This manuscript describes three complementary protocols for assessing the toxicity of polyglutamine (polyQ)-expansion proteins in the yeast Saccharomyces cerevisiae. These protocols can easily be modified to monitor the toxicity of other misfolded proteins in yeast.
Protein misfolding is associated with many human diseases, particularly neurodegenerative diseases, such as Alzheimer’s disease, Parkinson’s disease, and Huntington’s disease 1. Huntington’s disease (HD) is caused by the abnormal expansion of a polyglutamine (polyQ) region within the protein huntingtin. The polyQ-expanded huntingtin protein attains an aberrant conformation (i.e. it misfolds) and causes cellular toxicity 2. At least eight further neurodegenerative diseases are caused by polyQ-expansions, including the Spinocerebellar Ataxias and Kennedy’s disease 3.
The model organism yeast has facilitated significant insights into the cellular and molecular basis of polyQ-toxicity, including the impact of intra- and inter-molecular factors of polyQ-toxicity, and the identification of cellular pathways that are impaired in cells expressing polyQ-expansion proteins 3-8. Importantly, many aspects of polyQ-toxicity that were found in yeast were reproduced in other experimental systems and to some extent in samples from HD patients, thus demonstrating the significance of the yeast model for the discovery of basic mechanisms underpinning polyQ-toxicity.
A direct and relatively simple way to determine polyQ-toxicity in yeast is to measure growth defects of yeast cells expressing polyQ-expansion proteins. This manuscript describes three complementary experimental approaches to determine polyQ-toxicity in yeast by measuring the growth of yeast cells expressing polyQ-expansion proteins. The first two experimental approaches monitor yeast growth on plates, the third approach monitors the growth of liquid yeast cultures using the BioscreenC instrument.
Furthermore, this manuscript describes experimental difficulties that can occur when handling yeast polyQ models and outlines strategies that will help to avoid or minimize these difficulties. The protocols described here can be used to identify and to characterize genetic pathways and small molecules that modulate polyQ-toxicity. Moreover, the described assays may serve as templates for accurate analyses of the toxicity caused by other disease-associated misfolded proteins in yeast models.
1. Expression of Toxic PolyQ-expansion Proteins in Yeast
A systematic analysis has established the precise amino acid sequence of a polyQ-expansion protein that is required to produce toxicity in yeast 7. This toxic polyQ-expansion protein contains an amino-terminal FLAG-tag followed by 17 amino acids from the original sequence of the huntingtin protein, a polyQ region, and a carboxy-terminal fusion to a fluorescent protein (either GFP or CFP, see Figure 1 a). The expression of proteins with polyQ expansions of 46 glutamines or more (e.g. 72 and 103 glutamines) produces toxicity in yeast when expressed under the control of the inducible and relative strong GAL1 promoter 7, 9.
As described previously, polyQ toxicity in yeast is only apparent in cells that carry the protein Rnq1p in its prion conformation, [RNQ+], e.g. the yeast strain W303 9. The toxic polyQ-expansion protein recapitulates central aspects of polyQ-biology in yeast, such as polyQ length-dependent toxicity (see below) and aggregation (Figure 1 b). Notably, toxic polyQ-expansion proteins do not kill yeast cells immediately; they rather impair or arrest the cell cycle and cell devision, thereby slowing down or inhibiting the growth of yeast colonies on plates or liquid cultures (our unpublished data).
2. Potential Problems with Yeast Cells Expressing Toxic PolyQ-expansion Proteins
Yeast cells expressing the otherwise toxic polyQ expansion proteins may not show any growth defect 9. The genetic nature of these suppressors of polyQ-toxicity does not appear to be based on simple Mendelian mutations and thus can occur with relatively high frequency (our unpublished results). We speculate that these spontaneous suppressors are caused by the curing of unidentified prions that function similarly to [RNQ+] in determining polyQ toxicity.These spontaneous suppressors of polyQ toxicity can jeopardize the successful outcome of any experiment that aims to characterize polyQ toxicity or to identify and characterize modifiers of polyQ toxicity.
