External ID: HMS1303

Protocol: Prior to screening, lyophilized FITC-GIV peptide (a.a. 1671-1701) is dissolved in 100% DMSO to a concentration of 0.5 mM, then further diluted with water to a final concentration of 50 microM. One stock was prepared in batch for the entire screen and aliquoted. All protein stocks were stored at -80C. Thawing of aliquots on the day of the assay was performed rapidly at room temperature.

Prior to transfer of screening compounds, experimental wells of 384-well assay plates were pre-filled with 37.8 nM of FITC-GIV peptide (for 25 nM final concentration) in a 10 microL volume of assay buffer (50 mM Tris [pH 7.4], 100 mM NaCl, 10 mM MgCl2, 5 mM EDTA, 0.4% NP-40, 30 microM GDP, 1 mM DTT). Control wells were pre-filled with 37.8 nM of FITC-GIV peptide in 10 microL of assay buffer containing either 1% DMSO (vehicle negative control) or 45.3 microM AlCl3+15.1 mM NaF (AlF4- positive control). Each plate contained 16 negative control and 16 positive control wells. Plates were centrifuged at 1000xg for 2 minutes to eliminate bubbling on the surface. Following, 100 nL of experimental compound was pin-transferred into individual wells for each of two replicate assay plates (A&B). After pin-transfer, 5 microL of 3 microM His-tagged rat Galphai3 purified from E. coli (for 1 microM final concentration) was added to each well, and assay plates were shaken at low speed for 5 seconds. Final assay volume was 15 microL. Plates were centrifuged at 1000xg for 2 minutes to eliminate bubbling on the surface and shaken for 10 seconds prior to an incubation of 90 minutes at room temperature. Plates were protected from prolonged light exposure. All manipulations were performed at room temperature.

Assay plates were read with a PerkinElmer Wallac EnVision Multilabel 2103 plate reader using the fluorescent polarization measurement technology with a D505fp/D535 dual mirror and a 480 nm excitation filter. 535 nm emission filters were used for P and S channel reads at 11 mm measurement height with 30 flashes and detector gains of 319 and 563. G-factor 0.96.

Negative control: DMSO
Positive control: Aluminum tetrafluoride (AlF4-, composite of NaF and AlCl3)

Comment: Calculated result values and scoring of active compounds:

Z-score: Indicates the number of standard deviations an experimental condition is from the mean. It is calculated based on plate average (u) and standard deviation (s) of experimental well fluorescence polarization (FP) measurements for each replicate. Z-score (z) is defined as the quotient of the mean of the experimental well population subtracted from an individual well's FP (x), divided by the standard deviation within the experimental wells; z = (x-u)/s.

Normalized percent control: calculated by subtracting plate average positive control FP from experimental well FP, dividing by the difference between plate average negative and positive control FP, and multiplying by 100.

STD: the number of standard deviations from the plate average negative control FP for each replicate experimental well

Compound wells with FP values exceeding +5 standard deviations from the negative control sample average were flagged as potential aggregators or fluorescence quenchers due to their high polarization signal (HighPol) and were excluded from further consideration. Activity scores were calculated by averaging the replicate normalized percent of control values and subtracting from 100 (score based on single replicate if data for other replicate was excluded for any reason). Result values < 0 were set to 0 (no activity) and > 100 were set to 100 (100% activity). Compounds with activity score >= 15 were considered active.

External ID: WaveGuideAssay:441

Protocol: For initial screening, TetON human embryonic kidney cells (HEK293) expressing GluA2 and stargazin, gamma-2, were plated in 384-well BD purecoat amine-coated plates at 16k cells/well. Cells were induced with 8 ug/mL doxycycline in the presence of 30 uM NBQX and incubated O/N at 37deg in 5% CO2. They were washed with Hank's Balanced Salt Solution (HBSS) containing 20 mM HEPES, pH 7.4 and FLIPR VSD dye solution was added and incubated on the cells at room temperature for 45 minutes. The cell plate was loaded into the Hamamatsu Functional Drug Screening System 6000 (FDSS), 10 seconds of background were collected before addition of compound at 30 uM final concentration. Data collection continued at 1Hz sampling and experimentally determined EC50 glutamate stimulus was added at 300 seconds. Controls of 1 mM glutamate (maximal effect) and NBQX (antagonist) were used to calculate a Z' score per plate.

Data were processed using the Vanderbilt screening analysis platform, WaveGuide. Data were normalized by dividing the raw fluorescence for all time points by an average of the baseline fluorescence. For each well, 4 values were calculated as follows:

1. CMPDslope was the slope of the normalized value within the timeframe 11-18 seconds,
2. CMPDmaxmin was the difference in maximal and minimal normalized value within the timeframe 8-100 seconds
3. GLUslope was the slope of the normalized value within the timeframe 310-315 seconds, and
4. GLUmaxmin was the difference in maximal and minimal normalized value within the timeframe 301-380 seconds.

