Printer Friendly

Antibody arrays: technical considerations and clinical applications in cancer.

Two main proteomic strategies, targeted or untargeted, can be used to investigate the cancer proteome. The terminology refers to whether the proteins to be measured are known and considered in the experimental design (targeted) and the number of proteins that can be detected and characterized (decided at front in targeted approaches). The main features of the proteomic techniques are summarized in Table 1 (1-7). Untargeted architecture platforms are best-suited for first-pass comparisons of proteomes to identify relatively few, novel, and known proteins that exhibit the greatest differences in abundance. The 2 most commonly used untargeted technologies are 2-dimensional electrophoresis and low- and high-resolution mass spectrometry (2-5). Targeted architecture proteomic platforms suit analyses of quantitative differences in abundance among known protein families and pathways. The versatility of targeted platforms allows control and estimation of the reproducibility, scalability, and quantification precision, leading to high sensitivity and coverage. This approach allows experimental designs to address specific hypotheses and enables biological interpretation of the results. However, the number of proteins amenable for these analyses depends on the availability of antibodies with high affinity and specificity to bind a target protein. The main targeted techniques used for large-scale analysis of many samples and proteins include tissue arrays, multiplexed Western blots, and protein microarrays. Protein arrays represent the most versatile among the proteomics techniques available, because they allow immobilization of antigens, peptides, complex protein solutions, or antibodies to capture and quantify specific antibodies or proteins (1, 6, 7).

Antibody and Protein Array Formats

Innovation in immobilization surfaces and detection strategies has increased the number of planar arrays and bead-based technologies. Planar antibody arrays are the most common type of protein arrays, and they are the major focus of this review. This section describes the main formats of planar arrays and the differences between planar arrays and bead-based assays (Fig. 1).

The main planar label-based assays are 1-antibody assays, which use 1 antibody to capture the target molecule, and sandwich assays, which use 2 antibodies to capture the target protein (1, 6, 7). One-antibody and sandwich assays both have advantages and pitfalls. In 1-antibody label-based assays, the targeted proteins are captured by an immobilized antibody and detected through labeling with a tag (Fig. 1A). In direct labeling, proteins are labeled with a fluorophore such as cyanine (Cy3 or Cy5) (1). In indirect labeling, proteins are labeled with a tag that is later detected by a labeled antibody. One-antibody label-based assays allow the incubation of 2 different samples, each labeled with a different tag on the arrays. These types of assays, therefore, allow the use of a reference sample that is coincubated with a test sample and facilitates normalization (1, 6, 7). Another advantage of these types of assays is that they are competitive-the analytes in the test and reference solutions compete for binding at the antibodies (1, 6, 7), leading to improvement in linearity of response and dynamic range compared with noncompetitive assays (7). The main disadvantage of these types of assays is the disruption of the analyteantigen interaction by the label, which may limit detection as well as sensitivity and specificity.


In the sandwich label-based format, immobilized antibodies capture unlabeled proteins, which are detected by another antibody, with the signal for detection generated by several methods (Fig. 1B). The use of 2 antibodies targeting each analyte confers greater specificity than label-based assays. The reduced background of these assays also increases the detection limit. The sandwich format allows only noncompetitive assays, because only 1 sample can be incubated on each array (1, 6, 7). Non-competitive assays have sigmoidal binding responses, which are linear in competitive formats, and require standard curves of known concentrations of analytes to achieve accurate calibration of concentrations (7). Sandwich assays are more difficult to develop in a multiplexed manner than label-based assays, because matched pairs of antibodies and purified antigens may not be available for each target, and the potential cross-reactivity among detection antibodies increases with additional analytes (6, 7). The size of multiplexed sandwich assays is limited to 30 to 50 different targets (1, 6, 7), in contrast to 1-antibody assays, for which only the availability of antibodies and the space on the substrate limit the number of targets analyzed.

Suspension or bead-based arrays use different fluorescent beads. Each bead is coated with a different antibody, and all beads are spectrally resolvable from each other (8-12). The beads are incubated with a sample to allow protein binding to the capture antibodies, and the mixture is incubated with a cocktail of detection antibodies, each corresponding to one of the capture antibodies. The detection antibodies are tagged to allow fluorescent detection. The beads are passed through a flow cytometer system, and each bead is probed by 2 lasers, 1 to read the color or identity of the bead, and another 1 to read the amount of detection antibody on the bead (8-12). Multiplexed bead-based flow-cytometey assays represent an active area of development. Differentially identifiable beads coated with proteins, autoantigens, or antibodies use a cytometer system to identify a variety of bound antibodies or proteins (8-12). Advances in instrumentation and bead chemistries will probably make this approach valuable for the detection of circulating cancer cells in clinical practice. In another version of this method, suspensions of cells are incubated on antibody arrays, and the number of cells that bind each antibody is quantified by dark-field microscopy. These arrays enable characterization of multiple membrane proteins in specific cell populations or changes in cell surfaces induced by drug therapies.

