Function and Regulation of the Heat Shock Proteins
Chapter
88
CAROL A. GROSS
When cells, from bacteria to humans, are shifted to high temperature, the synthesis of a small number of proteins, called the heat shock proteins (hsps), is selectively and rapidly induced. This response is termed the heat shock response and was first discovered in Drosophila sp. In this organism, changes in gene expression can be monitored in the light microscope by examining changes in the puffing pattern of salivary gland chromosomes. In an elegant series of studies in the early 1960s, Ritossa (173, 174, 175) found that exposure to heat led to the transient appearance of several new puffs, suggesting that cells respond to heat by the transient induction of a small number of proteins. Ten years later, Tissieres and colleagues provided a molecular demonstration that this was indeed the case by showing that cells induce the synthesis of at least seven proteins after heat stress (201). Until this time, research on the heat shock response was confined to Drosophila, raising the question as to whether it was a universal phenomenon. However, in the late 1970s and early 1980s, a tremendous explosion of studies exploring the effects of heat on a wide variety of organisms indicated that the heat shock response is universal. Moreover, starting with the pioneering studies of Bardwell and Craig in the 1980s (9, 10), it has become clear that many of the hsps themselves are universally conserved.
That Escherichia coli had heat-responsive genes was discovered independently in 1978 by the Neidhardt and Yura groups, who monitored the rate of synthesis of individual proteins after a temperature upshift using either one- or two-dimensional gels (123, 230). Within the normal growth range, most proteins exhibit very little change (less than twofold) in their rate of synthesis after temperature change. In contrast, the rate of synthesis of one group of about 20 proteins is quite responsive to temperature. After temperature upshift, this group of proteins exhibits a large (10- to 20-fold) but transient increase in synthetic rate. Conversely, after temperature downshift, the rates of synthesis of the hsps show an immediate 20-fold decrease before returning to the low-temperature steady-state rate of synthesis (123, 157, 195, 196, 230). This group of proteins comprises the E. coli hsps, and their expression is regulated at the transcriptional level (35, 197, 231). The amount and activity of σ 32, the alternative sigma factor that guides RNA polymerase to promoters for the heat shock genes, controls expression of this regulon (124, 185, 193). Recently, it has become clear that E. coli contains a second heat-controlled regulon, controlled by another alternative sigma factor, σ E (σ 24) (49, 218). Most members of this regulon have yet to be identified. Finally, the heat induction of several additional genes may occur by other mechanisms (for example, see studies on the psp operon [19]).
What might be the functions of this highly conserved group of proteins? At first, the answers to this question were confusing. Expression of hsps was responsive to a variety of inducers in addition to temperature, including ethanol, viral infection, and unfolded proteins (reviewed in reference 157). In addition, many of the hsps were required during normal cell growth (88, 117, 232, 238). Taken together, this information suggested that the function of the hsps was more general than first anticipated. Moreover, mutations in a variety of heat shock genes seemed to affect every aspect of cell growth including DNA replication, protein and RNA synthesis, and cell division (reviewed in reference 157). As a result of intensive investigations on a limited number of hsps, an understanding of their function is beginning to emerge. A general function of the hsps is to monitor and respond to the state of protein folding in the cell. Many of the hsps are molecular chaperones, whose function is to bind to newly synthesized, partially folded or unfolded proteins and promote their folding and refolding by limiting the nonproductive interactions that lead to aggregation or misfolding (reviewed in references 40, 48, and 86). Other hsps are proteases that function to degrade misfolded or abnormal proteins. From this point of view it is quite understandable that the hsps are required for cell viability, that their synthesis is induced by a variety of treatments that alter proteins, and that mutations in these genes exhibit defects in a wide variety of processes.
