Balance between folding and degradation for Hsp90-dependent client proteins: a key role for CHIP

Abstract

Cells must regulate the synthesis and degradation of their proteins to maintain a balance that is appropriate for their specific growth conditions. Here we present the results of an investigation of the balance between protein folding and degradation for mammalian chaperone Hsp90-dependent client proteins. The central players are the molecular chaperones Hsp70 and Hsp90, the cochaperone HOP, and ubiquitin ligase, CHIP. Hsp70 and Hsp90 bind to HOP, thus forming a ternary folding complex whereas the binding of CHIP to the chaperones has previously been shown to lead to ubiquitination and ultimately to degradation of the client proteins as well as the chaperones. To understand the folding/degradation balance in more detail, we characterized the stoichiometries of the CHIP-Hsp70 and CHIP-Hsp90 complexes and measured the corresponding dissociation constants to be approximately 1 muM and approximately 4.5 muM, respectively. We quantified the rate of ubiquitination of various substrates by CHIP in vitro. We further determined that the folding and degradation machineries cannot coexist in one complex. Lastly, we measured the in vivo concentrations of Hsp70, Hsp90, HOP, and CHIP under normal conditions and when client proteins are being degraded due to inhibition of the folding pathway. These in vivo measurements along with the in vitro data allowed us to calculate the approximate cellular concentrations of the folding and degradation complexes under both conditions and formulate a quantitative model for the balance between protein folding and degradation as well as an explanation for the shift to client protein degradation when the folding pathway is inhibited.

FIGURE 1

Schematic of the Hsp70/Hsp90 folding pathway. Hsp70 and Hsp90 are brought into spatial proximity by binding to separate HOP domains. An unfolded client protein first interacts with Hsp70 and partially folded, is then passed to Hsp90, where its folding process is completed.

FIGURE 2

Complex formation between Hsp70 or Hsp90 and CHIP. (A) CHIP-peptide interactions monitored by surface plasmon resonance (SPR). Plot of the average response at equilibrium (RU_eq) versus concentration of CHIP. Open circles show binding to immobilized C-terminal peptide of Hsp70, solid circles show binding to immobilized C-terminal peptide of Hsp90. The solid lines show fits to a one-site binding equation, with dissociation constants of ~ 2 µM and ~ 5 µM, for CHIP binding to Hsp70 and Hsp90, respectively. (B and C) Isothermal Titration Calorimetry (ITC). Upper panels: thermogram of the titration of the C-terminal peptide of Hsp70 (B) or Hsp90 (C) into a solution of CHIP. Lower panels: plot of integrated areas under the peaks of heat in the upper panel as a function of molar ratio. Fitting the data gives a dissociation constant of ~ 1 µM with a stoichiometry of ~ 0.8 C-terminal Hsp70 peptide per CHIP monomer and a dissociation constant of ~ 4.4 µM with a stoichiometry of ~ 0.9 Hsp90 C-terminal peptide per CHIP monomer. (D – G) Complex formation between CHIP and Hsp70 or CHIP and Hsp90, all full-length. Size-exclusion chromatograms of purified CHIP dimer alone, purified Hsp70 monomer alone and an equimolar mixture of the two proteins (D). Coomassie-stained SDS-polyacrylamide gels of the indicated fractions (E). Size-exclusion chromatography analysis of purified CHIP dimer alone, purified Hsp90 dimer alone and an equimolar mixture of the two proteins (F). Coomassie-stained SDS-polyacrylamide gels of the indicated fractions (G). Cartoons of the proposed CHIP-Hsp70 and CHIP-Hsp90 complexes, based on the analysis of the composition of each fraction (see experimental procedures), are shown in (D) and (F), respectively. CHIP does not elute in the same fractions in (D) and (F), because the volume of the fractions is different in the two experiments.