Other antibody and protein array approaches are modifications of 1-antibody and sandwich label-based arrays. These alternative strategies of protein arrays allow detection of proteins on whole cells without protein isolation (Fig. 1C) (6,12). A growing area of cancer research that uses protein arrays on serum specimens entails the development and design of tumor-associated antigen (TAA) arrays to enhance detection of autoantibodies against TAAs for cancer diagnosis (Fig. 1D). The rationale is related to the presence of antibodies in the cancer sera that react with a unique group of autologous cellular antigens or TAAs (13,14). Complex protein extracts can also be spotted onto membranes and probed with antibodies targeting specific proteins on the so-called reverse-phase arrays (15,16) (Fig. 1E). Proteins in suspension can also be detected by use of bead arrays (Fig. 1F) (8-12).

Technical Issues Using Planar Antibody Arrays


The main bottleneck to the development of highly multiplexed antibody arrays is the requirement for ligands with specific affinity for each analyte. Polyclonal antibodies may have higher background and lower specificity and detection limit than monoclonal antibodies. For sandwich arrays, monoclonal/polyclonal pairs are more readily available than matched monoclonal antibody pairs targeting different epitopes of a given protein. In vitro selection of antibodies by phase ribosome or mRNA display technologies and engineered binding molecules have increasingly important roles in generating ligands with specific affinity for analytes, for which antibodies are unavailable (17). A novel strategy to produce specific antibodies has been validated, optimizing the design of protein subfragments of a selected size with a minimal sequence similarity to other proteins. The fragments are selected with an alignment scanning procedure based on a principle of lowest sequence similarity to other human proteins, optimally to generate antibodies with high selectivity (18).

Because antibodies cannot be manufactured with known affinity and specificity, it is advisable to validate the specificity and sensitivity of each antibody before used as a probe for protein arrays. Identification of a single band at the specified molecular weight on Western blotting and immunoprecipitation followed by mass spectrometry, is a common strategy for validating the specificity and sensitivity of the proposed antibody (1,12). Recombinant antigens can be used as positive and negative controls for printing (depositing the antibodies onto the slides), calibration, and detection (1, 6,12).

The linearity range of the assay depends on the antibody-antigen affinity. Linearity can be achieved only when the concentration of the analyte and antibody matches the affinity constant. When antibody arrays are used, it is advisable to include dilution and recovery experiments evaluating the specificity and affinity of the antibodies for their ligands. Multiplexed formats that include multiple antibodies with various affinities might not achieve high linearity for all analytes under study. In sandwich assays, it is necessary to have 2 distinct antibodies that are able to bind 2 available nonoverlapping epitopes on the same protein. The use of 2 different affinity constants for each tagging reaction doubles the stringency of the detection system (6,12).


It is necessary to immobilize antibodies in such a way that the functional component will be efficiently deposited without interfering with the subsequent binding. Conditions such as humidity, temperature, dust, and pin washing should also be stringently controlled during the printing step. Various immobilization and detection strategies are devised according to which target molecules are to be measured and which are used to capture them. The attributes of an ideal substratum for antibody arrays include limited nonspecific binding, high surface area-to-volume ratio, inertness to biological molecules, minimal autofluorescence, and compatibility with available detection methods. A variety of surfaces and immobilization chemistries have been described for antibody arrays. Derivatized supports in which capture antibodies are immobilized include surfaces such as polyvinylidene difluoride, nitrocellulose, agarose, polyacrylamide, and hydrogels. Glass slides are frequently coated with 1-, 2-, or 3-dimensional structured surface modifications that can be activated with aldehyde, polylysine, or a homofunctional cross-linker as part of the initial optimization experiments (6,12,19). The advantages of using a distinct coating or surfaces under different blocking, pH buffering, or ultraviolet cross-linking conditions for specific applications have been described (19). Glass slides coated with silane or acrylamide hydrogel can provide good reproducibility from day to day, efficient immobilization of antibodies, and low background when used in conjunction with fluorescence detection. Once the antibodies are immobilized, it is necessary to block nonspecific protein-binding sites on the printed microarrays. Suitable blocking solutions include diluted bovine serum albumin and casein solutions (1, 6,12,15,16). The use of bead-based arrays makes it possible to suppress nonspecific serologic binding by preincubating serum specimens with background inhibitors, such as polyvinylalcohol and polyvinylpyrrolidone and others. These treatments can considerably decrease nonspecific background that affects Luminex assays (11).



Several labeling and detection methods can be used for 1-antibody and sandwich label-based planar arrays (Fig. 2). The signal can be generated by a fluorescently labeled detection antibody (Fig. 2A), the approach used in typical sandwich arrays that requires chemical labeling of all secondary detection antibodies. The assay is a simple 2-step procedure that does not require a separate staining step (20, 21).

The 2nd possibility is the use of a species-specific fluorescently labeled tertiary antibody (Fig. 2B). This option avoids the use of large, chemically modified, detection antibodies but limits the species of the capture antibodies. The 3rd option uses available biotinylated detection antibodies (Fig. 2C) (21). In these assays, detection occurs after staining the sandwich complex with Cy3-labeled streptavidin or other streptavidin variants such as Texas Red conjugates or streptavidin-R-phycoerythrin (SAPE) (21). The 4th possibility is based on further amplification of the fluorescent signal by use of a second layer of SAPE coupled to the 1st layer via an anti-SAPE antibody (Fig. 2D). Alternatively, in the 5th option, the number of biotin labels can be increased via tyramide signal amplification (Fig. 2E) (6). An antibiotin horseradish peroxidase (HRP) will generate a tyramide radical that cross-links a biotin or a fluorophore to all exposed tyrosine residues of any protein near the recognition event (6).