The evolution of the idea that many hsps function as molecular chaperones to facilitate protein folding deserves some comment. Early views on protein folding were primarily derived from in vitro studies on the refolding of small polypeptides after denaturation. The pioneering work of Anfinsen on the renaturation of RNase established that the information required for folding was encoded within the primary sequence of the protein (5). His studies showed that neither input of energy nor extrinsic factors were required for refolding. Studies on the refolding of a wide variety of proteins allowed generalization of the principle that proteins were capable of self-assembly (99). Thus, it was initially assumed that auxiliary factors would not be required for proteins to fold in vivo. This turns out not to be the case. Instead, rapid folding of many protein molecules in vivo is dependent on transient interactions with molecular chaperones. These interactions are rapidly reversible by various effectors, permitting the molecule to resume folding after dissociation. By competing with various nonproductive "off-pathway" reactions like aggregation, and by sequestering folding intermediates, the chaperone-polypeptide interaction allows both concerted folding of a protein domain and multiple tries at achieving the native configuration. In retrospect, it is not surprising that such a machinery is necessary. The in vitro refolding studies were carried out on small, full-length proteins in dilute solution. In real life, both small and large proteins in various states of synthesis are folded in very concentrated solutions. All of these conditions enhance the possibility of nonproductive interactions that are either irreversible or very slow on the time scale of the protein folding reaction. Chaperones are necessary to rapidly achieve a high yield of correctly folded proteins.
A second group of hsps functions as proteases. It is a fascinating observation that, to date, all proteases that are part of the σ 32 regulon are ATP dependent. In these proteases, ATP provides energy for conformational changes and for processivity, in addition to being required for peptide bond cleavage in some cases. Indeed, in many respects, the chaperones and heat shock proteases represent two sides of the same coin. This issue will be more fully considered below.
The current review will focus both on the function of hsps in the cell and on the mechanisms by which their expression is regulated. Related information can be found in this volume in chapter 61 by Mayhew and Hartl on chaperones and chapter 62 by Miller on proteases. In addition, I refer the interested reader to many excellent recent and older reviews on this subject (23, 40, 48, 66, 82, 86, 157, 234). Readers should take particular note of an excellent review by Gottesman and Maurizi (76), which heavily influenced my discussion of proteases.
Two global approaches have proved to be particularly useful in identifying hsps. The first is a protein-based approach. By monitoring the rate of synthesis of individual proteins before and after heat shock on two-dimensional gels, Neidhardt and colleagues identified a number of proteins whose rates of synthesis dramatically increased after temperature upshift (123). Initially, these proteins were named by their positions on two-dimensional gels; later, many of these spots were correlated with known genes (for example, see Georgopoulos et al. [68], Neidhardt et al. [155], and Tilly et al. [200]). The second is an RNA-based approach pioneered by Blattner and colleagues, which allows global transcriptional mapping. In this method, radioactively labeled cDNA, made to total E. coli RNA, is hybridized to membrane filters containing an ordered E. coli genomic library carried in λ clones (known as the Kohara library) (113). The relevant loci and proteins are identified by subsequent subcloning and expression studies. Bacteriophage carrying genes whose transcription increases dramatically with temperature will show increased hybridization at high temperature (32). In theory, this method has the disadvantage that baseline transcription from the remainder of the E. coli genes in the insert will dilute the signal. In practice, almost all of the known E. coli heat shock genes and a large number of new heat shock genes have been detected by this method (30). Finally, as more and more of the sequence of the E. coli genome is entered in the database, the search for sequences resembling promoters recognized either by σ 32 or by σ E (see below) will undoubtedly identify additional members of these two regulons. A compendium of the proteins whose rates of synthesis increase upon temperature upshift is presented in Table 1.
Table 1Heat-inducible proteins in E. coli |
In 1975, Cooper and Ruettinger (34) isolated a nonsense mutation (Tsn-K165) that affected synthesis of GroE, an hsp. It was originally assumed that this mutation identified the groE structural gene. However, subsequent analysis indicated that cells containing this mutation failed to exhibit a transient increase in the rates of synthesis of many hsps after shift to high temperature (156, 232), indicating that the mutation affected a global regulator of the heat shock response. The nature of this regulator was clarified when the gene was sequenced and the regulator was purified. The sequence of the gene revealed strong homology to σ 70 (120, 235), and the regulator proved to be the first alternative sigma factor identified in E. coli (83). In view of its function, this regulatory locus, which had initially been called htpR (39) or hin (232), was renamed rpoH (83). rpoH encodes a 32-kDa sigma factor, σ 32, which directs core RNA polymerase (E) to promoters that differ considerably from the majority of those in E. coli, which are recognized by Eσ 70 (35). There is no indication that Eσ 32 promoters can be recognized by Eσ 70 either in vivo or in vitro or vice versa (35, 61, 238). The fact that these heat shock promoters are recognized by Eσ 32 allows their regulation to be distinct from that of most cellular proteins.
