Advertisement
Articles
Advertisement

Cell Sorting Based on RNA Detection in Living Cells

Wed, 05/01/2013 - 12:10pm
Don Weldon, Yuko Williams, Alex Ko, EMD Millipore Corporation

LISTED UNDER:

Cell sorting enables the isolation of highly pure cell subpopulations, increasing the statistical significance of observed relationships between gene expression and phenotype. Live cell sorting has traditionally been accomplished by detecting the presence of unique cell surface proteins, identified through the use of fluorescently-labeled antibodies. However, live cells cannot be sorted based on endogenous intracellular protein markers, because cells have to be fixed and permeabilized for antibody staining. Sometimes cells can be sorted by transfected reporter constructs; however, this treatment compromises cell integrity and may perturb cellular pathways, confounding downstream analyses.

Identifying cell types and sorting cells based on RNA expression levels without any transfection reagents or intrusive sample preparation can drastically improve live cell sorting efficiency, physiological relevance, and post-sorting survival rate. SmartFlare RNA detection probes (EMD Millipore) can detect levels of RNA inside living cells, providing the ability to sort and propagate live cell populations based on gene expression levels or in combination with surface markers detected with antibodies. This technology eliminates the need for permeabilization or transfection reagents, leaving the cells intact and viable after sorting. Because the particles leave the cells unharmed, the cells can be used in downstream assays.

The ability to separate live cells based on the level of a specific RNA target provides a new opportunity to study cellular functions and identify rare cell types such as tumor cells and cancer stem cells. This technology enables the sorting of cell populations that were previously difficult to sort, and improves sorting accuracy by using biologically relevant intracellular markers.

 

Methods

Determining cell viability

Mouse PBMCs and splenocytes were incubated with either RPMI 1640 medium alone or medium containing 1000X target-Cy5 SmartFlare probe, after which cell viability was measured using Trypan Blue staining.

Gene expression profiling

MDA-MB-231 cells were seeded and then incubated with EGFR SmartFlare probe following a 1:1000 dilution of stock solution, Scramble Control SmartFlare probe, or an equivalent volume of phosphate-buffered saline (PBS). Cells were lysed, total RNA was harvested, and whole genome expression analysis was performed using Illumina Human HT-12 v4 Expression BeadChips and the Illumina iScan microarray scanning system.

Figure 1. Twist expression levels in Hs578t and MCF-7 cells are distinguishable by intracellular detection as well as by RT-PCR. A. Confirmation of Twist expression levels by RT-PCR.B. Twist expression in Hs578t and MCF-7 cells as determined using SmartFlare technology and analyzed by flow cytometry.

mRNA detection and sorting

Hs578t and MCF-7 cells were mixed in a 1:1 ratio. SmartFlare Twist mRNA detection probe was added, and the cell population was sorted by fluorescence-activated cell sorting using a MoFlo XDP cell sorter (Beckman Coulter). Sorted populations were then returned to cell culture. Both high and low Twist-expressing populations were separated and collected for further gene expression analysis by quantitative reverse transcription-polymerase chain reaction (qRT-PCR). RT-PCR was performed for Twist, EGFR, and ESR1 on both the high and low sorted products.

microRNA detection

HUVEC and HeLa cells were mixed in a 1:1 ratio and incubated with the miR-155 SmartFlare RNA detection reagent. The cells were sorted based on the expression levels of miR-155. Isolated populations of high and low miR-155-expressing cells were retained and used for a TNFα treatment expected to upregulate Vascular Cell Adhesion Molecule (VCAM) in HUVEC cells and not affect HeLa cells, as they do not express VCAM.1 High and low miR-155-expressing, sorted populations were both stimulated with TNFα followed by VCAM antibody staining. Cells were detached using Accutase reagent, resuspended in culture medium and analyzed on a guava easyCyte 8HT flow cytometer (EMD Millipore).

 

Figure 2. Follow-up RT-PCR on Twist-sorted cells. Hs578T and MCF-7 cells have distinct mRNA expression profiles as shown in A. RT-PCR assays for Twist1 B, EGFR C, and ESR1 D were used to analyze high and low Twist-expressing cells and displayed profiles that were consistent with expected results.

Results

To confirm the probes are nontoxic, we tested the viability of cells that were either incubated in culture medium only, or in culture medium plus SmartFlare probe. No significant change in viability was observed. 

One drawback of traditional methods of RNA detection is that transfection is often required, which can alter gene expression in ill-characterized, unpredictable ways2. In contrast, SmartFlare probes do not require transfection. To study their effects on cellular processes, whole genome expression profiling was conducted on MDA-MB-231 cells treated with EGFR Ms-Cy5 SmartFlare RNA Probe, Scramble Control SmartFlare probe, or an equivalent volume of PBS. No significant gene expression changes were observed.

Twist expression was previously determined by RT-PCR to be higher in Hs578t cells than MCF-7 cells. Our intracellular RNA detection technology showed the same trend (Figure 1); however, because we were able to analyze expression in individual, intact cells, we obtained additional information about population distribution. Specifically, the broader peak observed for MCF-7cells indicated that they exhibited a wider range of Twist expression than the Hs578 cells, which produced a narrower, slightly bimodal peak. In contrast, the RT-PCR data merely provided the average level of gene expression.3


Figure 3. Sorting based on miR-155 expression level differences in HeLa and HUVEC cells. miR-155 is expressed at a higher level in HUVEC cells than in HeLa cells, allowing for sorting of the mixed population of cells using SmartFlare technology. Gates are shown drawn around miR-155 high and miR-155 low populations, which were sorted accordingly.Since the probe technology leaves cells unharmed, the same cells sorted based on Twist mRNA expression were reused to detect expression of other genes of interest using RT-PCR (Figure 2). Results were consistent with expected gene expression profiles.

miR-155 has been implicated in the development and pathologic progress of hypertension and is highly expressed in the vascular endothelium.4 Detection of miR-155 was used to sort a mixed population of HUVEC and HeLa cells; HUVEC cells expressed higher levels of miR-155 (Figure 3).

Because the probes are nontoxic, sorted products can be returned to culture, essentially unaltered, for further downstream applications. In this study, both miR-155-positive and miR-155-negative sort products were returned to culture and tested for their ability to upregulate VCAM expression after TNFα treatment. As predicted, the HUVEC cells, which were high in miR-155 expression, upregulated the expression of VCAM following the TNFα treatment, while the HeLa cells did not (Figure 4).5


Figure 4. TNFα treatment of miR-155 sort products followed by VCAM antibody staining and detection using flow cytometry. A. Stimulation of the miR-155 low population with TNFα showed no significant increase in VCAM staining. B. TNFα-stimulation of miR-155 high sort product showed significant increase in VCAM staining. C. Overlay of TNFα-stimulated miR-155 low and high sort products showed an 8.7 fold increase in mean fluorescence intensity (MFI).Conclusion

This article demonstrated the ability to sort live cells based on intracellular gene expression. The sorted products were returned to culture, viable and unchanged, enabling downstream analyses such as antibody staining, flow cytometry and RT-PCR. We have shown that SmartFlare probes enable the simultaneous detection of multiple RNAs and allow the researcher to reuse those same live cells for further protein analysis, providing a link between the transcriptome and the proteome that was missing until now. The ability to sort cells and obtain a highly enriched cell population based on gene expression greatly increases the sensitivity of cell analysis, making analysis of the molecular roles of rare events possible.


References

  1. Zhu N et. al., Endothelial enriched microRNAs regulate angiotensin II-induced endothelial inflammation and migration. Atherosclerosis. 2011 Apr;215(2):286-93.
  2. Manakov SA et al. Neuronal miRNAs stabilize the neuronal transcriptome. EMBL-EBI database, Experiment E-MTAB-686. 2011.
  3. Vesuna F et al. Twist contributes to hormone resistance in breast cancer by downregulating estrogen receptor-α. Oncogene. 2012 Jul 5;31(27):3223-34.
  4. Kurowska-Stolarska M et. al., MicroRNA-155 as a proinflammatory regulator in clinical and experimental arthritis. Proc Natl Acad Sci U S A 2011 Jul 05;108(27):11193-8.
  5. Mackay F et al. Tumor necrosis factor alpha (TNFα-alpha)-induced cell adhesion to human endothelial cells is under dominant control of one TNFα receptor type, TNF-R55, J Exp Med. 1993 May 1; 177(5): 1277–1286.



Advertisement

Share this Story

X
You may login with either your assigned username or your e-mail address.
The password field is case sensitive.
Loading