Cdc34-mediated degradation of ATF5 is blocked by cisplatin

cells treated with different concentration of cisplatin (+ 10 μ M; ++ 20 μ M, 12 h). The expression of endogenous Cdc34 and ATF5 in the cytoplasma ( lanes 1-3 ) and the nucleus ( lanes 4-6 ) were analyzed by western blot. β -Tubulin and Histone were used for the internal control of the cytoplasmic and nuclear fractions as well as to determine possible cross-contamination between the fractions. E . The half-life of endogenous ATF5 in the nucleus was prolonged by cisplatin. The cytoplasm and the nucleus protein were isolated from HeLa cells treated without or with cisplatin. The half-life of endogenous ATF5 was determined as described in ‘ EXPERIMENTAL PROCEDURES ’ ( upper panel ). The half-life of endogenous ATF5 in the nucleus was prolonged from approximately 45 min to 80 min by cisplatin treatment ( lower panel ). The graph represented ATF5/Histone ratio as measured by densitometry. The value of ATF5/Histone at 0 h was taken as 100%. F . Cisplatin inhibited the ubiquitination of ATF5 in vitro . ATF5 protein purified from bacteria BL21 using nickel agarose was used as a substrate in an in vitro ubiquitination assay in the presence of Flag-ubiquitin and 50 μ g of nuclear fractions from HeLa cells which was pretreated without ( lanes 1 , 3 , 5 and 7 ) or with ( lanes 2 , 4 , 6 and 8 ) cisplatin (10 μ M, 12 h). The reactions were carried out with an ATP-regenerating system ( lanes 5-8 ) or an ATP-depleting condition ( lanes 1-4 ) as described in ‘ EXPERIMENTAL PROCEDURES ’. The ubiquitin-conjugated form of ATF5 was detected using

2 glioblastoma. Consistent with this, ATF5 was widely expressed in carcinomas and interference with its function caused apoptotic cell death of neoplastic breast cell lines (8). Furthermore, ATF5 functioned as an anti-apoptotic role in an interleukin 3 (IL-3)-dependent cell line (9). And, we have showed that ATF5 was up-regulated by cisplatin and its over-expression increased cisplatin-induced apoptosis in HeLa cells (5), indicating the pro-apoptotic role of ATF5 in DNA damage-induced apoptosis. Previously, Debanda et al. showed that ATF5 interacted with E2 ubiquitin-conjugating enzyme Cdc34 (10). Cdc34 and RAD6 had been reported to be important components of the ubiquitin-proteasome system in the nucleus which was responsible for the degradation of some transcriptional factors (11).
Although the progresses in ATF5 interacting partners, function and its relation to some diseases have been made in recent years, the precise molecular mechanisms of ATF5 protein regulation during apoptosis are largely unknown.
Cisplatin is a well-known DNA-damaging agent and is widely used in the treatment of solid tumors, including testicular, ovarian, bladder, cervical, head and neck, and small-cell and non-small-cell lung cancers (12)(13)(14). Recent studies provided evidence that the stability of many proteins including Bax, Bak and p53 contributed to cisplatin cytotoxicity in cancer cells (15)(16)(17), indicating the contribution of proteolysis in cisplatin-induced apoptosis. Consistent with this, the proteasome inhibitor PS-341, the representative of a new class of chemotherapeutic drugs, was capable of inducing apoptosis in cisplatin-resistant squamous cell carcinoma cell (18). In addition, cisplatin-resistant ovarian cancer cells were defective in 26S proteasome (19). These data indicated that cisplatin might induce cell apoptosis by regulating the ubiqutin-dependent degradation of protein.
In this study, we provided evidence that ATF5   Immunofluorescence and confocal assay were performed as previously described (22).

Reverse transcription (RT)-PCR and
co-immunoprecipitations-RT-PCR was performed as described previously (5). Primers used for PCR were as follows: ATF5 sense,

Cisplatin increases ATF5 protein expression via inhibiting its ubiquitin-mediated degradation-
Our previous study showed that ATF5 protein was up-regulated during cisplatin-induced apoptosis in HeLa cells (5). To investigate the mechanism of cisplatin-induced ATF5 expression, ATF5 mRNA expression was examined using RT-PCR in HeLa We found that all these mutants were expressed at low levels similar to that of the wild-type protein by fluorescence microscopy and flow cytometry (supplementary S1). Consistent with this, immunoblotting analysis showed that expression levels of all these mutants were increased by MG132 significantly (Fig. 4A), suggesting that none of the lysine was essential for ATF5 degradation in vivo. Next, we measured the half-lives of these mutants by CHX chase experiments. Both wild-type ATF5 and its lysine mutants were rapidly degraded, with half-lives of approximately 45 min (Fig. 4B, lanes 1-5). And, their half-lives were prolonged to approximately 90 min by MG132 treatment (Fig. 4B, lanes 6-10).
Although the DBD domain of ATF5 was not Followed by MG132 treatment, the EGFP-tagged ATF5 was immunoprecipitated using anti-EGFP antibody from cellular extracts and analyzed by immunoblotting using anti-HA antibody. As expected, we detected a high-molecular-weight of HA-marked ubiquitin conjugated ATF5 and its mutants (Fig. 4D). On the basis of these results, we we constructed an N-terminally EGEP-tagged ATF5 plasmid (Fig. 5A, (a)). We found that the N-terminal EGFP-tag dramatically increased ATF5 expression compared to the C-terminal EGFP-tag by fluorescence microscopy observation (Fig. 5A, (b)), flow cytometry analysis (Fig. 5A, (c)) and western blot assay (Fig. 5A, (d)). CHX chase experiments clearly showed that N-terminal EGFP-tag dramatically increased the stability of ATF5 protein compared to the C-terminal EGFP-tag (Fig. 5B).
Since a large N-terminal EGFP tag could stabilize ATF5 protein, we investigated whether the tag blocked its ubiquitinylation. ATF5-EGFP or EGFP-ATF5 was co-transfected with HA-Ub into HeLa cells followed by MG132 treatment or not and immunoprecipitated using anti-EGFP antibody.
As expected, immunoblotting using anti-HA antibody detected the ubiquitin-ATF5 conjugates whose intensity was increased after MG132 treatment whatever the EGFP tag was at the C-terminus or the N-terminus (Fig. 5C) we did observe the ubiquitinylated HA-ATF5 and HA-Xa-ATF5 without treatment of factor Xa protease (Fig. 6, upper panel, lanes 3 and 5). After factor Xa proteasome treatment, the ubiquitinylated HA-Xa-ATF5 was eliminated completely but the ubiquitinylated HA-ATF5 was not diminished (Fig.   6, upper panel, lanes 4 and 6).  (16,18,19