Protein microarray technology

Abstract

Protein microarrays are increasingly utilized to better understand the expression patterns and function of proteins in various disease states. Additionally, their use in diagnostics holds great promise for applications in clinical medicine. The term "protein array" is used loosely to describe a technology founded on a number of classic protein assays that have been modified to function in a miniaturized environment with a common goal of enabling sensitive and reproducible, high throughput, multiplexed sample analysis. Similar to a gene array, a protein array is produced by immobilizing many (up to hundreds) of individual biomolecules in a defined pattern onto a solid surface for parallel analysis of samples in a high-throughput fashion. Generally, arrays consist of multiplexing on a planar surface, in contrast to multiplexing on beads, which is the basis for technologies such as xMAP® (Luminex). However, in contrast to DNA or RNA microarrays, the inherently diverse nature of proteins in biological systems makes it more difficult to achieve the same level of reproducibility for protein arrays as for gene arrays. A number of preanalytical and analytical variables must be identified and addressed to obtain meaningful and reproducible protein array data. In addition, because proteomes are characterized by protein expression across a large dynamic range, a common problem for most protein array technologies is sample complexity, which is being addressed via sample preparation methodologies performed before array binding. Frequently, the use of arrays in proteomic studies may include any of the following strategies: 1) antibody recognition of sample proteins, 2) chromatographic profiling of sample proteins, 3) expression of cDNA libraries, 3) in-situ tissue immuno-recognition, 4) protein function analysis (e.g., kinase), 5) protein-protein, nucleic acid or other molecule interaction or 6) protein domain-protein interaction. In this chapter, we will introduce the more common protein microarray technologies as well as challenges and considerations necessary for successful proteomic array studies. The design of current protein arrays ranges from the utilization of a substrate with immobilized, spatially addressed biomolecules (a flat surface such as a coated microscope glass slide, microwells or arrays of beads) to chemically modified surfaces (e.g., ProteinChip arrays). The immobilized biomolecules can include oligonucleotides or photoaptamers (single-stranded nucleic acids with high affinity to proteins), antibodies, proteins, peptides, carbohydrates, and other small molecules; whereas, a chromatographic substrate provides binding environments for proteins based on pH, hydrophobicity, or metal affinity. Since their inception in the late 1990s, protein array methods have continually undergone developmental changes and are substantially improved. Although technical challenges remain, the focused work of numerous laboratories has greatly advanced the understanding of many of the critical variables of array design and production. In general, protein array methodologies continue to be improved upon in four main categories: (1) formats of the protein array in terms of their applications, (2) sources of proteins used to generate the array, (3) surface and immobilization chemistry used to generate the protein array and (4) different methods used for the detection of protein activities on the array (1). Automation of many steps of both microarray production and use is essential for reproducibility. The general scheme of a typical protein array experiment is shown in Fig. 29.1. A set of capture ligands is arrayed on a solid support. Following buffer washes and blocking of any unreacted surface sites, the array is then probed with a complex sample containing the counterparts of the capture molecules bound to the array. When an interaction occurs, a signal is revealed on the surface by a variety of detection techniques such as fluorescence, chemiluminescence or direct detection of the capture molecule (15). The collection of molecules arranged (or "arrayed") on the substrate can easily contain all the negative and positive controls associated with each specific probe that is also needed for thorough data analysis. Various sample types, including serum (or other bodily fluids), cell culture or tissue lysate, and conditioned culture media, can be incubated with a single slide containing the multiple antibodies or other probes. As with standard protein assaying methodologies, minimal nonspecific binding of biomolecules to the surface is one of the most important criteria for high quality micro array experiments. Detection of bound analytes is achieved by standard visualization methods (fluorescent or enzyme-linked reaction) that can employ direct labeling of the sample proteins, hapten molecule(s) or a secondary and/or tertiary antibody schema. The best choice of detection method is dependent on the particular array in question. Antibody microarrays are common protein arrays in use today and are, in essence, a more high-throughput and efficient array-based version of a standard enzyme-linked immunosorbent assay (ELISA). Although retaining the specificity and quantitative characteristics of an ELISA, this technology also has the ability to make rapid, multiple, parallel, and more sensitive measurements of numerous analytes from a small volume of a single sample. Multiple antibodies can be attached onto a slide in a specified pattern via "printing" with a robot and normally contain as few as 20 or 30 up to several hundred different antibodies. Again, as with an ELISA, an antibody microarray will only be as good as each antibody/antigen pair. Therefore careful considerations must be taken into account concerning the quality (level of purification, concentration, specificity, etc.) of each antibody selected to measure the proteins of interest in each sample. Likewise, use of proper controls and generation of concentration curves are standard protocol. The tissue microarray is a widely used, high-throughput platform for the analysis of proteins in fixed tissue specimens. Simply put, up to 500 sections (0.6-2.0 mm cores) of a fixed tissue (s) are attached to a substrate (slide) and incubated with a single probe (antibody, DNA, etc.) (3). They are often used for target verification of results from cDNA micro arrays or expression profiling of tumors and tissues. Archival material can be used as well as freshly collected samples and both also yield histologic and cytologic detail not possible with other protein arrays. Although a tissue array does not replace the basic microscopic analysis of tissue histology or pathology, a well designed tissue array can replace the need to perform the same experiment over and over while also reducing the variability of experiments performed in multibatch mode. Mastering the method of tissue arraying is quite easy; however, getting the most out of a tissue microarray requires thoughtful planning and attention to detail before construction of the array. The goal of a tissue array is to present the pertinent tissue on the array. Therefore, different tissue types and disease states require different levels of accuracy in selecting the appropriate tissue from a donor block. Although some tissues can be cored from unmapped blocks, the optimal method to select the tissue for arraying from a donor block is to first map a hematoxylin and eosin (H &E) slide from the donor block (3). Although an antibody-based approach of sample "arraying" provides the means to measure the level of any list of known proteins, it is also possible to generate expression profiles of unknown protein species in a likewise high throughput, reproducible manner using chromatographic ProteinChip arrays. Depending on the array chemistry used, proteins can be retained on the array surface according to inherent protein characteristics (e.g., pI, hydrophobicity). Direct detection of the noncovalently bound analytes is made in a time of flight mass spectrometer (TOF-MS). This method allows for discovery of novel biomarkers of disease that might not be hypothesized to have a significant disregulation in the disease state and therefore, never measured by an antibody capture method. © 2008 Humana Press.