Monoubiquitination of CCTα K⁵⁷ masks its NLS. (A) Schematic diagram of five CCTα constructs: CFP-CCT, CFP-CCT_ΔN40, CFP-CCT_K57-Ub, CFP-CCT_ΔN40NLS_CTerm, and CFP-CCT_ΔN40K₅₇-Ub-NLS_CTerm. (B) All constructs shown were transfected into MLE cells. At 24 h after transfection, cells were fixed by paraformaldehyde and counterstained with TOPO-3 to visualize the nucleus. (C) Diagram illustration of the proposed mechanism of CCTα nuclear export.

TNF-α degradation of CCTα is mediated by the endosome-lysosomal pathway. (A and B) MLE cells were exposed to TNF-α (0.75 μg/ml) with or without NH₄Cl (20 mM) or lactacystin (20 μM) treatment for 24 h. Cells were then harvested for analysis of CCT activity (A) or for CCTα and β-actin immunoblotting (B). (C and D). MLE cells were transfected with 0.2 pmol of scramble siRNA or TSG101 siRNA for 24 h. Cells were then treated with or without TNF-α (0.75 μg/ml) for an additional 24 h. Cells were then harvested for analysis of CCT activity (C) or for CCTα, TSG101, and β-actin immunoblotting (D). The data represent the results of three independent experiments. *, P < 0.05 versus the control.

Model for CCTα nuclear activation, exclusion, and proteolytic processing mediated by protein monoubiquitination. A schematic diagram illustrates the proposed pathways by which CCTα is regulated by monoubiquitination. (Arrow 1) CCTα monoubiquitination at K⁵⁷ leads to masking of its NLS, thus mislocalizing the enzyme to the lysosome for degradation. (Arrow 2) Under conditions when the demand for PtdCho synthesis is high, editing of monoubiquitinated CCTα by the action of DUBs may result in free CCTα. Deubiquitinated CCTα can either dock to endoplasmic reticulum membranes (arrow 3) or translocate to the nuclear envelope by NLS-directed recruitment of karyopherins, such as importin-α (Iα), to augment PtdCho synthesis (arrow 4). Finally, when the demand for membrane phospholipid synthesis is low, nuclear CCTα can undergo monoubiquitination and export for proteolytic processing via the lysosomal pathway outlined in the present study.

A K^57R CCTα mutant exhibits greater protein stability and nuclear retention. (A) MLE cells were transfected with CFP-CCT_WT CFP-CCT_K57R and then exposed to cycloheximide (15 μg/ml) in DMEM with 0% FBS for 2 to 8 h prior to paraformaldehyde fixing. Cells were also counterstained with TOPO-3 to visualize the nucleus. A total of 60 random cells were counted at each time point, and CFP fluorescent signals within the cytosol and nucleus were calculated and plotted below in panel A and also in panel C. (B) MLE cells were transfected with V5-CCT_WT (left) or V5-CCT_K57R (right) and then exposed to cycloheximide (15 μg/ml) in DMEM with 0% FBS for 2 to 8 h prior to harvesting for V5 immunoblotting. (D) The densitometric results of CCTα protein were calculated and plotted. The data in each panel represents the results of three independent experiments. *, P < 0.05 versus the control.

The K^57R ubiquitin acceptor site within CCTα is functionally relevant. (A) MLE cells were transfected with V5-CCT_WT or V5-CCT_K57R to assess the physiologic relevance of the ubiquitin acceptor site (K⁵⁷). At 12 h after transfection, cells were exposed to TNF-α (0.75 μg/ml) for an additional 36 h. In some studies, cells were pulsed with [³H]choline (1 μCi/dish) for the final 3 h of incubation. Cells were then harvested for analysis of CCT activity (A), PtdCho synthesis (inset), or for CCTα and β-actin immunoblotting (B). In panel B, the overexpressed CCT mutant levels are better visualized after light exposure of immunoblots. The data in each panel represents the results of three independent experiments, and endogenous and overexpressed bands from the upper part of panel B are quantitated densitometrically below. *, P < 0.05 versus the control. NT, not transfected.

Mapping of a ubiquitination site within CCTα. (A) Map illustrating full-length GST-CCTα and GST-CCTα mutants devoid of the NH₂-terminal sequence (aa 1 to 40) (CCT_ΔN40), the membrane-binding domain (aa 236 to 315) (CCT_MEM), the catalytic region (aa 40 to 236) (CCT_CAT), the carboxyl-terminal phosphorylation domain (aa 315 to 367) (CCT₃₁₅), or the membrane and carboxyl terminus (aa 210 to 367) (CCT₂₁₀). (B) GST-tagged CCT mutants were copurified by using His-tagged ubiquitin in pulldown assays. Thus, GST-tagged CCT mutants were coexpressed in MLE cells, with or without His-tagged ubiquitin. At 24 h after transfection, cells were harvested with lysis buffer. Cell lysate was used for His purification; after stringent washing, the products were eluted and resolved by SDS-PAGE prior to GST immunoblotting. (C) The CCTα primary sequence within the catalytic core, which indicates (by vertical arrows) all of the candidate ubiquitin acceptor sites that were mutated. (D) GST-tagged CCT variants harboring point mutations within the catalytic core were copurified by His-tagged ubiquitin in pull-down assays as in panel B. GST-CCT_K57R, GST-CCT_K100R, GST-CCT_K122R, and GST-CCT_K183/186R were cotransfected with His-tagged ubiquitin. At 24 h after transfection, the cells were harvested with lysis buffer, and cell lysates were used for His pull-down assays. After stringent washing, products were eluted and resolved by SDS-PAGE prior to GST immunoblotting. The data in panels B and D represent the results of three independent experiments.