Compounds were marked as hits where any of the above calculated results for a compound differed by at least three standard deviations from the mean for that calculation across all compounds in the plate.

Compounds denoted as 'hits' were given the score of '100' and outcome of 'active'. All other compounds were denoted as 'inactive' and given a score of '0'.

Comment:

External ID: VANDERBILT_HTS_GIRK2_HPP

Protocol: To screen for modulators of the G protein-gated inwardly-rectifying potassium channel subunit 2 (GIRK2) homomeric channels, HEK293 cells were engineered to overexpress GIRK2 and the neuropeptide Y receptor type 4 (NPY4). The purpose of NPY4 overexpression was to promote modest activation (~30% activity) of GIRK2 channels upon NPY4 activation with human pancreatic polypeptide (hPP). These cells were screened using a high-throughput thallium-flux assay in a 384-well format. High-throughput thallium flux assays are based on three principles: (1) extracellular thallium can permeate through potassium channels into cells, (2) the fluorescent dye Thallos enters and stays in the cytoplasm of HEK293 cells, and (3) the fluorescence of Thallos increases dramatically when it interacts with thallium ions. Compounds that modulate the activity of an overexpressed channel, in this case GIRK2, are identified by the change in fluorescence that results from changes in thallium influx into cells through the overexpressed channel.

The screening protocol was conducted as follows. One hour prior to screening, cells were loaded with 1.5 micromolar (muM) Thallos in assay buffer (1x Hank's Balanced Salt Solution and 20 millimolar (mM) HEPES). The liquid handling and imaging was conducted using a kinetic-imaging plate reader, Panoptic (WaveFront Biosciences). Before screening, Thallos-containing buffer was removed from cells and replaced with 20 microliters (muL) per well of assay buffer. Compounds were dissolved in assay buffer at 20 muM. After 8 seconds of imaging, 20 muL of compound solution was added per well and allowed to incubate for 150 seconds. Next, 10 muL of a solution containing 2.0 mM thallium and 3.5 nanomolar (nM) hPP in Base Buffer (125 mM NaHCO3, 1 mM MgSO4, 1.8 mM CaSO4, 5mM Glucose, and 20 mM HEPES) was added to the wells. 30 seconds later, 12 muL of base buffer containing 2.0 mM thallium and 1 muM hPP, a maximally-activating concentration, was added to each well. Data was collected for 60 seconds after this final addition. The positive control for this screen was 16.6 muM eprinomectin, a natural product previously identified as a GIRK2 activator as part of a small-scale pilot screen. Microsoft Excel was utilized for data analysis. Data were normalized, F/Fo, on a well-to-well basis to account for differences in cell number. Compound activity was analyzed by comparing the sums of control-subtracted amplitudes of fluorescence intensity at 20 seconds after each thallium addition. The top 0.5% compounds corresponding to wells that demonstrated the largest increases in fluorescence were selected as "hits" for counter screening, as described below.

All 34,131 compounds screened using the method described above are listed in column OUTCOME_SCREEN_HITS. The 167 compounds selected as hits based on the above criteria are labelled ACTIVE in this column. Fluorescent compounds were included in the inactive compounds and marked as INACTIVE.

The 167 active compounds from column OUTCOME_SCREEN_HITS were tested for non-specific activity in untransfected HEK293 cells using assay conditions otherwise identical to the primary screen. Column OUTCOME_HEK_COUNTERSCREEN lists the 39 compounds that displayed an increase in fluorescence upon thallium addition, labelled with ACTIVE.

The 128 compounds which were identified as hits in the primary screen but inactive when tested in untransfected HEK293 cells were tested for activity in GIRK2-overexpressing HEK293 cells that did not express NPY4 receptor. Column OUTCOME_GIRK2_COUNTERSCREEN lists the 8 compound that displayed an increase in fluorescence upon thallium addition, labelled with ACTIVE.

The 8 active compounds from column OUTCOME_GIRK2_COUNTERSCREEN were tested at various concentrations to determine if compound activity was concentration dependent. Column OUTCOME_GIRK2_DOSE_RESPONSE provides the efficacy of these compounds; activity of all compounds at the provided concentrations was normalized to the activity of VU0529331 at 37.5 muM, which was the highest efficacy observed during this screening. For each compound concentration listed, the efficacy of that compound provided refers to the normalized, control-subtracted fluorescence intensity at 15 seconds after the addition of thallium. Compounds that demonstrated >5% efficacy and concentration-dependent activity were labelled as ACTIVE.

Finally, column Outcome combines all of the information from the above counterscreens together, indicating the 6 hits from this GIRK2 screen.

Comment: