|Home | About | Journals | Submit | Contact Us | Français|
Senescence-associated heterochromatin foci (SAHF) are specialized domains of facultative heterochromatin that contribute to silencing of proliferation-promoting genes (such as E2F target genes) in senescent cells. Thus, methods for detection of SAHF are invaluable tools in the field of senescence to identify these distinct changes in heterochromatin that are associated with the senescence-associated cell cycle exit. Here, we describe fluorescent-based methods for the detection of SAHF in senescent cells, including examining reorganization of DNA structure by 4′,6-diamidino-2-phenylindole (DAPI) staining and using specific antibodies to components of SAHF such as the histone H2A variant macroH2A, heterochromatin protein 1 (HP1) proteins and lysine 9 di-or tri-methylated histone H3 (H3K9Me2/3). Together, these markers identify SAHF, the distinct facultative heterochromatin structures that are associated with senescent cells.
Cellular senescence is a state of irreversible cell growth arrest. Cellular senescence can be triggered by a variety of stimuli such as critically shortened telomeres, which occurs after extensive cell division, or activation of certain oncogenes (such as H-RASG12V or BRAFV600E) (1–3). By definition, senescent cells are irreversibly growth arrested, and one of the necessary steps towards this irreversible cell cycle exit is the suppression of E2F target genes (4), which are mainly involved in promoting cell proliferation and S phase cell cycle progression (5). Promoters of E2F target genes typically acquire heterochromatic features during senescence (4). The heterochromatin associated with this process is specialized domains of facultative heterochromatin that often form in senescent human cells, senescence-associated heterochromatin foci (SAHF) (4, 6–12).
SAHF were first described in 2003 by Narita et al. who observed that the nuclei of senescent cells contain 30–50 bright, punctate DNA-stained dense foci that can be readily distinguished from chromatin in normal cells (4). Importantly, SAHF are not associated with cells undergoing quiescence, indicating that SAHF formation is not associated with reversible cell cycle exit (4). In addition, SAHF have also been shown to be distinct from constitutive heterochromatin because centromeres, telomeres, and other constitutive heterochromatin regions are not included in SAHF (4, 7, 13). Further, SAHF are also different from other facultative heterochromatin such as inactivated X chromosomes (Xi) in female human cells. For example, histone modifications such as lysine 27 trimethylated histone H3 (H3K27Me3) are associated with Xi but not SAHF (4).
SAHF play a role in sequestering proliferation-promoting genes (4), including E2F target genes such as cyclin A (7), which is required for the progression through S-phase of the cell cycle (14). Indeed, SAHF do not contain any active transcription sites (4), demonstrating their role in contributing to the senescence-associated cell cycle exit. Significantly, disruption of SAHF formation can cause cell transformation (15), which infers that SAHF contribute to the tumor suppressive function of senescence. Recently, there is evidence to suggest that SAHF may limit the extent of DNA damaging signaling which may prevent senescent cells from undergoing apoptosis induced by high DNA damage signaling, thereby maintaining the viability of senescent cells (12). Finally, there is emerging evidence to suggest that SAHF may play a role in the senescence phenotype in vivo (16–19).
A number of different inducers of senescence cause the formation of SAHF, including activated oncogenes such as H-RASG12V and BRAFV600E (4, 20, 21), extensive passaging (4), chemotherapeutics such as etoposide (4) and hydroxyurea (10), and bacterial toxins (10). However, SAHF formation and senescence are not always coupled. Indeed, a number of studies have shown that senescence can occur in the absence of SAHF formation. For instance, activation of AKT and knockdown of PTEN do not cause SAHF formation (22, 23). It is also important to note that SAHF formation is cell-line dependent (10). For example, senescence induced by extensive passaging in the primary human embryonic fibroblasts cell lines IMR90 and WI38 cells is associated with SAHF, while senescence triggered by extensive passaging in BJ cells (primary human foreskin fibroblasts) is not associated with SAHF formation (4). The difference between these cell lines correlates with a variation in activation of the p16/pRb pathway after extensive passaging (10). Indeed, senescence induced by activated oncogenes (such as H-RASG12V and BRAFV600E) in BJ cells triggers SAHF formation, which is associated with activation of the p16/pRb pathway (24, 25) Notably, mouse cells do not form robust SAHF, although they do display a marked increase in staining of certain components of SAHF such as macroH2A (26).
