Department of Biochemistry and Molecular Biology, School of Biomedical Sciences, Monash University
Mijaljica, D., Prescott, M., Devenish, R. J. A Fluorescence Microscopy Assay for Monitoring Mitophagy in the Yeast Saccharomyces cerevisiae. J. Vis. Exp. (53), e2779, doi:10.3791/2779 (2011).
Autophagy is important for turnover of cellular components under a range of different conditions. It serves an essential homeostatic function as well as a quality control mechanism that can target and selectively degrade cellular material including organelles1-4. For example, damaged or redundant mitochondria (Fig. 1), not disposed of by autophagy, can represent a threat to cellular homeostasis and cell survival. In the yeast, Saccharomyces cerevisiae, nutrient deprivation (e.g., nitrogen starvation) or damage can promote selective turnover of mitochondria by autophagy in a process termed mitophagy 5-9.
We describe a simple fluorescence microscopy approach to assess autophagy. For clarity we restrict our description here to show how the approach can be used to monitor mitophagy in yeast cells. The assay makes use of a fluorescent reporter, Rosella, which is a dual-emission biosensor comprising a relatively pH-stable red fluorescent protein linked to a pH-sensitive green fluorescent protein. The operation of this reporter relies on differences in pH between the vacuole (pH ~ 5.0-5.5) and mitochondria (pH ~ 8.2) in living cells. Under growing conditions, wild type cells exhibit both red and green fluorescence distributed in a manner characteristic of the mitochondria. Fluorescence emission is not associated with the vacuole. When subjected to nitrogen starvation, a condition which induces mitophagy, in addition to red and green fluorescence labeling the mitochondria, cells exhibit the accumulation of red, but not green fluorescence, in the acidic vacuolar lumen representing the delivery of mitochondria to the vacuole. Scoring cells with red, but not green fluorescent vacuoles can be used as a measure of mitophagic activity in cells5,10-12.
Selecting suitable control yeast strains
The assay relies on the experimenter visually evaluating the accumulation of red fluorescence in the yeast vacuole. As such the assay is subjective and relies on the selection of suitable control strains and growth conditions. A wild type control is used to indicate the levels of autophagy normally expected. As a negative control a strain null for expression of one of the core autophagy genes (e.g., ATG3) is suitable; such strains cannot deliver material (e.g., mitochondria, nucleus) to the vacuole
1. Transformation of yeast cells with plasmid DNA
Yeast cells can be easily made competent for transformation by treating with LiAcetate. Commercial kits may be purchased (e.g., EasyComp, Invitrogen) and published protocols are available13.
In this protocol we use wild type strain BY4741 (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0), and some members of a comprehensive library of deletion mutant strains (e.g., Δatg3) derived from it, each lacking expression of a different non-essential gene. The deletion strain library represents a valuable tool for investigating autophagy using Rosella-based probes. The reporter employed in this protocol is mt-Rosella for mitochondria (Fig. 2A), encoded by pAS1NB:mt-Rosella12.
The following steps are based on the use of cells made competent using the EasyComp Transformation Kit and stored frozen at -80°C for convenient use.
2. Confirming expression of Rosella in yeast cells
Before embarking on detailed experiments the correct localization and efficient expression of Rosella should be confirmed.
3. Cell growth and induction of autophagy
Autophagy can be induced using a number of different experimental conditions, including entry into stationary phase, change in carbon source, administration of rapamycin, or nitrogen starvation. We routinely use nitrogen starvation as the method to induce mitophagy.
Note: Labeling of the vacuole with CMAC-Arg
When first establishing the assay it is useful to confirm under the fluorescence microscope delivery of Rosella to the vacuole. Although the yeast vacuole is a large organelle whose location can often easily be determined by reference to transmitted light images, in some strains of yeast and under some growth conditions the vacuole can be fragmented and difficult to locate.
The vacuole can be readily labeled using a coumarin-based vacuole dye such as CMAC-Arg (7-amino-4-chloromethylcoumarin, L-arginine amide). The action of vacuolar proteases on the dye results in bright blue fluorescence labeling of the vacuole. The blue emission can be readily distinguished from the red and green emissions of Rosella11. Labeling with CMAC-Arg does not need to be performed routinely, but it is recommended for those who are not familiar with the location and appearance of the vacuole in yeast.
4. Labeling of the vacuole with CMAC-Arg
5. Mounting live yeast cells for confocal laser scanning microscopy (CLSM)
6. Visualizing autophagy using CLSM
The level of autophagy in a population of cells is routinely assessed by determining the number of cells showing red vacuolar fluorescence before and after the induction of autophagy. Ideally the progression of autophagy is monitored at selected time points following induction of autophagy (Fig. 3). This is best performed by visualizing through the microscope binocular. It is important that the correct controls are employed (e.g., Δatg3 mutant) and that the microscope settings remain unchanged between observing different samples (e.g., use of neutral density filters, filter selection).