In order to avoid the frequent occurrence of these spontaneous suppressors, follow the precautionary measures outlined below, which have proven to be very effective:
3. Growth Assays
3.1 Growth on plates
3.2 Spotting assays
This assay is more quantitative than the plating assay described above and can thus reveal even subtle differences in polyQ toxicity with the same experiment on the same plate.
3.3. Monitoring polyQ toxicity in yeast by growth of liquid cultures
This protocol is the most quantitative (OD600 numbers) of the three assays described here and can even detect very subtle differences in polyQ toxicity. The aforementioned occurrence of spontaneous suppressors, however, can potentially produce misleading results. I therefore recommend combining this assay with at least one of the two plating assays described above. We propose to use the BioscreenC instrument for these experiments. The BioscreenC is an instrument that automatically measures the optical density of yeast cultures in 100-well plates while incubated at defined temperatures with defined agitation. Other methods for growing yeast cells and measuring their optical density may also apply.
4. Representative Results
Figure 1. The yeast polyQ model. a) Schematic representation of the toxic polyQ-expansion protein. b) Fluorescence microscopy showing yeast cells expressing a short, non-toxic polyQ expansion protein (25Q, left panel) and yeast cells expressing a long, toxic polyQ-expansion protein (103Q, right panel).
Figure 2. Representative results of growth assays of yeast cells expressing polyQ-expansion proteins. a) Plating assay. Approximately 700 yeast cells were spread on plates and incubated for three days at 30 °C. The upper panel shows a plate containing glucose medium, i.e. the expression of the toxic polyQ-expansion protein (103Q) is not induced. The lower panel shows a yeast plate containing galactose medium, i.e. the expression of the toxic polyQ-expansion protein is induced. Note that in the experiment shown here, no spontaneous suppressor occurred. b) Spotting assay. Five serial five-fold dilutions of yeast cells harboring either a non-toxic polyQ protein (25Q) or a toxic polyQ-expansion protein (103Q) were spotted on glucose plates that repress the expression of the proteins (left panel) or on galactose plates that induce their expression. The plates were then incubated for three days at 30 °C. c) BioscreenC experiment. The growth of cultures of yeast cells expressing either a non-toxic polyQ protein (25Q) or a toxic polyQ-expansion protein (103Q) were monitored by the BioscreenC instrument. The experimental conditions and the analysis of the BioscreenC experiments are described in the main text.
This manuscript outlines three complementary experimental approaches to measure polyQ-toxicity in the model organism yeast based on reduced growth of yeast cells expressing toxic polyQ-expansion proteins. Work in yeast has offered profound insights into the basic cellular and molecular mechanisms of protein misfolding and its ensuing toxicity, including the misfolding and toxicity of polyQ-expansion proteins 9,11,12. Experiments based on the protocols presented here have already helped to identify and characterize genetic enhancers or suppressors of polyQ toxicity, small molecule modifiers of polyQ toxicity, and cellular mechanisms underpinning polyQ toxicity 5-8,13,14.
The presented protocols can easily be adapted to explore other modifiers of polyQ toxicity such as different growth conditions or genetic mutations that either reduce or exacerbate polyQ toxicity. Furthermore, the presented experimental protocol can easily be adjusted to test the toxicity of other toxic misfolded proteins in yeast.
The authors have nothing to disclose.
Work in the Duennwald laboratory is supported by grants from the American Federation for Aging Research (AFAR), the Hereditary Disease Foundation (HDF) and the William Wood Foundation.
Name of instrument | Company | Catalogue number |
Frogger (6×8 pins) | V&P Scientific, San Diego | VP 407 AH |
BioscreenC | Growthcurves USA | 5101370 |
100-well Honeycomb plates | Growthcurves USA | 9602550 |