Specific Binding of Tetratricopeptide Repeat Proteins to Heat Shock Protein 70 (Hsp70) and Heat Shock Protein 90 (Hsp90) Is Regulated by Affinity and Phosphorylation

Abstract

Heat shock protein 70 (Hsp70) and heat shock protein 90 (Hsp90) require the help of tetratricopeptide repeat (TPR) domain-containing cochaperones for many of their functions. Each monomer of Hsp70 or Hsp90 can interact with only a single TPR cochaperone at a time, and each member of the TPR cochaperone family brings distinct functions to the complex. Thus, competition for TPR binding sites on Hsp70 and Hsp90 appears to shape chaperone activity. Recent structural and biophysical efforts have improved our understanding of chaperone-TPR contacts, focusing on the C-terminal EEVD motif that is present in both chaperones. To better understand these important protein-protein interactions on a wider scale, we measured the affinity of five TPR cochaperones, CHIP, Hop, DnaJC7, FKBP51, and FKBP52, for the C-termini of four members of the chaperone family, Hsc70, Hsp72, Hsp90α, and Hsp90β, in vitro. These studies identified some surprising selectivity among the chaperone-TPR pairs, including the selective binding of FKBP51/52 to Hsp90α/β. These results also revealed that other TPR cochaperones are only able to weakly discriminate between the chaperones or between their paralogs. We also explored whether mimicking phosphorylation of serine and threonine residues near the EEVD motif might impact affinity and found that pseudophosphorylation had selective effects on binding to CHIP but not other cochaperones. Together, these findings suggest that both intrinsic affinity and post-translational modifications tune the interactions between the Hsp70 and Hsp90 proteins and the TPR cochaperones.

Fig 1

Binding of full-length TPR co-chaperones to the C-termini of cytosolic Hsp70s and Hsp90s. (A) Summary of affinity values, measured by FP. Experiments are the average of the results from at least two independent experiments performed in triplicate each. Error bars represent the standard error of the mean (SEM) of all measurements. Representative binding curves are shown for (B) DnaJC7 (C) FKBP52 and (D) the negative control Hip.

Fig 2

The Ile residue of Hsp70 (GSGPTIEEVD) and Met residue of Hsp90α (DDTSRMEEVD) strongly influence binding preferences. (A) CHIP preferentially binds the C-terminus of Hsp72 (GSGPTIEEVD) over the C-terminus of Hsp90α (DDTSRMEEVD). A mutant C-terminal Hsp72 probe (GSGPTMEEVD) has decreased affinity for CHIP. (B) A mutant C-terminal Hsp90α probe (DDTSRMEEVD) has increased affinity for CHIP. (C) The binding affinities of TPR proteins for mutant Hsp72 (GSGPTMEEVD) and Hsp90α (DDTSRMEEVD) C-terminal tracers. Affinities were measured by FP using full-length TPR proteins. Experiments are the average of the results from at least two independent experiments performed in triplicate each. Error bars represent SEM of all measurements.

Fig 3

The binding of TPR co-chaperones to Hsp70/90 involves polar contacts in the EEVD motif. (A) The Hsp70/90 binding interface of TPR co-chaperones has a strong electropositive character. Surface representation of Hop’s TPR1 domain (PBD code = 1ELW) is shown as an example. Cationic residues (Lys and Arg) are highlighted in gray. Images were prepared using PyMOL. (B) Switching a glutamic acid in the EEVD motif to an alanine slightly decreased the affinity of TPR co-chaperones for Hsp70/90, while replacement with a lysine greatly decreased binding. (C) Affinities of TPR co-chaperones for Hsp90α mutant tracers DDTSRMEAVD and DDTSRMEKVD. (D) Binding of TPR co-chaperones to C-termini of Hsp70s and Hsp90s is pH dependent. Representative results are shown of CHIP binding GSGPTIEEVD. There was no change in the intrinsic fluorescence of the Fam fluorophore under these pH conditions (data not shown). All affinities were measured by FP using full-length TPR proteins. Experiments are the average of the results from at least two independent experiments performed in triplicate each. Error bars represent SEM of all measurements.

Fig 4

Hop’s TPR1 and TPR2A domains selectively interact with the C-termini of Hsp70 and Hsp90. (A) Schematic of the domain architecture of Hop. Gray lines indicate point mutations made in Hop’s TPR1 and TPR2A domains. (B) Structures of Hop’s TPR1 domain (PBD = 1ELW) and TPR2A domain (PBD = 1ELR). Residues that were mutated in these domains are highlighted in Gray. Hsp70/90 C-terminal peptides are shown in black. Structures were prepared using PyMOL. (C) Table summarizing binding affinities of Hop point mutants (K8A, R77A, N223A, and R305A) for chaperone tracers. Affinities were measured by FP using full-length Hop. The results are the average of triplicates and the error bars are SEM. Representative data are shown from two independent replicates. (D) Hop R77A binds specifically to Hsp90s. (E) Hop R305A binds specifically to Hsp70s.

Fig 5

The TPR and J domain regions of DnaJC7 are modular. (A) Addition of DnaJ, a prototypical J protein, stimulated refolding of firefly luciferase by either Hsp72 or Hsp72ΔEEVD. (B) Likewise, DnaJC7 worked with both Hsp72 and Hsp72ΔEEVD to refold denatured luciferase. The results are the average of triplicates and the error bars are SEM. Representative data are shown from two independent replicates.

Fig 6

Mimicking phosphorylation of Hsp70/90 selectively weakens binding to CHIP. (A) CHIP has decreased affinity for Hsp70/90 C-termini that contain phosphomimetic residues. (B) Binding affinities of other TPR co-chaperones for mutant Hsp70/90 C-termini. All affinities measured by FP using full-length TPR proteins. Experiments are the average of the results from at least two independent experiments performed in triplicate each. Error bars represent SEM of all measurements.