Expression profiling on a large scale, as is the case in drug discovery, is often accomplished through use of sophisticated solid-phase protein microarrays or multiplex bead technologies. While offering both high-throughput and high-content analysis, these platforms are often too cost prohibitive or technically challenging for many research settings. Capitalizing on the favorable attributes of the standard ELISA and slot blotting techniques, we developed a modified dot blot assay that provides a simple cost-effective alternative for semiquantitative expression analysis of multiple proteins across multiple samples. Similar in protocol to an ELISA, but based in a membrane bound 96-well microplate, the assay takes advantage of vacuum filtration to expedite the tedious process of washing in between binding steps. We report on the optimization of the assay and demonstrate its use in profiling temporal changes in phosphorylation events in the well-characterized EGF-induced signaling cascade of A431 cells.
Signaling through receptor tyrosine kinases (RTKs) is a highly conserved cellular mechanism, controlling fate determination, proliferation, survival, and migration [
The EGFR signaling cascade. The binding of EGF to the EGF receptor (EGFR) results in receptor dimerization and conformational changes triggering autophosphorylation. Under proper conditions, phosphorylated EGFR activates any number of three downstream signaling pathways through Ras, PI3K, and JAK, respectively. The work performed in this study focuses on the Ras cascade (highlighted in green). The three inhibitors (blue) are shown acting at their specific sites of signal interruption.
Due to the inherent complexity of the global signaling network and the involvement of their constituents in malignancy, these pathways have been extensively studied by researchers looking for insight into the mechanisms underlying both normal and aberrant growth. Detecting alterations in phosphorylation patterns within signaling profiles is a technique commonly employed to map the dose dependence and specificity of small-molecule inhibitors targeting upstream components. Traditionally, compound screening was performed in cell-free assays using purified enzymes as the target. More recently, a greater significance has been placed on the use of cell-based approaches where multiple components in a single pathway and multiple signaling cascades can be monitored simultaneously. Such analyses require complex and expensive detection platforms such as flow cytometers or high-content imaging systems. While well suited for the high throughput needs of large-scale screens at the industrial level, such platforms may not fit the workflow or demand of smaller research groups. Alternative detection systems offering semiquantitative measurement of multiple proteins in parallel include the enzyme-linked immunosorbent assay (ELISA), multiplex bead arrays, western blots, and slot blots. Although plate based, a fact that simplifies setup and signal detection, ELISAs require a pair of protein-specific antibodies which identify unique epitopes and are also quite time consuming due to multiple binding steps and extensive washing. Slot blotting apparatuses have been developed that offer increased throughput over standard western blots yet retain the same overall labor-intensive protocol that is not amenable to automation. In addition, signal quantitation for both blotting techniques is limited by the dynamic range of the developing film and method of densitometric analysis.
In this paper, we present a modified dot blotting technique for protein detection where purified proteins or cell lysates are applied directly to membrane-based 96-well microplates. This dot blot assay combines the plate-based ease of handling offered by ELISAs with vacuum filtration to greatly expedite the process of semiquantitative analysis of protein expression. Following the addition of protein sample, a two-step antibody binding process using an HRP-conjugated secondary detection antibody is performed. In the final step, the conversion of chemiluminescent substrate provides the signal in each well that can be quantified using a standard plate reader. For a 96-well plate, the entire process, from sample addition to data acquisition, requires less than 90 minutes. Our assay was validated using lysates derived from the extensively studied EGFR signaling cascade of the A431 epidermal carcinoma cell line. Temporal changes in phosphoactivation of three proteins (EGFR, MEK1/2, and ERK1/2) were measured at five-minute intervals across a twenty-minute time course of EGF stimulation. Pathway mapping was further interrogated using a set of three site-specific inhibitors of the EGFR cascade.
A431 (CRL-1555, ATCC, Manassas, VA), a human skin carcinoma cell line, was maintained in complete media (DMEM + 10% FBS) and passaged routinely by trypsinization (TrypLESelect, GIBCO/Life Technologies, Grand Island, NY) to ensure log phase growth. For induction experiments, 100 K live cells were seeded per well in 6-well plates, cultured for 2 days, and then serum-starved for 20 HR. Following synchronization, cells were exposed to 100 ng/mL human EGF (Cell Signaling Technology, Danvers, MA) for 5–20 minutes. For inhibitor studies, cultures were pre-incubated with 10
10
Lysis was performed using two buffers: (1) Cytobuster Protein Extraction Reagent (EMD Millipore) and a modified RIPA buffer (25 mM Tris-HCl pH 7.6, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, and 0.1% SDS). All buffers were supplemented with protease inhibitors and phosphatase inhibitors (EMD Millipore). All buffers were chilled on ice prior to use. Following induction, cells were washed twice with ice-cold PBS. 400
Proteins were quantified using the Direct Detect assay-free sample card and Direct Detect Spectrometer (EMD Millipore). Each card contained four hydrophilic polytetrafluoroethylene (PTFE) membrane positions, each surrounded by a hydrophobic ring to retain analyzed sample within the device’s IR beam. All measurements were performed using 2
One of the limiting factors in biochemistry is the availability and quality of antibodies. Prior to use in the dot blot assay, each candidate antibody was subjected to a stringent validation procedure [
The dot blot protocol. Depicted in the diagram is the assay for one representative well from a 96-well microplate. After protein binding, a blocking step is performed to reduce nonspecific Ab binding to any unoccupied region of membrane. The remaining portion of the assay is similar to ELISA except that all wash steps are performed via vacuum filtration.
