Fig. 1.

The HB complex mediates LATS1/2 suppression in response to proteasome inhibition. The antibodies for immunoblots are indicated on the right. (A) Bag3 depletion reverses the effects of proteasome inhibition on JNK and p38 activity. Control MCF10A cells and MCF10A cells depleted of Bag3 were treated with MG132 under the indicated conditions. Here and throughout the figures, the results of immunoblotting of total cellular lysates were independently reproduced at least three times. (B) The effects of the deletion of various functional domains in Bag3 on its association with LATS1. HeLa cells were transfected with plasmids encoding His-tagged full-length Bag3, with Bag3 deletion mutants, or with empty plasmid (vector, V). Here and throughout this work HeLa cells were utilized because, unlike MCF10A cells, they could be efficiently transfected with a plasmid DNA. On the next day the cells were treated for 75 min with 5 μM MG132 or were left untreated. Then the cells were incubated for 10 min with 1.2% formaldehyde, and His-tagged constructs were isolated using cobalt affinity resin. To calculate relative amounts of LATS1 associated with each His-tagged construct, the quantified LATS1 signal in each lane was normalized by a FLAG signal representing the amount of pulled-down Bag3 in the same sample. No detectable signal was seen in the isolate from the cells transfected with empty vector (V). The results of this pull-down are typical of three independent experiments. ΔC, mutant with a deleted C terminus containing the Bag domain critical for interaction with Hsp70; ΔM, mutant with a deleted M-domain; ΔP, mutant with a deleted PxxP motif; ΔWW, mutant with a deleted WW domain; WT, full-length Bag3. (C) Proteasome inhibition leads to the suppression of LATS1/2 activity as monitored by phosphorylation of YAP. MCF10A cells were incubated for 3.5 h with the indicated concentrations of MG132. (D) Bag3 depletion reduces the effects of proteasome inhibition on LATS1/2 activity. Control MCF10A cells and cells depleted of Bag3 were treated with MG132 under the indicated conditions. Quantification (expressed in relative units) of the phospho-YAP signal normalized by total YAP in the same samples is shown for each of the series. (E) Hsp70 depletion blocks the effects of proteasome inhibition on LATS1/2 activity. MCF10A cells were infected with HSF1 shRNA, briefly selected, and 3 d after the infection were transfected with either HspA8 or control siRNA. Two days later the cells were treated with MG132. Quantification (expressed in relative units) of the phospho-YAP signal normalized by total YAP in the same samples is shown for each of the series. (F) The Bag domain of Bag3 is critical for the regulation of LATS1/2 by proteasome inhibition. MCF10A cells were infected with retroviruses encoding full-length Bag3 (WT) or its ΔC-truncated form (both with mutations making them resistant to the Bag3 siRNA used in this experiment) or with an empty vector. After selection, the cells were transfected with either Bag3 (Lower), or control (Upper) siRNA and 2 d later were treated for 3.5 h with 0.6 μM MG132. Quantification (expressed in relative units) of the phospho-YAP signal normalized by total YAP signal in the same samples is shown for each of the series. Results of an analogous experiment are shown in SI Appendix, Fig. S1G.