I found this article to be useful background.  It provides a pretty thorough and recent review of what heat shock proteins are, and how some of them work, which I’ve attempted to summarize.

“The Heat Shock Response: Life on the Verge of Death,” Richter et al., Molecular Cell (2010)


The Damaging Effects of Heat

  • Cells may not recognize temperature per se.  Instead the heat shock response may be triggered by unfolded proteins that are a result of a variety of stresses (oxidative stress, heavy metals, ethanol, other toxic substances.)
  • Cross-protection is possible – Heat shock proteins (Hsps) induced by one kind of stress can still protect against other kinds of stresses.

Seven Classes of Heat Shock Proteins

  • Roughly 50-200 genes are significantly stress-induced in model organisms.
  • These can be functionally grouped into 7 classes.

1. Molecular chaperones: the initially discovered Hsps (and the focus of this article)

2. Components of the proteolytic system, needed to clear irreversibly damaged/aggregated proteins from the cell

3. RNA- and DNA-modifying enzymes, needed to cure nonphysiological covalent modifications of nucleic acids and other kinds of stress-induced DNA damage/ processing failures

4. Metabolic enzymes – these show the most variation between species

5. Regulatory proteins (like transcription factors or kinases) needed to further initiate stress response pathways or inhibit expression cascades

6. Proteins involved in sustaining cellular structures (like the cytoskeleton)

7. Transport, detoxifying, and membrane-modulating proteins needed to maintain/restore membrane stability and function.

Molecular Chaperones

  • All molecular chaperones interact with a broad range of unfolded proteins.
  • These chaperones recognize over-exposure of hydrophobic amino acids (a feature which differentiates  a native protein from a badly folded counterpart)
  • Molecular chaperones bind to hydrophobic patches, specific peptide sequences, or structural elements of the nonnative proteins
  • Binding is usually accomplished by a change of the affinity of the chaperone for its substrate.  This change between at least two affinity states is controlled by the binding and hydrolysis of ATP in most chaperone families.
  • Small Hsps (sHsps) are optimized for binding to nonnative proteins, making them an efficient first line of defense.
  • “Holdases” (eg sHsps): often only expressed upon stress
  • vs. “foldases” (eg Hsp 70 and Hsp 90): exist in both stress-induced and constitutively expressed forms


  • Chaperonins are ring-shaped chaperones that encapsulate nonnative proteins in an ATP-dependent manner
  • The most prominent chaperonin in bacteria is the GroE machinery
  • In eukaryotic cells, GroE is replaced by a distant relative, the CCT or TRiC machinery
  • “Interestingly, the substrate spectrum of the eukaryotic chaperonin may be more limited, and more importantly and stunningly, it is not a heat shock protein in these organisms.  TRiC is actually downregulated under stress conditions in yeast.  This conundrum waits to be resolved: Why remove a protein-folding machine under conditions where it seems to be needed most?  There is certainly more to the picture than meets the eye.”


  • One of the most highly conserved chaperones
  • Involved in de novo folding of proteins under physiological conditions
  • Under stress, they prevent the aggregation of unfolding proteins and can even refold aggregated proteins
  • 2 domains: ATPase domain and protein binding domain
  • Activity regulated by cofactors – largest group is the Hsp40/J-domain containing proteins
  • “While many features of this hydrolysis cycle have been determined, the contribution of Hsp70 to the folding process of a protein or to dissolving aggregates (Goloubinoff and De Los Rios, 2007) remains an important issue to be addressed.”


  • Present in high concentrations even under physiological conditions; upregulated under stress
  • Not as promiscuous in substrate binding as GroE or Hsp70
  • Binds native-like proteins, not unfolded proteins
  • “Whether the substrate spectrum of Hsp90 changes under stress conditions is an important open issue.  What happens to the Hsp90-bound substrates upon restoration of physiological conditions also remains to be determined.”


  • Two different classes, both with dynamic hexameric structures
  • Thought to function by pulling misfolded proteins through the central pore of the hexameric ring in an unfolded state, allowing them to become refolded
  • The exact mechanism at work here is still under debate
  • “Importantly, Hsp100, class 1 proteins are able to support protein disaggregation.  The disaggreagation system composed on ClpB, Hsp70 and Hsp40-like proteins can extract substrates efficiently from aggregates and fold them in a mechanistically still to be defined way to the native state (Goloubinoff et al., 1999).  It is strange that some higher eukaryotes (e.g., nematodes, arthropods, and mammals) lack cytosolic Hsp100, class 1 proteins.  In contrast to the situation with CCT, there seems to be no related protein complex with comparable disaggregation properties in the genomes of these organisms.”

Small Hsps

  • ATP-independent chaperones
  • Interact with large numbers of partially folded proteins to prevent aggregation upon stress-induced misfolding
  • May serve as a storage depot for unfolded proteins, which are then refolded by other proteins like Hsp70 or Hsp 100
  • “It seems that sHsps are not only able to form soluble complexes with their unfolding clients but sometimes, especially when protein unfolding is massive the cell, they are sequestered into the aggregates.  This seems to be a special trait related to their passive holdase function which affects the structure of the aggregates and their remodeling by ATP-dependent chaperones.”

Protein Degradation as Tool for Homeostasis Control

  • It takes less energy to refold/repair a damaged protein than to synthesize a new one.  Nevertheless, protein degradation machineries are expressed as part of the stress response, especially in bacteria – sometimes the damage is irreversible and proteins just need to be cleared.
  • In yeast, only four components of the proteolytic system are among the gene products highly induced by stress.  Two are vacuolar proteases – this hints at the involvement of the autophagic system in protein clearance.
  • Multicellular systems such as human cell lines don’t show much upregulation of proteosome machinery in response to stress. Higher eukaryotes seem to rely more on refolding/repair than on degradation.

Regulation of the Heat Shock Response

  • In humans, the critical transcription factor is heat shock factor 1 (Hsf1).  The name of the transcription site to which it binds on DNA is heat shock element (HSE.)
  • It is believed that Hsf1 is activated by a disturbance in protein homeostasis.
  • The chaperones Hsc70, Hsp90 and Hsp40 have the potential to inhibit Hsf1.
  • Hsf1 is usually kept in an inactive complex along with components of the Hsp90 chaperone system.  A complicated series of regulatory steps including phosphorylation, other posttranslational modifications, and oligomerization regulate Hsf1 activity.
  • “Chaperone titration model” – In the presence of unemployed chaperones, heat shock transcription factors are inactive.  If all of the heat shock proteins are busy with misfolded proteins, heat shock transcription is dramatically activated.

Evolutionary Conservation of Chaperone Networks

  • The different kinds of chaperones interact as a team
  • The fact that different species contain a different subset of chaperones implies that some represent different evolutionary solutions to the same problems
  • Other chaperone networks work primarily in de novo protein folding; there are some overlapping components, but some strikingly differ
  • Ribosome-associated chaperones are mainly involved in de-novo folding and cold shock response
  • Chaperones are just one segment of the heat shock response – protein degradation and metabolic modulation also play a role


  • Stress response in multicellular organisms takes a multi-level approach; “master cell” neurons specialize in sensing heat shock and other toxic stresses, and signal stress with well-known neurotransmitters; autonomous decisions are also made on the cell level.  Both are necessary.
  • We’re just beginning to understand the interplay between cellular, tissue, and organismal stress responses.