A summary of the known Eσ 32 promoters is presented in Fig. 1. Note that this list includes promoters for rRNA genes. At rRNA genes, an Eσ 32 promoter overlaps the Eσ 70 P1 promoter (160). Both Eσ 70 and Eσ 32 initiate RNA at the same nucleotide, making it impossible to accurately judge the amount of transcription originating from the Eσ 32 promoter in vivo. Presumably, the Eσ 32 promoter ensures a high rate of rRNA synthesis at high temperatures. In addition to consensus at the –10 and –35 regions of the promoter, these promoters also share limited sequence similarity upstream of the –35 region. These similarities may be important in the upstream activation that occurs at the rRNA Eσ 32 promoter and presumably at other Eσ 32 promoters.
The overall role of the Eσ 32 regulon has been examined by investigating the effects of mutations in the rpoH gene on cell growth. The general conclusion of these studies is that some σ 32 regulon hsps are required at all temperatures. Moreover, as the temperature is increased, a higher hsp concentration is required for viability. The circumstantial correlation of rate of hsp synthesis with increasing growth temperature was first documented by Herendeen et al. (88). For example, the rate of synthesis of GroEL is less than 1% of total cell protein at 30°C but becomes as much as 20% of total cell protein at 46°C. This correlation is not fortuitous. Systematically varying the concentration of σ 32 in the cell indicated that more σ 32 results in more hsps and permits growth at higher temperatures (232). Finally, studies on strains lacking σ 32 indicate that members of the σ 32 regulon are required for viability throughout the normal growth range. When rpoH is inactivated, cells are viable only at temperatures below 20°C and are defective in growth even at these very low temperatures (238). Moreover, it is possible that viability at low temperature reflects low expression of members of the σ 32 regulon from weak promoters recognized by different sigma factors.
Which members of the regulon are most crucial for cell growth? Strains lacking σ 32 have been used to answer this question. These studies indicate that overexpression of the GroEL-GroES chaperone team (see below) permits growth of Δ rpoH strains up to 40°C and that simultaneous overexpression of the GroEL-GroES and DnaK-DnaJ chaperone teams permits growth as high as 42°C (117). This finding suggests that these two chaperone groups are uniquely important members of the σ 32 regulon. One caveat is that this approach will not identify those regulon members essential for cell viability that have additional strong promoters recognized by different sigma factors.
Why is the σ 32 regulon required for growth at normal temperatures? To uncover this answer, the phenotypes of a variety of strains that are deficient in σ 32 have been characterized by numerous investigators. One role for these chaperones is to promote normal folding of most cellular proteins in vivo. In rpoH mutant strains, most newly synthesized proteins aggregate extensively and are recovered in the cell pellet fractions (80). A second role for the regulon is to promote proteolysis. rpoH mutant strains are defective in the degradation of abnormal peptides and some specific proteins (7, 70). Part of this defect reflects a deficit of σ 32 regulon proteases. However, this is not the whole story. Chaperones themselves are important for the proteolytic process, as dnaK, dnaJ, and groEL mutant strains are defective in this process (111, 194). Finally, rpoH mutant strains are defective in cell division (207), plasmid replication (214), and growth of some temperate bacteriophages (T. Yura, personal communication).
A great deal of information has been obtained about the structure and function of specific members of the regulon. The next section is a summary of the most important findings about each of these proteins.
The DnaK-DnaJ-GrpE Chaperone Team.
In this section I first describe the salient features of DnaK, DnaJ, and GrpE, then discuss how they function together as a chaperone team, and end with a description of their roles in vivo.
DnaK. DnaK is a member of the hsp70 family, which is the largest and best-studied group of hsps. Eukaryotic cells have as many as 10 members of the hsp70 family, allowing for functional diversity, expression in every cellular compartment, and alternative regulation (38). E. coli has at least two family members, DnaK and HscA, both identified by their phenotypic characteristics. dnaK mutants are defective in the replication of bacteriophage λ (reviewed by Friedman et al. [60]), and hscA mutants affect the activity of H-NS, a small protein that is a component of the bacterial nucleoid (110, 182). DnaK is a high-abundance protein, whose gene is transcribed from an Eσ 32 promoter. The mode of regulation of HscA has not yet been established (110, 182). Even though HscA is of low abundance compared to DnaK, inactivating hscA partially reverts the defects of an hns mutant strain. This suggests that HscA and DnaK are functionally distinct. However, the nature of these differences and their structural basis are currently unknown. In addition to DnaK and HscA, a potential third family member, YegD, was recently identified; its relationship to other hsp70s is somewhat distant, and no functional information is currently available (16, 17).
