QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online March 29, 2004
doi: 10.1242/10.1242/jcs.01008

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Smillie, D. A. Articles by Sommerville, J. Search for Related Content PubMed PubMed Citation Articles by Smillie, D. A. Articles by Sommerville, J.

Nuclear import and activity of histone deacetylase in Xenopus oocytes is regulated by phosphorylation

Division of Cell and Molecular Biology, School of Biology, University of St Andrews, St Andrews, Fife KY16 9TS, UK

View larger version (34K): [in a new window]   Fig. 1. Sequence motifs present in the charged C-terminal domain of HDACm. (A) DNA encoding the C-terminal 167 amino acid residues of HDACm was expressed as a fusion protein with glutathionine S-transferase (GST-V). The fusion site is indicated by an arrow and trypsin sensitive sites are indicated by arrowheads. Potential CK2 phosphorylation sites are boxed by dashed lines, with the phosphorylated residue indicated by an asterisk. Groups of basic (K+R) residues, which might contribute to a nuclear import signal (NLS), are boxed by solid lines. (B) The proposed bipartite NLS of HDACm is aligned with the reported NLS of the histone H3/H4 chaperone N1/N2. Numbers indicate the residue positions in the protein sequences.

View larger version (63K): [in a new window]   Fig. 2. Phosphorylation of fusion proteins that contain different regions of HDACm by CK2 activity extracted from oocytes. (A) Diagram showing the relative positions of potential CK2 phosphorylation sites (P) within the complete HDACm protein (+) and various deletions and mutations. The enzymatic core site is hatched and the C-terminal hydrophilic region is black. All six sites are contained in the GST-R deletion, the five sites present in the hydrophilic region are deleted in GST-R/H and the single N-terminal site is deleted from GST-V. Deletion of 10 kDa from the C-terminus removes all five of the marked sites in the hydrophylic region, whereas deletion of 5 kDa removes only the most C-terminal site. In the GST-V mutant 3m, the three sites most towards the center are mutated (Ser to Ala) and in the mutant 4m the most C-terminal site is also mutated (Thr to Ala). (B) Phosphorylation of GST-fusion proteins expressed in a K12 strain of E. coli. The stained gel (left) shows that partial hydrolysis of GST-V generates fragments that have lost a molecular mass of 5 kDa, 10 kDa and 15 kDa from the C-terminus. Abundant proteins from an added nuclear extract (GV), N1/N2 and nucleoplasmin (Npl) are also seen as stained bands. Phospholabeling with 32PATP by a GV extract (middle) shows that labeling is restricted to the hydrophobic region but is excluded from the C-terminal 10 kDa. Phospholabeling of GST-V by CK2 isolated from GVs (right) removes labeling due to N1/N2 and nucleoplasmin and is almost completely sensitive to 10 µg/ml heparin. The known CK2 substrate Xenopus FRGY2 (Y2) is shown as a positive control. (C) Phosphorylation of GST-V and 3m and 4m mutants expressed in a BL21 strain of E. coli by oocyte nuclear (GV) and cytoplasmic extracts (D). Whereas GV extract labels neither 3m nor 4m (C), a small, but significant, labeling of 3m is detected with cytoplasmic extract in the presence of 10 nM okadaic acid (OA, D). The stained gels (top of each panel) indicates equal loading of GST-fusion proteins (GST-FP).

View larger version (36K): [in a new window]   Fig. 3. Mapping of phosphorylation sites in HDACm. GST-V was phospholabeled in vitro using oocyte nuclear CK2. (A) Only two phospholabeled tryptic peptides (1 and 2) were recovered by HPLC. (B) Sequencing of peptide 1 reveals two phospholabeled residues at positions S421 and S423. (C) Sequencing of peptide 2 reveals a single phospholabeled residue at positions S393. The potential site S388 (asterisk) is not labeled.

View larger version (63K): [in a new window]   Fig. 4. Both the C-terminal region of HDACm and phosphorylation are required for nuclear uptake. (A) Phospholabeled GST-V (10 ng) was injected into the cytoplasm of stage V oocytes and incorporation of label into the GV was monitored from isolated nuclei and cytoplasms. Whereas most of the full-length fusion protein was imported within 8 hours, the GST-V-5 degradation product remained in the cytoplasm, again indicating that the C-terminal 5 kDa region is required for nuclear import. (B) Quantitation of the amounts and distribution of GST-V and GST-V-5 in GVs and cytoplasms up to 48 hours post-injection not only confirms the requirement for the C-terminal 5 kDa region but also indicates that it is required for stability of the protein. (C) The requirement of phosphorylation for nuclear import of GST-V is shown by comparing the amounts of fusion protein present in GVs at 24 hours post-injection with the amount of endogenous nuclear HDACm. Import is severely restricted in the presence of the inhibitors of CK2 activity, quercetin (Q, 60 nM) and 5,6-dichloro-1ß-D-ribofuranosylbenzimidazole (D, 50 µM), whereas the inactive analogue of quercetin, 3ß-D-rutinoside (R, 60 nM), has little inhibitory effect. Immunoblot using antibody directed against the C-terminal peptide of HDACm. (D) Quantitation of three repeat experiments confirm the requirement of phosphoryation for efficient nuclear import of GST-V. Density scan of each GST-V band is expressed as a fraction of the density scan of the corresponding endogenous HDACm band.