Chemiluminesce, the 6th possibility, can be used in multiplexed sandwich assays (Fig. 2F) with a streptavidin-horseradish peroxidase (HRP) or a species-specific antibody conjugated with HRP or alkaline phosphatase and chemiluminescence substrates. Chemiluminescence is typically more sensitive than standard fluorescence. A polymer decorated with streptavidin and europium chelates is used not only for microplate but also for microarray measurements. Evanescence wave-guide can be used as an alternative to ultrasensitive fluorescence (22). Rolling circle amplification can be applied as the 7th option for signal generation (Fig. 2G). The 5' end of an oligonucleotide primer is attached to an antibiotin antibody (23). After the antibiotin antibody is bound to the biotinylated detection antibody of the sandwich, the oligonucleotide is enzymatically extended with a circular DNA sequence as template. Fluorescently labeled short oligos are then hybridized to the extended DNA, decorating each bound antibody with thousands of fluorophores (21). An alternative 8th staining method, with sensitivity similar to evanescence wave technology and rolling-circle amplification, involves the use of colloidal gold particles coated with an antibiotin antibody (24). Because of resonance light scattering (RLS), these particles scatter white light very intensely, and quantitative readouts of miniaturized sandwich assays can be obtained with a simple charge-coupled device camera-based imaging system (Fig. 2H) (24). Unlike fluorescence and chemiluminescence, resonance light scattering does not photobleach particles (17, 20-24).


The main detection methods used for antibody arrays include radioactivity, fluorescence, and chemiluminescence. Radioactivity is not frequently used because of safety concerns and its longer exposure times (up to 10 h). Fluorescence is one of the most frequently used detection methods. Fluorophores, like chromogens, are present in many formulations and have defined emission spectra. Fluorescenn, rhodamine (Texas Red), phycobiliproteins, nitrobenzoxadiazole (NBD), acridines, Cy3, Cy5, and bodipy compounds are commonly used for protein labeling (20, 21). Selection of fluorophores for use with microarrays depends on sample type, substrate, emission characteristics, and the number of analytes to be assayed. Because of the inherent autofluorescence of some materials, which considerably decreases the signal-to-noise ratios, not all substrates are compatible with fluorescent detection strategies (19). Nitrocellulose-coated slides cause light scatter and higher background than aldehyde-treated slides do, and laser scanner detection methods limit the use of nitrocellulose substrata for fluorescent detection methods (19). The sample may also have components that interfere with a selected fluorophore. Flavoproteins autofluoresce, emitting light in the same region as fluorescein and limiting the use of this fluorophore in samples rich in flavoproteins (e.g., liver and kidney tissues). Photobleaching and quenching of fluorophores can decrease the total signal observed on an array. Because they overcome these effects, the Cy3 and Cy5 dyes are commonly used for fluorescent detection. These dyes are suitable for fluorescence detection because of their decreased dye interactions, increased brightness, and the ability to add charged groups to the molecules (19-21). Fluorescent-tagged proteins, including antibodies, can be used in both indirect and sandwich strategies to detect immobilized molecules on a microarray. Applied to high-throughput screening for drug discovery, fluorescent labels, such as Cy3, can also be coupled to enzyme inhibitors such as fluorophosphonate, vinyl sulfone, and phosphatase, and used for enzyme classification and testing of potential inhibitors. Streptavidin-biotin amplification chemistries can also be applied to fluorescence detection strategies (20, 21). Fluorescence detection can be coupled to signal amplification, such as rolling circle amplification, to provide sufficient sensitivity for most applications (20, 21, 23).

Chemiluminescent detection methods are based on Western blotting protocols for detection of antigen-bound antibodies with secondary antibodies conjugated to alkaline phosphatase or HRP (20, 21). Chemiluminescence is highly sensitive and can be applied to any of the label detection methods, but it may pose limitations because of its dynamic range and compatibility with multiplexing. Amplification strategies such as biotinyl-tyramide can be applied to chemiluminescence. A useful application consists of total protein determinations made directly on arrays using a ruthenium organic complex, which interacts noncovalently with proteins immobilized on nitrocellulose (25). The dye is applicable to arrays printed on nitrocellulose membranes. This type of total protein analysis is useful for minute sample volumes in which a protein spectrophotometric analysis would not be feasible.

In the field of bio- and chemosensors, many detection principles have been developed based either on the observation of fluorescence labeled systems or on direct optical detection in the heterogeneous phase. Newly developed fluorescent systems, such as quantum dot technology, together with immunofluorescent semiconductor materials and continued research in nanotechnology, are key to optimizing microarray detection strategies for subcellular multianalyte analyses (26, 27). Direct optical detection can be determined by measuring remission (absorption of reflected radiation), microrefractivity, or interference. In the latter case of optical interference, either interferometers or measurement of changes in the physical thickness of the layer (measuring microreflectivity), for example those caused by swelling effects in polymers (from interaction with analytes) or in bioassays (from affinity reactions), also play an important role (27). Examples of interference arrays are optically coated silicon wafers that capture antibodies adsorbed to their surface. Secondary antibodies can be conjugated to HRP, which catalyzes the formation of a thin film on the surface in the presence of a precipitating substrate. The film changes the color of the microarray by altering the interference pattern of reflected light. This assay, which represents the concept of biosensors as low multiplexed antibody arrays (26, 27), yields qualitative results that can be determined visually and quantitative results that can be obtained with a charge-coupled device camera. An alternative to the optical interference-based array is the use of surface plasmon resonance imaging to develop label-free protein arrays. SPR imaging makes it possible to fabricate arrays with controlled surface density of antigens and provides qualitative and quantitative evaluation of antibody binding in a label free format (28).


