Fluorescent in situ hybridization (FISH) allows the detection of nucleic acids in their native environment within cells. We here describe a protocol for the combined, simultaneous detection of RNA and DNA by means of FISH, which can be used to study X chromosome inactivation in mouse embryonic stem cells.
Fluorescent in situ hybridization (FISH) is a molecular technique which enables the detection of nucleic acids in cells. DNA FISH is often used in cytogenetics and cancer diagnostics, and can detect aberrations of the genome, which often has important clinical implications. RNA FISH can be used to detect RNA molecules in cells and has provided important insights in regulation of gene expression. Combining DNA and RNA FISH within the same cell is technically challenging, as conditions suitable for DNA FISH might be too harsh for fragile, single stranded RNA molecules. We here present an easily applicable protocol which enables the combined, simultaneous detection of Xist RNA and DNA encoded by the X chromosomes. This combined DNA-RNA FISH protocol can likely be applied to other systems where both RNA and DNA need to be detected.
Studying cells and tissues by means of fluorescent in situ hybridization (FISH) has, since its introduction in the late 1970s1, allowed researchers to study genes, chromatin organization and gene expression at the subcellular level. DNA FISH is frequently used in cytogenetics, karyotyping2, cancer diagnostics3 and pre-implantation genetic screening4, and has an important role in molecular research5-6 since it allows the detection of nucleic acids in their native environment. Single cell expression analysis by RNA FISH can detect native primary transcripts and noncoding RNAs being transcribed from chromosomes, and offers advantages to other techniques which assess gene expression at a population level, including for example quantitative RT-PCR, genome wide expression analysis or Northern blotting. By visualizing RNA transcripts originating directly from their sites of origin, it has for example been noticed that gene expression is stochastic7, and can sometimes be allele specific8. Improvements in the technique have even allowed the detection and quantification of single mRNA molecules within cells9-11.
The basic principle of FISH consists of hybridization of nucleic acids within the cell to a nucleic acid probe by means of highly specific Watson and Crick base-pairing. The probe can be either directly or indirectly detected, resulting in a signal which can be microscopically visualized. Initial attempts consisted of radioactively labeled probes, which had drawbacks based on safety issues, limited spatial resolution and the ability to detect only one target at a time12-14. The subsequent development of nonradioactive labels, including fluorochromes, haptens, and enzymes, has allowed the wide spread use of FISH as a routine molecular biology technique. The FISH probe can either be directly labeled with fluorochromes by chemical linking of fluorescent molecules to nucleic acid sequences15 and integration of fluorescently labeled nucleotides16-19, or the probe can be indirectly visualized after integration of haptens (including biotin and digoxigenin) and immunological detection of haptens by hapten specific antibodies conjugated to fluorescent reporter molecules20-21. The latter approach allows signal amplification by using several layers of fluorescently labeled antibodies which are used to enhance the original signal, and enables the detection of RNA species which are expressed at low levels. By combining direct and indirect labeling techniques and various haptens, several targets can be simultaneously visualized within the same cell.
One of the most important steps in FISH protocols is the hybridization of the probe to its target. Although in theory, a probe will specifically bind only to its target, in practice, this specificity is not always achieved, as probes may bind to homologous regions, and hybridization conditions will not always be ideal for a certain DNA region or RNA species. Post-hybridization washes are therefore of particular importance, as they can increase the stringency of the FISH procedure, and can prevent nonspecific binding of FISH probes, which would result in a high level of background noise. As RNA molecules are single stranded they can easily be hybridized to a FISH probe. In contrast, the double stranded DNA molecule first needs a denaturation step, after which a probe can hybridize. This is usually achieved by heating of the specimen, which results in denaturation of the DNA. However, under these harsh conditions, fragile, single stranded RNA molecules might be lost. Therefore, combined DNA-RNA FISH requires significant optimization of the conditions, and is more technically challenging compared to the separate detection of only RNA or DNA.
