YODA: Software to facilitate high-throughput analysis of chronological life span, growth rate, and survival in budding yeast
BMC Bioinformaticsvolume 11, Article number: 141 (2010)
The budding yeast Saccharomyces cerevisiae is one of the most widely studied model organisms in aging-related science. Although several genetic modifiers of yeast longevity have been identified, the utility of this system for longevity studies has been limited by a lack of high-throughput assays for quantitatively measuring survival of individual yeast cells during aging.
Here we describe the Yeast Outgrowth Data Analyzer (YODA), an automated system for analyzing population survival of yeast cells based on the kinetics of outgrowth measured by optical density over time. YODA has been designed specifically for quantification of yeast chronological life span, but can also be used to quantify growth rate and survival of yeast cells in response to a variety of different conditions, including temperature, nutritional composition of the growth media, and chemical treatments. YODA is optimized for use with a Bioscreen C MBR shaker/incubator/plate reader, but is also amenable to use with any standard plate reader or spectrophotometer.
We estimate that use of YODA as described here reduces the effort and resources required to measure chronological life span and analyze the resulting data by at least 15-fold.
The ability to accurately monitor survival and growth rate of cells is essential for many assays employed in studies of the budding yeast. Changes in growth rate and survival over time are often monitored in response to a chemical treatment, environmental change (e.g. temperature, starvation, etc.), or genetic variant. For example, the yeast ORF deletion collection, which consists of >5000 unique single-gene deletion strains in an isogenic background, has been queried for more than 100 unique phenotypes by monitoring growth or viability under different conditions . Growth rate (doubling time) of yeast cells can be quantified by monitoring the change in optical density at 600 nm (OD600) of a yeast culture under specified conditions. Survival of yeast cells has traditionally been quantified by plating the cells onto rich growth medium (yeast peptone dextrose, YPD) and counting colony forming units (CFUs) before and after treatment.
One important assay that involves monitoring survival of yeast cells over time is measurement of chronological life span (CLS), which is defined as the length of time a yeast cell is able to maintain viability during post-diauxic growth arrest . Yeast CLS has emerged as a useful paradigm in aging-related research and has led to the identification of several dozen genetic modifiers of yeast longevity, some of which play a conserved aging-related function in multicellular eukaryotes . CLS has traditionally been performed by culturing cells in a synthetically defined medium and monitoring survival over time (every 2-3 days) by periodically removing an aliquot of the aging culture, serially diluting that aliquot, plating the cells onto YPD and counting CFUs .
We have recently developed a high-throughput method of measuring CLS that involves quantifying survival based on measuring the outgrowth of a defined number of cells by monitoring OD420-580 using a combined shaker/incubator/plate reader, the Bioscreen C MBR machine (Growth Curves USA) [5, 6]. This method is based on the fact that the optical density of a culture after a fixed period of growth will be proportionate to the number of viable cells present in the culture initially. We have used a qualitative version of this method in which only a single outgrowth optical density measurement was taken to screen the yeast ORF deletion collection for long-lived single-gene deletion strains , and more recently, the quantitative Bioscreen C MBR method has been used to quantify the effects of media composition on CLS, and to define a molecular mechanism of chronological aging in yeast [8, 9]. Briefly, this method involves culturing cells in individual 5 mL cultures in tubes on a rotating drum at 30°C (the aging cultures). At each age-point (e.g. day 2, 4, 6, etc.), 5 μL of each aging culture is inoculated into 145 μL of YPD in one well of a Bioscreen Honeycomb plate. Outgrowth of the cells from each aging culture is then determined by taking OD420-580 measurements for each well every 30 minutes for 24 hours using the Bioscreen C MBR machine. The Bioscreen C MBR machine has a maximum capacity of two 100-well Honeycomb plates, allowing for simultaneous measurement of outgrowth for up to 200 individual aging cultures. Survival at each age-point is determined by quantifying the rightward shift of the outgrowth curve along the time axis for each aging culture relative to the outgrowth curve for the aging culture at the initial age-point (day 2 of culture) using the formula:
where sn is the survival percentage, Δtn is the time shift, and δn is the doubling time. Detailed methodology and a video protocol describing the CLS assay have been published, and we refer the interested reader to these references [5, 6].