To date, a number of molecular markers of SAHF have been described [reviewed in (6, 11, 27)] including: macroH2A (9), a histone variant known to contribute to X chromosome inactivation and gene silencing (28); high mobility group A (HMGA) proteins, which coordinate with p16INK4a to induce SAHF formation and are required for maintaining SAHF (15); and di- or tri-methylated lysine 9 histone H3 (H3K9Me2/3) and bound HP1 proteins (4, 7), two common markers of heterochromatin (29). Together with DAPI, co-staining for these markers is a simple and reliable method to determine the presence of SAHF in senescent cells. Here, using oncogenic-RAS (H-RASG12V) as an inducer of senescence and SAHF, we describe a method for the immunofluorescent detection of SAHF using DAPI and specific antibodies to components of SAHF such as macroH2A, H3K9Me2/3, and HP1 proteins.
The infectious retrovirus is generated by transfecting the plasmid DNA into a packaging cell line, for example, Phoenix cells (http://www.standford.edu/group/nolan/protocols/pro_helper_dep.html). Transfection-quality DNA of the retrovirus plasmid is made using a Qiafilter Plasmid Maxi kit, according to the manufacturer’s instructions. After transfection, the Phoenix cells reverse transcribe the plasmid DNA into an RNA that is packaged into infectious virus and expelled from the cells. Then, the tissue culture supernatant containing the virus is applied to the cells of interest to deliver the activated oncogenes (e.g., H-RASG12V) to the target cells. The infected cells can be selected in puromycin to enrich for cells that have been infected. To assess the efficiency of puromycin to kill uninfected cells, perform a mock virus infection. To assess the efficiency of infection, infect one plate with a virus known to have a high titer (e.g., pBABE-puro vector, Addgene).
To visualize SAHF, DAPI is used to stain for DNA, while specific antibodies to components of SAHF (e.g., macroH2A, H3K9Me2/3, and HP1 proteins) should be used for indirect immunofluorescence staining. To assess the specificity of the antibody staining pattern, an isotype matched control primary antibody and/or omission of the primary antibody should be included as negative controls for immunofluorescence staining. In addition, IMR90 cells infected with vector control-encoding retrovirus should be included as a negative control.
After immunofluorescence staining, the processed slides can be examined under a fluorescent microscope. DAPI staining should reveal punctate staining in oncogenic H-RASG12V infected cells, while vector control cells should display diffuse staining across the cell nuclei (Fig. 1). The number of cells with SAHF staining depends upon the duration of RAS expression. For example, ~50% of RAS-infected cells display SAHF by DAPI staining 7–9 d post-drug selection (4, 7). Components of SAHF such as macroH2A, HP1 proteins, and H3K9Me2/3 form foci that co-localize with DAPI-stained SAHF in RAS-infected cells. In vector control cells, HP1 proteins and H3K9Me2/3 staining is largely diffuse. MacroH2A staining marks the single foci in female mammalian cells (such as IMR90 and WI38 primary human fibroblast cells) that is the inactivated X chromosome (Fig. 1). As negative controls, staining using isotype matched IgG controls as the primary antibody or using the secondary antibody only (without the primary antibody) should not show foci that co-localize with DAPI-stained SAHF. The controls herein will obviously be specific to each component of SAHF. Importantly, regardless of the specific components, not all cells with DAPI-stained SAHF-positive cells will display foci for components of SAHF. This is because relocalization of components of SAHF into DAPI stained foci is typically a later event compared with formation of DAPI foci (9).
This work was funded by a NIH/NCI grant (R01CA160331), a Liz Tilberis Scholar Award from the Ovarian Cancer Research Fund and the Department of Defense Ovarian Cancer Academy Award (OC093420). Support of Core Facilities used in this study was provided by Cancer Center Support Grant (CCSG) CA010815 to The Wistar Institute.
1As discussed in the introduction, other oncogenes (BRAF/MEK/ERK) and other cellular stressors can cause the formation of SAHF. We recommend using H-RASG12V as a positive control because activation of RAS induces distinct SAHF formation in a variety of cell types.
