BioTechniques

ISSN: 0736-6205 (Print) 1940-9818 (Online) Journal homepage: www.tandfonline.com/journals/ibtn20

Protein Microarrays

Chien-Sheng Chen & Heng Zhu

To cite this article: Chien-Sheng Chen & Heng Zhu (2006) Protein Microarrays, BioTechniques,
40:4, 423-429, DOI: 10.2144/06404TE01
To link to this article: https://doi.org/10.2144/06404TE01

© 2006 Author(s)

Published online: 21 May 2018.

Submit your article to this journal

Article views: 31108

View related articles

Citing articles: 11 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ibtn20
Protein Microarrays
Chien-Sheng Chen and Heng Zhu
Department of Pharmacology and Molecular Sciences/High-Throughput Biology Center, Johns Hopkins
University School of Medicine, Baltimore, MD, USA

Introduction following sections, we will provide examples of various ap-

plications of both types of microarray, with an emphasis on
Protein microarrays, an emerging class of proteomic tech- functional protein microarrays. Given the large volume of
nologies, are fast becoming critical tools in biochemistry papers related to protein microarray technology, we regret
and molecular biology. Two classes of protein microarrays that we are unable to cite all the published work in the field.
are currently available: analytical and functional protein
microarrays. Analytical protein microarrays, mostly antibody
Analytical Microarrays
microarrays, have become one of the most powerful multi-
plexed detection technologies. Functional protein microarrays Perhaps the most representative class of analytical micro-
are being increasingly applied to many areas of biological arrays is the antibody microarray, in which antibodies
discovery, including studies of protein interaction, biochemi- are arrayed on glass surfaces at high density. The biggest
cal activity, and immune responses. Great progress has been challenge associated with antibody microarrays is that of
achieved in both classes of protein microarrays in terms of producing antibodies that are able to identify the proteins
sensitivity, specificity, and expanded application. of interest with high specificity and affinity in a high-
Protein microarrays, also known as protein chips, are minia- throughput fashion. Because the traditional method for
turized and parallel assay systems that contain small amounts generating monoclonal antibodies is time-consuming and
of purified proteins in a high-density format (1). They allow laborious, researchers have recently sought alternative ap-
simultaneous determination of a great variety of analytes proaches. For example, phage antibody-display, ribosome
from small amounts of samples within a single experiment. display, systematic evolution of ligands by exponential enrich-
Protein microarrays are typically prepared by immobilizing ment (SELEX), messenger RNA (mRNA) display, and affibody
proteins onto a microscope slide using a standard contact display have been developed to expedite the production of
spotter (1,2) or noncontact microarrayer (3–5). A variety of antibodies with high specificity (11–14). All of these methods
slide surfaces can be used. Popular types include aldehyde- involve the construction of large repertoires of viable regions
and epoxy-derivatized glass surfaces for random attachment with potential binding activity, which can be selected by mul-
through amines (2,6), nitrocellulose (7,8), or gel-coated slides tiple rounds of affinity purification. The binding affinity of
(9,10) and nickel-coated slides for affinity attachment of the resulting candidate clones can be further improved using
His6-tagged proteins. The last type was reported to provide maturation strategies. However, the ideal selection system
10-fold better signals than those obtained with other random is yet to be fully developed: one that is not only fast, robust,
attachment methods (1). After proteins are immobilized on sensitive, and of low cost, but also automated and minimized
the slides, they can be probed for a variety of functions/ (13,14).
activities. Finally, the resulting
signals are usually measured by
detecting fluorescent or radio-
isotope labels. The typical image
of protein microarrays is shown
as Figure 1.
Analytical protein arrays can
be used to monitor protein
expression levels or for bio-
marker identification, clinical
diagnosis, or environmental/
food safety analysis. Func-
tional protein microarrays have
many uses: (i) to probe for
various types of protein activi-
ties, including protein-protein,
protein-lipid, protein-DNA,
protein-drug, and protein-pep-
tide interactions; (ii) to identify
Figure 1. A typical protein microarray image. A yeast
enzyme substrates; and (iii) protein microarray is probed with anti-GST antibodies followed
to profile immune responses, by detection with Cy5-conjugated secondary antibodies. An
among many others. Applica- enlarged image of one of the 48 blocks is depicted below the
tions of both the analytical and protein chip.
functional protein microarrays
are depicted in Figure 2. In the

Vol. 40, No. 4 (2006) BioTechniques 423

 Functional Protein Microarrays
Functional protein microarrays have recently been applied to
many aspects of discovery-based biology, including protein-
protein, protein-lipid, protein-DNA, protein-drug, and
protein-peptide interactions. Although we have attempted
to describe all the major applications of functional protein
microarrays, it is impossible to cover all the instances in which
they have been used. Therefore, we have chosen to focus most
of our examples on yeast proteome microarrays (Figure 4).

Protein-Protein and Protein-Lipid Interactions

Zhu and coworkers (1) reported the construction and ap-
plication of the first proteome microarrays, which contained
>5800 individually purified yeast proteins or 85% of the yeast
proteome. These proteome chips were first used to study pro-
tein-protein interactions, in which the chips were incubated
with biotinylated calmodulin in order to identify its new bind-
Figure 2. Applications of protein microarrays. Antibody arrays can be ing partners. In addition, protein microarrays were also used
used for clinical diagnosis or environmental/food safety analysis. Functional
to examine interactions with various phospholipids, which are
protein arrays are mainly used to study various types of protein activities,
known to act as secondary messengers. When biotinylated
including protein-protein, protein-lipid, protein-DNA, protein-drug, and
protein-peptide interactions, to identify enzyme substrates and to profile liposomes containing various phospholipids of interest were
immune responses. used as probes, more than 150 phospholipid binding proteins
were identified, and a wide range of proteins was found to
bind to the lipid vesicles; over 50% of these were previously
Despite the challenge involved in obtaining specific anti- known to be associated with membranes.
bodies, many studies using antibody microarrays have recently
been reported. In a pioneer work by Haab and colleagues Protein-DNA Interactions
(15), the first high-density antibody microarrays were used to In a later report, the same research group also used the chips
test whether a linear relationship could be detected between to screen for novel DNA binding activities using fluorescently
an antibody and antigen pair in an array format. They investi- labeled yeast genomic DNA (18). A total of 200 proteins that
gated the ability of 115 well-characterized antibody-antigen reproducibly bound DNA were identified. Half of them had
pairs to react in high-density microarrays on modified glass not previously been shown to have DNA binding activity;
slides: 30% of the pairs showed the expected linear relation- these new proteins fell into a wide variety of functional cat-
ships, indicating that a fraction of the antibodies were suit- egories. The most surprising discovery in this category was the
able for quantitative analysis. Sreekumar and coworkers (16) identification of Arg5,6 as a DNA binding protein. The ARG5,6
created antibody arrays with 146 distinct antibodies against gene encodes two mitochondrial enzymes that mediate two
proteins involved in the stress response, cell cycle progression, key steps in the biosynthesis of ornithine (a precursor to
and apoptosis and used these arrays to monitor the altera- arginine). Follow-up experiments revealed that this enzyme
tions in protein quantity in LoVo colon carcinoma cells. The associates with specific mitochondrial loci in vivo, and this
reference standards and samples were labeled separately information was used to define a DNA binding motif for this
using either Cy™5 or Cy3 dyes, and the fluorescent signals of protein. Thus, a novel DNA binding activity was found to be
the bound proteins were detected with a confocal microarray associated with a well-characterized protein, thereby identify-
scanner. These investigators were able to obtain differential ing a novel function for that protein.
expression profiles, with radiation-induced up-regulation of
apoptotic regulators, such as p53, DNA fragmentation factors,
and tumor necrosis factor-related ligand.
Protein-Drug Interactions
In order to increase affinity and specificity, analytical micro- Protein microarrays also have great potential for drug dis-
arrays usually employ a signal amplification system and sand- covery and the identification of drug targets. Because the
wich assay format, in which the first antibody is spotted on binding profile of a drug of interest can be simultaneously
the array and then a captured antigen on the chip is detected obtained across an entire proteome using this approach, the
with a second antibody that recognizes a different part of the specificity or side effects of a drug can be monitored. This
antigen (Figure 3). A highly sensitive antibody microarray sys- information should also provide important clues about how
tem combining both methods has been shown to be capable to improve drug design (19). To demonstrate that protein
of simultaneously detecting 75 cytokines with high specific- microarrays can be used to identify drug targets, Huang and
ity, femtomolar sensitivity, a 3-log quantitative range, and coworkers (20) probed yeast proteome chips with biotinylated
economy of sample consumption (17). Although the sandwich small-molecule inhibitors of rapamycin (SMIRs) to find genetic
format dramatically increases the specificity of the antigen modifiers of the target of rapamycin (TOR) signaling network.
detection, it requires at least two high-quality antibodies for They identified candidate drug targets of the SMIRs and vali-
each antigen that is to be detected. dated a previously unknown protein as the bona fide target
of the SMIRs. Interestingly, in an independent study, Claudio
De Virgilio and colleagues (21) also identified the same pro-
tein in the TOR signaling pathway using a different approach.

Vol. 40, No. 4 (2006) BioTechniques 425

Protein (Domain)-Peptide Interactions overexpressed in many human cancers, the authors suggested
that this potential for increased promiscuity might contribute
In a most recent report, Jones and colleagues (3) demonstrated to the oncogenic potential of receptor tyrosine kinases.
that they could measure quantitative interactions between
proteins and peptides in an array format. They cloned, expressed,
Identification of Kinase Substrates on Protein Chips
and purified almost all the human Src homology 2 (SH2) and
phosphotyrosine binding (PTB) domains. The total 159 proteins Since phosphorylation is known to be involved in almost
were then printed on the aldehyde-modified glass substrates, every aspect of cell processes, identification of the down-
and 61 peptides representing physiological sites of tyrosine stream substrates of protein kinases is a critical step toward
phosphorylation on the four ErbB receptors were incubated understanding the effects of phosphorylation on protein
with the protein chips. For quantitative measurement, eight functions. To demonstrate that the protein chip approach is
concentrations of each peptide, ranging from 10 nM to 5 suitable for such investigations, Zhu and coworkers (22) first
µM, were used in the assay, allowing the binding affinity of analyzed the substrate specificity of 119 yeast kinases on 17
each peptide to be measured. With this microarray, 43 of the different substrates using nanowell protein chips. Recently,
65 previously reported interactions were detected, and 116 as an extension of the same idea but on a much larger scale,
new interactions were identified. Also, ErbB1 and ErbB2 were the so-called “Phosphorylome Project” was tested using the
found to become more promiscuous with increasing concen- yeast proteome microarrays (23). The goal was to identify all
tration, whereas ErbB3 did not. Because ErbB1 and ErbB2 are the potential protein substrates of each yeast kinase. In vitro
kinase reactions were carried out on the yeast proteome chips
using 87 individually purified kinases/kinase complexes in the
presence of [33P]ATP. Phosphorylation events (4129), involv-
ing 1325 different proteins, were identified. To ensure that
the signals resulted from phosphorylation events, 5% sodium
dodecyl sulfate (SDS) was used to denature proteins on chips
to remove signals from binding of kinase proteins or [33P]ATP.
Those phosphorylation results have been assembled into a
first-generation, global kinase signaling network in yeast.

Profiling Immune Responses

The microarray-based identification of the autoantigens
targeted by autoantibodies during the immune response has
considerable potential for use in diagnosis, classification, and
prognosis (24). Robinson and colleagues (25) published the
first report of the simultaneous analysis of multiple human
Figure 3. A sandwich assay format. A multivalent antigen is first caught
disease sera via protein microarray. They arrayed 196 distinct
by a capture antibody immobilized on the surface and then detected by a
detection antibody. The label is usually tagged on the detection antibody biomolecules involved in eight distinct human autoimmune
and can be further amplified. diseases, including proteins, peptides, enzymes complexes,
ribonucleoprotein complexes, DNA, and posttranslationally
modified antigens, onto glass slides to form the autoantigen
microarrays. These arrays were incubated with
patient serum samples as a means of defining
the pathogenesis of autoantibody responses in
human autoimmune diseases. Recently, Cahill
and colleagues (24) constructed a protein
array consisting of polypeptides translated
from 37,200 random human cDNA clones in
Escherichia coli and used this array to identify
Protein Lipid DNA Drug potential autoantigens involved in the patho-
genesis of alopecia areata. Eight autoantigens
were identified and successfully confirmed
by Western blot analysis. Likewise, a human
protein chip containing 2413 nonredundant
human fusion proteins was constructed for
serum profiling and antibody screening by the
same group (26).
Zhu and coworkers (27) also fabricated protein
chips that allowed them to rapidly and sen-
Kinase Antibody Serum ATP sitively distinguish the immune responses of
Figure 4. Examples of different assays on functional protein chips. Different types severe acute respiratory syndrome (SARS)-in-
of biochemical assays were carried out on chips, including assays of (A) protein-protein, fected and healthy people. These protein chips
(B) protein-lipid, (C) protein-DNA, (D) protein-drug, (H) protein-small molecule, and (F) harbored all the SARS-coronavirus (CoV) pro-
protein-antibody interactions. The chips can also be used to monitor immune responses in teins as well as proteins from five additional
patients (G) and posttranslational modifications of proteins, such as phosphorylation (E). corona viruses that can infect humans (HCoV-
These assays achieved high signal-to-noise ratios and were very informative for elucidating 229E and HCoV-OC43), cows [bovine corona-
the function of previously uncharacterized genes.
Vol. 40, No. 4 (2006) BioTechniques 427
virus (BCV)], cats [feline infectious peritonitis virus (FIPV)], 11. Haab, B.B. 2001. Advances in protein microarray technology for
and mice [mouse hepatitis coronavirus (MHVA59)]. The pres- protein expression and interaction profiling. Curr. Opin. Drug Discov.
ence of human immunoglobulin G (IgG) and immunoglobulin Devel. 4:116-123.
M (IgM) antibodies against SARS-CoV was detected on the 12. Cahill, D.J. 2001. Protein and antibody arrays and their medical
chips with labeled anti-human IgG and IgM antibodies. Sera applications. J. Immunol. Methods 250:81-91.
from patients could quickly be clustered as SARS positive or 13. Templin, M.F., D. Stoll, M. Schrenk, P.C. Traub, C.F. Vohringer,
SARS negative on the basis of the serum-probing signals. and T.O. Joos. 2002. Protein microarray technology. Trends
Comparison to other methods [e.g., enzyme-linked immuno- Biotechnol. 20:160-166.
sorbent assay (ELISA) and immunofluorescence assay (IFA)]
14. Stoll, D., M.F. Templin, M. Schrenk, P.C. Traub, C.F. Vohringer,
indicated that the origin of 94% of the sera could be correctly and T.O. Joos. 2002. Protein microarray technology. Front. Biosci.
predicted on the microarrays. The chip-based assay was at 7:c13-c32.
least 100-fold more sensitive than the ELISA/IFA assays and
15. Haab, B.B., M.J. Dunham, and P.O. Brown. 2001. Protein
required a smaller amount of sample.
microarrays for highly parallel detection and quantitation of
specific proteins and antibodies in complex solutions. Genome Biol.
Conclusion 2:RESEARCH0004.
Protein microarray technology has been shown to be a use- 16. Sreekumar, A., M.K. Nyati, S. Varambally, T.R. Barrette, D.
ful tool for multiplexed detection and proteomics studies. Ghosh, T.S. Lawrence, and A.M. Chinnaiyan. 2001. Profiling of
Femtomolar sensitivity has been achieved in analytical protein cancer cells using protein microarrays: discovery of novel radiation-
microarrays, and the number of applications of functional regulated proteins. Cancer Res 61:7585-7593.
protein microarrays has grown dramatically. It appears that 17. Schweitzer, B., S. Roberts, B. Grimwade, W. Shao, M. Wang, Q.
protein microarrays will prove to be one of the most powerful Fu, Q. Shu, I. Laroche, et al. 2002. Multiplexed protein profiling on
tools in the field of diagnostics and high-throughput biology. microarrays by rolling-circle amplification. Nat. Biotechnol. 20:359-
Improvements in our ability to generate large sets of high- 365.
quality proteins and antibodies or their mimetics will play a 18. Hall, D.A., H. Zhu, X. Zhu, T. Royce, M. Gerstein, and M. Snyder.
key role in quantitative analysis and promote the extension 2004. Regulation of gene expression by a metabolic enzyme. Science
of this technology to other model organisms. 306:482-484.
19. Huang, Y.H., D. Li, A. Winoto, and E.A. Robey. 2004. Distinct
Acknowledgments transcriptional programs in thymocytes responding to T cell recep-
We thank the financial support from the National Institutes of tor, Notch, and positive selection signals. Proc. Natl. Acad. Sci. USA
Health (U54RR020839-01). 101:4936-4941.
20. Huang, J., H. Zhu, S.J. Haggarty, D.R. Spring, H. Hwang, F. Jin,
References M. Snyder, and S.L. Schreiber. 2004. Finding new components of
the target of rapamycin (TOR) signaling network through chemical ge-
1. Zhu, H., M. Bilgin, R. Bangham, D. Hall, A. Casamayor, P. netics and proteome chips. Proc. Natl. Acad. Sci. USA 101:16594-16599.
Bertone, N. Lan, R. Jansen, et al. 2001. Global analysis of protein
21. Dubouloz, F., O. Deloche, V. Wanke, E. Cameroni, and C. De
activities using proteome chips. Science 293:2101-2105.
Virgilio. 2005. The TOR and EGO protein complexes orchestrate mi-
2. MacBeath, G. and S.L. Schreiber. 2000. Printing proteins as croautophagy in yeast. Mol. Cell 19:15-26.
microarrays for high-throughput function determination. Science
22. Zhu, H., J.F. Klemic, S. Chang, P. Bertone, A. Casamayor, K.G.
289:1760-1763.
Klemic, D. Smith, M. Gerstein, et al. 2000. Analysis of yeast pro-
3. Jones, R.B., A. Gordus, J.A. Krall, and G. Macbeath. 2006. A tein kinases using protein chips. Nat. Genet. 26:283-289.
quantitative protein interaction network for the ErbB receptors using
23. Ptacek, J., G. Devgan, G. Michaud, H. Zhu, X. Zhu, J. Fasolo, H.
protein microarrays. Nature 439:168-174.
Guo, G. Jona, et al. 2005. Global analysis of protein phosphoryla-
4. Delehanty, J.B. 2004. Printing functional protein microarrays using tion in yeast. Nature 438:679-684.
piezoelectric capillaries. Methods Mol. Biol. 264:135-143.
24. Lueking, A., O. Huber, C. Wirths, K. Schulte, K.M. Stieler, U.
5. Delehanty, J.B. and F.S. Ligler. 2003. Method for printing func- Blume-Peytavi, A. Kowald, K. Hensel-Wiegel, et al. 2005. Profil-
tional protein microarrays. BioTechniques 34:380-385. ing of alopecia areata autoantigens based on protein microarray tech-
6. Kusnezow, W., A. Jacob, A. Walijew, F. Diehl, and J.D. Hoheisel. nology. Mol. Cell. Proteomics 4:1382-1390.
2003. Antibody microarrays: an evaluation of production parameters. 25. Robinson, W.H., C. DiGennaro, W. Hueber, B.B. Haab, M. Ka-
Proteomics 3:254-264. machi, E.J. Dean, S. Fournel, D. Fong, et al. 2002. Autoantigen
7. Stillman, B.A. and J.L. Tonkinson. 2000. FAST slides: a novel microarrays for multiplex characterization of autoantibody responses.
surface for microarrays. BioTechniques 29:630-635. Nat. Med. 8:295-301.
8. Kramer, A., T. Feilner, A. Possling, V. Radchuk, W. Weschke, L. 26. Lueking, A., A. Possling, O. Huber, A. Beveridge, M. Horn, H.
Burkle, and B. Kersten. 2004. Identification of barley CK2alpha Eickhoff, J. Schuchardt, H. Lehrach, et al. 2003. A nonredundant
targets by using the protein microarray technology. Phytochemistry human protein chip for antibody screening and serum profiling. Mol.
65:1777-1784. Cell. Proteomics 2:1342-1349.
9. Angenendt, P., J. Glokler, D. Murphy, H. Lehrach, and D.J. 27. Zhu, H., S. Hu, G. Jona, X. Zhu, N. Kreiswirth, G. Liu, Q. Song,
Cahill. 2002. Toward optimized antibody microarrays: a comparison P. Chen, et al. 2006. Severe acute respiratory syndrome diagnostics
of current microarray support materials. Anal. Biochem. 309:253-260. using a coronavirus protein microarray. Proc. Natl. Acad. Sci. USA (In
10. Charles, P.T., E.R. Goldman, J.G. Rangasammy, C.L. Schauer, press).
M.S. Chen, and C.R. Taitt. 2004. Fabrication and characterization of To purchase reprints
3D hydrogel microarrays to measure antigenicity and antibody func-
tionality for biosensor applications. Biosens. Bioelectron. 20:753-764. of this article, contact
Reprints@BioTechniques.com
Vol. 40, No. 4 (2006) BioTechniques 429