In order to automate analysis of the data generated during a CLS experiment, we developed a software package called YODA, the Yeast Outgrowth Data Analyzer. YODA accepts as input single or multiple text files containing OD values as a function of time (one file for each age-point) and returns several useful parameters, including maximal growth rate for each well, survival at each age-point for each aging culture, and the survival integral (SI) for each strain, which is defined as the area under the survival curve. YODA also has the capacity to group data from replicate cultures and perform simple statistical analyses of each group of replicates both individually and relative to experiment-matched control replicates. YODA is provided to the scientific community as a freely available utility on our website at http://www.kaeberleinlab.org/yoda and on the SAGEWEB website at http://www.sageweb.org/yoda. We have also provided the YODA source code for download along with an issue tracker at http://code.google.com/p/sageweb-yoda/.
Here we describe the key features of YODA and provide a demonstration of how YODA can be used to analyze data from a typical CLS experiment. We also provide examples of how YODA can be used for additional types of experiments in yeast where quantitation of cell survival or growth rate are desired and describe how data generated from sources other than a Bioscreen C MBR machine can be analyzed with YODA.
Uploading outgrowth data from a CLS experiment
Detailed instructions for performing chronological aging experiments using the Bioscreen C MBR machine are published [5, 6] and are also available on our website. At each age-point, the Bioscreen "EZExperiment" software produces a comma-delimited file containing OD420-580 values for each of the 200 wells corresponding to the maximal loading capacity of the machine (2 × 100 well plates). The first column of the resulting table contains the "Time" data (30 minute intervals by our protocol [5, 6], although this can be varied) and each subsequent column contains the data for a single well. In order to upload your data into YODA, the outgrowth data for each age-point should be saved in a separate comma-delimited (.csv) or Microsoft Excel (.xls) file. It is important that well position be maintained for each aging culture throughout the entire experiment (e.g. wild type replicate #1 always in well #1, mutant A replicate #1 always in well #2, wild type replicate #2 always in well #3, etc.). This is ensured by loading the wells in the same order at each age-point. Save each file with an appropriate identifying name, such as 'name_date.csv'.
Once your files are formatted and named appropriately, they can be uploaded to YODA by choosing the 'Upload' link in the Analyzer Menu. This will display the Upload Experiment form (Fig. 1). Fill in the experiment name, start date, and description of your experiment. If you want to use named wells for grouping, fill in the well info input (click the question mark next to the Well Info text box on the Upload Experiment form for detailed information about the format).
Upload each of your age-point outgrowth data files. Assign each run a day (for example, corresponding to the age of the culture at that age-point) so that YODA can determine which point in the survival curve belongs to which run (survival curve integrals automatically take into consideration days between runs). If you have named your wells using the well info input, type the name of the media-only well in the background field to compute a background reading for the run. The background reading defines the optical density value for a well with media only. Otherwise, select 'value' from the background type drop-down and type in a custom background (the default is 0.15, which is approximately the OD reading for 150 μL of YPD in the Bioscreen). When all of the data files for the experiment have been selected, click the 'Upload' button. YODA will tell you what the computed backgrounds are and if any errors have occurred in the results window (Fig. 2).
Quantifying growth rate and survival with YODA
To analyze your CLS data and export it for graphical presentation, click the "Export" link in the Analyzer Menu. This will display the Export form (Fig. 3). Select the desired experiment, data sets (age-points), and well positions. YODA's default settings will select all data sets and well positions (1-200) for a particular experiment, unless otherwise selected.
Under the Export window, set the "output" drop-down to either "runs" or "lineages" (Fig. 4). By default, selecting 'runs' will output well name, run name, run day, doubling times, and background normalized OD420-580 readings (click the "show options" link to adjust what parameters are exported and how they are calculated. See Table 1). To export the OD420-580 readings for a single well position across all data sets for a given experiment, select just the desired well in the select box and set the output drop-down to 'runs'. Selecting "lineages" will output doubling time, survival values at each age-point, and survival integrals for each well across age-points. If replicates have been grouped using the "Well Info" option, the average or median values for a well group can be output by selecting the desired option under "grouping". In order to calculate survival values at each age-point, the first age-point is defined as maximal survival (1.0) and subsequent survival values are calculated as the relative fraction surviving at each age-point. To graphically present survival curves from YODA, plot the fractional survival (y-axis) as a function of age (x-axis) for each well or replicate.
Using YODA for non-aging assays
YODA can be easily adapted for analysis of data for a variety of assays in which growth rate or survival is the parameter of interest. To quantify growth rate from OD420-580 readings obtained with a Bioscreen C MBR machine, simply upload a single comma-delimited "EZExperiment" output file as described above and export the data using the "runs" option from the "output" drop-down menu under the Export window. Quantification of survival in response to different experimental stimuli can be performed in a manner analogous to that described for CLS above, with the control treatment assigned as the first age-point in the CLS experiment and each experimental group assigned as a subsequent age-point. Specific examples are provided in the Results section below.
Using YODA with alternative methods of optical density determination
Although YODA is designed for simplified analysis of OD data obtained from the Bioscreen C MBR machine, OD measurements obtained from alternative sources, such as a standard plate reader or spectrophotometer, can also be analyzed with YODA. For such uses, simply ensure that the OD data is entered into .csv or .xls files in the manner described above for data obtained from the Bioscreen machine. All subsequent steps in the analysis are identical.
Quantitation of yeast CLS with YODA
In this exemplary yeast CLS experiment, the aging potential of a control and four single-gene deletion mutant strains was determined using a Bioscreen C MBR machine. Outgrowth analysis was performed as described above by inoculating 5 μL of each aging culture into 145 μL of YPD in individual wells of a Bioscreen Honeycomb plate. Each strain was assayed in triplicate using biological replicates (3 independently derived aging cultures per genotype). Outgrowth was determined at the following age-points: days 2, 4, 6, 9, 11, and 13. Outgrowth data for each age-point was retrieved as comma-delimited text files from the EZExperiment software (Additional File 1) and uploaded to YODA as described above. 'Average groups' was selected under the 'Grouping' options in the Export screen of YODA and the data was exported as a Microsoft Excel Worksheet. Survival curves were plotted from the data generated by YODA along with the standard deviation at each age-point for each strain (Fig. 5).
Quantitation of yeast survival following heat shock
In addition to determining the CLS of yeast strains, YODA can also be used to quantify cell viability following an environmental stressor such as heat shock. As an example experiment, an overnight culture of wild type cells was split into 100 μL aliquots and subjected to varying lengths of 55°C heat shock (0, 5, 10, 15, and 20 minutes). Following heat shock, 5 μL of each treatment was inoculated into 145 μL YPD in individual wells of a Bioscreen Honeycomb plate. Each treatment was assayed in triplicate.
To quantify viability following heat shock, the data output by the Bioscreen was split into individual comma-delimited (.csv) files, based on the length of time the cells were subjected to heat shock. For example, the first three wells of the Bioscreen plate were inoculated with cells subjected to 0 minutes of heat shock, the fourth through sixth wells were inoculated with cells subjected to 5 minutes of heat shock, and so forth. The data from the first three wells were put into a new comma-delimited file along with the data from one blank well; the data from the fourth through sixth wells were put into a second comma delimited file along with the data from the same blank well, and so on until all treatments were accounted for. Once each treatment had been placed into a separate .csv file, all files were then uploaded to YODA (see Additional File 2, for the original Bioscreen output file and the modified .csv files loaded into YODA). In the 'Upload Experiment' form, "0 minutes, 5 minutes, 10 minutes, etc." were filled in the 'Day' field next to the corresponding .csv files. 'Average groups' was selected under the 'Grouping' options in the Export screen of YODA and the data was exported as a Microsoft Excel Worksheet. Outgrowth and survival curves were generated from two independent experiments (Fig. 6). For comparison, viability following heat shock was also quantified using the traditional labor-intensive method of plating cells onto solid media following serial dilution and counting CFUs. CFU counts showed a similar loss in viability due to heat shock as was observed using the Bioscreen.
Quantitation of growth inhibition by rapamycin
As an alternative to survival, it is often desirable to monitor changes in growth kinetics in response to a stress or stimulus. To demonstrate how to use YODA for these types of assays, we examined the response of wild type, tor1Δ, and fpr1Δ cells to rapamycin, a chemical inhibitor of the target of rapamycin kinase and a putative anti-aging compound [10, 11]. For this experiment, cells were grown overnight in YPD and 5 μL of each culture was inoculated into either 145 μL of YPD or YPD containing 10 ng/mL rapamycin in individual Honeycomb plate wells. Each strain was assayed using five technical replicates. Outgrowth data was obtained using the Bioscreen C MBR machine, as described above. The resulting comma-delimited text file was uploaded to YODA with 'average groups' selected under 'grouping' as well as 'media' under the 'append properties' category. Growth curves were plotted using the data generated by YODA in Microsoft Excel (Fig. 7). As predicted from previously published data , the tor1Δ strain was sensitive to rapamycin, while the fpr1Δ strain was resistant.
YODA provides a web-based platform for quantifying survival and growth inhibition in the budding yeast. YODA was designed to facilitate high-throughput studies of yeast CLS, but is equally suitable for identifying genetic variants or environmental interventions that modify cell survival or growth rate. In addition to providing information not attainable using the traditional CFU method, such as growth rate of each strain, the use of YODA with a machine such as the Bioscreen C MBR enhances efficiency in the laboratory. For example, the method described here can accommodate up to three biological replicates of 66 strains in a single experiment (198 out of 200 wells used). We estimate that an entire experiment of this nature requires approximately 4 hours of effort for an experienced researcher, including preparation time, strain handling, and data analysis. In contrast, we estimate that the traditional CFU method with serial dilutions would require at least 15-fold longer for the same researcher to obtain equivalent data. In terms of resources, both methods require equal volumes of media for aging cultures; however, the method described here requires only 30 mL of liquid media and two multi-well plates per age point, whereas the traditional CFU method requires approximately 5 L of agar-based media and 198 Petri dishes per age-point (25 mL per plate × 198 strains). Thus we conclude that YODA in combination with the methods described here results in a significant reduction in time and resources required to measure CLS or perform other survival-based assays in yeast.
Availability and Requirements
YODA is provided to the scientific community as a freely available utility on our website at http://www.kaeberleinlab.org/yoda and on the SAGEWEB website at http://www.sageweb.org/yoda. YODA can be used anonymously without a username and password. We have also provided the YODA source code for download along with an issue tracker at http://code.google.com/p/sageweb-yoda/. Installation requires a web server with PHP 5.2 support and MySQL 5.1 or above.
Winzeler EA, Shoemaker DD, Astromoff A, Liang H, Anderson K, Andre B, Bangham R, Benito R, Boeke JD, Bussey H, et al.: Functional characterization of the S. cerevisiae genome by gene deletion and parallel analysis. Science 1999, 285(5429):901–906. 10.1126/science.285.5429.901
Fabrizio P, Longo VD: The chronological life span of Saccharomyces cerevisiae. Aging Cell 2003, 2(2):73–81. 10.1046/j.1474-9728.2003.00033.x
Steinkraus KA, Kaeberlein M, Kennedy BK: Replicative aging in yeast: the means to the end. Annu Rev Cell Dev Biol 2008, 24: 29–54. 10.1146/annurev.cellbio.23.090506.123509
Kaeberlein M: Longevity and aging in the budding yeast. In Handbook of models for human aging. Edited by: Conn PM. Boston: Elsevier Press; 2006:109–120. full_text
Burtner CR, Murakami CJ, Kaeberlein M: A genomic approach to yeast chronological aging. Methods Mol Biol 2009, 548: 101–114. full_text
Murakami C, Kaeberlein M: Quantifying yeast chronological life span by outgrowth of aged cells. J Vis Exp 2009., (27): pii: 1156 pii: 1156
Powers RW, Kaeberlein M, Caldwell SD, Kennedy BK, Fields S: Extension of chronological life span in yeast by decreased TOR pathway signaling. Genes Dev 2006, 20(2):174–184. 10.1101/gad.1381406
Burtner CR, Murakami CJ, Kennedy BK, Kaeberlein M: A molecular mechanism of chronological aging in yeast. Cell Cycle 2009, 8(8):1256–1270.
Murakami CJ, Burtner CR, Kennedy BK, Kaeberlein M: A method for high-throughput quantitative analysis of yeast chronological life span. J Gerontol A Biol Sci Med Sci 2008, 63(2):113–121.
Kaeberlein M: Resveratrol and rapamycin: are they anti-aging drugs? Bioessays 2010, 32(2):96–99. 10.1002/bies.200900171
Stanfel MN, Shamieh LS, Kaeberlein M, Kennedy BK: The TOR pathway comes of age. Biochim Biophys Acta 2009, 1790(10):1067–1074.
Xie MW, Jin F, Hwang H, Hwang S, Anand V, Duncan MC, Huang J: Insights into TOR function and rapamycin response: chemical genomic profiling by using a high-density cell array method. Proc Natl Acad Sci USA 2005, 102(20):7215–7220. 10.1073/pnas.0500297102
This work was supported by NIH grant R21AG031965, an infrastructure award from the Ellison Medical Foundation supporting SAGEWEB: the Science of Aging Electronic Resource Center, and a Pilot Project Grant from the University of Washington Nathan Shock Center of Excellence in the Basic Biology of Aging (NIH Grant P30AG013280) to MK. MK is an Ellison Medical Foundation New Scholar in Aging.
BO, CJM, and MK conceived of the project. MK wrote the manuscript. BO developed the software. CJM performed the yeast experiments. All authors read and approved the final manuscript.