Based upon studies of structure and function and homology arguments, a picture is emerging for the construction of hsp70. Hsp70 is a two-domain protein connected by a protease-sensitive linker. The N-terminal domain binds and hydrolyzes ATP (54) and also binds GrpE (21), whereas the C-terminal domain binds peptides (81). It has been established that a monomer is the species active in ATP hydrolysis (163); however, the active species during chaperone function has not yet been determined. The N-terminal ATPase domain of hsc70 has been crystallized and shown to resemble the ATPase domains of actin and hexokinase, proteins in which ATP binding or hydrolysis is associated with large conformational changes (53, 54). This also seems to be true for hsp70 family members. Their spectral properties, susceptibility to proteases, and affinity for target proteins all change upon ATP binding (130, 146, 165). The crystal structure of the C-terminal domain has not yet been reported; however, homology arguments suggest that the peptide-binding site resembles that of the major histocompatibility complex class I antigen (20, 27, 55, 59, 136, 146, 172). Hsp70 family members clearly bind a great many peptides. To identify features of the peptide sequence that promote binding, several groups have used affinity selection from peptide libraries. These studies indicate that the peptide must be at least a heptamer and that there is a preference for a hydrophobic residue in every other position. This finding is consistent with experiments suggesting that hsp70 proteins bind to extended chains and early folding intermediates that might have exposed hydrophobic residues (13, 58, 81). Interestingly, three different hsp70 family members (BiP, DnaK, hsc70) exhibited differences in their relative affinity for a set of heptapeptides (79). This difference could be one basis for the functional distinction among different hsp70s.
DnaJ. DnaJ is a member of the hsp40 family. As was the case for the hsp70s, eukaryotic cells contain many hsp40 family members (41). The same is true in E. coli, where four DnaJ-like proteins have been identified. DnaJ itself, like DnaK, was discovered because it is a host protein required for λ replication (reviewed by Friedman et al. [60]). This was the first example of many instances in which mutations in either dnaK or dnaJ have similar phenotypes, a finding that is simply explained by the fact that these two proteins work together. A DnaJ homolog called CbpA was initially identified as a protein that bound curved DNA (209). Cloning and sequencing of the gene revealed that it was homologous to dnaJ. Moreover, CbpA appears functionally similar to DnaJ by two criteria. First, when present in multicopy, it complements several mutant properties of a dnaJ allele, restoring growth at high temperature and the ability to replicate λ bacteriophage and F-plasmids. Second, deletion of cbpA together with dnaJ exhibits a more severe phenotype than deletion of dnaJ alone: whereas the double mutant strain cannot grow at 37°C, strains with Δ dnaJ alone are not temperature sensitive until 43°C. The significance of the DNA-binding properties of CbpA is not known. The other two DnaJ homologs, an open reading frame immediately upstream of hscA at 55 min (110) and Orf81 at 1.3 min (233) on the E. coli chromosome, have been identified by sequence similarity only.
Much less is known about DnaJ than DnaK, both structurally and functionally. DnaJ is about 40 kDa, functions as a dimer, and is a chaperone. In addition, DnaJ binds to DnaK, altering its activity in two respects: it accelerates the rate of ATP hydrolysis by DnaK and promotes stable binding of DnaK to some substrates. The relative importance of these varied activities of DnaJ has not been assessed. Sequence comparisons among DnaJ family members have indicated four regions of the protein. From N- to C-terminal, DnaJ contains: (i) the J domain, a highly conserved region of 70 to 80 amino acids (aa); (ii) the G/F linker region of about 30 aa, rich in glycines and phenylalanines; (iii) a cysteine-rich region with the repeated motif CXXCXGXG; and (iv) a less conserved distal region (41). All DnaJ family members contain the J domain. Some family members, for example Sec63 in Saccharomyces cerevisiae and Orf81 in E. coli, contain only this domain. Both genetic and functional evidence suggests that this domain is necessary for binding to DnaK and stimulating its ATPase activity (181, 216, 217; R. McMacken, submitted for publication). Both the J domain and the G/F linker region seem to be necessary for enhancing the binding of DnaK to substrate (217). The role of the cysteine-rich region is unclear at this time. CbpA, which substitutes functionally for DnaJ when present in multicopy, does not have this region. Fragments of DnaJ containing only the C-terminal half of the molecule contain the substrate-binding domain. This suggests that substrate binding is the function of the most distal, least conserved region of DnaJ (217; R. McMacken, submitted). No systematic studies have been reported on the nature of the peptides bound by DnaJ.
