Search Thermo Fisher Scientific
An Optimized Gene Expression Workflow for Single Cells
Typical column methods for RNA purification are not amenable to analysis of RNA from extremely small samples such as individual cells due to sample loss and large elution volumes. Alternate procedures, such as boiling, have been used to prepare RNA from small samples. Boiling does not efficiently free RNA from its cellular matrix, can induce RNA loss due to hydrolysis, and does not remove genomic DNA which can interfere with accurate RNA analysis. Applied Biosystems Cells-to-CT™ Kits lyse cultured mammalian cells, free and stabilize their RNA and remove genomic DNA to maximize the sensitivity of real-time RT-PCR experiments. The TaqMan PreAmp Cells-to-CT Kit includes reagents for preamplification, which enable quantitation of many targets from extremely limited samples. Applied Biosystems scientists have improved the reliability of gene expression analysis in single cells by optimizing the protocol for the TaqMan PreAmp Cells-to-CT Kit to prepare single cells for preamplification and real-time PCR analysis. This enables gene expression analysis of samples as small as a single cell.
Studying Individual Cells: Worthwhile Despite the Technical Challenges
The study of single cells has emerged as an important area in life science research. Complex tissues may contain very small cell subpopulations with unique gene expression profiles, and averaging gene expression results from many cells can mask subtle, important changes occurring within single cells. Microenvironments around identical cell types can cause significantly different gene expression profiles, and furthermore, mutations can result in gene expression changes at the single cell level. Even genetically identical cells exposed to identical conditions display heterogeneity in gene expression [1].
Technical advances in single cell isolation, such as laser capture microdissection and fluorescence-based cell sorting, provide the means to capture and recover individual cells, but the detection of minute starting amounts of RNA remains a significant challenge to studying gene expression in these tiny samples, particularly when analysis of multiple targets is required.
Technical advances in single cell isolation, such as laser capture microdissection and fluorescence-based cell sorting, provide the means to capture and recover individual cells, but the detection of minute starting amounts of RNA remains a significant challenge to studying gene expression in these tiny samples, particularly when analysis of multiple targets is required.
Overcoming the Problem of RNA Isolation from Single Cells
For routine gene expression studies using real-time RT-PCR, researchers typically prepare samples by purifying RNA using organic solvents, glass fiber filters, or magnetic beads. A disadvantage of these methods is that material can be lost due to incomplete RNA precipitation, binding, and elution. This loss may be insignificant with ordinary-sized samples, but can be catastrophic when the starting material consists of very small samples or single cells, such as those obtained by FACS or microdissection. Additionally, scaling down elution volumes is unsatisfactory because it can result in sample loss.
One solution that has been used with some success is to omit nucleic acid purification, and to lyse cells by boiling in water or buffer. However, simple boiling may not liberate RNA from its cellular matrix as reproducibly as RNA purification. In contrast, Applied Biosystems Cells-to-CT™ Kits effectively and reproducibly lyse cells for gene expression analysis with minimal processing, no organic solvents or filters, and no sample transfers—directly in the plates in which the cells are grown or collected. Cells-to-CT lysates are fully compatible with Applied Biosystems reverse transcription and amplification reagents. The problem of sample loss is overcome by eliminating RNA purification steps, and the optimized lysis solution results in improved target detection compared to sample boiling.
One solution that has been used with some success is to omit nucleic acid purification, and to lyse cells by boiling in water or buffer. However, simple boiling may not liberate RNA from its cellular matrix as reproducibly as RNA purification. In contrast, Applied Biosystems Cells-to-CT™ Kits effectively and reproducibly lyse cells for gene expression analysis with minimal processing, no organic solvents or filters, and no sample transfers—directly in the plates in which the cells are grown or collected. Cells-to-CT lysates are fully compatible with Applied Biosystems reverse transcription and amplification reagents. The problem of sample loss is overcome by eliminating RNA purification steps, and the optimized lysis solution results in improved target detection compared to sample boiling.
Optimizing the TaqMan Cells-to-CT Protocol for Single Cells
To optimize the TaqMan Cells-to-CT procedure for single cell analysis, we tested whether lysis and stop incubation times could be increased. Flexibility at these steps allows more time for complex and time-consuming single cell isolation procedures such as microdissection, micromanipulation, and cell sorting.
In the first phase of this study, 8 biological replicates of 10 HeLa cells were lysed for either 5 or 30 min, with subsequent stop incubation times of 5, 15, or 60 min. The samples were then evaluated using the TaqMan Gene Expression Assay for ß-actin following the procedure in the TaqMan Cells-to-CT Protocol. Consistent CT values were obtained across the different lysis and stop times (Figure 2). This result demonstrates that with these limited sample sizes, the Cells-to-CT workflow is robust to changes in sample preparation time, allowing the flexibility needed to accommodate single cell sample capture procedures.
Figure 2. Consistent CT Values Despite Varied Lysis and Stop Incubation Times. HeLa cells (10 cells per reaction) were lysed for 5 or 30 minutes, with subsequent reaction stop times of 5, 15, or 60 minutes using the TaqMan Cells-to-CT™ Kit. Consistent CT values for real-time PCR of ß-actin were obtained. Error bars represent the standard deviation of 8 biological replicates.
In the first phase of this study, 8 biological replicates of 10 HeLa cells were lysed for either 5 or 30 min, with subsequent stop incubation times of 5, 15, or 60 min. The samples were then evaluated using the TaqMan Gene Expression Assay for ß-actin following the procedure in the TaqMan Cells-to-CT Protocol. Consistent CT values were obtained across the different lysis and stop times (Figure 2). This result demonstrates that with these limited sample sizes, the Cells-to-CT workflow is robust to changes in sample preparation time, allowing the flexibility needed to accommodate single cell sample capture procedures.
Figure 2. Consistent CT Values Despite Varied Lysis and Stop Incubation Times. HeLa cells (10 cells per reaction) were lysed for 5 or 30 minutes, with subsequent reaction stop times of 5, 15, or 60 minutes using the TaqMan Cells-to-CT™ Kit. Consistent CT values for real-time PCR of ß-actin were obtained. Error bars represent the standard deviation of 8 biological replicates.
Comparing Current Methods for Single Cell Analysis
For single cell analysis, preamplification of cDNA prior to real-time PCR is especially valuable. With the TaqMan PreAmp Cells-to-CT Kit, an intermediate amplification step between reverse transcription and real-time PCR is performed in which cDNA is enriched for up to 100 gene targets of interest without amplification bias. The resulting preamplified reaction is diluted and serves as the starting material for the subsequent real-time PCR of individual targets using the TaqMan Gene Expression Master Mix included in the kit, and the TaqMan Gene Expression Assays of choice. The TaqMan PreAmp process effectively extends samples up to 64-fold, enabling analyses previously not possible from single cells or limited sample material.
The TaqMan PreAmp Cells-to-CT cell lysis protocol was compared to boiling cultured samples in reverse transcription (RT) buffer, a method commonly used to release RNA from small numbers of cells (Figure 3). The PreAmp Cells-to-CT procedure resulted in significantly higher sensitivity compared to boiling samples in buffer. Additionally, Cells-to-CT reagents remove genomic DNA which can interfere with accurate analysis of RNA in single cells.
Figure 3. High Sensitivity and Efficient Genomic DNA Removal with the Cells-to-CT™ Procedure as Compared to Boiling. Genomic DNA was detected in the “–RT” reactions (faded bars). Experimental details are similar to those in Figure 2, with unmodified lysis and stop times.
The TaqMan PreAmp Cells-to-CT cell lysis protocol was compared to boiling cultured samples in reverse transcription (RT) buffer, a method commonly used to release RNA from small numbers of cells (Figure 3). The PreAmp Cells-to-CT procedure resulted in significantly higher sensitivity compared to boiling samples in buffer. Additionally, Cells-to-CT reagents remove genomic DNA which can interfere with accurate analysis of RNA in single cells.
Figure 3. High Sensitivity and Efficient Genomic DNA Removal with the Cells-to-CT™ Procedure as Compared to Boiling. Genomic DNA was detected in the “–RT” reactions (faded bars). Experimental details are similar to those in Figure 2, with unmodified lysis and stop times.
Expression Analysis from Single Cells
HeLa cells were sorted by FACS and the modified TaqMan PreAmp Cells-to-CT procedure was used to analyze b-actin expression levels from single cells, 10 cells, and 100 cells. CT values corresponded linearly to cell number (Figure 4). The larger standard deviation observed for single cell samples is due to expression level variations from cell to cell. This is demonstrated by the fact that single cell equivalents (e.g., one-hundredth of the lysate from 100 cells) do not display this variation (data not shown).
Figure 4. Linear Correlation between CT Values and Cell Number. HeLa cells were sorted into 82 single cells, 4 sets of 10 cells, and 4 sets of 100 cells and lysed according to the TaqMan PreAmp Cells-to-CT™ protocol. Real-time PCR was performed for ß-actin. There is a linear correlation between CT values and cell number. The standard deviation increases with decreasing cell numbers due to biological variation among single cells.
Figure 4. Linear Correlation between CT Values and Cell Number. HeLa cells were sorted into 82 single cells, 4 sets of 10 cells, and 4 sets of 100 cells and lysed according to the TaqMan PreAmp Cells-to-CT™ protocol. Real-time PCR was performed for ß-actin. There is a linear correlation between CT values and cell number. The standard deviation increases with decreasing cell numbers due to biological variation among single cells.
Conclusions
By optimizing conditions for sample preparation and reverse transcription, a reliable method for gene expression analysis from extremely small samples is demonstrated. Most importantly, the workflow incorporates the flexibility in timing that enables the use of FACS and micromanipulation, two of the most common methods for obtaining single cells. Furthermore, the entire sample lysate can be used in reverse transcription and preamplification if desired, maximizing sensitivity and expanding precious samples.
More details on the optimized TaqMan PreAmp Cells-to-CT protocol will be available soon.
Scientific Contributors:
Richard Fekete, Alexis Lennart, Ron Abruzzese, Laura Chapman, David Keys • Applied Biosystems
More details on the optimized TaqMan PreAmp Cells-to-CT protocol will be available soon.
Scientific Contributors:
Richard Fekete, Alexis Lennart, Ron Abruzzese, Laura Chapman, David Keys • Applied Biosystems
References1. Kaern M, Elston TC, Blake WJ, Collins JJ (2005) Stochasticity in Gene Expression: From Theories to Phenotypes. Nat Rev Genet 6(6):451–64.