Bookmark and Share
BioAssay: AID 1614

Screen for compounds that reduce polyglutamine (polyQ) inclusions in nerve growth factor (NGF)-treated PC12 cells (GFPQ80)

Intracellular inclusions are a common pathological hallmark of polyglutaminediseases, and are believed to reflect conformational abnormalities of the protein. Some cell-based models, including the differentiated PC12 neural cells used in our assays, show inclusion formation that correlates with cytotoxicity. Moreover, genes that block toxicity in our cell-based system also reduce inclusion more ..
_
   
 Tested Compounds
 Tested Compounds
All(1033)
 
 
Unspecified(1033)
 
 
 Tested Substances
 Tested Substances
All(1040)
 
 
Unspecified(1040)
 
 
AID: 1614
Data Source: NINDS Approved Drug Screening Program (GFPQ80)
Deposit Date: 2009-03-24

Data Table ( Complete ):           View All Data
Tested Compounds:
Description:
Intracellular inclusions are a common pathological hallmark of polyglutaminediseases, and are believed to reflect conformational abnormalities of the protein. Some cell-based models, including the differentiated PC12 neural cells used in our assays, show inclusion formation that correlates with cytotoxicity. Moreover, genes that block toxicity in our cell-based system also reduce inclusion formation. In addition, genetic suppressors of polyQ toxicity in a fly model of disease increase the solubility of polyQinclusions in the denaturing detergent, sodium dodecyl sulfate (SDS). The green fluorescent protein (GFP)-PolyQ fusion proteins in our cell-based system can easily be screened for formation of intracellular inclusions. We designed the screen to identify compounds that reduce inclusion formation and/or increase SDS solubility of inclusions.
Protocol
Screen: Our primary screen involved two six-day assays. The first identified compounds that reduced polyQ inclusions in our NGF-treated, stably transfected, inducible GFP-Q80 PC12 cell line. The first assay also identified compounds that increase SDS solubilization of inclusions as an additional outcome measure. The second assay analyzed hits from the first assay in a similar inducible PC12 cell line (under identical experimental conditions) that expresses non-pathogenic GFP-Q19, which does not form inclusions. Compounds that also reduced GFP fluorescence in this cell line were presumed to act in a doxycycline-like manner and thus were excluded from further analyses. (The underlying assumption is that compounds reducing inclusions via a doxycycline-like repressor effect would reduce GFP expression in the GFP-Q19 cell line as well as the GFP-Q80 line, while those compounds reducing inclusion formation by different mechanisms would retain GFP expression levels in the GFP-Q19 cell line.) 96-Well Costar tissue culture plates (Corning, Inc; 35966) were coated with 100 mcg/ml Type I Rat Tail Collagen (BD Biosciences; 35-4236) diluted in Dulbecco's Phosphate Buffered Saline (Gibco; 12377-016) for at least one hour and then aspirated prior to cell plating. PC63 tet-off cells were split by Trypsin/EDTA (Gibco; 25200-056) treatment and removal followed by resuspension in RPMI 1640 supplemented with 10% equine serum, 5% fetal bovine serum, and 1% penicillin/streptomycin. Cells were plated at a concentration of 2500-3000 cells per well and allowed to adhere to the plate for 24 hours. After 24 hours, the cell media was changed to 200uL RPMI 1640 supplemented with 2% equine serum, 1% fetal bovine serum, 1% penicillin/streptomycin, and 90ng/ml nerve growth factor. Negative control wells were treated with the same medium that also contained 1.5mcg/ml doxycycline to repress GFP-Q80 expression. After 24 hours, the medium was changed and compounds were added in 50 #M concentrations (duplicate wells per compound) by adding 1.0 #L of the 10mM stock solutions to 200uL cell media. The GFP-expressing cells were incubated in 50uM concentration of the compounds for 48 hours, at which time the media was changed and 50uM of the compounds were added a second time. Following 48 additional hours, the wells were analyzed for polyQ inclusion levels, and subsequently for SDS solubility after a 15-minute incubation in the 2% SDS solution. Compounds eliciting a reduction in inclusion formation were then analyzed in the PC12 tet-off GFP-Q19 cell line (identical protocol as previously discussed with the GFPQ80 cell line) in an effort to eliminate those suppressor compounds that modulated inclusion formation merely by reducing transgene expression.