Here we present a detailed protocol of combined, simultaneous DNA-RNA FISH which has allowed us to study X chromosome inactivation (XCI) in differentiating female mouse embryonic stem cells22-24. XCI is a crucial epigenetic mechanism for female embryonic development25, and results in heterochromatinization and hence silencing of one of the two X chromosomes in female individuals26-27. Essential to this process is the noncoding RNA Xist28-30, which is regulated by the RNF1222-23 and REX124 proteins. Xist expression becomes upregulated on the future inactive X chromosome (Xi) during embryonic development or upon ES cell differentiation in vitro, and can spread along the X chromosome and thereby attract chromatin remodeling enzymes which result in the transcriptional shutdown of the X chromosome31. This spreading of Xist RNA can be visualized by RNA FISH as a coating of the X chromosome, which is also referred to as an Xist cloud. Since female ES cells can lose one of their X chromosomes due to genomic instability, we and others have employed combined DNA-RNA FISH to study XCI, to make sure that only karyotypically stable cells are assessed in the analysis of this important process22,32-34. As with every molecular biology technique, several different excellent protocols have been published35-38. Here we present our method, starting from the induction of differentiation in mouse ES cells, fixation of cells, labeling of FISH probes by Nick-translation, pre-treatment of fixed cells to allow permeabilization and subsequent probe uptake, hybridization of the probe to the target, and finally detection of the probe by fluorescently labeled antibodies. The herein presented protocol allows the faithful detection of Xist RNA and the X chromosome within a period of two days, and the basics of this technique can likely be adapted to other systems and areas of research.
1. Induction of Differentiation in Female Embryonic Stem Cells to Induce X Chromosome Inactivation
NOTE: Female mouse embryonic stem cells (available upon request) are grown under standard ES cell conditions on gelatinized culture dishes coated with mouse embryonic fibroblasts (MEFs). We here assume that the reader is familiar with standard cell culture techniques39-41. To induce differentiation, ES cells grown in T25 dishes will be separated from the MEFs, and will be plated in differentiation medium.
2. Fixation of Differentiated ES Cells for Subsequent DNA-RNA FISH Analysis
3. Labeling of DNA Probes for FISH Experiments
NOTE: Obtain a DNA fragment of the gene of interest (minimal length >2 kb for RNA FISH, or >20 kb for DNA FISH), or a BAC covering the gene of interest. For RNA detection, a smaller probe can be sufficient compared to DNA detection, since most likely transcription will result in multiple transcripts, which can be simultaneously detected. For a strong signal on the single copy DNA strand, a longer probe is required, to be able to detect a sufficient number of incorporated haptens. Preferably, a non-transcribed genomic region is chosen for designing the DNA probe. In addition, when using a smaller probe to detect the RNA, it will be prevented that the genomic loci from which the RNA is transcribed is being detected in the combined DNA-RNA FISH approach, although this cannot be fully excluded. To detect mouse Xist, a 5.5 kb cDNA probe was used. For detection of the X chromosome, several BAC probes can be used. Good results have been obtained with a mixture of BAC RP23-100E1 and BAC CT7-474E4, which are located in close proximity to the Xist locus. Depending on the gene or chromosome of interest, several different probes might be required to obtain optimal results. Whenever working with materials which will be used for RNA-FISH, use sterile filter tips, RNAse free H2O, and work in a clean environment.
4. Permeabilization, Pre-treatment and Hybridization of Fixed Cells for Combined DNA-RNA FISH
5. Post-hybridization Washes and Antibody Mediated Detection of Probe
Using the above mentioned protocol for combined DNA-RNA FISH, we have been able to visualize X chromosome inactivation in differentiating female embryonic stem cells. Figure 1 shows a representative example of an DNA-RNA FISH experiment, where we detected both Xist (which is visible as an so-called Xist RNA cloud on the inactive X chromosome and a basal transcription pinpoint on the active X chromosome), and a region of the X chromosome, which is visible as a pinpoint signal. Note that one of the pinpoint signals is located within the Xist cloud, thereby showing that the Xist cloud is indeed localizing along, and coating the X chromosome. In more than 99% of cells, we detect using our protocol a correct DNA signal (data not shown). Compared to RNA-FISH, Xist clouds visualized using the combined DNA-RNA FISH procedure look sometimes larger, as the denatured DNA seems to occupy a larger area. We have obtained similar results using human ES cells, human lymphocytes and various fibroblast cell lines of various species, and the same protocol has been applied to study gene expression of various X-linked loci, including Rnf12. In addition, when this protocol was applied to undifferentiated female ES cells, basal Xist expression from both X chromosomes could be detected, which shows that using this protocol also genes expressed at low levels can be visualized (data not shown).