Prewet PVDF membrane Multiscreen plates (EMD Millipore) with 100 uL 70% Ethanol for 15 seconds then immediately wash 2X with 100
Initial feasibility studies and optimization of the dot blot assay were performed using purified Glyceraldehyde-3-Phosphate Dehydrogenase (GAPDH) protein in PBS buffer. Representative results from a serial dilution of GAPDH are presented in Figure
Dot blot assay feasibility—titration of pure GAPDH. (a) The titration curve demonstrates a range 4–100 ng input protein whereby linear signal could be detected. Each point represents the mean of 4 replicates. (b) Coomassie staining of membrane wells after protein loading. Four replicates are shown for each concentration point.
For the dot blot assay to have broader application, it must perform in the context of total cell lysates where relative protein concentration, buffering conditions, and lysate clarity may impact not only membrane binding but also antibody detection characteristics. To assess such issues, we first extracted lysates from A431 cells using two distinct lysis buffers, a modified RIPA buffer and a nondetergent-based commercial extraction reagent. Following extraction, protein samples were quantified using the Direct Detect IR-based quantitation system. On average, the RIPA buffer liberated five times greater total protein than the nondetergent buffer (data not shown); this may be due to the presence of harsher detergent in the RIPA buffer resulting in greater protein solubilization.
Total lysates from each extraction condition were used to assess the potential effects of protein concentration and buffer components on dot blot assay performance. The results of this experiment are outlined in Figure
Impact of lysis buffer components on dot blot performance—Detergents. Lysates (pink bars) or buffers alone (blue) were diluted in PBS, spiked with 100 ng GAPDH and assayed for changes in GAPDH detection. The green bars show detection of the native GAPDH present in A431 lysates (no GAPDH was spiked into these samples). The CPS signal for GAPDH in PBS is displayed by the orange Bar. Each bar represents the average of 3 individual replicates. Protein concentrations for the lysates used were determined to be (a) nondetergent buffer, 494 ng/
We sought to determine the assay’s linear range and define the optimal concentration of primary antibody required for protein detection. Given the wide variability in relative protein expression within a cell and between different cells, such optimization may be required for each protein (and each cell type’s lysate) to be measured. The graphs in Figure
Native protein detection—primary antibody optimization. Serum-starved A431cells were stimulated with EGF for 20 min and harvested and lysates prepared as described previously. Dot blot assays were performed to optimize detection by (a) GAPDH and (b) phospho-ERK1/2 antibodies. A series of seven lysate concentrations (100–10000 ng/well) was used to assess six different dilutions of each primary antibody. All points were run in triplicate.
We next applied the dot blot assay in a proof-of-concept study to track changes in protein phosphorylation for the EGFR signaling cascade of A431 cells cultured with EGF. Ligand binding activates a chain of signaling events, which includes successive phosphorylation of the EGFR, MEK, and ERK proteins. Given the temporal nature of phosphoactivation within the cascade, synchronized A431 cells were exposed to EGF and harvested at 5 minute intervals for a total of 20 minutes. The time course was performed with EGF stimulus alone and in the presence of three pathway inhibitors. In the absence of inhibition, a clear temporal order of phosphorylation events is seen (Figure
Temporal mapping of the EGFR cascade using site-specific inhibitor molecules. The four bar graphs represent results for the following: (a) EGF stimulation, no inhibitor, and EGF stimulation in the presence of (b) neutralizing EGFR Ab, (c) GW5074—an inhibitor of Raf kinase activity, and (d) U0126—a highly selective inhibitor of MEK1/2. For each inhibitor, the arrow indicates the point of pathway inhibition. Each graph contains the time course (0–20 minutes) of expression profiles for each of the four proteins analyzed. Each bar value represents the average of three replicates.
Standard dot blotting is a method of protein detection similar to the western blot technique but differing in that protein samples are not initially separated electrophoretically on polyacrylamide gels but simply applied directly onto the membrane’s surface. Once applied, proteins are driven to bind the membrane through either active pressure (vacuum) or gentle agitation and passive absorption. Dot and slot blotting techniques have been used extensively by molecular biology researchers to affix proteins and nucleic acids on membranes for the purpose of quantitation, DNA homology assessment, protein-DNA/RNA interactions, enzymatic activity, and the study of ligand-receptor binding. The standard device is comprised of three main parts: the upper block with an array of slots for sample loading, a middle component that holds the inserted membrane, and a bottom block with a connector permitting vacuum filtration. A set of screws clamps the assembled device in place thereby minimizing sample bleed-over. Given this format, slot blots offer greater throughput capacity than the standard western blot and are therefore ideal for screening applications. However, the device’s main value is in sample loading; all subsequent steps are performed in the same labor-intensive manner as a western blot. To expedite the process without sacrificing throughput, our modified dot blot takes advantage of 96-well microplates equipped with PVDF membrane. The plate-based format minimizes sample bleed-over and permits easy reagent loading at each step. Since the plate is membrane based, all wash steps can be performed via vacuum filtration. The plate format and simple reaction steps are also well suited for automation and expanded screening needs. A final benefit is being able to use a standard plate reader for chemiluminescent signal detection; this format offers greater dynamic range than film densitometry enhancing quantitative capacity of the assay.
The work presented here clearly demonstrates the feasibility of the plate-based dot blot application for semiquantitative detection or comparative analyses of multiple proteins and/or multiple samples in parallel. The assay performed well on pure protein samples but more importantly worked for total cell lysates although the linear ranges of detection were considerably different. The assay is, however, quite sensitive to detergent interference, an important consideration when choosing extraction reagents. As well, samples with high viscosity or large amounts of debris had a tendency to cause a reduction in filtration flow rate and, in more severe cases, complete clogging of the membrane. Dilution of viscous samples or precentrifugation to clear particulates ameliorated clogging issues. In summary, the dot blot assay offers a cost-effective protein expression screening tool for researchers with moderate throughput needs.
We would like to thank Dr. Elene Chemokalskaya for manuscript review, advice, and assistance with the submission process.