Cell Lines: Stably transfected, clonal PC12 lines that express expanded polyQ (Q80) as a green fluorescent fusion protein (GFP-Q80) were created. Clones were generated in the PC12 Tet-off cell line (Clontech), in which transgene expression is repressed in the presence of doxycycline (dox) or related antibiotics. Clone #29 was chosen for this screen because of its robust, tightly dox-regulated expression and efficient formation of cytoplasmic and nuclear inclusions. As a nonpathogenic control, clones were also generated that inducibly express GFP-Q19 (clone #27) fusion protein, which is not toxic and does not form inclusions. This cell line was used in the second assay of the primary screen, in which compounds acting through an apparently doxycycline-like manner (i.e., reducing transgene expression) were identified and eliminated from further studies.
The growth medium for the PC12 tet-off GFP-Q80 and GFP-Q19 cell lines was RMPI 1640 (Gibco; 11875-093), supplemented with 10% equine serum (HyClone; SH30074.03), 5% fetal bovine serum (HyClone; SH30071.03), 1% penicillin/streptomycin (Gibco 15140-122), 0.2mg/ml G418 sulfate (Cellgro; 61-234-RG), 0.1mg/ml hygromycin B (Roche; 843-555), and 1.5mcg/ml doxycycline HCl.
Assay Medium: RPMI 1640 supplemented with 2% equine serum, 1% fetal bovine serum, 1% penicillin/streptomycin, and 90ng/ml nerve growth factor (Promega; G514A).
Assay Reagent: 2% Sodium dodecyl sulfate (BDH; 4424444) in sterile water was used to lyse cells to analyze inclusion resistance to SDS denaturation.
Data Analysis: Cell wells were visually screened at the conclusion of the six-day assay by fluorescence microscopy for the presence of inclusions prior to and after a 10 minute incubation in 2% SDS. Compound-treated wells were scored relative to untreated GFP-Q80 expressing cells. The wells were first analyzed for drug-induced toxicity, and then scored on a four-point scale relative to control wells (3=inclusion formation > control, 2=inclusion formation approximately equal to control, 1=inclusion formation approximately less than 50% control, 0=no inclusion formation). This scale was used for the analysis both prior to and after SDS treatment. To ensure that the duplicate wells for a given compound were evaluated independently, one of each duplicate well was scored for all tested compounds first (from right to left and top to bottom of 96-well plate columns), after which the second duplicate well for all compounds was analyzed (from right to left and bottom to top). Scores were recorded on separate sheets for each duplicate and later consolidated in a final record. The subsequent test of compound effects on GFP-Q19 expression levels was performed in a similar fashion, with the four-point scale representing levels of GFP expression rather than inclusion burden. Normalization of the scores for the "final data sample" was done by taking the sum total score of polyQ inclusion formation (before SDS treatment) for duplicate wells for each compound and dividing this by the sum total control score, which by definition is 4. Thus, a compound with no observable effect on inclusion formation would yield duplicate scores of 2, resulting in a total score of 4 which, when divided by the control value of 4, yields a normalized score of 1.
Dilution Series: 37 Compounds met criteria for potential suppressors and were retested in a dose-response analysis. Compounds were serially diluted in RPMI 1640 supplemented with 2% equine serum, 1% fetal bovine serum, 1% penicillin/streptomycin, and 90ng/ml nerve growth factor to achieve concentrations of 200uM, 50uM, 5uM, and 0.5uM. Untreated wells served as positive and negative controls (+ and - doxycycline, respectively). This was used to establish a minimum effective concentration (MEC), not an EC50, for our assay. Serial dilutions were done manually using pipettmen to dilute the compounds in 1.5mL Eppendorf tubes.
Literature Reference Chai, Y., S. Koppenhafer, N. Bonini, and H. Paulson. 1999.Analysis of the role of heat shock protein (Hsp) chaperones in polyglutaminedisease. J. Neurosci 19:10338-10347Chan, H.Y.E., J. M. Warrick, G. L. Gray-Board, S. L.Koppenhafer, H. L. Paulson, and N. M. Bonini. 2000. Mechanisms of chaperonesuppression of polyglutamine disease: selectivity, synergy, and modulationof protein solubility in Drosophila. Hum Mol Genet 9:2811-2820
Cell Lines We created stably transfected, clonal PC12 lines that express expandedpolyQ (Q80) as a green fluorescent fusion protein (GFP-Q80). Clones weregenerated in the PC12 Tet-off cell line (Clontech), in which transgeneexpression is repressed in the presence of doxycycline (dox) or relatedantibiotics. Clone #29 was chosen for this screen because of its robust, tightly dox-regulated expression and efficient formation of cytoplasmic and nuclear inclusions. As a nonpathogenic control, clones were also generated that inducibly express GFP-Q19 (clone #27) fusion protein, which is not toxic and does not form inclusions. This cell line was used in the second assay of the primary screen, in which compounds acting through an apparently doxycycline-like manner (i.e., reducing transgene expression) were identified and eliminated from further studies.
