Biosensor Array Strategies for Cancer Biomarker Proteins James

Biosensor Array Strategies for Cancer Biomarker Proteins James F. Rusling Professor of Chemistry and Cell Biology University of Connecticut and University Health Center Collaboration with F. Papadimitrakopoulos (IMS) and Drs. Gutkind and Patel (NIH, Bethesda) Website: http: //web 2. uconn. edu/rusling/ Email: James. Rusling@uconn. edu

Cancer Biomarkers: National Institutes of Health – molecules that can be objectively measured and evaluated as indicators of normal or disease processes and pharmacologic responses to therapeutic intervention Cancer Biomarker Proteins: increase in serum concentration At the onset of cancer, even before tumor develops • Multiple proteins must be measured for reliable predictions • Excellent hope for early detection and treatment monitoring • May also be able to monitor inflammation • May facilitate new therapies

Key aspects and needs: • Ultrasensitive detection of multiple cancer biomarkers • Long term objectives 1. early cancer detection and monitoring 2. tools for cancer research and surgical decisions • point-of-care (POC) clinical assays – need to be cheap, simple, fast, accurate, multiplexed Expensive, and complex methodologies such as LCMS/MS, some automated optical-based methods are currently not competitive for POC Reviews:

Biomarker Targets: 1. Prostate Specific Antigen Ø PSA - Single chain glycoprotein , MW 33 k. Da Sensitive, specific biomarker for detection of prostate cancer years before clinical signs of disease ØDetection of PSA in serum: clinical detection of prostate cancer: 4 -10 ng/m. L Adapted From Brookhaven Protein Databank ØLed to less invasive treatment protocols, avoid surgery 2. Interleukin 6 (IL-6) - prostate and oral cancer biomarker - human plasma conc. normal < 6 pg/m. L; cancer 20 -1000 pg/m. L

ELISA- enzyme linked immuno-sorbent assay 96 -well plate Well in plate Antigen = Protein, pathogen Secondary antibody enzyme label and labels Primary antibody Antibodies capture The antigen Detection by optical absorbance plate reader after running enzyme reaction that gives a colored product • Reliable method for over 30 years • Best DL ~ 3 pg/m. L in serum • many commercial assay kits for single proteins • limitations in sample size, speed, multiplexing

Possible approaches: • fluorescence labels • surface plasmon resonance, SPR • Electrochemiluminescence (ECL) –Ru(bpy)32+ labels • bead based assays, ECL or Fluorescence S Slope = sensitivity Detection limit = blank signal + 3 x avg. noise Conc. protein

Multilabel Strategies – high sensitivity • detection by fluorescence, amperometry, ECL • non-specific binding must be minimized in any immunoassay

Possible multilabel strategies Dissolve, measure M+n Also used on particles Up to 400, 000 labels

$SPR arrays – measure refractive index at interface – potentially label free -More susceptible$

SPR arrays – measure refractive index at interface – potentially label free -More susceptible to Non-specific binding (NSB)

Bead-based protein assays Label – enzyme; Fluorophor; DNA, Ru. BPY(ECL) + Protein in sample multi-enzyme. Magnetic bead-Ab 2 Bead captures protein, Magnetic separation, wash Multiple labels Detection: • ECL – Ru. BPY label • Fluorescence – Fl. Label • 1 -10 pg/m. L DL • $200, 000 for machine Magnet under well in 96 -well plate, wash to remove NSB

2. 1. A conventional Single electrode Au. NP film B Electrochemical Immunosensors Ab 2 -enzyme Au. NP Capture Antibody Multilabel magnetic particle, on-line capture Multi-enzyme. Protein analyte Magnetic particle-Ab 2

Au. NP-based immunosensors – single sensors nanostructures sensor + massive multilabel strategies AFM, carboxylated-Au. NPs on polycation underlayer Antibodies on Au. NPs Rotating disk amperometry (A) and calibration for PSA on Au. NP platform Au. NPs: DL for PSA = 0. 5 pg/m. L (8 x)

Accurate PSA detection in cell lysates and patient serum Using Au. NP-based immunosensors

SWNT array for 4 prostate cancer proteins in human serum vs. ELISA 8 -sensor array Bhaskara V. Chikkaveeraiah, Ashwin Bhirde, Ruchika Malhotra, Vyomesh Patel, J. Silvio Gutkind, and James F. Rusling, Single-Wall Carbon Nanotube Forest Immunoarrays For electrochemical measurement of 4 Protein Biomarkers for Prostate Cancer, Anal. Chem. , 2009, 81, 9129– 9134.