Monoubiquitinated CCTα accumulates in the cytosol after lysosomal inhibition and effects of a CCTα-ubiquitin mimic. (A) MLE cells were first transfected with CFP-CCTα for 12 h and then exposed to NH₄Cl (20 mM/ml) with or without TNF-α (0.75 μg/ml) or vehicle for an additional 36 h. Cells were permeabilized with 0.001% digitonin-PBS and then fixed and counterstained with To-Pro-3 (to visualize the nucleus, top panel). Cells were also immunostained with an antibody recognizing M6PR (late endosome marker, left panel). Cells were incubated with a 1:10,000 dilution of LysoTracker Red (lysosomal marker), followed by PBS washing and 4% paraformaldehyde fixation (right panel). (B) Expression of CCTα-ubiquitin fusion proteins. WT CCTα (CFP-CCT) and two monoubiquitinated CCTα mimics, where one copy of WT ubiquitin or ubiquitin with all six lysines mutated (CFP-UbiΔK-CCT) was fused to the CCTα NH₂ terminus (CFP-Ubi-CCT) and was transfected into MLE cells. At 24 h after transfection, cells were analyzed as in panel A. The CCTα-ubiquitin fusion proteins that mimic monoubiquitinated CCTα accumulate predominantly in the cytoplasm and colocalize with M6PR and LysoTracker Red. The data represents the results of three independent experiments.

CCTα is monoubiquitinated in vitro and in vivo. (A) Purified CCTα (0.1 μg) was incubated with increasing amounts of conjugation enzyme mixture for 120 min. Reactions were terminated by adding 20 μl of SDS loading buffer, boiled for 5 min, and then resolved by SDS-PAGE prior to CCTα immunoblotting. (B) Purified CCTα (0.1 μg) was incubated with 10 μg of conjugation enzyme mixture for 0 to 120 min. Reactions were terminated and processed by SDS-PAGE prior to CCTα immunoblotting as described above. (C) On the left, MLE cells transfected with ubiquitin (Ubi) were incubated for 24 h and then exposed to 20 mM NH₄Cl for additional 24 h to impair endosomal function, thereby allowing monoubiquitinated protein to accumulate. Cells were lysed, and 10 μg of lysate was resolved by SDS-PAGE prior to CCTα immunoblotting. The presence of faint upper bands represents CCT phosphorylation variants. Alternatively 200-μg lysates were immunoprecipitated with a CCTα rabbit polyclonal antibody, as shown on the right. Immunoprecipitants were resolved by SDS-PAGE prior to ubiquitin immunoblotting. As a negative control, 200-μg lysates were immunoprecipitated with a rabbit immunoglobulin G. Immunoprecipitants were resolved by SDS-PAGE prior to ubiquitin immunoblotting. Below are the results of a densitometric analysis of bands on immunoblots from four independent experiments.

Monoubiquitination of CCTα at K⁵⁷ disrupts its interaction with importin-α. (A) Coimmunoprecipitation. CCTα was immunoprecipitated from MLE cells, and samples were processed for immunoblotting for β-actin (negative control), importin-α, calmodulin (CaM), or Erk. The relative association is presented densitometrically (right graph). (B) Importin interaction assay. Purified CCT_WT or CCT_K57R (4 μg) were reacted in an in vitro ubiquitination reaction or incubated in buffer (top row). After incubation on cobalt beads and washing (middle row), products were reincubated with purified importin-α (2 μg), washed, and processed for CCTα or importin-α immunoblotting (lower row). (C) Densitometric analysis of immunoblots was performed showing relative amounts of importin-α associated with unmodified or monoubiquitinated CCTα after correcting for loading. (D to J) FRET analysis. Cells were cotransfected with YFP-importin-α and one of six constructs: CFP-CCT, CFP-CCT_K57R, CFP-Ub-CCT, CFP-CCT_K57Ub, CFP-CCT_ΔN40K₅₇-Ub-NLS_CTerm, and CFP alone. Importin-α-CCT interaction at the single cell level was imaged by using laser scanning microscopy before and after photobleaching. Shown in the upper sets of panels is single cell imaging showing that, after acceptor photobleaching, the fluorescence intensity of YFP decreases, and CFP increases in panels D, E, and H, confirming the protein interaction between importin-α and CCTα. In panels F, G, and I, after acceptor photobleaching, the fluorescence intensity of YFP decreases, and yet CFP does not change, indicating a lack of protein interaction between importin-α and CFP-Ub-CCT or CFP-CCT_K57Ub. (J) The same FRET in each panel was confirmed quantitatively by graphing of fluorescence intensities. FRET efficiencies (E%) were calculated and graphed in panel J from three experiments and >12 randomly selected cells for each condition that was analyzed.