GrpE. GrpE, a 21-kDa protein that functions as a dimer, is the third member of the DnaK chaperone team (reviewed in reference 66). To date, only a single grpE locus has been found in E. coli, and it is essential (6). As was the case for DnaJ and DnaK, GrpE was discovered as a host protein required for λ replication (reviewed by Friedman et al. [60]). A eukaryotic grpE-related gene has recently been discovered in S. cerevisiae (15, 97, 119, 213). It is an essential nuclear gene encoding a mitochondrial protein that functions in import in concert with mitochondrial hsp70 and hsp40. It is not clear whether eukaryotic cells will have cytosolic equivalents of GrpE. Sequence homology searches indicate that GrpE family members have five conserved regions spread throughout the length of the protein (15, 229).
GrpE is an accessory factor whose role is to accelerate the release of nucleotides bound to DnaK, thus permitting rapid recycling of DnaK. GrpE is a general nucleotide exchange factor, accelerating the release of both ADP and ATP (104, 129). To perform this function, GrpE must bind to DnaK, and ample evidence exists that this is the case. Genetic evidence from suppression studies, as well as cross-linking and immunoprecipitation, indicates that the two proteins interact (102, 161). In fact, GrpE can be purified on a DnaK column and can be eluted with ATP (241). Bukau and collaborators have recently found that an exposed loop in DnaK, consisting of residues 28–33, is involved in binding GrpE (21). The authors speculate that binding to this loop disrupts the intramolecular interactions within DnaK that inhibit nucleotide release.
Mechanism of action of the DnaK-DnaJ-GrpE chaperone team. The basic function of the DnaK-DnaJ-GrpE chaperone team is to bind to and release target proteins. Exactly how this is performed is currently a subject of great debate. However, the following facts are relevant. (i) The intrinsic ATPase activity of DnaK is very low (k cat, <1 molecule per 10 min) (129, 163, 242). (ii) Physiologically relevant modulators increase the k cat for ATP; DnaJ and GrpE together increase k cat >100-fold (104, 129) and substrate proteins increase k cat about 10-fold (22, 57, 104). (iii) Binding of ATP without hydrolysis induces a conformational change in DnaK (164, 177). (iv) In general, ATP leads to the release of target proteins bound to DnaK. However, examination of the kinetics reveals that ATP speeds up both binding and release but has a larger effect on the release phase (177).
Based on these facts, the following picture emerges. Initial binding to substrate is probably carried out by DnaK·ATP. DnaJ stabilizes this interaction, possibly by several different mechanisms. DnaJ accelerates the conversion of ATP to ADP, may promote a conformational change in DnaK, and also independently binds to substrate (242; Y. Jubete, M. R. Maurizi, and S. Gottesman, submitted for publication). Substrate also stabilizes this interaction by accelerating the conversion of ATP to ADP. Release of the peptide will be accelerated by GrpE, which dissociates the bound ADP, thus permitting rebinding of ATP and peptide release. The detailed study of this reaction is just beginning, and there is little doubt that new insights about chaperone function will emerge from these investigations.
Functions of the DnaK-DnaJ-GrpE chaperone team in vivo. The available evidence suggests that DnaK, DnaJ, and GrpE play a significant role in cellular folding reactions. They are involved in folding of nascent chains, in maintaining proteins destined to be secreted in a translocation-competent form, and in refolding proteins after thermal damage (39, 62, 80, 87, 159, 168, 180, 226, 227, 239). Furthermore, these chaperones are a participant in general proteolysis (111, 194). In addition to their general role in protein folding, DnaK, DnaJ, and GrpE are utilized in certain specific reactions including regulation of σ 32 (discussed below); initiation of flagellar synthesis (184); mutagenesis (167); ribosome assembly at high temperature (3); cell division (207); and replication of bacteriophages λ and P1, the F episome, and possibly cellular DNA as well (24, 44, 45, 66, 82, 91, 98, 128, 224, 225).
The GroEL-GroES Chaperone Team.
In this section, I present a brief overview of the GroEL-GroES chaperone team. The mechanism and function of GroEL-GroES action are treated in great detail in chapter 61.
As was the case for the DnaK-DnaJ-GrpE chaperone team, the groEL and groES loci were initially identified as host genes required for bacteriophage growth, in this case specifically for the assembly of the head structures of bacteriophage particles (67, 69, 189; reviewed in reference 66). GroEL occurs in cells as a ribosome-sized tetradecameric complex of GroEL subunits arranged as a double-ringed structure with sevenfold symmetry (66). This structure interacts with GroES, which consists of a single seven-membered ring (66). As far as is known, there is only one groEL/groES locus in E. coli; however, that situation is not universal. Streptomyces coelicolor has two groEL genes (47), and Bradyrhizobium japonicum has five (52). In the case of B. japonicum, some of the groEL genes show differential regulation, suggesting that regulatory requirements, possibly related to symbiosis, may underlie the proliferation of the gene family. Interestingly, bacteriophage T4 encodes a protein (Gp31) that can functionally substitute for GroES, despite very limited amino acid homology (114, 210). Eukaryotic homologs of GroEL and GroES, variously called hsp60 and hsp10 or chaperonin 60 and chaperonin 10, respectively, function in chloroplasts and mitochondria but not in the cytosol. Instead, a heterooligomeric complex composed of at least six different proteins, which appears to be the functional analog of the GroEL-GroES chaperone team, works in the eukaryotic cytoplasm (reviewed in reference 86).
The available evidence suggests that GroEL and GroES play a significant role in cellular folding reactions. They are involved in folding of nascent chains, in the translation process itself (80, 92, 212), in maintaining proteins destined to be secreted in a translocation-competent form (14, 118), and in refolding proteins after thermal damage (239). Furthermore, these chaperones participate in general proteolysis (194). In addition to their general role in protein folding, GroEL and GroES are involved in certain specific processes such as mutagenesis (46) and bacteriophage head morphogenesis (66, 67, 69, 189). Although GroEL and GroES are involved in the same general processes as DnaK, DnaJ, and GrpE, from the very beginning it was clear that there was a distinction in their mode of action. This distinction was clearly evident in the specialized reactions carried out by each chaperone group: whereas DnaK participated in λ replication, GroEL participated in λ head morphogenesis. However, this distinction in mode of action is also evident upon consideration of their general role in protein folding. In a Δ rpoH strain, most newly synthesized proteins are aggregated. When physiological levels of GroEL, GroES, DnaK, and DnaJ are supplied, protein aggregation is prevented. However, vast overexpression of each separate chaperone team is required to prevent aggregation. These results suggest that the GroEL chaperone team and the DnaK chaperone team play distinct roles in the protein folding pathway (80). One early idea about the difference was that GroEL primarily affected oligomeric protein assembly whereas DnaK affected monomers. That does not seem to be the case, as GroEL participates in folding many monomeric proteins (72, 92, 237). Instead, the difference seems to lie in the kinds of structures recognized by each of these chaperone families. Whereas the preferred substrate of DnaK is an extended chain with hydrophobic residues that is likely to be a very early folding intermediate, a molten globule state that is a later intermediate is preferred by GroEL (13, 56, 58, 84, 121). These substrate preferences suggest that DnaK and GroEL, at least in some cases, may act sequentially in the protein folding pathway. Indeed, a number of different kinds of studies, primarily carried out by Hartl and collaborators, suggest that some unfolded proteins interact initially with DnaK and subsequently with GroEL (122, 167).
HtpG (C62.5).
The HtpG (C62.5) protein was initially identified on two-dimensional gels as an hsp (123) and then shown to be an hsp90 homolog by a low-stringency hybridization approach using the corresponding Drosophila gene as a probe (10). In contrast to the severe phenotypes exhibited by mutations in genes encoding the other chaperones, a Δ htpG strain can grow throughout the normal growth range, albeit more slowly than the wild type at high temperature (11). In addition, Δ htpG exhibits a normal response to carbon starvation (186). Multicopy clones containing htpG have been obtained as suppressors of a secY mutant strain (208), possibly implicating this chaperone in protein translocation. The role of hsp90 in eukaryotic cells has also been something of an enigma, although it is the most abundant cytosolic hsp and has an essential function in S. cerevisiae (100). In addition to associating with a variety of steroid receptors, hsp90 has also been found in combination with hsp70 and peptidyl prolyl isomerases, suggesting that together they may form a cytosolic machine dealing with protein folding (94). The demonstrated chaperone activity of hsp90 is ATP independent (100).
IbpA and IbpB.
