Here we describe a protocol for simultaneous detection of histone modifications by immunofluorescence and DNA sequences by DNA FISH followed by 3D microscopy and analyses (3D immuno-DNA FISH).
Fluorescent in situ hybridization using DNA probes on 3-dimensionally preserved nuclei followed by 3D confocal microscopy (3D DNA FISH) represents the most direct way to visualize the location of gene loci, chromosomal sub-regions or entire territories in individual cells. This type of analysis provides insight into the global architecture of the nucleus as well as the behavior of specific genomic loci and regions within the nuclear space. Immunofluorescence, on the other hand, permits the detection of nuclear proteins (modified histones, histone variants and modifiers, transcription machinery and factors, nuclear sub-compartments, etc). The major challenge in combining immunofluorescence and 3D DNA FISH is, on the one hand to preserve the epitope detected by the antibody as well as the 3D architecture of the nucleus, and on the other hand, to allow the penetration of the DNA probe to detect gene loci or chromosome territories 1-5. Here we provide a protocol that combines visualization of chromatin modifications with genomic loci in 3D preserved nuclei.
Epigenetic mechanisms trigger establishment and inheritance of developmental and cell-type specific transcriptional profiles. At one level this involves modulation of chromatin packaging that defines active or silent genomic regions. On a larger scale, global 3D organization of the genome and nuclear architecture also play a role in the control of transcriptional patterns. Thus, dissection of these epigenomic features is essential for a full understanding of how genes are regulated 6-11.
Combined immunofluorescence and 3D DNA FISH provide a unique opportunity to complement molecular and biochemical analyses by assessing specific interactions/associations of DNA sequences and/or proteins within the nucleus. Furthermore, while genome-wide high throughput techniques such as chromatin immunoprecipitation (ChIP-seq) or chromosome capture conformation coupled with deep sequencing (4C-seq, 5C, Hi-C) provide global data on cell populations12, immunofluorescence/DNA FISH techniques enable analyses at the single cell level.
Here we describe a protocol for simultaneous detection of histone modifications by immunofluorescence and DNA sequences by DNA FISH followed by 3D microscopy and analyses (3D immuno-FISH). The advantage of this protocol is the combined visualization of DNA and preservation of protein structures. Our experience in this field has enabled us to improve and simplify existing protocols. Although we have used this protocol to detect DNA double-stranded breaks in lymphocytes undergoing recombination, this method can be applied to other proteins and other cell-types.
1. DNA Probe Labeling with Fluorophores: Nick Translation (~ 6 hr)
2. Probe Precipitation/Denaturation/Pre-annealing (3-4 hr)
3. 3-dimensional DNA FISH – First Day (~ 3 hr)
4. 3-dimensional Immuno-FISH – First Day (~ 6-7 hr)
5. 3-dimensional DNA FISH / Immuno-FISH – Second Day (~ 2 hr)
6. 3D Microscopy and Analysis
Statistical analysis of the entire distributions of distances between two alleles. We first construct cumulative distribution frequency curves over the entire range of measured distances between the two alleles.To assess the significance of differences in the distributions of empirical inter-allelic distances we use the nonparametric two-sample Kolmogorov-Smirnov (KS) test 13,14 .
Statistical analysis of close association (i.e. pairing) of two alleles using an optimal distance cut-off. To identify the range of most robust differences in distances between two alleles or loci, we use a series of two-tail Fisher exact test 15 s. Namely, at each measured distance we test whether one condition is significantly over-represented at shorter distances between the two samples. The minimum in the distribution of P-Values indicates the range of distances that should be considered as a cut-off for close association (i.e. pairing) of the two alleles. Subsequently the two-tail Fisher exact test is used to assess the significance of pairing of two alleles at the identified cut-off value15. Multiple testing corrections are applied to account for the total number of Fisher tests performed 16.
Statistical analysis of association of the locus of interest with sub-nuclear compartments or proteins. The two-tail Fisher-exact test is used to analyze the significance of association of the locus of interest with different compartments or proteins 15.
DNA and immuno-FISH are used in the Skok lab to study the changes in nuclear organization associated with the process of V(D)J recombination of antigen receptor loci during B and T lymphocyte development. The techniques detailed above enable us to i) measure distances between the two ends of a locus (contraction) ii) measure distances between alleles or loci (pairing), iii) analyze the DNA damage occurring within loci, iv) assess the location of alleles and loci relative to nuclear sub-compartments (repressive pericentromeric heterochromatin in our study), and v) assess the association of nuclear proteins on alleles and loci (association of the phosphorylated histone γ-H2AX in our study). In particular, we analyzed the role of locus dynamics in the regulation of V(D)J recombination of the Immunoglobulin loci in developing B cells 17-30, and in the regulation of lineage commitment between CD4+ and CD8+ T cells 14. We also assessed DNA damage at one of the T-cell receptor locus, Tcra/d, in developing T cells carrying a mutant form of the recombinase protein RAG2 31.