FIGURE 3

Ubiquitination of various substrates by CHIP. (A) In vitro ubiquitination of Hsp90, Hsp70, and Hsp70 T13G by CHIP. Coomassie-stained SDS-polyacrylamide gel analysis of the reaction at various times after initiation. Protein bands corresponding to unmodified protein and its ubiquitin conjugates are indicated. (B) Time course of ubiquitination of Hsp90 (diamonds), Hsp70 (circles), and Hsp70 T13G (triangles). Ubiquitination was quantified by determining the amount of protein in the unmodified band as a function of time. Data were fit to first-order rate equation. CHIP ubiquitinates Hsp90, Hsp70, and Hsp70 T13G at the rates of 0.063, 0.074, and 0.071 nmole of ubiquitinated protein/minute, respectively. Linear line at 0.1 nmole of ubiquitinated protein indicates 100% substrate ubiquitination.

FIGURE 4

Model client protein binds Hsp70 and CHIP, is ubiquitinated by CHIP, and affects the rate of Hsp70 ubiquitination. (A) Interactions of the model client protein with Hsp70, Hsp70 T13G, and CHIP monitored by SPR. Plot of the average response at equilibrium versus concentrations of proteins. Binding of immobilized client protein to Hsp70 (circles), Hsp70 T13G (triangles), and CHIP (diamonds) is depicted. The solid lines show fits to a one-site binding equation with dissociation constants of ~ 50 µM for Hsp70, ~ 30 µM for Hsp70 T13G, and ~ 6.7 µM for CHIP. (B) In vitro ubiquitination of the model client protein by CHIP. Coomassie-stained SDS-polyacrylamide gel analysis of 2 hour ubiquitination reactions in the presence of Hsp70, Hsp70 T13G, or in the absence of Hsp70. Protein band corresponding to the client protein conjugated to one ubiquitin is indicated with an arrow. Other components of the reaction are labeled. (C) In vitro ubiquitination of Hsp70 T13G in the presence and absence of the model client protein. Coomassie-stained SDS-polyacrylamide gel analysis of the reactions at various times after initiation. Time points are the same for both conditions. Unmodified Hsp70 T13G as well as its ubiquitin conjugate species are indicated. (D) Time course of Hsp70 T13G ubiquitination in the presence (solid circles) and absence (open circles) of the model client protein. Ubiquitination was quantified from 3 independent experiments as in (C) by determining the amount of ubiquitin attached to Hsp70 T13G as a function of time. The initial phase of the curve was fit to a linear equation and the obtained rates for Hsp70 T13G ubiquitination are ~ 0.07 and ~ 0.1 ubiquitin molecules attached to Hsp70 T13G/CHIP monomer/minute in the presence and absence of client protein, respectively.

FIGURE 5

HOP and CHIP binding to Hsp90 is mutually exclusive. (A) Schematic illustration of the experiment. An initial complex of two Hsp90 dimers bound to a CHIP dimer is attached to the glutathione resin via GST-tag on CHIP. As the concentration of HOP is increased, it competes with CHIP for binding to Hsp90. Two different scenarios are depicted. In the first, an Hsp90 dimer can bind to both CHIP and HOP simultaneously. In the second, an Hsp90 dimer can bind to either CHIP or HOP, but cannot bind to both simultaneously. Eventually, in both scenarios, at high concentrations of HOP, all the Hsp90 will be bound to HOP, and none will be bound to CHIP and the resin. (B) Coomassie-stained SDS-polyacrylamide gel of the resin-bound proteins as the concentration of HOP competitor is increased. In each experiment +/− 10 µM GST-CHIP and +/− 10 µM Hsp90 were incubated with indicated concentrations of HOP. Protein bands corresponding to Hsp90, HOP, and CHIP are indicated. The input lane contains 20% of CHIP and Hsp90 at the above concentrations and 10 µM HOP. Control experiments in lanes 2 and 3 show that neither Hsp90 nor HOP non-specifically bind to the resin and that HOP does not bind to CHIP. The decrease in the amount of Hsp90 bound to CHIP as HOP concentration increases indicates that HOP is competing with CHIP for binding to Hsp90. Because HOP is never pulled-down by CHIP through the Hsp90 dimer, it is evident that HOP competes with CHIP for binding to Hsp90 via scenario #2 as depicted in (A). (C) Hsp90 pulls-down HOP in a similar experiment as in (B). Coomassie-stained SDS-polyacrylamide gel of Ni-NTA resin elusions of protein mixtures: +/− 20 µM His₆-Hsp90 and 20 µM HOP with its His₆-tag cleaved off by TEV protease. Protein bands corresponding to Hsp90 and HOP are indicated. HOP does not bind non-specifically to the resin but binds to Hsp90. Because the shown lanes are not adjacent on the gel, other lanes between them were cut out.