To efficiently and reproducibly measure multiple proteins simultaneously with high sensitivity, specificity, and quantitative accuracy over large concentration ranges, it is necessary to consider quality control issues in the design of the arrays (1, 6,12). Optimal assessment of the technology through filtering and data analyses procedures will later address the linearity, calibration, and specificity of the antibodies, as well as whether labeling and/or hybridization protocols are optimized adequately to ensure high signal-to-noise ratios (29). The experimental designs should include replicates and controls to evaluate the infra- and interassay reproducibility of the measurements, and they should include means for normalization of the infra- and interassay results (29-31). The array should also include appropriate means to test for the presence of potential antibody interference and cross-reactivity. In this regard, the quantity of antibody spotted can be used to standardize the antigen concentration. It is possible to use an internally controlled system in which one color represents the amount of antibody spotted and the other color represents the amount of the antigen used to quantify protein expression. This normalization for antibody spot intensity can decrease variability and lower the limits of detection of antibody arrays (31). Such strategy allows scaling up the technology for high-throughput screening for hundreds of proteins in complex biofluids such as blood (31).


The establishment of a filtering process to assess the quality of the data is a critical step in using quantitative data obtained through antibody arrays. The conceptual similarity of label-based antibody arrays and 2-color competitive detection genomic arrays has allowed the application of normalization and data analysis tools classically used for cDNA arrays to be used for protein profiling with antibody arrays (29). The first rank of quality control deals with the experimental design of the printing of the antibody arrays, which should include various replicated spots dispersed along the complete surface of the array as well as controls in every experiment (1,6,12). The initial control of the scanned data is at the spot level, performed with the scanner software, e.g., GenePix (29). The created customized report can be used to analyze the quality of the spots, and spots of low quality can be flagged. The criteria to flag the spots may include the SDs away from background, the [R.sup.2], or the percentage saturation (29). At the array level of comparison, the quality control of the data includes normalization of the array, as well as calculation of mean and SD of the intensities of each antibody in its various replicates along the slide (29-32). Spots with high SDs between replicated spots can be filtered out. Normalization of the arrays can be performed with the mean intensity of each array (30), protein standards such as IgG (1, 32, 33), or internal controls based on antibody spot intensity (31).

In the next stage of filtering the data, each experiment set is compared, and results are calibrated to a dilution series of the antibodies by a best-fit line, removing data with high variability. The results can also be correlated to independent measurements obtained through ELISAs to quantify targets included in the antibody arrays. At this step, if the series for an antibody is bad, the antibody can be flagged. It is possible to set thresholds of expression for an antibody, specifying a maximum and minimum ratio for spots to be considered in further analyses (29). This is a critical step because it filters the input data based on the SD between replicate spots and the output data based on the SD of dilution experiments. The last stage of quality control is to compare independent sets of experiments based on internal controls that will allow comparison between experiments performed on different days. The combined use of unsupervised and supervised methods can identify protein patterns associated with disease progression and clinical outcome.

Applications in Cancer Research with Various Specimens


For cancer diagnostics, direct-labeling methods have been applied to the detection of proteins in the serum of patients with prostate cancer (32). The use of a 2-color rolling circle amplification method improves the detection of low-quantity proteins. This method has also provided adequate reproducibility and accuracy for protein profiling on serum specimens and clinical applications (23, 33, 34). Sandwich assays can also measure proteins in body fluids by using detection methods such as RLS (35), enhanced chemiluminescence (36), tyramide signal amplification (37), and fluorescence (38). Reverse protein arrays have also been optimized for use with spot serum specimens and to obtain high-throughput measurement of IgA in thousands of sera from a single experiment (39). Antibody arrays for bladder cancer have been designed by selecting antibodies against targets differentially expressed in bladder tumors identified by gene profiling (40). Serum protein profiles obtained by 2 independent antibody arrays provided comprehensive tools for bladder cancer diagnosis and clinical outcome stratification (40). Validation analyses with ELISA and immunohistochemistry on tissue microarrays confirmed the relevance of identified proteins for tumor progression. This strategy provides experimental evidence for the use of several integrated technologies, strengthening the process of biomarker discovery.

Serum specimens can be used to profile the humoral immune signature of cancer patients for detection of both auto-antibodies against tumor antigens and secreted cytokines. The combined detection of antibodies against a group of TAAs has provided high sensitivity for the diagnosis of prostate cancer (13). The use of phage display arrays can enhance the tumor subtype specificity of such measurements (13,14). Cytokine profiling of serum and plasma specimens is one of the most widely reported applications of protein array technology for autoimmune diseases and, to a lesser extent, neoplastic diseases. Studies in clinical materials and in vitro systems have revealed the potential of cytokine profiling with antibody arrays for characterizing hematologic neoplasias (8-10, 38). Cytokine profiles can support differentiation of cancer patients from controls and stratification of patients with leukemia based on clinical outcome. Several reports have also compared the reproducibility and differences among the several technologies available for multiplexing cytokine measurements, including planar and bead-based antibody arrays (8-10).

