Figure 1.
The C-terminal tail of canonical H2A.
(A) Structure of histone H2A. Top: Histone H2A extracted from the crystal structure of the nucleosome core particle (1kx5) [58]. Residues of H2A are depicted according to the following color code: H2A1-114 blue, H2A115-122 red, H2A123-129 yellow. The C-terminal tail can be present in a trans or cis conformation with respect to the βC part of the H2A histone. Transitions between the two conformations occur on the 100 ns time scale as inferred from MD simulations (Ettig & Rippe, data not shown). Bottom: Schematic respresentation of the H2A constructs used. (B) GFP-H2A fusion proteins in HEK293 cells stably expressing GFP-H2A1-129, GFP-H2A1-122 or GFP-H2A1-114 localize to the nucleus. Immunoflourescence images of representative cells are shown. The DNA was counterstained with DAPI. The scale bar represents 5 µm. (C) GFP-H2A fusions are present on metaphase chromosomes of HEK293. Metaphase chromosomes were prepared from cells stably expressing GFP-H2A1-129, GFP-H2A1-122 or GFP-H2A1-114. Representative examples are shown. The DNA was counterstained with DAPI. (D) Expression level of GFP-H2A fusion proteins is 5–10% of endogenous H2A. Quantitative Western blot analysis of the GFP-H2A signal compared to the endogenous H2A signal using an antibody against H2A. The percentage of GFP-H2A is shown, error bars represent s.d.
Figure 2.
Cells expressing C-terminally truncated GFP-H2A are stress sensitive.
(A) Synchronized cells expressing GFP-H2A1-122 and GFP-H2A1-114 show a reduced growth rate. Cells were seeded at a density of 5*103 cells ml−1 and subjected to a double thymidine block. Cell numbers were determined in triplicates approx. every 24 h for 3 d after release from the block. Growth curves of one representative experiment are shown with standard deviations. (B) Cells expressing GFP-H2A1-114 are more sensitive to CPT treatment. After treatment with 1 µM CPT for 30 min, a colony forming assay was performed and colonies were fixed and stained with crystal violet. The colony number was determined and the colony number for untreated cells was set to 1. The experiment was performed in triplicates and average relative colony numbers with standard deviation of one representative experiment are given. (C) Cells expressing GFP-H2A1-122 and GFP-H2A1-114 are more sensitive to HU treatment. Cells were treated with 1 mM HU for approx. 4 h. Colony numbers were determined as in 2B.
Figure 3.
Loss of the H2A C-terminal tail negatively affects nucleosome stability.
(A) Nucleosomes containing C-terminally truncated H2A elute at lower salt concentrations. Stepwise salt elution of GFP-H2A fusion proteins from HEK293 cells stably expressing GFP-H2A1-129, GFP-H2A1-122 or GFP-H2A1-114. Extracts from 2*105 cells per lane were analyzed by immunoblot with antibodies against endogenous H2A or GFP as indicated. (B) Deletion of the H2A C-terminus results in increased FRAP recovery kinetics. FRAP analysis with stable cell lines expressing GFP-H2A1-129, GFP-H2A1-122 and GFP-H2A1-114. Left panel: A 120×120 pixel spot was bleached and the fluorescence recovery was measured in 5.01 s intervals over 80 min. 10–15 cells were used for quantification. The recovery curve for GFP-H2A1-129 is in blue, for GFP-H2A1-122 in green and for GFP-H2A1-114 in red. Bottom panel: Calculated t1/2 and t70 recovery times. (C) C-terminally truncated H2A is enriched in accessible chromatin fractions. For chromatin fractionation, approx. 1*108 nuclei from stable cell lines expressing GFP-H2A1-129, GFP-H2A1-122 and GFP-H2A1-114 were digested with MNase for 1 min and fractionated into S1, S2 and P (pellet). The fractions were precipitated and 1/50 (S1 and S2) or 1/100 (P) was analyzed by immunoblotting with a GFP specific antibody (upper panel). As a control an immunoblot against the chromatin associated protein DEK was used (lower panel).
