Hook2 contributes to aggresome formation - PubMed (original) (raw)
Hook2 contributes to aggresome formation
Györgyi Szebenyi et al. BMC Cell Biol. 2007.
Abstract
Background: Aggresomes are pericentrosomal accumulations of misfolded proteins, chaperones and proteasomes. Their positioning near the centrosome, like that of other organelles, requires active, microtubule-dependent transport. Linker proteins that can associate with the motor protein dynein, organelles, and microtubules are thought to contribute to the active maintenance of the juxtanuclear localization of many membrane bound organelles and aggresomes. Hook proteins have been proposed to serve as adaptors for the association of cargos with dynein for transport on microtubules. Hook2 was shown to localize to the centrosome, bind centriolin, and contribute to centrosomal function.
Results: Here we show that overexpression of hook2 promotes the accumulation of the cystic fibrosis transmembrane regulator in aggresomes without altering its biochemical properties or its steady state level. A dominant negatively acting form of hook2 that lacks the centriolin binding C-terminal inhibits aggresome formation.
Conclusion: We propose that hook2 contributes to the establishment and maintenance of the pericentrosomal localization of aggresomes by promoting the microtubule-based delivery of protein aggregates to pericentriolar aggresomes.
Figures
Figure 1
Hook2 co-localizes with aggresomes at the centrosome and promotes aggresome formation. Vero cells were transfected to express Hook2. (A) Immunostaining revealed Hook2 either diffusely distributed (type I) or enriched at centrosomes (types II-IV) and [21]. Type IV refers to large aggresomes (> 4 μm). (B) Cells containing distinct Hook2 patterns were counted after the indicated days of expression. The fraction of cells with some hook2 at centrosomes (type II-IV combined) only modestly increased from 1 to 3 days of expression. The fraction of cells in which co-expressed CFTR accumulated in large juxtanuclear aggresomes (type IV) increased about ten-fold from day 1 to 3. (C-D) Cells were transfected to express hook2 and ΔF508-CFTR or (E) hook2 and CFTR-GFP. After 24 hours, 10 μM lactacystin was added and cells were incubated for an additional 16 hours before either being double-stained (C-D) for hook2 (red in merged images) and co-expressed CFTR (green) or triple stained (E) for hook2, CFTR and endogenous hsc70. (F-J) Hook2 was expressed in the absence of CFTR and lactacystin and tested for co-localization with the endogenous aggresome markers hsc-70, ubiquitin (FK1), the 20S proteasome subunit or cytoplasmic dynein, as indicated. Merged images in (I and J) indicate the accumulation of dynein in hook2-induced type II (I) and type IV (J) aggresomes.
Figure 2
Effects of hook2 on aggresomes formation. (A) Vero cells were transfected to express CFTR. Immunofluorescence staining revealed CFTR either diffusely distributed □, aggregated 
, or accumulated in aggresomes ■. (B-C) Cells were transfected to express ΔF508-CFTR with or without hook2. After 24 h 10 μM lactacystine was added for the indicated times, before cells were stained for CFTR and the fractions of cells exhibiting distinct patterns were counted. Pie charts show an increase in the percentage of cells with aggresomal CFTR in cells that co-express hook2, compared to cells that express ΔF508-CFTR alone. Even though lactacystin treatment increased the incidence of ΔF508-CFTR in aggresomes, hook2 promoted aggresome formation in the absence of lactacystin (0 h). (C) The aggresome-promoting effect of hook2 was even more pronounced when cells in which hook2 was enriched at centrosomes were considered separately from those in which hook2 was distributed diffusely. (D-E) Cells were transfected to express hook2 alone or with ΔF508-CFTR and the fraction of cells with type IV hook2 distribution (See Fig. 1A) was counted. (D) In contrast to the effect of hook2 on CFTR (shown in B and C), the presence of CFTR did not change the percentage of cells with type IV hook2. 6 or 12 hours of lactacystin exposure increased the percentage of cells with hook2 aggresomes, but CFTR had no additional effect. (E) CFTR distribution strongly correlated with that of hook2 since in 100% of cells with aggresomal CFTR hook2 was co-enriched at centrosomes.
Figure 3
Effect of hook proteins on CFTR distribution. Hook proteins and CFTR were co-expressed in Vero cells for 36 hours in the absence (A) or presence (B) of 10 μM lactacystin for the final 12 hours. Graphs show the fraction of cells with aggregated CFTR (top) or with CFTR in juxtanuclear aggresomes (bottom), as shown in Figure 2A. In the rest of cells, CFTR was distributed diffusely throughout the cell. At least 400 cells were counted for each construct in at least two separate assays. For hook2 and hook2 truncations more than 5000 cells from at least ten separate assays were counted and yielded similar results. Results shown are from one representative experiment in which all constructs were used in parallel. While absolute numbers changed significantly in different experiments, the relative effect of different constructs was highly consistent between different experiments.
Figure 4
Hook2 expression has no effect on the biochemical properties of CFTR. (A) Vero cells transfected with ΔF508-CFTR were treated (+) or not (-) with lactacystin (lact), lysed and immunoblotted (blot:) for CFTR (left) and ubiquitin (middle). Stars indicate bands specifically seen in ΔF508-CFTR-expressing cells, compared to untransfected controls (-). As shown previously [39, 40] the upper band of CFTR runs at the same position as the ubiquitin-rich band in ΔF508-CFTR transfected and lactacystin treated cells. To control for loading membranes were stained with MemCode after immunoblotting (right). (B) Anti-ubiquitin antibodies also recognized the upper band of immunoprecipitated CFTR. (C) CFTR was expressed in the presence of lactacystin and analyzed by differential detergent extraction. Little CFTR was extracted by 1% Triton X-100 (1). Whereas the lower band of CFTR is soluble in 1% SDS (ISS), a fraction of the upper band is detected in the pellet of the 1% SDS extract (ISP) and needed to be solubilized in 6% SDS, as previously observed for other proteins in aggresomes [4, 14]. (D) CFTR western blot of whole cell homogenates of untransfected cells (lane 1) or cells expressing ΔF508-CFTR-GFP alone (lane 2) or together with hook2 (lanes 3 and 4 are duplicates) showed that hook2 had no effect on the migration of ΔF508-CFTR in SDS-PAGE gels. The apparent reduced level of CFTR-GFP after hook2 co-expression was not consistently observed. (E) Hook2 co-expression did not significantly change the fraction of ΔF508-CFTR detected in Western blots of the 0.1% or 1% Triton-X 100 soluble or insoluble (IS) fractions.
Figure 5
Dominant-negative hook2 proteins do not alter the biochemical properties of CFTR. (A) Western blots indicate that co-expression of hook2 truncations did not significantly change the ratio of CFTR in the high (**) to low-molecular mass (*) bands or the detergent solubility of ΔF508-CFTR. (B) Western blots of homogenates of cells co-transfected with CFTR and various hook proteins or controls showed that steady state levels of wt CFTR or ΔF508-CFTR were not altered by the indicated hook proteins. Band C is the mature CFTR protein and band B the core glycosylated form [41]. Equivalent loading was assured by evaluation of protein concentrations in cell lysates using the DC Protein Assay from BioRad. 20 μg total protein was loaded per well.
References
Publication types
MeSH terms
Substances
Grants and funding
- R01 EY010199/EY/NEI NIH HHS/United States
- R01 DK049835/DK/NIDDK NIH HHS/United States
- DK49835/DK/NIDDK NIH HHS/United States
- R01 NS043406/NS/NINDS NIH HHS/United States
- NS43406/NS/NINDS NIH HHS/United States
- R37 DK049835/DK/NIDDK NIH HHS/United States
LinkOut - more resources
Full Text Sources
Other Literature Sources
Molecular Biology Databases