Figure 1. Combined DNA-RNA FISH in differentiated female ES cells. Representative results obtained after the herein described protocol for combined DNA-RNA FISH. The X chromosome (X chr.) is detected using a digoxigenin-labeled BAC probe, which is visualized using antibodies conjugated to FITC (green signal). In almost every cell, two X chromosomes are detected. Xist RNA is detected using a biotin labeled cDNA probe, which is visualized using antibodies conjugated to rhodamine (red signal). Both large Xist accumulations on the inactive X chromosome (Xist clouds) and basal Xist expression from the active X chromosome (Xist pinpoints) are detected (note: upon differentiation this basal Xist expression is ceased on the future active X chromosome, and therefore not detected in every nucleus). Nuclei are counterstained with DAPI (blue). Scale bar represents 10 µm. Click here to view larger image.
Combined DNA-RNA FISH can be technically challenging, as conditions suitable for DNA FISH can be too harsh for less stable RNA molecules. Several approaches have been applied to study both DNA and RNA in the same cell, circumventing the need for simultaneous incubation with probes detecting both the DNA and RNA. For example, in a superimposition approach, first RNA FISH is performed, and cells are imaged, and coordinates are taken. Subsequently, the same slides are used for DNA FISH, during which the signal of RNA FISH is lost. After DNA FISH, the cells are again imaged, and the pictures obtained from both the RNA and DNA FISH are superimposed. Although this approach will work in almost all cases, as the DNA of fixed cells is stable, and is not affected by the treatment needed for initial RNA FISH, this is a very laborious approach which will cost at least four days. In another approach, first the RNA FISH is performed, the RNA FISH signal is fixed by addition of fixatives, after which DNA FISH is applied. Although also this approach can work, some of the fluorescent signal derived from the RNA FISH can be lost upon fixation. This might especially hinder the detection of transcripts which are only expressed at a low level. The herein presented approach is optimized for the simultaneous detection of both one RNA target and one DNA target in a single FISH experiment, and has the advantage that both results can be obtained in a single experiment within two days, with reliable results.
Several steps in the FISH protocol are crucial to obtain optimal results. First, whenever dealing with RNA molecules, one should be careful not to introduce RNAse in any buffer or solution used. Hence, a clean working environment should be established, and filter tips should be used when pipetting reagents being used for FISH. Care should be taken to use RNAse free compounds (for example RNAse free water, cell culture grade PBS, absolutely pure ethanol etc.). When these simple rules are followed, it is generally not necessary to use additional RNAse inhibitors supplemented to the reagents used. Second, the permeabilization procedure using pepsin is a gentle equilibrium between too little, or too much permeabilization. Depending on the particular cell type used, it might be required to optimize the time allowed for permeabilization, or the pepsin concentrations. The latter also depends on your particular batch of pepsin, as the activity might vary. As an alternative, permeabilization using 0.1% Triton-X100/PBS for 5 min results in good results under certain conditions, and has the advantage that FISH using this method of permeabilization can be combined with immunostaining to detect nuclear or cytoplasmic proteins. As a third important aspect of FISH, the aging, denaturation, hybridization and washing temperatures should be kept constant. Even small deviations from the target temperature can result in less efficient denaturation, hybridization, or less stringent washes, which will result in more background staining. As an alternative to the 1 hr aging step at 65 °C followed by denaturation at 65 °C for 2 min, also an initial denaturation at 80 °C for 5 min, without aging, can give good results. We prefer the aging procedure, as in general this procedure gives more reliable DNA FISH signals. Lastly, certain DNA probes will be so specific, that they will not require three layers of antibodies for sufficient detection. This should be empirically tested for every newly generated probe.