FIGURE 6

Approximate cellular concentrations of Hsp70, Hsp90, HOP, and CHIP, the various complexes they form, and their effects on the folding/degradation balance. (A) Determination of the cellular concentrations of Hsp70, Hsp90, HOP, and CHIP under normal conditions and upon 17-AAG treatment using quantitative Western blot analysis. BT474 breast cancer cells were left untreated or treated with 0.178 µM 17-AAG for 7 or 14 hours. Total cell lysates from a known number of cells were loaded in each lane and blotted for various proteins. Increasing amount of purified recombinant Hsp70 (0.02, 0.06, 0.1, 0.3, 0.5 µg), Hsp90 (0.02, 0.06, 0.1, 0.14, 0.18 µg), HOP (0.005, 0.025, 0.045, 0.065, 0.085 µg), and CHIP (0.0003, 0.001, 0.002, 0.003 µg) were run along side for concentration calibration. Dashed line separates the endogenous from purified proteins. Cell lysates were also blotted for qualitative comparison of HER2 and GAPDH (loading control) levels. (B) A schematic depicting the approximate total cellular concentrations of Hsp70, Hsp90 dimer, HOP, and CHIP under normal conditions and the dissociation constants between interacting partners. K_d values for the interactions of CHIP with Hsp70 and Hsp90 are as determined here and those for the interactions of HOP with Hsp70 and Hsp90 are as reported (5). (C) Calculated approximate concentrations of the various protein complexes formed by Hsp70, Hsp90, HOP, and CHIP as well as free protein components (not bound to a co-chaperone). Concentrations at normal conditions were calculated based on the total protein concentrations and dissociation constants in (B). Concentrations after 17-AAG treatment were calculated the same way; only the total Hsp70 concentration was increased from 10 µM to 39 µM. (D) A schematic depiction of the folding/degradation balance under normal conditions based on the concentrations shown in (C). In the cell, 10 µM Hsp70 is a component of various complexes depicted in the schematic along with their approximate concentrations: Hsp70 can be free of a co-chaperone, bound only to HOP, bound to HOP-Hsp90, or bound to CHIP. The folding process of a client protein bound to Hsp70, which is free of a co-chaperone or only bound to HOP (~ 9.48 µM), cannot be completed. Client protein bound to Hsp70 within the Hsp70/HOP/Hsp90 folding complex (~ 0.43 µM) will be passed from Hsp70 to Hsp90 for final steps of maturation. Client protein bound to Hsp70 which is associated with CHIP (~ 0.083 µM) will be ubiquitinated. Thus, a client protein is ~ 5 times more likely to be folded (outcome highlighted in bold) than ubiquitinated. There is, however, a low level of client protein ubiquitination. (E) A schematic depicting the shift to client protein degradation upon 17-AAG treatment based on the concentrations shown in (C). When Hsp90 is inhibited, the amount of Hsp70 which is a component of the Hsp70/HOP/Hsp90 folding complex (~ 0.83 µM) and Hsp70 which is part of the Hsp70/CHIP degradation complex (~ 0.095 µM) does not differ much from the amounts at normal conditions in (D). The client protein thus has a similar probability of encountering the Hsp70/HOP/Hsp90 folding complex or the Hsp70/CHIP ubiquitination complex under both conditions. Because Hsp90 is inhibited upon 17-AAG treatment, however, even if the client protein is bound to Hsp70/HOP/Hsp90 complex, its folding process will be initiated, but not completed. Although the probability of the client proteins to encounter the degradation complex is low, all proteins will ultimately be degraded, because there is no alternative pathway.