Claudia Preininger

Quick and simple: quality control of microarray data

MOTIVATION Microarrays are high-throughput tools for parallel miniaturized detection of biomolecules. In contrast to experiments using ratios of signals in two channels, experiments with only one fluorescent dye cause special problems for data analysis. The present work compares algorithms for quality filtering on spot level as well as array/slide level. RESULTS Methods for quantitative spot filtering are discussed and new sets of quality scores for data preprocessing are designed. As measures of spot quality also reflect the quality of protocols, they were employed to find the optimal print buffer in an optimization experiment. In order to determine problematic arrays within a set of replicates we tested methods of outlier detection which can suitably replace the visual inspection of slides. CONTACT Ursula.Sauer@arcs.ac.at.

Protein chip for the parallel quantification of high and low abundant biomarkers for sepsis.

We present herein a protein chip for diagnosis of sepsis that combines both a sandwich and a binding inhibition format in order to quantify high (CRP) and low abundant proteins (cytokines, PCT, neopterin) in parallel. Using the combined assay format the lowest detectable concentrations for CRP, IL-6, IL-8, IL-10, TNFα, PCT, and neopterin are 3 mg/L, 15 ng/L, 26 ng/L, 65 ng/L, 40 ng/L, 78 ng/L, and 0.46 μg/L. Four different combined assay formats are tested, using separate or joint incubation steps of analytes and detection antibodies. Yet, low limit of detection (LOD) and short processing time are contradictory: while the combined assay performed in a multistep protocol is extremely sensitive (e.g., the LOD for IL-6 is 15 ng/L), but more time-consuming (4 h), the all-in-one protocol takes only 2.5 h, but suffers from lower sensitivity compared with the multistep protocol (e.g., the LOD for IL-6 is up to 40 times enhanced). Reproducibility is good in both cases (CV 5-20%).

Miniaturized protein arrays: Model and experiment

Nanobiolithography techniques have the ability to fabricate structures of biomolecules as small as ∼40 nm. However, very few examples of working biosensors of these sizes have been demonstrated. These examples use substrates like Gold and Silicon, that are advantageous for fabrication purposes, but present disadvantages as far as signal detection is concerned. The preferred and standard substrates used in microarray research are fabricated on glass. On these surfaces, the binding site density varies between and within individual samples, and is largely not characterized. We report here on the fabrication of a fully functional immunochip with spots of ∼1 μm diameter and a signal to noise ratio (SNR) above 10, using Nano-fountain pen (NFP). To achieve this, we analyze the dominant parameters influencing SNR, develop a model that enables us to compare various types of surfaces and choose the most appropriate ones. We show that a miniaturized immunochip is feasible, yielding detection limit as low as 1.3 ng/ml and dynamic range well above 10(5). Cross-reactivity of two different species is shown to be negligible. In addition, we study the binding mechanism of surfaces, show how to differentiate between 2D and 3D immobilization, and show that a hydrogel surface (using non-covalent immobilization strategy) yields higher intensities for the same target molecule concentrations, and higher dynamic range.

Immobilization of oligonucleotides on crosslinked poly(vinyl alcohol) for application in DNA chips.

In biotechnological applications there is an enormeous growth in the development of new miniaturized devices to reduce timescales, cost and amounts of reagents and starting materials. DNA chips represent miniaturized analytical tools that allow the simultaneous detection of different targets for high throughput screening. They consist of a small size support on which DNA probes are in-situ synthezised or immobilized. We present protocols for the effective immobilization of pre-synthezised 16S rRNA oligonucleotides on crosslinked poly(vinyl alcohol) (PVA). The polymeric gel consists of poly(vinyl alcohol) (PVA) crosslinked with poly(allylamin chloride) (PALAM) and monochlortriazinyl-beta-cyclodextrin (beta-CD) at pH 4,6.8,8 and 9. Mechanical characterization of the gels show that the immobilization capacity increases with increasing pH and increasing crosslinking within the gel. We will demonstrate that the hybridization efficiency on PVA chips is superior over commercially available chips based on nylon, nitrocellulose and aminoalkylsilane.

Miniaturized Protein Microarray with Internal Calibration as Point-of-Care Device for Diagnosis of Neonatal Sepsis

Neonatal sepsis is still a leading cause of death among newborns. Therefore a protein-microarray for point-of-care testing that simultaneously quantifies the sepsis associated serum proteins IL-6, IL-8, IL-10, TNF alpha, S-100, PCT, E-Selectin, CRP and Neopterin has been developed. The chip works with only a 4 μL patient serum sample and hence minimizes excessive blood withdrawal from newborns. The 4 μL patient samples are diluted with 36 μL assay buffer and distributed to four slides for repetitive measurements. Streptavidin coated magnetic particles that act as distinct stirring detection components are added, not only to stir the sample, but also to detect antibody antigen binding events. We demonstrate that the test is complete within 2.5 h using a single step assay. S-100 conjugated to BSA is spotted in increasing concentrations to create an internal calibration. The presented low volume protein-chip fulfills the requirements of point-of-care testing for accurate and repeatable (CV < 14%) quantification of serum proteins for the diagnosis of neonatal sepsis.

Microarray analysis of protein-protein interactions based on FRET using subnanosecond-resolved fluorescence lifetime imaging.