Figure 4.
The H2A C-terminal tail is important for nucleosome stability and chromatin remodeling in vitro.
(A) Differential positioning and thermal shift of mononucleosomes containing either full-length H2A, H2A1-122 or H2A1-114. Nucleosomes were reconstituted on a linear MMTV NucA DNA fragment and incubated at 45°C for increasing periods of time. Nucleosome positions were analyzed on a native 5% polyacrylamide gel and visualized by ethidium bromide staining (left panel). Quantification of the relative repositioning is shown in the right panel. (B) C-terminal deletions affect chromatin remodeling. In vitro chromatin remodeling assay with mononucleosomes containing H2A1-129, H2A1-122 or H2A1-114 that were assembled on linear DNA fragment (Drosophila Hsp70 promoter) and incubated with the indicated chromatin remodeling complex with or without ATP. Upper panel: ISWI. Middle panel: SNF2H. Lower panel: ACF. Lanes 1, 5, 9: reconstituted mononucleosomes alone; lanes 2, 6, 10: + remodeling factor, no ATP; lanes 3, 7, 11 and 4, 8, 12: + remodeling factor, + ATP. Nucleosome positions were analyzed by native gel electrophoresis and staining with ethidium bromide. Black arrowheads indicate positions to which nucleosomes were moved, white arrowheads indicate positions from which nucleosomes were removed. (C) Kinetics of nucleosome remodeling assayed with a 601 remodeling template. Nucleosomes containing the wildtype H2A and the indicated C-terminal truncations (300 ng), as indicated on the left, were incubated without (lane 1), or with 100 ng of the indicated remodeling enzyme and ATP (lanes 2 to 6). The remodeling reaction was incubated from 1 up to 40 min and analyzed as described above. Nucleosome positions are indicated. Black triangles indicate new nucleosome positions observed with H2A1-122 and H2A1-114.
Figure 5.
The H2A C-terminal tail binds linker histone H1.
(A) Schematic representation of the construct used to identify H2A C-terminus interacting proteins. Proteins interacting with the HA Flag 8x C-terminus were identified by Mass Spectrometry after Flag affinity purification. Histone H1 peptides that bind to the H2A C-terminal tail as identified by mass spectrometry. DeltaCN and XCorr values are shown. (B) H1 co-immunoprecipitates with the 8x repeat of the H2A C-terminal tail. The HA-Flag 8x H2A C-terminus was purified and the eluate from the Flag Sepharose probed by immunoblotting with an H1 specific antibody. 10% of the input was loaded (lanes 1 and 2). As control, the empty vector without the 8x H2A C-terminus was used. Purified H1 was used as an immunoblot control (lane 5). (C) Deletion of the H2A C-terminus reduces H1 binding. Binding of H1 to in vitro reconstituted mononucleosomes containing either full-length or C-terminally truncated H2A. Mononucleosomes containing wild type H2A (lanes 1–6), H2A 1-122 (lanes 7–12) or H2A 1-114 (lanes 13–14) were reconstituted on a DNA fragment containing the 601 positioning sequence. The nucleosomes were incubated with increasing amounts of H1 (H1 to nucleosome ratios 1∶1 to 6∶1) and then analyzed on native polyacrylamide gels. Nucleoprotein complexes were visualized by ethidium bromide staining. (D) Partial MNase digestion of H1-nucleosome complexes shows a chromatosome stop. Nucleosomes and H1-nucleosome complexes reconstituted on a 208 bp 601 DNA fragment at a molar ratio of 6∶1 were incubated with increasing MNase concentrations and the resulting DNA cleavage products were analyzed by polyacrylamide gel electrophoresis and ethidium bromide staining next to a DNA standard (M). The position of the undigested DNA fragment (208 bp), the protected nucleosomal DNA (147 bp) and the chromatosome stop (arrow) are indicated.