Figure 1. Analysis of distances between two alleles or loci. (A) Confocal sections showing representative unpaired (right) or paired (left) alleles in individual nuclei. The DNA FISH probe was generated from a BAC (~200 kb) spanning the locus of interest. Distances between alleles were measured between the center of mass of each BAC signal in 3D image stacks. Dashed circles outline nucleus shapes. Scale bars = 1 μm. (B) Cumulative frequency distribution curves showing the entire range of distances between the two alleles measured in two different biological conditions (wild-type vs. mutant). The nonparametric two-sample KS test was used to assess the significance of the differences between the two distributions. The range of most robust differences in inter-allelic distances was assessed using a series of two-tail Fisher exact tests (blue dots). The most robust P-values (higher dots) were found for the distances around 1 μm. Thus the cut-off for close association (i.e. pairing) of the two alleles was set up at 1 μm. (C) Cumulative frequency curves showing the distances between the two alleles from 0 to 1.5 μm in the two different biological conditions. The significance of pairing of the two alleles was assessed using the two-tail Fisher exact test for a cut-off at 1 μm. Click here to view larger figure.
Figure 2. Analysis of the location of a locus relative to sub-nuclear compartments. In our studies we analyze the relationship between our loci of interest and the repressive sub-nuclear compartment formed by the pericentromeric heterochromatin (PCH) 32. Alleles are considered as located at PCH when the BAC signal for the locus of interest is adjacent or overlapping with the γ-satellite signal that hybridizes to PCH (no pixel in between the edges of the BAC and γ-satellite DNA FISH signals). Confocal sections representative of the location of Tcra alleles relative to PCH: (Top) no allele located at PCH, (Middle) one allele at PCH, (Bottom) two alleles at PCH. Locus in red; and γ-satellite in white. Yellow arrows show the alleles located at PCH. Dashed circles outline nucleus shapes. Scale bar = 1 μm.
Figure 3. Analysis of the association of nuclear proteins and histone modifications on a locus. (A) In our studies we analyze the relationship between our loci of interest and the phosphorylated form of the histone H2AX (γ-H2AX), as read-out for DNA double-stranded breaks 33. Confocal sections show examples of no γ-H2AX association (Top) or association on one Tcra allele (Bottom) in individual nuclei. Alleles were defined as associated with γ-H2AX if the BAC signals and immunofluorescence foci were at least partially overlapped (at least one pixel of colocalization). Locus is shown in red and γ-H2AX in white. Dashed circles outline nucleus shapes. Scale bar = 1 μm. It is of note that colocalization analyses of the BAC signal with PCH or γ -H2AX foci are performed manually using Image J software without automated threshold. (B) Loss of part of the γ-H2AX signal after denaturation is one of the main technical questions raised by this immuno-DNA FISH protocol. Here we show an example of immunofluorescence before (Top) and after (Bottom) HCl-based denaturation treatment (Formamide-treatment gave similar results (not shown)). Although the signal was weaker, the main features of the staining, in particular the number of foci, remain after the treatment. Using an SP5 Leica confocal, the settings for the 561 laser before and after denaturation were the following: 35% power / smart gain at 730V, and 45% power / smart gain at 750V, respectively. Furthermore, we show here an example of an immunofluorescence staining for another histone modification, H3K27me3 (Right panels) showing that this protocol can be applied to other histone marks, as has previously been published 1,3 . γ-H2AX is shown in yellow and H3K27me3 in red. Dashed circles outline nucleus shapes. Scale bar = 2 μm. Click here to view larger figure.
