Multiplex HTS Assay for Inhibitors of MEK Kinase PB1 Domains, specifically MEK5 binding to MEK Kinase 2 Wildtype
PB1 (Phox/Bem1p) domains function as protein-protein interaction sites by forming PB1-PB1 domain heterodimers (Moscat, et al. 2006). There are at least 20 human PB1 domain-containing proteins. Different PB1 domains contribute to the formation of specific protein complexes critical for biological responses including proliferation, apoptosis, cell polarity, and angiogenesis. These proteins include more ..
BioActive Compounds: 3276
University of New Mexico Assay Overview:
Assay Support: 1R03MH084830-01
Project Title: TR-FRET HTS Assay for Inhibitors of MEKK2-MEK5 PB1 Domain Interaction
PI: Kazuhiro Nakamura, Ph.D
Center PI: Larry Sklar, Ph.D
Assay Implementatiion: Zurab Surviladze Ph.D, Mark Haynes Ph.D, Anna Waller Ph.D, Mark Carter MS
Assay Background and Significance:
PB1 (Phox/Bem1p) domains function as protein-protein interaction sites by forming PB1-PB1 domain heterodimers (Moscat, et al. 2006). There are at least 20 human PB1 domain-containing proteins. Different PB1 domains contribute to the formation of specific protein complexes critical for biological responses including proliferation, apoptosis, cell polarity, and angiogenesis. These proteins include the mitogen/extracellular signal regulated kinase kinases MEKK2 and MEKK3 (MAP3Ks), as well as the downstream regulated kinase MEK5 (a MAP2K), which solely govern the ERK5 MAPK pathway involved in angiogenesis, cell growth and inhibition of apoptosis.
ERK5 is important for protecting many cell types from stress-induced apoptosis (Nakamura, et al. 2007). ERK5 is critical for both development of the vasculature and vascular homeostasis in the adult. No other MAPK has been shown to be critical in vascular maintenance in the adult animal mouse model (Hayashi, et al. 2004). MEKK2 and MEKK3 are the only MAP3Ks that physically interact with and activate the MEK5-ERK5 signaling module. MEKK2 and MEKK3 PB1 domains are classified as type B due to the basic amino acids expressed in the binding pocket. The MEK5 PB1 domain contains acidic amino acids and is classified as a type A PB1. These proteins bind via a beta grasp topology where basic residues from one protein interact with clusters of acidic residues expressed in the binding partner. This unique interaction may be more susceptible to small molecule inhibition than beta sheet interactions observed in most protein-protein interactions. The MEKK2 mutant used in this study has a lysine to alanine mutation in the PB1 domain that disrupt binding to MEK5, resulting in a loss of activation of ERK5 (Nakamura, et al. 2003). The mutant is used as a counterscreen for the wild-type MEKK2 protein, but it could also be considered an interesting target for identifying small molecules that reconstitute its binding capacity.
This project is based on rapidly accumulating evidence that the PB1 domain is the crucial interface for the MEKK2-MEK5 interaction that is involved in various pathophysiological conditions. Inhibition of the MEKK2/3-MEK5 PB1-PB1 domain dimerization by small molecules would be a unique pharmacological mechanism that would selectively disrupt the activation of the EKR5 signaling network in cells and consequently may prove extremely useful for therapeutic intervention directed at inhibiting angiogenesis, tumorigenesis, and chronic inflammation.
The binding of fluorochrome-conjugated MEK5 to MEK kinase targets (MEKK2, MEKK2 mutant and MEKK3) is the basis for construction of a fluorescence-based, flow cytometric assay adapted from a TR-FRET HTS assay format. The assay described here is a bead-based high-throughput multiplex screen to identify small molecule regulators of protein interactions via the PB1 domains of MEK5 and three MEK kinase targets.
For each target, the components of the multiplex consist of a Streptavidin-functionalized polystyrene bead (Spherotech, USA), a biotinylated MEKK-fusion protein target (3 total, supplied by assay provider), and a fluorescent peptide probe (AlexaFluo488-GST-MEK5 WT, supplied by the assay provider). Beads for each target have been labeled with varying intensities of red color, such that each assay is built on a unique bead set, and each bead set is associated with a unique optical address. The similarly sized bead sets (~5 micron) are distinguised by distance emission characteristics at 750 nanometer with excitation at 635 nanometer.
Bead sets are coated with biotin-conjugated MEK kinase protein targets (final protein concentration 1 microM) in HPSN bead coupling buffer (30milliM HEPES, 100milliM KCl, 20milliM NaCl, 0,01% NP-40, pH 7.5) at 4 degrees C, and incubated overnight (16-18 hours). The three bead sets (each with bound protein) and an uncoated bead set (scavenger beads, described below) are centrifuged separately, re-suspended in 50 microL of flow buffer (HPSN with 0.1% BSA) and mixed for 15 minutes to block unbound sites with BSA. Blocked bead sets are centrifuged and re-suspended together in flow buffer and held on ice. The Alexa488-MEK5 is added at a final assay concentration of 1.25 nanoM. A streptavidin-only bead control,(no associated protein) is incorporated into each well as a fluorescence scavenger to determine inherent fluorescent properties of the test compounds (530 nanometer emission).
The assay is conducted in 384-well microplates in a total assay volume of 10.1 microL (5 microL of bead mixture, 0.1 microL of test compound, and 5 microL of 1.25 nanoM Alexa488-MEK5. Plates are incubated on a rotating platform (10-12 RPM) for 15 min at room temperature. Specificity of Alexa488-MEK5-binding is determined by competitive binding of the Alexa-MEK5 fluor (1.25 nanoM) with non-labeled MEK 5 protein (200 nanoM). This blocking control is conducted daily as a single tube assay.
Sample analysis is conducted with the HyperCyt(R) high throughput flow cytometry platform. The HyperCyt system interfaces a flow cytometer and autosampler for high-throughput microliter-volume sampling from 384-well microtiter plates [Kuckuck, et al. 2001]. Flow cytometric data of light scatter and fluorescence emission at 530 +/- 20 nanometer (FL1) and 750+ nanometer (FL9) are collected on a Cyan Flow Cytometer (Dako). Analysis is performed using time-resolved acquisition into a single data file and analysis using IDLeQuery/HyperView software to merge the flow cytometry data files with compound worklist files generated by HyperSip software. Gating based on forward scatter (FS) and side scatter (SS) parameters is used to identify singlet bead populations. Gating based on FL9 emission distinguishes the beads coated with different proteins, and the median channel fluorescence (MCF) associated with the Alexa-488 emission spectra (FL1) for each bead population is calculated. The raw data are parsed in IDLeQuery/HyperView to produce annotated fluorescence summary data for each well. The parsed data are then processed through an Excel template file constructed specifically for the assay to segregate data for each target and the fluorescence scavenger in the multiplex.
A minimum of 25 beads bound with a particular protein target was required for a significant measurement of a test compound's effect on that particular protein. When less than 25 beads are counted, the result for that protein is considered missing. Compounds from missing wells are given a PUBCHEM_ACTIVITY_OUTCOME = 4, and PUBCHEM_ACTIVITY_SCORE of 0 is assigned.
The potential innate compound fluorescence are accounted for by subtracting out the fluorescence of Scavenger bead in the calculation of percent response (%Response). Additionally, potential systemic trends in the data over the entire plate are dealt with by utilizing a linear fit of the controls to each position on the plate (LinFitPCntrl). The following equation was used for calculating %Response.
%Response = 100*([SampleFL-ScavFL] - NCntrl)/(LinFitPCntrl-NCntrl)
where all variables are Median Channel Fluorescence (MCF) Intensity associated with particular bead sets. SampleFL and ScavFL are the MCF measurements, from wells containing the test compound, of beads bound with protein target (for SampleFL) or non-protein bound beads (for ScavFL). NCntrl, Negative Control, is the MCF measurement of the protein target bound bead in the presence of 1.25 nanoM Alexa488-MEK5 and blocking 200 nanoM MEK5. LinFitPCntrl are the calculated MCF value of control well with only DMSO present. Baseline of %Response is 100% and represents test compound has no effect.
Test compounds were assessed for either increasing Alexa488-MEK5 binding ("Activator") or decreasing Alexa488-MEK5 binding ("Inhibitor"), thus the change in %Response from baseline was utilized in scoring the compounds (i.e., PUBCHEM_ACTIVITY_SCORE is the absolute difference of %Response and Baseline). However, due to some plate irregularities, the evaluation was segmented into two parts to allow accountibility of the plate position of the test compounds. One part included rows A, B, and C on the 384-well plate and the second part included rows D through P on the 384-well plate. Average %response (Ave%Resp) and standard deviation (Std) were calculated for each segment. These different average %response were utilized as the baseline for each segment. A compound was demeaned active as an "Activator" if
%Response > Ave%Resp + 3*(Std)
and active as an "Inhibitor" if
%Response < Ave%Resp - 3*(Std).
For the compounds in rows A, B, and C the Ave%Resp = 92% and Activators had %Response >136 and Inhibitors had %Response <48%. And both Inhibitors and Activators have PUBCHEM_ACTIVITY_SCORE >44. For the compounds in rows D through P, the Ave%Resp = 100% and Activators had %Response >136 and Inhibitors had %Response <64. For these D through P row compound, Inhibitors and Activators have PUBCHEM_ACTIVITY_SCORE >36. The type of active compound is notated in the PUBCHEM_ASSAYDATA_COMMENT column.
Additionally, compounds have been noted in the PUBCHEM_ASSAYDATA_COMMENT column with "InstrumentError" due to erroneous hits occuring in plate location B15. These compounds have been demeaned "inconclusive" and given PUBCHEM_ACTIVITY_SCORE = 0.
Average Zprime for this screening campaign was 0.71 +/- 0.13. Plates with Zprimes significantly outside of this averaged Zprime were not deamend accetable, i.e., Zprime < 0.32 (0.71-3*0.13).
Keywords: NIH Roadmap, NMMLSC, high throughput flow cytometry, multiplex bead-based screening, MEK5, MEKK2, MEKK3
Abbreviations used; milliM for millimolar, microM for micromolar, nanoM for nanomolar, microL for microliters, min for minutes
** Test Concentration.
Data Table (Concise)