Although the current protocol works robustly in the study of X chromosome inactivation and detection of Xist, combined detection of RNA and DNA might be more complicated in other settings. This might especially be the case when the RNA species which is aimed to be detected is only expressed at very low levels, is full of repeat sequences or is a relatively short transcript. In these situations, it might be rather difficult to design an optimal probe, which will allow the efficient hybridization to a limited, not optimal available amount of target. In such situations, it might be worth optimizing first conditions for RNA FISH. Several alternative probes could be tried, and the duration of probe labeling by Nick-translation could be titrated, to optimize the amount of incorporated haptens. Importantly, it should be verified that the aimed RNA is expressed in the cells to be investigated, by for example verifying expression by RT-PCR. Also when analyzing a different type of cells, it is essential to include cells in which expression has previously been detected as positive controls, as this might allow the distinction between a technical, FISH related problem, or the simple absence of expression in the new type of cells analyzed. When, despite verified expression, the use of various probes and optimization of labeling conditions, RNA FISH does not result in a clear signal, it might be worth changing the hybridization conditions, by varying the amount of formamide used, or the duration and temperature of probe hybridization. Alternatively, directly labeled oligonucleotide probes might be used9-10. The same optimization steps can subsequently be applied to improve DNA FISH, and the simultaneous detection of DNA and RNA.
Using the combined DNA-RNA FISH protocol, we have been able to study X chromosome inactivation in differentiating female ES cells. Similar results have been obtained in our studies using different cell types and embryos, and we are currently further optimizing conditions to apply combined DNA-RNA FISH on tissue sections. As the important role of non-coding RNAs in many biological processes becomes more and more appreciated, we foresee that the same protocol can likely be used to study other non-coding RNAs in the context of their specific chromatin.
The authors have nothing to disclose.
The authors thank all previous and present members of the department of Reproduction and Development of the Erasmus MC for helpful and stimulating discussions during the past years. We also want to thank the three anonymous reviewers for providing helpful feedback. Work in the Gribnau lab is supported by funding from the Dutch research council (NWO-VICI) and an ERC starting grant.
female mouse embryonic stem cells | will be made available upon request | ||
mouse embryonic fibroblasts | can be derived from E13.5 embryos, or via commercial sources e.g Millipore | ||
ES cell culture medium | DMEM, 15% foetal calf serum, 100 U ml-1 penicillin, 100 mg ml-1 streptomycin, non-essential amino acids, 0.1 mM ß-mercaptoethanol, and 1000 U ml-1 LIF. Filter sterilize and store at 4ºC for maximal 2 weeks. | ||
LIF | Chemicon | ESG1107 | |
DMEM | Invitrogen | 11960085 | |
Fetal Calf Serum | Invitrogen | 10099141 | |
Penicillin–streptomycin (100×) | Invitrogen | 15140-122 | |
Non-essential aminoacids | Invitrogen | 11140-050 | |
ß-mercaptoethanol | Sigma-Aldrich | M7522 | |
ES cell differentiation medium | IMDM-Glutamax, 15% heat inactivated foetal calf serum, 100 U ml-1 penicillin, 100 mg ml-1 streptomycin, non-essential amino acids, 37.8 μl/l monothioglycerol and 50 μg/ml ascorbic acid | ||