Name of Reagent/Material | Company | Catalogm Number | Comments |
H2O | Fisher | # BP2470 | |
RNase A | Sigma | # R4642 | |
dNTP | Sigma | # DNTP100 | |
Alexa dUTP | Invitrogen | # C11397 to C-11401 | |
Cy3 or Cy5 dUTP | Fisher | # 45-001-xxx | |
DNase I | Roche | # 04536282001 | |
DNA Pol I | Biolabs | # M0209 | |
0.025 μm filters | Millipore | # VSWP02500 | |
Cot-1 DNA 1 mg/ml | Invitrogen | # 18440 | |
Hybloc DNA 1 mg/ml | Applied Genetics | # MHB | |
Salmon sperm | Sigma | # D1626 | powder to be resuspended at 10 mg/ml in H2O |
NaAc (Sodium Acetate, pH 5.2, buffer solution) | Sigma | # S7899 | |
Ficoll 400 (Mol Biol grade) | Fisher | # 525 | |
Polyvinylpyrrolidone (Mol Biol grade) | Fisher | # BP431 | |
Dextran sulfate powder | Sigma | # D8906 | |
SSPE (Saline-Sodium Phosphate-EDTA) 20x solution | Fisher | # BP1328 | |
Formamide | Fisher | # BP227 | |
Coverslips | Fisher | # 12-548-B | |
Slides | Fisher | # 12-550 | |
6-well plates | Fisher | # 0720080 | |
PBS, 10x | Fisher | # MT-46-013-CM | |
Poly-L-lysine solution | Sigma | # P8920 | |
Paraformaldehyde, prills, 95% | Sigma | # 441244 | |
Triton-X-100, Mol Biol grade | Sigma | # T8787 | |
BSA (Bovine Serum Albumin) Fraction V | Fisher | # BP 1600 | |
Normal goat serum | Vector Labs | # S-1000 | |
Tween-20, Mol Biol grade | Sigma | # P9416 | |
SSC (Saline Sodium Citrate) 20x solution | Fisher | # BP1325 | |
ProLong Gold antifade reagent | Invitrogen | # P36930 | |
DAPI (4′,6-diamidino-2-phenylindole) | Sigma | # D9542 | |
Best test one coat rubber cement | Art or office supply stores |
Table 1. Specific reagents and small equipment.
NT with dUTP coupled with Alexa-fluorophores | Amount | Final Concentration |
10 μg DNA | X μl | |
RNase A – 0.17 μg/μl | 10 μl | |
H2O | Fill to 260 μl | |
TOTAL – 30 min at 37 °C | 260 μl | |
10x NT buffer | 50 μl | 1x |
100 mM β-mercaptoethanol | 50 μl | 10 μM |
dNTP mix | 40 μl | 40 μM each dNTP |
0.1 mM dTTP | 50 μl | 10 μM |
1 mM Alexa dUTP | 12 μl | 30 μM |
DNase I (working solution, see Table below) | 30 μl | |
DNA Pol I | 7.5 μl | 15 U/ml |
TOTAL – 2 hr at 16 °C | 500 μl | |
NT with dUTP coupled with Cy3 or Cy5 | Amount | Final Concentration |
10 μg DNA | X μl | |
RNase A – 0.17 μg/μl | 10 μl | |
H2O | Fill to 300 μl | |
TOTAL – 30 min at 37 °C | 300 μl | |
10x NT buffer | 50 μl | 1x |
100 mM β-mercaptoethanol | 50 μl | 10 μM |
dNTP mix | 50 μl | 50 μM each dNTP |
1 mM Cy3- or Cy5-dUTP | 12 μl | 30 μM |
DNase I | ~30 μl | ~1.2 U/μl |
DNA Pol I | 7.5 μl | 15 U/ml |
TOTAL – 2 hr at 16 °C | 500 μl |
Table 2. RNase A treatment and Nick Translation for 10 μg DNA.
Solutions | Stocks | |
RNase A | Working solution: 0.17 μg/μl | Made up fresh |
10x NT buffer: | 0.5M Tris-HCl, pH 8; 50 mM MgCl2; 0.5 mg/ml BSA | 1.5 ml tubes at -20 °C. |
dNTP mix: | dATP, dCTP, and dGTP: 0.5mM each in H2O | 1.5 ml tubes at -20 °C |
DNase I: | Stock solution: reconstituted at 20,000 U/ml in 20 mM Tris pH 7.5/50% glycerol/50 mM NaCl | 0.5 ml tubes at -20 °C |
Working solution: Each new batch of DNase I stock solution should be tested at different dilutions (10 U/ml, 20 U/ml and 40 U/ml in H2O). The dilution giving a DNA smear between 100 and 1,000 bp (see step 5) should be used as the working solution. | Made up fresh |
Table 3. Solutions for RNase A treatment and Nick Translation.
Hybridization buffer | Solutions | Stocks | |
10% | 50x Denhardt’s solution: | 1 g Ficoll 400, 1 g Polyvinylpyrrolidone and 100 ml H20 – Filtered | 50 ml falcon at -20 °C |
40% | 25% Dextran Sulfate: | Dextran sulfate powder is dissolved in 12.5x SSPE at 37-65 °C | 50 ml falcon at -20 °C |
50% | Formamide | Bottles at +4 °C | |
Aliquots in 1.5 ml Eppendorfs are stored at -20 °C |
Table 4. Hybridization buffer.
Antibodies | Supplier | Catalog number | Dilution |
γ-H2AX – phospho H2A.X, clone JBW301, mouse monoclonal | Millipore | # 05636 | 1/200 |
H3K27me3, clone , rabbit polyclonal | Millipore | # 07449 | 1/200 |
Alexa Fluor 555 Donkey Anti-mouse | Invitrogen | # A-31570 | 1/500 |
Alexa Fluor 488 Goat Anti-mouse | Invitrogen | # A-11001 | 1/500 |
Alexa Fluor 488 Goat Anti-rabbit | Invitrogen | # A-11008 | 1/500 |
Table 5. Primary and secondary antibodies.