View larger version (54K): [in a new window]   Fig. 5. Identification of phosphorylation sites required for efficient nuclear import of GST-V in vivo. (A) 10 ng of GST-V or mutants: 4m (S393A, S421A, S423A, T445A); 3m (S393A, S421A, S423A); and T445A alone, were injected into the cytoplasm of oocytes, with and without co-injection of 12 ng of importin-, and GVs and cytoplasms were recovered after 2, 4 and 6 hours. Extracted protein was immunoblotted using antibody directed against the C-terminal peptide of HDACm. (B) Quantitation of results from upper two panels of (A) normalized to ratios of nuclear:cytoplasmic concentration. GST-V plus importin- (filled circles), GST-V minus importin- (open circles); 4m plus importin- (filled squares), 4m minus importin- (open squares). (C) Quantitation of results from lower two panels of (A). 3m plus importin- (filled circles), 3m minus importin- (open circles); T445A plus importin- (filled squares), T445A minus importin- (open squares). (D) Nuclear import of anti-GST IgG on co-injection with GST, GST-V, 3m and T445A. GVs and cytoplasms were isolated after 4 hours and a slot blot of protein was developed using anti-mouse IgG. (E) Nuclear import of GST-V on co-injection of 5 ng of Ran or of the Ran mutant Q69L. (F) Quantitation of results from (E) showing GST-V plus Ran (filled circles), GST-V plus RanQ69L (open circles).

View larger version (54K): [in a new window]   Fig. 6. Interaction of GST-V with importin- in vitro is influenced by the presence of C-terminal phosphorylation sites. (A) Binding of endogenous importin- from an oocyte SN100 fraction to GST-V, pre-phosphorylated GST-V (Vp), GST-V with deletion of the C-terminal 5 kDa (V-5) and GST, all immobilized on glutathione-Sepharose. Immunoblot using anti-importin . (B) Elution of importin- from immobilized GST-V in the presence and absence of Xenopus GV extract, 1 mM ATP and DRB. Importin- was first bound as in A, track 2. (C) Dependence of release of importin- from immobilized GST-V by addition of ATP (filled circles) rather than GTP (open circles). As in B, track 3, but with different concentrations of ATP or GTP. Immunoblots were scanned and relative intensities of signal from the importin- bands were calculated and plotted as a percentage of input protein released. (D) Binding of pre-phosphorylated and non-phosphorylated GST-V, 3m, T445A, 4m and of GST to importin- immobilized on nitrocellulose filters. Immunoblot using anti-GST. (E) Release of pre-phosphorylated and non-phosphorylated GST-V and 3m from immobilized importin- after treatment of filters with GV extract plus ATP. Protein eluted (E) and retained (R) was detected using anti-GST.

View larger version (75K): [in a new window]   Fig. 7. Phosphorylated isoforms of both GST-V and endogenous HDACm can be detected in GVs. (A) Immunoblots of GST-V, T445A and 3m from GVs and cytoplasms isolated at 4 hours post-injection into the cytoplasm. Bars on the left indicate the positions of three possible isoforms. (B) Dephosphorylation of nuclear HDACm in vivo during oocyte maturation. Stage VI oocytes were treated with 0.5 µM progesterone to induce maturation and GVs and cytoplasms were isolated at 0, 2 and 4 hours. Samples taken at 4 hours post-hormone were also treated with 2 units/µl of alkaline phosphatase. Immunoblot using antibody directed against the C-terminal peptide of HDACm. GST-V is shown as a non-phosphorylated control for antibody detection. Bars on the left indicate the positions of four possible isoforms, the fastest migrating of which corresponds to non-phosphorylated HDACm at 57 kDa.

View larger version (66K): [in a new window]   Fig. 8. Effects of overexpression of HDACm and HDACm-2m on the endogenous chromatin of Xenopus oocytes. Vector expressing HDACm (+) or HDACm-2m (S421A, S423A) was injected into GVs of stage IV oocytes and after 24 hours nuclear spreads were made to view lampbrush chromosomes. Some oocytes were co-injected with 50 ng of bromo-UTP to detect endogenous RNA transcription. Chromosomes were immunostained using antibodies against HDACm (A,F), incorporated bromouridine (BrU) (C,H) and histone H4 acetylated at residue lysine12 (H4AcK12) (D,I). All preparations were counterstained with 4,6-diamidino-2-phenylindole (DAPI) to detect chromosomal DNA (B,E,G,J). Bars, 20 µm.