) were investigated. Increased assay S/N can be achieved through surface modification to decrease assay background noise, and also through increased density of biomarker capture agents that are immobilised on the assay platform. Non-specific biomolecule adsorption gives rise to low signal-to-noise and reduced assay confidence. In order to minimise assay background noise, the antifouling effectiveness of grafted poly(ethylene glycol) (PEG) layers was tested by a newly developed quantitative flow cytometric methodology. Particles were modified with various polymer morphologies using 2,000, 3,400, 6,000, 10,000 and 20,000 g/mol PEG, with grafted amounts ranging from 0.14 to 1.4 mg/m2 with corresponding thickness between 2 Å and 15 Å as determined by XPS. It was found that increasing the amount of grafted polymer decreased non-specific protein adsorption. This decrease in adsorption was such that the adsorption of two common serum proteins, BSA (68 kDa) and IgG (150 kDa) could be prevented when greater than 0.6 mg/m2 of PEG was grafted. However, greater than 1.3 mg/m2 of grafted PEG was required to prevent the adsorption of the smaller Protein G (6.5 kDa). The optimised 3,400 MW PEG modified low protein fouling particles were then used as an immunoassay platform. A novel and covalent biomolecule immobilisation methodology using ‘tresyl-activation’ was developed to enable the attachment of immunoassay capture antibodies onto the antifouling PEGylated particles to establish a model immunoassay. Tresyl activation time, particle activation stability, antibody immobilisation time and concentration were investigated and optimised. It was found that tresyl activation of the grafted PEG layers required a minimum reaction time of 1.5 hours. By increasing the antibody incubation time (24 hours) and concentration (10 μg/mL), a maximum antibody loading of 1.6×10-2 molecules per nm2 was achieved. This optimised low fouling high capture probe loading platform gave a ten-fold S/N ratio increase in a model anti-species immunoassay. This optimised model immunoassay platform was then used to validate its application as a serum-based immunoassay for ovarian cancer diagnosis. In order to build a model ovarian cancer diagnostic using the optimised PEGylated organosilica particles, full antibodies and single chain variable fragments (scFvs) towards three ovarian cancer biomarkers (MSLN, HE4 and SLPI) were used as capture probes. In particular, the use of biomarker-specific scFvs as capture agents were investigated as higher capture loading on assay surface could be potentially achieved due to the reduced size and molecular weight, hence improve assay S/N ratio. All three biomarkers gave a minimum assay detection limit of 0.4 ng/mL. This low detection limit is believed to be significant, as a positive diagnosis cut-off limit of 1.8 ng/mL for HE4 has been used in previous studies for ovarian cancer screening1, 2. The applicability of the developed ovarian cancer biomarker assay as a clinical diagnostic was validated using a panel of 12 healthy controls and 12 ovarian cancer patient clinical serum samples. The assay results were compared to that of a commercially available polystyrene particle assay system. It was found that the developed particle-based immunoassay using full antibodies as capture probes exhibited significant increase in assay signal and improved the number of positively identified clinical cases compared to that of the commercial kit. The ovarian cancer case diagnosis data achieved by the commercial kit for the detection of MSLN, HE4, and SLPI biomarkers was 5 (42%), 4 (33%), and 1 (8%) cancerous cases out of 12, respectively. In comparison, it was observed that the developed organosilica particle-based immunoassays coated with full antibodies as capture probes for the detection of MSLN, HE4 and SLPI biomarkers diagnosed 11 (92%), 4 (33%), and 9 (75%) cancerous cases out of 12, respectively. As for the developed scFv assay, 7 (58%) and 8 (67%) out of 12 cancerous cases were positively identified by screening MSLN and SLPI biomarkers, respectively. In comparison, the commercially available assay system was not able to successfully immobilise scFvs as antigen capture agents. These promising preliminary findings on the detection of ovarian cancer biomarkers from clinical serum samples support the potential application of the PEG modified organosilica particle assay as an early stage, molecular-based disease diagnostic. 1. Havrilesky, L. J.; Whitehead, C. M.; Rubatt, J. M.; Cheek, R. L.; Groelke, J.; He, Q.; Malinowski, D. P.; Fischer, T. J.; Berchuck, A., Evaluation of biomarker panels for early stage ovarian cancer detection and monitoring for disease recurrence. Gynecologic Oncology 2008, 110, (3), 374-382. 2. Walsh, C. S.; Karlan, B. Y., Molecular signatures of ovarian cancer: from detection to prognosis. Molecular Diagnosis and Therapy 2010, 14, (1), 13-22.
Keyword particle-based immunoassay
PEG grafting and characterisation
Proteomic biomarker
Ovarian cancer
Additional Notes 28, 31, 32, 34, 35, 38, 39, 42, 44, 47, 85, 113, 123, 128, 134, 135, 139, 154, 157

Citation counts: Google Scholar Search Google Scholar
Created: Tue, 28 Sep 2010, 12:55:47 EST by Miss Annie Chen on behalf of Library - Information Access Service