Yeast cells are relatively small requiring a good quality fluorescence microscope (e.g., Olympus Fluoview FV500) equipped with excitation/emission filters suitable for the separate visualization of GFP and/or pHluorin (green fluorescence emission) (FITC filter) and DsRed.T3 (red fluorescence emission) (TRITC filter).
7. Representative results:
Here, we show typical results obtained using mt-Rosella, expressed in wild type and Δatg3 mutant cells. Under growing conditions with ethanol as carbon source, both wild type and Δatg3 mutant cells exhibit a cellular distribution of fluorescence (red and green) typical of mitochondria in yeast cells. Red and green fluorescence emission is not detected in the vacuole (Fig. 2B and 2C). When subjected to nitrogen starvation for 6 h and beyond then, in addition to red and green fluorescence corresponding to mitochondria, wild type cells also exhibit the accumulation of red, but not green fluorescence in the vacuolar lumen (Fig. 2B; Fig. 3). However, in Δatg3 mutant cells neither red nor green fluorescence accumulates in the vacuolar lumen (Fig. 2C; Fig. 3).
Figure 1. Top, scheme of organelles and compartments in a yeast cell. Bottom, a vacuole-centric view of a yeast cell is presented indicating autophagic degradation of different organelles and compartments. Some of the differences between the various organelle-specific types of autophagy (e.g., mitophagy, nucleophagy) include the specificity (cargo selection) of the engulfed contents, various intracellular needs and extracellular cues (e.g., starvation, damage), specific signals, ATG genes and time dependency.
Figure 2. (A) Schematic representation of mt-Rosella. The mitochondrial leader sequence (in this case of citrate synthase) is used to target the Rosella biosensor to the mitochondrial matrix. Excitation and emission maxima for both red fluorescent protein (DsRed.T3) and green fluorescent protein (pHluorin) are indicated. At high pH (mitochondrial pH~8.2) the biosensor fluoresces both red and green, however, at low pH (vacuolar pH~5.5) it fluoresces only red. (B) Schematic representation of expected results in wild type and Δatg3 mutant cells expressing mt-Rosella. (C) Actual results obtained by CLSM for wild type cells under growing and nitrogen starvation conditions. From left to right: DIC (difference in contrast), RFP (DsRed.T3) fluorescence emission, GFP (pHluorin) fluorescence emission, CMAC-Arg fluorescence and merge image. (D) Actual results obtained by fluorescence microscopy for Datg3 mutant cells under growing and nitrogen starvation conditions. From left to right: DIC (difference in contrast), RFP (DsRed.T3) fluorescence emission, GFP (pHluorin) fluorescence emission, CMAC-Arg fluorescence and merge image.
Figure 3. The percentage of wild type and Δatg3 mutant cells, expressing mt-Rosella and grown under nitrogen starvation, showing red fluorescence in the vacuole over a time course of 48 hours. The percentage of cells showing accumulation of red fluorescence in the vacuole was recorded at 0, 6, 12, 24 and 48 hours after commencing nitrogen starvation.
In the representative images presented in Fig. 2B and 2C it can be clearly seen that the use of organelle-specific fluorescent reporters and fluorescence microscopy provides evidence of mitochondrial localization/distribution in both growing and nitrogen-starved cells, and allows mitophagy to be visualized.
It is to be emphasized that this method is not restricted to following mitophagy. Autophagy of other cellular components or organelles (e.g., ribosomes, lipid droplets, peroxisomes, endoplasmic reticulum and nucleus) may be monitored with suitable labeling. Details of the Rosella biosensor construct, cell culture conditions and typical results for turnover of the nucleus (nucleophagy) have been reported11,12 illustrating the broader application of the method.
Use of organelle-specific fluorescent reporters for monitoring autophagy requires several considerations of construct design to be kept in mind. These include: (1) selection and fusion to the appropriate targeting signal to achieve organelle-specific targeting; (2) selection of a suitable fluorescent protein(s); (3) strong fluorescence emission; (4) correct organelle localization and/or distribution within the cell. Once the organelle-specific reporter has been successfully expressed in the chosen yeast host strain a second group of considerations become: (1) yeast growth conditions; (2) cell sample preparation; (3) imaging parameters as it is important that the parameters selected for CLSM (e.g., laser intensity, scan rate and mode, pinhole value, zoom value, objectives used or exposure for CCD-camera) are maintained throughout an experiment ensuring that comparisons can be made between control (growing cells) and starved samples; (3) autophagy induction and successful delivery of the reporter into the vacuole.
Overall, this fluorescent microscopy method is easy, relatively fast, and has potential application for high throughput screening for novel drugs that enhance or inhibit autophagy, and also for genes that regulate or modulate autophagy.
No conflicts of interest declared.
This work was supported by Australian Research Council Grant DP0986937.