Growth Medium The growth medium for the PC12 tet-off GFP-Q80 and GFP-Q19 cell lines was RMPI 1640 (Gibco; 11875-093), supplemented with 10% equine serum (HyClone; SH30074.03), 5% fetal bovine serum (HyClone; SH30071.03), 1% penicillin/streptomycin (Gibco 15140-122), 0.2mg/ml G418 sulfate (Cellgro; 61-234-RG), 0.1mg/ml hygromycin B (Roche; 843-555), and 1.5mcg/ml doxycycline HCl.
Assay Medium RPMI 1640 supplemented with 2% equine serum, 1% fetal bovine serum, 1% penicillin/streptomycin, and 90ng/ml nerve growth factor (Promega; G514A).
Assay Buffer Not Applicable
Assay Reagent 2% Sodium dodecyl sulfate (BDH; 4424444) in sterile water was used to lyse cells to analyze inclusion resistance to SDS denaturation.
Liquid Handling Robot None
Concentration Primary Screen (#M) 50
Number Negative Controls 130
Negative Control Mean 3.0
Negative Control Standard Deviation 0.0
Number Positive Controls 52
Positive Control Mean 0.0
Positive Control Standard Deviation 0.0
Comment
This assay was a visual analysis that assessed reductions in polyQ inclusions relative to untreated control wells. Preliminary blinded analyses showed that inclusion levels in control wells were highly consistent, and never varied enough to warrant a change in the subjective scoring of those wells, meaning there was essentially no standard deviation for the control wells. In defining "hits" (inclusion suppressors) in our cell-based model, several confounding factors need to be considered and excluded. These include the cytotoxicity of compounds (eliminating cells with or without inclusions), and doxycycline-like effects on transgene expression. Thus, only compounds appearing to have reproducible effects on the number of inclusions/cell were selected for rescreening in the GFPQ19 cell line and for dose response analyses; these are the compounds with accompanying MEC values on the spreadsheet. This cell-based assay does not permit the generation of true EC50 values, as the visually assessed cellular changes we measured are not easily quantifiable beyond a four-point scale. Hence, the minimal concentration at which we detected a suppressor effect is what is listed in the dose-response analysis. In several cases during the dose-response analysis, the concentration at which inclusions were reduced was higher than the 50uM concentration used for the initial screen. In addition, some compounds identified in the first screen failed to show an effect in the subsequent dose-response analysis. Only those compounds that led to a dose-dependent reduction in inclusion formation in the dose-response analysis were designated with MEC values on the data spreadsheet. For those compounds with MECs higher than the concentration seen in the initial assay, the 50uM concentration used in the first screen is recorded as the MEC value. Note that two groups of compounds could not be fully analyzed in our assay: compounds that were toxic in our cellular model (A) at the 50 uM concentration used in the primary screen, and compounds that reduced GFP expression levels in the control GFP-Q19 cell line and thus may act through a doxycycline-like manner (B). The letters "A" and "B" are used to denote these respective groups on the data spreadsheets. There was also a group of compounds that proved to be toxic in only one of the two duplicated test wells. The well with dead cells was not used in averaging the inclusion score on the Raw Assay Data spreadsheet, thus a standard deviation for these compounds is not listed. Furthermore, these compounds were denoted with an "A" (meaning dead) to avoid confusion in the interpretation of the final normalized score.
A Note about the Activity Matrix Rankings:
The data for each assay has been converted into a numerical ranking of all 1040 compounds for that assay. Thus, a compound with a rank of 1 was the top scoring compound in that assay and a compound with a rank of 1040 was the lowest scoring compound in that assay. Compounds with equal scoring were given equal rank.
PubChem scores were derived as (1040-NINDS activity rank). Substances with reported EC50 values were considered active.
PI Name Henry L. Paulson
Affiliation University of Iowa
Email henry-paulson@uiowa.edu
Investigator Jeffrey R.Bishop
Categorized Comment - additional comments and annotations
From PubChem:
Assay Cell Type: PC-12
Result Definitions
Show more
TIDNameDescriptionHistogramTypeUnit
OutcomeThe BioAssay activity outcomeOutcome
ScoreThe BioAssay activity ranking scoreInteger
1DRUG_NAMEString
2PLATEInteger
3ROWString
4COLUMNInteger
5GFPQ80_R1Inclusion formation at 50 uM concentration (first test)Float
6GFPQ80_R2Inclusion formation at 50uM concentration (second test)Float
7GFPQ80_avgAverage inclusion formation at 50uM concentration Float
8GFPQ80_SDStandard DeviationFloat
9GFPQ80_EC50Effective ConcentrationFloat
10GFPQ80_finalFinal dataFloat
11NINDS activity rankInteger

Data Table (Concise)
Data Table ( Complete ):     View All Data
PageFrom: