doi:10.1016/j.jim.2005.07.008 
Copyright © 2005 Elsevier B.V. All rights reserved.
Research paper
Development of a cluster of differentiation antibody-based protein microarray
Michael Abdoa, 
, 
, Bob Irvinga, b, Peter Hudsona, b and Heddy Zolab, c
aCSIRO Division of Health Sciences and Nutrition, 343 Royal Parade, Parkville, Victoria 3052, Australia
bCRC for Diagnostics, Queensland University of Technology, Gardens Point Campus, Brisbane, Queensland 4001, Australia
cChild Health Research Institute, Women's and Children's Hospital, North Adelaide, South Australia 5006, Australia
Accepted 8 June 2005.
Available online 18 August 2005.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
Protein microarrays combine aspects of DNA microarrays and ELISA for the parallel interrogation of a biological sample using a multiplex of protein biomarkers. Here we report the development of a protein microarray consisting of a subset of CD antibodies and CRP. Several preparations (culture supernatant, ascites fluid and purified Ig) of each antibody were used in a forward phase protein microarray. Microarrays were fabricated using a non-contact printer delivering 300 pL (± 30 pL) to specific locations on polyacrylamide gel-based substrates. Following production, microarrays were blocked for non-specific binding and incubated with sera conjugated directly with Cy3. Using CRP as a control biomarker, 12 clinical samples (inflammatory conditions and controls) were interrogated using the protein microarray format and results compared to CRP measured by conventional immunoassay. The data obtained from the microarray correlated with CRP assessed by immunoassay. Subsequently CRP ‘positive’ samples were interrogated for CD antigen expression; which revealed CD25 and CD45RO expression in all samples. Whilst this study focussed on a subset of CD antibodies, it is anticipated that this array could be expanded to include a larger number of CD antibodies and allow screening of sera from multiple conditions in order to identify disease markers.
Keywords: Antibody microarray; Inflammation; ELISA; C-reactive protein
Abbreviations: BSA, bovine serum albumin; CD, cluster of differentiation; CRP, C-reactive protein; DDW, double distilled water; ELISA, enzyme-linked immunosorbent assay; HLDA8, Human Leukocyte Differentiation Antigen Workshop 8; HTP, high-throughput; LOD, limit of detection; PBST, phosphate-buffered saline/Tween 20
Fig. 1. The sensitivity of the protein microarray format was first assessed using a monoclonal anti-human CRP antibody microarray. Shown are data collected from all repeats (mean ± S.E.M.) (a) and a representative image of the array scanned at 595 nm from a single experiment (b). CRP antibody, BSA (negative control) and mouse IgG (signal control) were printed as the probe at 30 pg/spot and used to interrogate a purified human CRP sample of varying concentration (analyte). There was a significant difference in signal to noise ratio from as little as 0.16 ng/mL CRP (P < 0.05; one-way ANOVA). The spot intensity of CRP at 0.4 ng/mL was equivalent to that generated by the mouse IgG (signal to noise = 7.6 ± 0.6) reference at the same concentration. Values without shared notations differ at P < 0.05 (LSD test).
Fig. 2. The accuracy of the protein microarray format was assessed using an anti-human CRP antibody microarray and compared to results from a conventional immunoassay (Table 1). Shown are data collected from all repeats (mean ± S.E.M.) (a) and a composite image from each array scanned at 595 nm (b). CRP antibody and controls were printed as the probe at 30 pg/spot and used to interrogate a human serum sample (50 μg/mL). CRP levels in samples 6, 8, 9, 10 and 11 differed significantly from CRP controls and remaining samples (P < 0.05; one-way ANOVA). Values without shared notations differ at P < 0.05 (LSD test).
Fig. 3. CD antigen expression was assessed using a CD antibody microarray. Shown are data collected from each CRP positive sample (n = 6) for all repeats (mean ± S.E.M.) (a) and a representative image (single repeat) of the array scanned at 595 nm (b). CD antibodies (all preparations) and controls were printed at 30 pg/spot and used to interrogate a serum samples 6, 8, 9, 10, 11 and 12 (50 μg/mL). CD25 (ascites) and CD45RO (purified and ascites) were elevated in all samples. Incubation with CRP negative samples resulted in no significant signal for all CD antibodies (data not shown).
Table 1.
Quantitation of CRP expression using conventional immunoassay
Abstract
Purchase PDF (401 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (5)
Purchase PDF (298 K)