2Correct pH of the 2X BBS solution is necessary for efficient plasmid transfection using this method.
3Never let cells reach confluence as this may decrease transfection efficiency. Passage Phoenix cells in 300 μg/mL hygromycin (Roche) and 1 μg/mL diptheria toxin (Sigma) for 1 week every month to maintain transfection efficiency. For more information, please visit Gary Nolan’s website: (http://www.stanford.edu/group/nolan/retroviral_systems/retsys.html).
4For oncogene-induced SAHF formation (or to study other types of stressors), low population doubling (PD) IMR90 cells should be used (PD<35). However, if you wish to study replicative senescence-induced SAHF, high PD IMR90 cells (e.g., PD65) should be used instead.
5Other cell lines and cell types form SAHF (10); however, this must be determined on a cell type by cell type basis. It is important to note that mouse cells do not form distinct SAHF structures that are visible by DAPI staining alone. Instead, macroH2A may server as an marker for SAHF in senescent mouse cells (26).
6To make 4% paraformaldehyde, weigh 2 g of paraformaldehyde (Sigma) and add 25 mL of ddH2O, 25 mL 2X PBS, pH 7.3, and 100 μL 1 N NaOH. Vortex and incubate this mixture at 65°C for 10–15 min to dissolve the paraformaldehyde (vortex the tube every 5 min). Cool the solution to room temperature before use.
7Adding 0.02% (w/v) sodium azide to the 3% (w/v) BSA in PBS will decrease potential contamination and increase the shelf life.
8Any of these markers can be used in addition to DAPI staining to identify SAHF. See Table 1 for antibody catalog numbers, dilutions, and references.
9Secondary antibodies to the appropriate species can be purchased from Jackson Immunolabs. See Table 2 for details.
10Mounting media should aliquoted and stored at −80°C. Once the media changes to an orange color, replace with a fresh batch.
11Although the Phoenix cells are easy to remove from the plate, use 0.25% (w/v) trypsin + 1 mM EDTA to to fully separate the cells and avoid cell aggregation.
12Be very gentle when adding fresh media as Phoenix cells do not adhere very well to the plate. It is typical that Phoenix cells change shape after changing them into fresh medium. They will fully recover after 6–8 h of culture in the newly changed medium.
13SAHF should start to appear approximately 3–4 d after starting selection and typically reach the highest percentage with ~50% RAS-infected cells positive for DAPI-stained SAHF around 7–9 d after staring drug selection.
14Protein lysates can be made by directly adding 1X Laemmli sample buffer followed by boiling for 5 min. Samples can be stored at −80°C. Ectopically expressed oncogenic-RAS should be confirmed by Western blotting. Cell extracts should be fractionated by SDS-PAGE, transferred to a PVDF membrane, and then Western blotted with antibody to RAS (BD Biosciences, 610001). An antibody to a house-keeping gene (e.g., β-actin) should be used as a loading control.
15Senescent cells do not adhere very well to coverslips; therefore, rinsing coverslips with PBS before fixation should be done very carefully. In instances where rinsing coverslips with PBS may lead to loss of cells, this step can be skipped, and coverslips can be directly immersed in 4% (w/v) paraformaldehyde for 10 minutes.
16At this step, soluble proteins can be removed to only retain tightly-bound chromatin proteins. To do so, cells can be pre-extracted with 5 mM MgCl2 in 0.2% (v/v) Triton-X in PBS for 5 min to preserve SAHF while removing the soluble proteins. Extra care should be take when performing pre-extraction and the following steps of washing and fixation as these pre-extracted cells can be easily washed off the coverslips.
17After fixation, coverslips can be washed 3X with PBS and stored at 4°C for up to 2 weeks before processing.
18Higher DAPI concentratins and longer incubation periods will make it harder to distinguish DAPI-stained SAHF.
19Avoid air bubbles when mounting coverslips.
20Slides can be stored at −20°C for up to a month. However, it is important to note that the fluroescent signal may fade over time.
21The same methods can be applied to other cell lines/cell types to determine SAHF formation. As not all cell lines or all causes of cellular senescence induce SAHF formation (see Introduction for details), we recommend the use of IMR90 cells expressing oncogenic H-RASG12V as a positive control.