5 nm Au nanoparticle vs. single wall nanotube electrodes • head to head comparison on IL-6 detection show better Detection limits and better linearity with Au. NPs • Au. NPs are easier to handle and produce highly reproducible electrodes Bernard S. Munge, Colleen E. Krause, Ruchika Malhotra, Vyomesh Patel, J. Silvio Gutkind, and James F. Rusling, Electrochemical Immunosensors for Interleukin-6. Comparison of Carbon Nanotube Forest and Gold Nanoparticle platforms, Electrochem. Comm. , 2009, 11, 1009– 1012

Off-line capture magnetic particle microfluidic strategy A Conventional, single label array electrode B Electrical contacts Multilabel magnetic Particle, off line capture -0. 2 V + H 2 O 2 + HQ signal 8 electrodes In channel Ab 2 -enzyme Au. NP Capture Antibody Protein analyte 1 m multi-enzyme. Magnetic particle-Ab 2 7000 -400, 000 labels

8 -electrode PDMS microfluidic array electrodes Electrical contacts

Components of microfluidic device made of micro-machined polymethylmethacrylate), soft PDMS microfluidic channel + screen printed 8 electrode carbon array

Microfluidic protein assay system

Microfluidic array detection of PSA in serum samples Using off-line capture with 1 m multilabel magnetic particle; Detection limit ~100 fg/m. L PSA in 10 L serum; 5 -fold better than manual assay without off-line capture

Microfluidic array data for mixture of PSA and IL 6 in serum Flow rate : 100 µL/min H 2 O 2 : 100 µM Hydroquinone : 1 m. M Limit of detection of PSA: 225 fg/m. L Limit of detection of IL 6 : 300 fg/m. L

Low cross reactivity of PSA and IL-6 Ab 1 PSA IL 6 PSA 2 -5% IL 6 1 -3%

Immunosensor assay validation on human serum PSA IL-6

Serum Protein Biomarkers for Oral Cancer ØInterleukin-6 [IL-6] ØInterleukin-8 [IL-8] ØVascular Endothelial Growth Factor [VEGF] ØVascular Endothelial Growth Factor C [VEGF-C] Trikha, M. ; Corringham, R. ; Klein, B. ; Rossi, J. Clin. Cancer Res. 2003, 9, 4653 -4665 Hebert, C. A. ; Baker, J. B. Cancer Invest. 1993, 11, 743 -750 O-charoenrat, O. ; Rhys-Evans, P. ; Eccles, S. A. Cancer 2001, 92, 556 -568

Microfluidic Immunoarray: Oral Cancer Biomarkers Off-line capture using magnetic particles with 400, 000 HRPs IL-6 DL: 10 fg m. L-1 VEGF DL: 8 fg m. L-1 IL-8 DL: 15 fg m. L-1 VEGF-C DL: 60 fg m. L-1

Protein Array using Ru. BPY ECL label Immunoassays in 10 m. L wells Carbon (PG) chip (no microelectronics) SWCNT forest ECL = Electrochemiluminescence Detection labels are 100 nm d. silica with internal Ru. BPY

Forster and Voss, 1980 s (synthesis) Ru. PVP Solid PG chip Spots contain capture antibody on Ru. PVP ECL polymer

ECL arrays for detection of PSA and IL-6 DL ~ 0. 1 pg/m. L 2 ng/m. L 0. 1 pg/m. L Control 0

Surface plasmon resonance (SPR) detection of protein biomarkers using superparamagnetic beads labels

SPR response to PSA in Serum with Magnetic and Silica labels Using off-line capture with 1 m magnetic particle label; Detection limit ~10 fg/m. L PSA

Ultrasensitive multiple protein arrays: • Combining nanostructured sensors or SPR with (multi-label) magnetic particles gives ultrahigh sensitivity in fg/m. L range • Microfluidics with off-line analyte protein capture gives very low S/N, semiautomated • May open door to new ultralow abundance biomarkers • ECL provides simpler array for protein detection, no microelectronic chip needed

Cancer Biomaker Protein Measurements • as yet, limited POC or clinical use except for PSA • ELISA, commercial kits, one protein, 3 pg/m. L DL • bead based methods, up to 10 proteins, equip. and kits expensive, 1 -10 pg/m. L DLs • LC-MS, great for discovery, emerging for routine tests • new experimental methods promise ultrasensitivity, detection in fg/m. L range, multiplexing - microfluidic amperometric arrays, multilabel - ECL arrays, simplicity - SPR arrays with magnetic particle labels - fiber optic microwell arrays – D. Walt (Tufts) - DNA label “bar-codes” - C. Mirkin (Northwestern)