The interstitial fluid that perfuses the tumor environment has also been used for protein profiling with antibody arrays, not only for biomarker discovery of tumor-secreted proteins for tumor diagnosis but also for the development of targeted therapies, as reported for breast cancer (41). Interestingly, labeling and hybridization methods have been optimized for multiple protein detection on specimens of cerebrospinal fluid, characterized by low protein concentrations. This approach would have diagnostic applications in cancer and infectious diseases (42). Noninvasive body fluids such as saliva, sputum, and urine can be used for clinical application of antibody arrays, but labeling and hybridization protocols must be optimized to the sensitivities required for such specimens. The development of diagnostic and prognostic markers by using these samples may provide improved, noninvasive approaches to clinical management of cancer patients.


In addition to diagnostic, prognostic, and response-to-therapy studies, multiplexed studies of intracellular or secreted proteins with differential expression in tumor specimens are applications of antibody arrays. The use of antibody arrays to measure protein concentrations, phosphorylation states, and signaling pathways is another approach to characterizing the biology underlining tumorigenesis and progression (16, 43).

Comprehensive gene profiling analyses can be used to identify tumor targets relevant to specific neoplasias for antibody-array design (40). Such approach can be applied in antibody-based proteomics to generate protein-specific affinity antibodies to functionally explore the human proteome. Specific protein epitope signature tags (PrEST) can be identified and used to raise monospecific, polyclonal antibodies and can be subsequently analyzed on paraffin-embedded sections of malignant and healthy tissue. Genome-based affinity proteomics, using antibodies induced by PrEST, is an efficient way to rapidly identify several disease-associated protein candidates of known and unknown identity (44). A descriptive and comprehensive protein atlas for tissue distribution and subcellular localization of human proteins in both healthy and cancer tissues is being created (45). The antibodies generated can be used to analyze corresponding proteins in a wide range of assay platforms, including: (a), immuno-histochemistry for detailed tissue profiling; (b), specific affinity reagents for various functional protein assays; and (c), capture (pull-down) reagents for purification of specific proteins and their associated complexes for structural and biochemical analyses (45).

It is possible to study protein profiles of frozen resected tumor specimens using antibody and reversed-phase protein arrays. Laser capture microdissection can be used to remove selected tumor and/or stromal counterparts within a tissue specimen (16, 43, 46). This is relevant in heterogeneous tumors such as breast or prostate neoplasias (43, 46). In analyses, using isolated proteins from tissue specimens, validation studies may include not only Western blotting but also immunohistochemical analyses (31, 40). The use of 2-color, comparative fluorescence assays, such as Cy3 and Cy5, allows comparison of 2 disease status determinations, enabling clinical applications for biomarker discovery, for example by comparing protein profiles of tumor specimens with their respective healthy counterparts, as has been reported for breast cancer (43).

Protein analyses with antibody arrays allows monitoring of multiple cell-signaling endpoints and thus mapping of specific cellular networks to reverse-phase protein arrays by use of a parallel approach, as described for prostate cancer (46). Changes in phosphorylation status or cleaved states of key signaling proteins can be evaluated with antibody arrays. It is possible to test whether a pathway might become blocked by chemotherapeutic agents. Analyses of these pathways might reveal relevant information for designing individual targeted therapies and/or combinatorial strategies directed at multiple nodes in a cell-signaling cascade. The use of strategy with protein extracts from tumors may be used to test and predict responses to novel drug therapies (16).


In vitro studies using antibody arrays allow in-depth analyses of cancer biology. Two-color comparative fluorescence analyses can compare protein changes of 2 experimental conditions associated with disease progression or drug effects (43, 44). Cytokine profiles of cell lysates have also been compared with those obtained from body fluids and tissue extracts (47). The use of antibody arrays for high-throughput profiling of cultured cells can be useful to evaluate signaling pathways including tyrosine kinase networks (45). Forward and reverse arrays can assess the presence of phosphorylated and unphosphorylated forms of proteins, if adequate antibodies are available to address specific posttranslational modifications of target proteins (10,14, 46, 47). Antibody arrays can profile enzyme activities in protein extracts and cell culture supernatants and can provide a convenient platform to evaluate activity-based protein profiling with high sensitivity and specificity and less sample consumption than initial gel-based strategies for assessing the functional state of enzymes. Complex protein samples are treated with fluorescent activity-based probes, and the labeled enzymes are captured and detected on antibody arrays targeting the enzymes (48).


In summary, antibody arrays can be used for the following applications: (a) discovery of candidate disease biomarkers; (b) characterization of signaling pathways, disease progression, clinical subtypes, and outcomes; (c) measurement of changes in posttranslational modifications or expression of disease-related proteins; (d) identifying binding partners to proteins, especially in functional studies for drug discovery; (e) epitope mapping for determining regions of proteins that bind specific antibodies.