The techniques detailed above were used in our lab to analyze the regulation of V(D)J recombination of the Immunoglobulin and Tcra/d loci in developing lymphocytes 30,31 . We are confident that this technique can be adapted for detection of various nuclear proteins, nuclear compartments and loci, in different cell-types. Modifications of the protocol may be necessary, and in this case the major steps to focus on are the following. First, the length of permeabilization can be adjusted depending on the cell-type. Second the length of the primary antibody incubation can also be adjusted according to the quality of the antibody and the abundance of the protein of interest (for example in some cases, incubation may be performed overnight at 4 °C). Importantly, the denaturation step can be adapted, and HCl treatment may be replaced by denaturation in Formamide 3,4,34 and NaOH for DNA FISH 20. Also repeated freeze-thaw cycles of liquid nitrogen may be applied to the cells after fixation / permeabilization for DNA FISH alone 4,34 . Finally, while several histone modifications, RNA Polymerase II and co-factors have been successfully detected using this method 30,35 , immuno-staining for other nuclear proteins may be performed after the DNA FISH hybridization 4.
3D image analysis can involve the use of different types of microscopes, including wide-field epifluorescence (coupled with deconvolution) and confocal laser microscopy 36. The choice of the imaging method will depend on the resolution required, the number of fluorophores, the depth of the samples and the quality of the DNA and protein signals. The development of new high-resolution microscopes opens an entirely new avenue in the accurate visualization of nuclear and cellular processes 37-39. Furthermore, new techniques for the generation of custom-made probes that contain no repetitive sequence enable the clean detection of genomic regions, from a 15-kb specific locus, to complex pools of exons 40. This increases the accuracy and sensitivity of high-resolution microscopy in tracking the dynamics of nuclear processes.
The authors have nothing to disclose.
We would like to thank the members of the Skok lab, especially Susannah Hewitt, for discussions and comments. This work is supported by the National Institute of Health grants R01GM086852, RC1CA145746 (J.A.S.). J.A.S. is a Leukemia & Lymphoma Society scholar. J.C. is an Irvington Institute Fellow of the Cancer Research Institute. M.M. is supported by a National Science Foundation Grant Integrative Graduate Education and Research Traineeship (NSF IGERT 0333389).
Name of Reagent/Material | Company | Catalogue Number | Comments |
H2O | Fisher | # BP2470 | |
RNase A | Sigma | # R4642 | |
dNTP | Sigma | # DNTP100 | |
Alexa dUTP | Invitrogen | # C11397 to C-11401 | |
Cy3 or Cy5 dUTP | Fisher | # 45-001-xxx | |
DNase I | Roche | # 04536282001 | |
DNA Pol I | Biolabs | # M0209 | |
0.025 μm filters | Millipore | # VSWP02500 | |
Cot-1 DNA 1 mg/ml | Invitrogen | # 18440 | |
Hybloc DNA 1 mg/ml | Applied Genetics | # MHB | |
Salmon sperm | Sigma | # D1626 | powder to be resuspended at 10 mg/ml in H2O |
NaAc (Sodium Acetate, pH 5.2, buffer solution) | Sigma | # S7899 | |
Ficoll 400 (Mol Biol grade) | Fisher | # 525 | |
Polyvinylpyrrolidone (Mol Biol grade) | Fisher | # BP431 | |
Dextran sulfate powder | Sigma | # D8906 | |
SSPE (Saline-Sodium Phosphate-EDTA) 20x solution | Fisher | # BP1328 | |
Formamide | Fisher | # BP227 | |
Coverslips | Fisher | # 12-548-B | |
Slides | Fisher | # 12-550 | |
6-well plates | Fisher | # 0720080 | |
PBS, 10x | Fisher | # MT-46-013-CM | |
Poly-L-lysine solution | Sigma | # P8920 | |
Paraformaldehyde, prills, 95% | Sigma | # 441244 | |
Triton-X-100, Mol Biol grade | Sigma | # T8787 | |
BSA (Bovine Serum Albumin) Fraction V | Fisher | # BP 1600 | |
Normal goat serum | Vector Labs | # S-1000 | |
Tween-20, Mol Biol grade | Sigma | # P9416 | |
SSC (Saline Sodium Citrate) 20x solution | Fisher | # BP1325 | |
ProLong Gold antifade reagent | Invitrogen | # P36930 | |
DAPI (4′,6-diamidino-2-phenylindole) | Sigma | # D9542 | |
Best test one coat rubber cement | Art or office supply stores | ||
Table 1. Specific reagents and small equipment. |