Reverse phase protein lysate microarray

Reverse-phase protein lysate microarray (RPA) is a micro-cell lysate dot-blot that allows measurement of protein expression levels in a large number of biological samples simultaneously in a quantitative manner when high-quality antibodies are available . Technically, miniscule amount of cellular lysates are immobilized on individual spots on a microarray that is then incubated with a single specific antibody to detect expression of the target protein across many samples. One microarray, depending on the design, can accommodate hundreds of samples that are printed in dilution series and replicates. Detection is performed using either a primary or a secondary labeled antibody by chemiluminescent, fluorescent or colorimetric assays. At the end an image is obtained and the data is quantified. Multiplexing is achieved by probing multiple arrays spotted with the same lysate with different antibodies simultaneously. In addition, since RPA uses the whole-cell or tissue lysate, it can provide access to post translationally modified proteins that are not accessible with other high-throughput techniques. Thus, RPA provides a high dimensional proteomic data in a high throughput, sensitive and quantitative manner. However, since RPA does not account for antibody specificity and performance, the signal from a single spot could be due to antibody binding to an off-target protein or a sum of antibody binding to the target and off-target proteins. Thus, the antibodies used in RPA must be validated for specificity and performance against cell lysates by Western blot (WB).

RPA has various uses such as quantitative analysis of protein expression in cancer cells or tissues for biomarker discovery and clinical diagnosis. This is possible as a RPA with lysates from different cell lines and or tissue biopsies of different disease stages from various organs of one or many patients can be constructed for determination of relative abundance or differential expression of a protein marker level in a single experiment. It is also used for monitoring protein dynamics in response to various stimuli or doses of drugs at multiple time points. Some other applications that RPA is used for include exploring and mapping protein signaling pathways, evaluating molecular drug targets and understanding a candidate drug’s mechanism of action. It has been also suggested as a potential early screen test in cancer patients to facilitate or guide therapeutic decision making.

Other protein microarrays include protein microarrays (PMAs) and antibody microarrays (AMAs). PMAs immobilize individual purified and sometimes denatured recombinant proteins on the microarray that are screened by antibodies and other small compounds. AMAs immobilize antibodies that capture analytes from the sample applied on the microarray. The target protein is detected either by direct labeling or a secondary labeled antibody against a different epitope on the analyte target protein (sandwich approach). Both PMAs and AMAs can be classified as forward phase arrays as they involve immobilization of a bait to capture an analyte. In forward phase arrays, each array is incubated with one test sample such as a cellular lysate or a patient’s serum, but multiple analytes in the sample are tested simultaneously. Figure 1 shows a forward (using antibody as a bait in here) and reverse phase protein microarray at the molecular level.

Experimental Design and Procedure:

Depending on the research question or the type and aim of the study, RPA can be designed by selecting the content of the array, the number of samples, sample placement within micro- plates, array layout, type of microarrayer, correct detection antibody, signal detection method, inclusion of control and quality control of the samples. The actual experiment is then set up in the laboratory and the results obtained are quantified and analyzed. The experimental stages are listed below:

Sample collection: Cells are grown in T-25 flasks at 370C and 5% CO2 in appropriate medium. Depending on the design of the study, after cells are confluent they could be treated with drugs, growth factors or they could be irradiated before lysis step. For time course studies, a stimulant is added to a set of flasks concurrently and the flasks are then processed at different time points (1). For drug dose studies, a set of flasks are treated with different doses of the drug and all the flasks are collected at the same time.

If a RPA containing cell fraction lysates of a tissue/s is to be made, laser capture microdisection (LCM) or fine needle aspiration methods is used to isolate specific cells from a region of tissue microscopically.

Cell lysis: Pellets from cells collected through any of the above means are lysed with a cell lysis buffer to obtain high protein concentration.

Antibody screening: Part of the lysates collected are mixed and run on a 2D-single lane prep Wb. The membrane is cut into four millimeter strips and each strip is probed with a different antibody. Strips with single band indicate specific antibodies that are suitable for RPA use. Antibody performance should be also validated with a smaller sample size under identical condition before actual sample collection for RPA.

RPA construction: The remaining cell lysates are collected individually that are usually serially twofold diluted around six to ten times in to a 384- or a 1536-well microtiter plate depending on the experiment. The lysates are then printed onto either nitrocellulose or PVDF membrane coated glass slides by a microarrayer such as Aushon BioSystem 2470 or Flexys robot (Genomic solution). At the beginning, RPAs were produced by GMS417 (Affymetrix), a pin-in-ring arrayer that is no longer available in the market. Aushon 2470 with a solid pin system is the ideal choice as it can be used for producing arrays with very viscous lysates and it has humidity environmental control and automated slide supply system. The membrane coated glass slides are commercially available from different companies such as Whatman’s (www.whatman.com) and Schleicher and Schuell Bioscience.

Immunochemical signal detection: After the slides are printed, non-specific binding sites on the array are blocked using blocking buffer such as I-Block and the arrays are probed with primary antibody and later with secondary antibody. Detection is usually conducted with DakoCytomation catalyzed signal amplification (CSA) system. For signal amplification, slides are incubated with streptavidin-biotin-peroxidase complex followed by biotinyl-tyramide/hydrogen peroxide and streptavidin-peroxidase. Development is completed using hydrogen peroxide and scans of the slides are obtained (1). Tyramide signal amplification works as following: immobilized HRP (Horse Radish Peroxidase) converts tyramide into reactive intermediate in the presence of hydrogen peroxide. Activated tyramide binds to neighboring proteins in fact close to a site where activating HRP enzyme is bound. This leads to more tyramide molecule deposition at the site; hence the signal amplification (7).

• For a detailed protocol of the technique, refer to: Spurrier, S. Ramalingam, S. Nishizuka. (2008). Reverse-phase protein lysate microarrays for cell signaling analysis. Nature Protocols. 3 (11): 1796- 1808

Data Quantification/Analysis:

Once immunostaining has been performed protein expression must then be quantified. As most detection methods employ signal amplification by colourimetric assays, the signal levels can be obtained by using the reflective mode of an ordinary optical flatbed scanner. Two programs available on line (P-SCAN and ProteinScan) can then be used to convert the scanned image into numerical values. These programs quantify signal intensities at each spot and use a dose interpolation algorithm (DI25) to compute a single normalized protein expression level value for each sample. Normalization is necessary to account for differences in total protein concentration between each sample and so that antibody staining can be directly compared between samples. This can be achieved by performing an experiment in parallel in which total proteins are stained by Colloidal Gold total protein staining or Sypro Ruby total protein staining. When multiple RPAs are analyzed, the signal intensity values can be displayed as a heat map, allowing for Bayesian clustering analysis and profiling of signaling pathways.

Strengths:

RPAs allow for high throughput detection of protein, which cannot be done by conventional Western blotting or ELISA. The small spot size on the microarray, ranging in diameter from 150 to 200 microns, enables the analysis of thousands of samples with the same antibody in one experiment. RPAs have increased sensitivity and are capable of detecting proteins in the picogram range. Some researchers have even reported detection of proteins in the attogram range. This is a significant improvement over protein detection by ELISA which requires microgram amounts of protein. The increase in sensitivity is due to the miniature format of the array, which leads to an increase in the signal density (signal intensity/area). The high sensitivity of RPAs allows for the detection of low abundance proteins or biomarkers from very small amounts of starting material such as biopsy samples, which are often contaminated with normal tissue. Using laser capture microdissection lysates can be analyzed from as few as 10 cells. A great improvement of RPAs over traditional forward phase protein arrays is a reduction in the number of antibodies needed to detect a protein. Forward phase protein arrays usually use a sandwich method to capture and detect the desired protein. This implies that there must be two epitopes on the protein (one to capture the protein and one to detect the protein) for which specific antibodies are available. Other forward phase protein microarrays directly label the sample, however there is often variability in the labeling efficiency for different protein. This problem is overcome by RPAs as sample need not be labeled directly. Another strength of RPAs over forward phase protein microarrays and Western blotting is the uniformity of results, as all samples on the chip are probed with the same antibody and the same concentration of amplification reagents for the same length of time. This allows for the quantification of differences in protein levels across all samples. Furthermore, printing each sample on the chip in serial dilution provides an internal control. A problem that is encountered with tissue microarrays is antigen retrieval. Antibodies often detect linear peptide sequences that may be masked due to the 3-dimensional conformation of the protein. This problem is overcome with RPAs as the samples can be denatured, revealing any concealed epitopes.

Weaknesses:

The biggest limitation of RPA is its dependence on antibodies for detection of proteins. Currently there is a limited number of signaling proteins for which antibodies exist that give an analyzable signal. In addition, finding the appropriate antibody requires extensive screening of hundreds of antibodies by Western blotting prior to beginning RPA analysis. Two open resource databases have been created to display Western blot results for antibodies that have good binding specificity within the expected range (http://discover.nci.nih.gov/abminer and http://mtt.uscancer.org). Furthermore, RPAs, unlike Western blots, do not resolve protein fractions by molecular weight. Thus, the antibody binding specificity is lower for RPA than for Western blots. This also limits the information which can be learned by antibody specificity and performance.