Multiple challenges that remain in the design and application of antibody arrays include (12, 53, 54): (i) poor understanding of protein immobilization; (ii) limited dynamic ranges of 2 or 3 orders of magnitude; (iii) lower accuracy and reproducibility than clinical immunoassays; (iv) molecular protein complexity and denaturation affecting immunoreactivity; (v) lack of standards and calibrators; and (vi) need for high-affinity and specific antibodies for target antigens. Such challenges are being addressed by the multi-institutional efforts of the Human Proteome Organization toward the standardization of critical variables in serum and plasma proteomic analyses. Initial studies provide guidance on preanalytic variables that can alter the analysis of blood-derived samples, including choice of sample type, stability during storage, use of protease inhibitors, and clinical standardization (53). As part of the Human Proteome Organization approach, it is also critical to standardize statistical strategies for high-confidence protein identification and data analysis. These efforts and strategies toward integrating proteomic datasets will lead to accurate and comprehensive representation of human proteomes (54).

In conclusion, the methods and applications of antibody arrays are increasing in scope and effectiveness. The use of several proteomic methods represents a strategy that not only provides complementary information but also produces added benefits for cancer diagnostics and characterization of the biology underlining tumor progression. Antibody arrays and new formats that may be developed in the near future are likely to markedly accelerate the rate of biomarker discovery and characterization of cancer-specific pathways that will eventually lead to the development of individualized therapies that take into account markers of disease predisposition and therapeutic response.

Received August 22, 2005; accepted June 5, 2006.

Previously published online at DOI: 10.1373/clinchem.2005.059592


(1.) Haab BB, Dunham MJ, Brown P0. Protein microarrays for highly parallel detection and quantitation of specific proteins and antibodies in complex solutions. Genome Biol 2001;2:RESEARCH0004.

(2.) Hortin GL, Jortani SA, Ritchie JC Jr., Valdes R Jr., Chan DW. Proteomics: a new diagnostic frontier. Clin Chem 2006;52:121822.

(3.) Gygi SP, Corthals GL, Zhang Y, Rochon Y, Aebersold R. Evaluation of two-dimensional gel electrophoresis-based proteome analysis technology. Proc Natl Acad Sci U S A 2000;97:9390-5.

(4.) Hortin GL. The MALDI-TOF mass spectrometric view of the plasma proteome and peptidome. Clin Chem 2006:52:1223-37.

(5.) Caldwell RL, Caprioli RM. Tissue profiling by mass spectrometry: a review of methodology and applications. Mol Cell Proteomics 2005;4:394-401.

(6.) Chan SM, Ermann J, Su L, Fathman CG, Utz PJ. Protein microarrays for multiplex analysis of signal transduction pathways. Nat Med 2004;10:1390-6.

(7.) Barry R, Diggle T, Terrett J, Soloviev M. Competitive assay formats for high-throughput affinity arrays. J Biomol Screen 2003;8:25763.

(8.) Pang S, Smith J, Onley D, Reeve J, Walker M, Foy C. A comparability study of the emerging protein array platforms with established ELISA procedures. J Immunol Meth 2005;302:1-13.

(9.) Lash GE, Scaife PJ, Innes BA, Otun HA, Robson SC, Searle RF, et al. Comparison of three multiplex cytokine analysis systems: Luminex, Searchlight, and FAST Quant. J Immunol Meth 2006; 309:205-8.

(10.) de Jager W, Rijkers GT. Solid-phase and bead-based cytokine immunoassay: a comparison. Methods 2006;38:294-303.

(11.) Waterboer T, Sehr P, Pawlita M. Suppression of non-specific binding in serological assays. J Immunol Methods 2006;309: 200-4.

(12.) Kingsmore SF. Multiplexed protein measurement: technologies and applications of protein and antibody arrays. Nat Rev Drug Discov 2006;5:310-21.

(13.) Wang X, Yu J, Sreekumar A, Varambally S, Shen R, Giacherio D, et al. Autoantibody signatures in prostate cancer. N Engl J Med 2005;353:1224-35.

(14.) Anderson KS, LaBaerJ. The sentinel within: exploiting the immune system for cancer biomarkers. J Proteome Res 2005;4:1123-33.

(15.) Nishizuka S, Charboneau L, Young L, Major S, Reinhold WC, Waltham M, et al. Proteomic profiling of the NCI-60 cancer cell lines using new high-density reverse-phase lysate microarrays. Proc Natl Acad Sci U S A 2003;100:14229-34.

(16.) Petricoin EF III, Bichsel VE, Calvert VS, Espina V, Winters M, Young L, et al. Mapping molecular networks using proteomics: a vision for patient-tailored combination therapy. J Clin Oncol 2005;23: 3614-21.

(17.) Stich N, Gandhum A, Matyushin V, Raats J, Mayer C, Alguel Y, et al. Phage display antibody-based proteomic device using resonance-enhanced detection. J Nanosci Nanotechnol 2002;2:37581.

(18.) Lindskog M, Rockberg J, Uhlen M, Sterky F. Selection of protein epitopes for antibody production. Biotechniques 2005;38:723-7.

(19.) Angenendt P, Glokler J, Murphy D, Lehrach H, Cahill DJ. Toward optimized antibody microarrays: a comparison of current microarray support materials. Anal Biochem 2002;309:253-60.

(20.) Espina V, Woodhouse EC, Wulfkuhle J, Asmussen HD, Petricoin EF III, Liotta LA. Protein microarray detection strategies: focus on direct detection technologies. J Immunol Methods 2004;290: 121-33.

(21.) Levit-Binnun N, LindnerAB, Zik 0, EshharZ, Moses E. Quantitative detection of protein arrays. Anal Chem 2003;75:1436-41.

(22.) Pawlak B, Gordon R. Density estimation for positron emission tomography. Technol Cancer Res Treat 2005;4:131-42.

(23.) Schweitzer B, Roberts S, Grimwade B, Shao W, Wang M, Fu Q, et al. Multiplexed protein profiling on microarrays by rolling-circle amplification. Nat Biotechnol 2002;20:359-65.

(24.) Pasternack RF, Collings PJ. Resonance light scattering: a new technique for studying chromophore aggregation. Science 1995; 269:935-9.

(25.) Berggren K, Steinberg TH, Lauber WM, Carrol JA, Lopez MF, Chernokalskaya E, et al. A luminescent ruthenium complex for ultrasensitive detection of proteins immobilized on membrane supports. Anal Biochem 1999;276:129-43.

(26.) Wu X, Liu H, Liu J, Haley KN, Treadway JA, Larson JP, et al. Immunofluorescent labeling of cancer marker Her2 and other cellular targets with semiconductor quantum dots. Nat Biotechnol 2003; 21:41-6.

(27.) Gauglitz G. Direct optical sensors: principles and selected applications. Anal Bioanal Chem 2005;381:141-55.

(28.) Kanda V, Kariuki JK, Harrison DJ, McDermott MT. Label-free reading of microarray-based immunoassays with surface plasmon resonance imaging. Anal Chem 2004;76:7257-62.

(29.) Eckel-Passow JE, Hoering A, Therneau TM, Ghobrial I. Experimental design and analysis of antibody microarrays: applying methods from cDNA arrays. Cancer Res 2005;65:2985-9.

(30.) Hamelinck D, Zhou H, Li L, Venveij C, Dillon D, Feng Z, et al. Optimized normalization for antibody microarrays and application to serum-protein profiling. Mol Cell Proteomics 2005;4:773-84.

(31.) Olle EW, Sreekumar A, Warner RL, McClintock SD, Chinnaiyan AM, Bleavins MR, et al. Development of an internally controlled antibody microarray. Mol Cell Proteomics 2005;4:1664-72.

(32.) Miller JC, Zhou H, Kwekel J, Cavallo R, Burke J, Butler EB, et al. Antibody microarray profiling of human prostate cancer sera: antibody screening and identification of potential biomarkers. Proteomics 2003;3:56-63.

(33.) Zhou H, Bouwman K, Schotanus M, Venveij C, Marrero JA, Dillon D, et al. Two-color, rolling-circle amplification on antibody microarrays for sensitive, multiplexed serum-protein measurements. Genome Biol 2004;5:R28.

(34.) Shao W, Zhou Z, Laroche I, Lu H, Zong Q, Patel DD, et al. Optimization of rolling-circle amplified protein microarrays for multiplexed protein profiling. J Biomed Biotechnol 2003;5:299307.

(35.) Saviranta P, Okon R, Brinker A, Warashina M, Eppinger J, Geierstanger BH. Evaluating sandwich immunoassays in microarray format in terms of the ambient analyte regime. Clin Chem 2004; 50:1907-20.

(36.) Huang R, Lin Y, Shi Q, Flowers L, Ramachandran S, Horowitz IR, et al. Enhanced protein profiling arrays with ELISA-based amplification for high-throughput molecular changes of tumor patients' plasma. Clin Cancer Res 2004;10:598-609.

(37.) Varnum SM, Woodbury RL, Zangar RC. A protein microarray ELISA for screening biological fluids. Methods Mol Biol 2004;264:16172.

(38.) Li Y, Schutte RJ, Abu-Shakra A, Reichert WM. Protein array method for assessing in vitro biomaterial-induced cytokine expression. Biomaterials 2005;26:1081-5.

(39.) Sanchez-Carbayo M, Socci ND, Lozano JJ, Haab BB, Cordon-Cardo C. Profiling bladder cancer using targeted antibody arrays. Am J Pathol 2006;168:93-103.

(40.) Janzi M, OdIingJ, Pan-Hammarstrom Q, Sundberg M, LundebergJ, Uhlen M, et al. Serum microarrays for large-scale screening of protein levels. Mol Cell Proteomics 2005;4:1942-7.

(41.) Celis JE, Gromov P, Cabezon T, Moreira JM, Ambartsumian N, Sandelin K, et al. Proteomic characterization of the interstitial fluid perfusing the breast tumor microenvironment: a novel re source for biomarker and therapeutic target discovery. Mol Cell Proteomics 2004;3:327-44.

(42.) Romeo MJ, Espina V, Lowenthal M, Espina BH, Petricoin EF III, Liotta LA. CSF proteome: a protein repository for potential biomarker identification. Expert Rev Proteomics 2005;2:57-70.

(43.) Hudelist G, Pacher-Zavisin M, Singer CF, Holper T, Kubista E, Schreiber M, et al. Use of high-throughput protein array for profiling of differentially expressed proteins in normal and malignant breast tissue. Breast Cancer Res Treat 2004;86:281-91.

(44.) Ek S, Andreasson U, Hober S, Kampf C, Proeen F, Uhlen M, et al. From gene expression analysis to tissue microarrays: a rational approach to identify therapeutic and diagnostic targets in lymphoid malignancies. Mol Cell Proteomics 2006;[E-pub ahead of print].

(45.) Uhlen M, Bjorling E, Agaton C, Szigyarto CA, Amini B, Andersen E, et al. A human protein atlas for normal and cancer tissues based on antibody proteomics. Mol Cell Proteomics 2005;4:1920-32.

(46.) Grubb RL, Calvert VS, Wulkuhle JD, Paweletz CP, Linehan WM, Phillips JL, et al. Signal pathway profiling of prostate cancer using reverse phase protein arrays. Proteomics 2003;3:2142-6.

(47.) Lin Y, Huang R, Cao X, Wang SM, Shi Q, Huang RP. Detection of multiple cytokines by protein arrays from cell lysate and tissue lysate. Clin Chem Lab Med 2003;41:139-45.

(48.) Sreekumar A, Nyati MK, Varambally S, Barrette TR, Ghosh D, Lawrence TS, et al. Profiling of cancer cells using protein microarrays: discovery of novel radiation-regulated proteins. Cancer Res 2001;61:7585-93.

(49.) Nielsen UB, Cardone MH, Sinskey AJ, MacBeath G, Sorger PK. Profiling receptor tyrosine kinase activation by using Ab microarrays. Proc Natl Acad Sci U S A 2003;100:9330-5.

(50.) Gembitsky DS, Lawlor K, Jacovina A, Yaneva M, Tempst P. A prototype antibody microarray platform to monitor changes in protein tyrosine phosphorylation. Mol Cell Proteomics 2004;3: 1102-18.

(51.) Ivanov SS, Chung AS, Yuan ZL, Guan YJ, Sachs KV, Reichner JS, et al. Antibodies immobilized as arrays to profile protein posttranslational modifications in mammalian cells. Mol Cell Proteomics 2004;3:788-95.

(52.) Sieber SA, Mondala TS, Head SR, Cravatt BF. Microarray platform for profiling enzyme activities in complex proteomes. J Am Chem Soc 2004;126:15640-1.

(53.) Rai AJ, Gelfand CA, Haywood BC, Warunek DJ, Yi J, Schuchard MD, et al. HUPO Plasma Proteome Project specimen collection and handling: towards the standardization of parameters for plasma proteome samples. Proteomics 2005;5:3262-77.

(54.) States DJ, Omenn GS, Blackwell TW, Fermin D, Eng J, Speicher DW, et al. Challenges in deriving high-confidence protein identifications from data gathered by a HUPO plasma proteome collaborative study. Nat Biotechnol 2006;24:333-8.

(55.) Zirrolli JA, Bradshaw EL, Long ME, Gustafson DL. Rapid and sensitive LC/MS/MS analysis of the novel tyrosine kinase inhibitor ZD6474 in mouse plasma and tissues. J Pharm Biomed Anal 2005;39:705-11.

[1] Nonstandard abbreviations: Cy, cyanine; TAA, tumor-associated anfigen; HI2P, horseradish peroxidase; PrEST, protein epitope signature tags; SAPE, streptavidin-R-phycoerythrin.


Tumor Markers Group, Centro Nacional de Invesfigaciones Oncologicas, Madrid, Spain.

* Address correspondence to the author at: Tumor Markers Group, 208A, Centro Nacional de Invesfigaciones Oncologicas, Melchor Fernandez Almagro 3, E-28029 Madrid, Spain. Fax 34-91-224-6972; e-mail
Table 1. Main characteristics of untargeted and targeted proteomic

Technology Description Limitations

Mass spectrometry (a) Separates a complex The identities of the
 (MALDI-TOF, SELDI) mixture of ionized peaks generated are
 proteins or not immediately known
 peptides on the
 bases of their mass
 to charge ratio

Two-dimensional gel Separates a complex Not easily
 electrophoresis mixture of proteins reproducible; not all
 on the bases of classes of proteins can
 molecular mass and be resolved; limited
 isoelectric point dynamic range/

Whole proteome Proteins coded by Huge investment in
analyses all known genes are resources for the

 expressed, purified production of purified
 and deposited on proteins, which might
 microarray slides not be corrected
 folded or modified; not
 realistic for complex

Multiplexed Multiple fluorescent Degree of multiplexing
 FISH-IS-IHC-tissue in situ hybridization limited by number of
 microarrays immunostainings; differentially
 multiple specimens identifiable

Multiplexed ELISA 96-well format Degree of multiplexing
 limited by number of

Multiplexed 800 analytes Mw Cross-reactivity
Western blots

Protein arrays Proteins or Require experimental
 antibodies design;

Technology Applications

Mass spectrometry (a) Diagnostic, prognostic,
 (MALDI-TOF, SELDI) and stratification patterns

Two-dimensional gel Detection of aberrant
 electrophoresis protein expression

Whole proteome Yeast autoantigens

Multiplexed Biomarker discovery;
 FISH-IS-IHC-tissue signaling; modification

Multiplexed ELISA Biomarker discovery

Multiplexed Biomarker discovery;
Western blots signaling

Protein arrays Biomarker discovery;
 signaling; modification

(a) Mass spectrometry using multiple reaction monitoring could be
considered an adapted targeted approach (55).
COPYRIGHT 2006 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Sanchez-Carbayo, Marta
Publication:Clinical Chemistry
Date:Sep 1, 2006
Previous Article:Lp-PL[A.sub.2]: a new kid on the block.
Next Article:Immunochemistry of lysosomal storage disorders.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters