DnaK/DnaJ Proteins 
 
These proteins were originally discovered because mutations in their encoding genes block the replication of lambda phage in Escherichia coli (1). Mutations in these genes were later shown to exert pleiotropic effects on bacterial metabolism, including defects in DNA and RNA synthesis, proteolysis, cell division, temperature-sensitive growth, and the overproduction of heat shock proteins. All these effects are believed to result from changes in protein–protein interactions mediated by complexes of DnaK protein with DnaJ protein and a third protein encoded by the grpE gene that acts as a nucleotide-exchange factor (2-5). The DnaK and DnaJ proteins act as molecular chaperones in the folding (6), export (7), and degradation (8, 9) of both newly synthesized and stress-denatured polypeptide chains, and in the dissociation of oligomeric complexes essential for the initiation of phage and plasmid DNA replication (10). This multifaceted property stems from the ability of both DnaK and DnaJ proteins to bind and release hydrophobic segments of an unfolded polypeptide chain in an ATP-driven reaction cycle. Unlike the chaperonins, the DnaK and DnaJ proteins release their polypeptide substrates in unfolded states, and thus their binding serves only to shield such unfolded polypeptides transiently from premature folding during translation and from aggregation under conditions producing a stress response. Both the DnaK and the DnaJ proteins also have the unusual property of altering the conformations of proteins that are native, and thus seemingly fully folded; such proteins may expose chaperone recognition elements that are shielded in other proteins. The ability of both DnaK and DnaJ proteins to bind to the heat shock transcription factor sigma 32 is believed to be part of the autoregulation of stress gene expression in E. coli (3, 11).
 The DnaK protein is a weak ATPase and is about 50% identical in primary structure to the eukaryotic family of hsp70 proteins (see BiP (Hsp70)), while the DnaJ protein has homologues called hsp40 proteins in the cytosol, mitochondria, chloroplasts, and endoplasmic reticulum of eukaryotic cells (6). Genes for homologues of both DnaK and DnaJ proteins have been identified in E. coli (12, 13); the expression of these genes is induced by exposure to low, rather than high, temperatures (14).
Like all the hsp70 proteins, DnaK possesses an N-terminal ATPase domain and a C-terminal
peptide-binding domain; the crystal structure of the latter is known (15), as is that of the ATPase domain of a homologous mammalian hsc70 protein (16). Polypeptides bind in extended conformations to a compact b-sandwich containing two four-stranded antiparallel b-strands. Optimal binding is produced by runs of seven hydrophobic residues; the most important determinant of peptide binding is the central binding pocket for a leucine residue. The ATPase domain transmits ATP-dependent conformational changes to the peptide-binding domain. All the DnaJ proteins possess a conserved J domain of 70 residues, which interacts with hsp70 proteins. The NMR structure of this J domain reveals a scaffolding of four a-helices bearing at its exposed end a conserved tripeptide in a loop region (17); amino acid substitution in this tripeptide prevents binding to DnaK (18). The J domain is followed by a glycine-phenylalanine-rich region, and then a cysteine-rich domain resembling a zinc finger that is involved in binding to unfolded polypeptides (19). Thus the binding of DnaJ to DnaK combines the functions of two chaperones that have rather different specificities for binding to hydrophobic amino acid residues.
References
1. D. E. Friedman, E. R. Olson, C. Georgopoulos, K. Tilly, I. Herskowitz, and F. Banuett (1984(Microbiol. Rev. 48, 299–325. 
2. C. Georgopoulos, K. Liberek, M. Zylicz, and D. Ang (1994) In The Biology of Heat Shock Proteins and Molecular Chaperones (R. I. Morimoto, A. Tissieres, and C. Geogopoulos, eds.), Cold Spring Harbor Press, New York, pp. 209–249. 
3. B. Bukau (1993) Mol. Microbiol. 9, 671–680. 
4. J. Rassow, O. von Ahsen, U. Borner, and N. Pfanner (1997) Trends Cell Biol. 7, 129–133. 
5. D. M. Cyr, T. Langer, and M. G. Douglas (1994) Trends Biochem. Sci. 19, 176–181. 
6. F. U. Hartl (1996) Nature 381, 571–580. 
7. J. Wild, E. Altman, T. Yura, and C. A. Gross (1992) Genes Develop. 1165–1172. 
8. S. A. Hayes and J. F. Dice (1996) J. Cell Biol. 132, 255–258. 
9. D. B. Straus, W. A. Walter, and C. A. Gross (1988) Genes Develop. 2, 1851–1858. 
10. M. Zylicz (1993) Phil. Trans. Roy. Soc. B 339, 255–373. 
11. J. Gamer, H. Bujard, and B. Bukau (1992) Cell 69, 833–842. 
12. T. H. Kawula and M. J. Levivelt (1994) J. Bacteriol. 176, 610–619. 
13. B. L. Seaton and L. E. Vickery (1994) Proc. Natl. Acad. Sci. USA 91, 2066–2070. 
14. M. J. Lelivelt and T. H. Kawula (1995) J. Bacteriol. 177, 4900–4907.
15. X. Zhu, X. Zhao, W. F. Burkholder, A. Gragerov, C. M. Ogata, M. E. Gottesman, and W. A. Hendrickson (1996) Science 272, 1606–1614. 
16. K. M. Flaherty, C. Deluca-Flaherty, and D. B. McKay (1990) Nature 346, 623–628. 
17. R. B. Hill, J. M. Flanagan, and J. H. Prestegard (1995) Biochemistry 34, 5587–5596. 
18. D. Wall, M. Zylicz, and C. Georgopoulos (1994) J. Biol. Chem. 269, 5446–5451. 
19. A. Szabo, R. Korszun, F. U. Hartl, and J. Flanagan (1996) EMBO J. 15, 408–417.