The detection of protein-protein binding on microarrays using the fluorescence lifetime as a dynamic analytical parameter was investigated in a model system. The assay is based on Förster resonance energy transfer (FRET) and carried out with biotinylated Bovine Serum Albumin and streptavidin, labeled with the commonly used microarray dyes Alexa 555 and Alexa 647, respectively. This efficient FRET donor/acceptor pair was employed in a competitive assay format on three different microarray surfaces. The fluorescence was excited by 200ps laser pulses from a mode-locked and cavity-dumped argon-ion laser, adapted to an intensified CCD camera as detection unit allowing time resolution with subnanosecond precision. Lifetime maps were recorded according to the Rapid Lifetime Determination (RLD) scheme. Interaction between the proteins could clearly be detected on all formats and resulted in almost complete quenching on CEL Epoxy surfaces upon addition of excess streptavidin labeled the FRET acceptor dye. In this case, the fluorescence lifetimes dropped by 90%, whereas on ARChip Epoxy and ARChip Gel the reduction was 54% and 47%, respectively. Good linearity of the quenching curve was obtained in all cases. The method is applicable to all types of protein interaction analysis on microarrays, particularly in cases where evaluation of fluorescence intensity is prone to erroneous results and a more robust parameter is required.

Quality control of chip manufacture and chip analysis using epoxy-chips as a model

Abstract Biochips are miniaturized, highly ordered analysis systems which offer the unique advantage of highly parallel analysis of thousands of analytes at the same time. Although this technique has been enthusiastically developed and has promised to improve and speed up numerous biological assays, the quality control of chip manufacture, chip analysis and data management has received less attention. The following article compares three epoxy-containing chip surfaces (ARChip Epoxy, 3D-Link™, and EasySpot) with respect to their autofluorescence, immobilization capacity, background fluorescence and hybridization efficiency. Since data collected from biochip experiments are random snapshots with errors, inherently noisy and incomplete, we tried to evaluate technical factors causing variability and to set up quality control procedures for chip manufacture and chip analysis. Variabilities caused by arraying, glass substrate and polymer coating, fluorescent label and experimental conditions are discussed in details.

Critical role of the sample matrix in a point-of-care protein chip for sepsis.

Both highly specific antibodies and appropriate assay buffers are key elements in the development of sensitive multi-analyte diagnostic tests and essential assay components to minimize interferences from the sample matrix. Herein, we investigate the influence of 0.1 M Tris (pH 7.4)/0.1 M NaCl/10 mM CaCl(2)/0.1% Tween-20 used as assay buffer and diluent for serum, plasma and saliva samples in a protein biomarker chip for the diagnosis of sepsis. In detail, on-chip sandwich assays for detection of IL-6 and PCT are established using pure assay buffer and serum, plasma, and saliva, each diluted by a factor of 10 and 100 with assay buffer. The dilution linearity as well as the cross-reactivity to immobilized IL-8, IL-10 and TNF-α antibodies (<1.8% in plasma and serum) is investigated; furthermore the influence of immunoglobulin G, fibrinogen and lysozyme, highly abundant proteins in serum, plasma and saliva. This effect is two times more pronounced in serum than in plasma and saliva and strongly decreases with increasing analyte concentration. Though the matrix proteins bind unspecifically to the immobilized receptors, they do not prevent the analyte binding; on the contrary, the analyte is reliably detected with high sensitivity, featuring limits of detection of 16 ng/L and 0.31 μg/L, and coefficients of variation of 18% and 29% for IL-6 and PCT in 10% serum.

Evaluation of substrate performance for a microbial diagnostic microarray using a four parameter ranking

The impact of substrates for probe immobilization was studied using a microbial diagnostic microarray consisting of probes designed against the pmoA genes of methanotrophs and functionally related bacteria. The proprietary ARChip Epoxy was compared to seven epoxy-modified competitor slides, and one porous 3D, furthermore two substrates recommended for oligoprobes without description of the surface chemistry. The diagnostic microarray on Cel silylated aldehyde (CSS) slide was used as a reference for this comparison as it was actually used for assay development and validation. The suitability of binding chemistries for microarrays was evaluated by specificity, signal, and inter- and intra-slide precision and ranked accordingly. The performance of four epoxy substrates (ARChip Epoxy, Cel Epoxy, Corning Epoxy, sciChip) and Spot On slides has proved satisfactory and comparable to the reference Cel CSS in that inter-slide precision was between 8% and 18% CV, intra-slide precision below 30%, respectively. The four parameter ranking shows great promise of providing deeper insight in the performance of materials and protocols tested.

Thermometric determination of copper(II) using acid urease

A thermometric assay for the measurement of copper(II) is reported that makes use of acid urease immobilized on controlled-pore glass (CPG) with coupling to a thermometric continuous-flow sensor system. Enzyme thermistor analysis was performed by a flow injection technique. Urea was injected into the system to give a temperature change corresponding to 100% enzyme activity. After inhibition by copper(II), the enzyme activity and the corresponding temperature change of the reaction were reduced. The percentage inhibition was calculated from the peak heights before and after copper(II) injection. The typical response time was 3–5 min and the dynamic range was from 5 × 10–6 to 1 × 10–4 mol l–1(0.3–6.4 ppm) coper(II). The main advantages of the thermometric biosensor assay are (a) a 20-fold higher sensitivity of acid urease to copper(II) than urease (from jack beans), which is commonly used for metal sensing, (b) the fact that regeneration of the enzyme does not require any metal chelating agent, (c) no decrease in enzyme activity because of irreversible inhibition and (d) the possibility of performing intermittent monitoring of Copper(II) using a thermistor device.