IMDM-Glutamax | Invitrogen | 31980-030 | |
Ascorbic acid (50 mg/ml) | Sigma-Aldrich | A4403 | |
monothioglycerol | Sigma-Aldrich | M6145 | |
T25 cell culture dishes | Greiner Bio-one | 690160 | |
6-well cell culture dishes | Greiner Bio-one | 662160 | |
cell culture grade PBS | Sigma-Aldrich | D8537 | |
Trypsin-EDTA 0.25% (v/v) trypsin/0.2% (w/v) EDTA in PBS | Gibco | 25200-056 | |
falcon tubes | Falcon | 352196 | |
24×24 mm coverslips | Menzle | 780882 | |
gelatin | Sigma-Aldrich | G1890-100G | prepare a 0.2% gelatin/cell culture PBS solution, and autoclave |
paraformaldehyde | Sigma-Aldrich | P6148 | prepare 4% PFA/PBS solutions, by dissolving PFA in PBS. Stir at 70 ᴼC until completely dissolved and cool down to room temperature. Aliquots can be stored at -20ᴼC |
absolute 100% ethanol | Sigma-Aldrich | 32221-2.5L | use absolute 100% ethanol and RNAse free water to make 70% and 90% ethanol solutions |
RNAse free water | Baxter | TKF7114 | |
sterile filter tips | Rainin Bio Clean | GP-L10F, GP-L200F, RT-1000F | |
digoxigenin-labeling kit (DIG-Nick translation) | Roche | 11745816910 | |
biotin-labeling kit (Biotin Nick translation) | Roche | 11745824910 | |
Sephadex G50 | Sigma-Aldrich | G5080 | |
2 ml syringe | BD Plastipak | 300013 | |
sterilized cotton | |||
eppendorf tubes | Eppendorf | 30120086 | |
yeast tRNA 10 mg/ml | Invitrogen | 15401-029 | |
Salmon Sperm DNA 10 mg/ml | Invitrogen | 15632-011 | |
mouse cot-1 DNA 1 mg/ml | Invitrogen | 18440-016 | |
eppendorf bench top microcentrifuge | Eppendorf | Model 5417R | |
Eppendorf Refrigerated Centrifuge | Eppendorf | Model 5810R | |
2M NaAc, pH 5.6 | Sigma-Aldrich | S7670 | |
50+ hybridization solution | 50% formamide, 2x SSC, 50 mM phosphate buffer pH 7, 10% dextran sulfate pH 7 | ||
formamide | Acros organics | 3272350000 | |
dextran sulfate | Fluka | 31403 | |
pepsin | Sigma-Aldrich | P6887 | prepare a 0.2% solution in RNAse free water, and add 84 µl 37% HCL per 100 ml |
HCL, 37% solution | Fluka | 84422 | |
object glasses | Starfrost | 8037/1 | |
20x SSC | Invitrogen | AM9763 | dilute to 2x SSC in RNAse free water |
10 mM Phosphate buffer, pH 7 | add 57.7 ml of 1M Na2HPO4 and 42.3 ml 1M NaH2PO4; use DEPC treated water and autoclave | ||
denaturation buffer | 70% Formamide/2xSSC/10mM phosphate buffer pH 7 | ||
Tween-20 | Sigma-Aldrich | P2287 | |
2M Tris pH 7.5 solution | |||
Tris-Saline-Tween washing solution (TST) | 100 ml 10x Saline, 50 ml 2M Tris, 500 µl Tween-20, add RNAse free water till 1 liter | ||
10x Saline solution | dissolve 85 g NaCl in 1 liter RNAse free H2O, and autoclave | ||
Tris-Saline-BSA (TSBSA) | 500 μl 10x Saline, 250 μl 2 M Tris, 1 ml 100x BSA, 3,25 ml RNAse free H2O | ||
BSA, purified, 100x | New England Biolabs | B9001S | |
sheep-anti-dig antibody | Roche diagnostics | use 1:500 diluted in TSBSA | |
rabbit-anti-sheep antibody conjugated with FITC | Jackson labs | use 1:250 diluted in TSBSA | |
goat-anti-rabbit antibody conjugated with FITC | Jackson labs | use 1:250 diluted in TSBSA | |
mouse-anti-biotin antibody | Roche diagnostics | use 1:200 diluted in TSBSA | |
donkey-anti-mouse antibody conjugated with Rhodamine | Jackson labs | use 1:250 diluted in TSBSA | |
goat-anti-horse antibody conjugated with Rhodamine | Jackson labs | use 1:250 diluted in TSBSA | |
Tris-Saline washing solution (TS) | 10 ml 10x Saline, 5 ml 2 M Tris, 85 ml RNAse free H2O | ||
Vectashield mounting medium with DAPI | Vectorlabs | H-1200 | |
nail-varnish | |||
fluorescent miscroscope |