IbpA and IbpB were identified as E. coli proteins that associate with inclusion bodies formed after high-level induction of heterologous proteins (4) and exhibit the greatest heat shock of any known E. coli proteins (30). ibpA and ibpB are in an operon and encode proteins that are about 50% homologous to each other (4, 31) and to a set of rather heterogeneous proteins called the small hsps (sHsps) that share conservation of sequence in certain localized regions (4, 31, 100). The α-crystalline lens protein belongs to this group. Several members of the group have been shown to function as chaperones (93, 101), and the location of IbpA and IbpB in inclusion bodies provides circumstantial evidence that they perform a chaperone function (although in this case, the intervention of chaperones was not sufficient to prevent aggregation).
Lon (La).
Lon, the first ATP-dependent protease isolated from E. coli, was initially identified as the product of the capR gene (28, 33). Mutations in capR result in mucoidy. Lack of the Lon protease causes mucoidy because the RcsA protein, a positive regulator of capsular polysaccharide synthesis, is a natural substrate for Lon (190, 206). In the absence of Lon, RcsA builds up to high levels and induces high-level synthesis of capsule, which results in a mucoid phenotype. A second natural substrate, SulA (SfiA), is inducible following damage to DNA and functions to reversibly inhibit cell septation by antagonizing the cell division protein FtsZ (12, 103, 134, 135). In the absence of Lon, SulA levels build up in normal cells and even more in UV-damaged cells. Continuing DNA replication and cell growth in the absence of septation leads not only to sensitivity to DNA-damaging agents but also to long filaments, which give lon (for "long") its name (148, 179). Three other important targets of Lon are known, and other specific substrates are being identified. Lon degrades Tn903 transposase, thus limiting its cellular concentration and possibly contributing to the fact that it is a cis-acting protein (42). Finally, lon mutants have decreased frequency of lysogenization primarily because the bacteriophage λCII protein, which promotes lysogenic growth, is destabilized (75) (see HflB, below).
Lon plays an important role in general protein degradation, removing abnormal proteins from the cell. Lon degrades puromycyl fragments or canavanine-containing proteins (33, 71, 78). The fact that cells lacking Lon show a 50% decrease in the ATP-dependent degradation of such abnormal proteins indicates that Lon is a significant participant in this process (141). In fact, the idea that Lon could be a protease originally came from the finding that lon mutants were identified in a selection that required stabilization of an unstable protein in E. coli (25). Interestingly, a number of hsp chaperones are also involved in degrading abnormal proteins, possibly in conjunction with Lon (111, 194). Lon homologs are probably present in the mitochondria of all eukaryotic cells (215). At least in S. cerevisiae, the function of Lon homolog PIM-1 is strikingly similar to that in E. coli. It is induced by heat shock, is required for the degradation of abnormal proteins, and works in conjunction with DnaK, DnaJ, and possibly GrpE homologs (126, 215). PIM-1 is essential for the maintenance of respiratory-proficient mitochondria (215).
Lon is a serine protease (220, 221) that functions as an endopeptidase, releasing peptides between 5 to 20 aa in length (138, 145). Lon is a tetramer (221), and each monomer contains an ATP-binding domain and a separable proteolytic active site domain (reviewed in reference 76). This arrangement is reminiscent of DnaK, which has a C-terminal peptide-binding domain and an N-terminal ATPase (60, 73). Moreover, as was the case with DnaK, binding either ATP or peptides leads to allosteric changes in Lon (reviewed in reference 76). Binding of ATP (or a nonhydrolyzable analog) alters the proteolytic domain so that there is increased hydrolysis of peptides, increased binding of larger proteins, and greater processivity in some cases. The reciprocal is also true: binding of peptides increases the basal rate of ATP hydrolysis by Lon. Finally, large proteins that bind to Lon have effects that differ from binding peptides: they promote dissociation of ADP and activate Lon peptidase. It is not known whether the protein-binding site is one of the peptide-binding sites in the tetramer or a distinct site.
The Clp Family of Proteases.
The fact that Δ lon strains still exhibit significant ATP-dependent degradation of abnormal proteins led to a search for additional ATP-dependent proteases and resulted in the discovery of a two-component protease, ClpAP (also called Ti). ClpP is the proteolytic subunit. ClpA, the regulatory subunit, binds to the substrate and both binds and hydrolyzes ATP (95, 96, 109). ClpP is an hsp under σ 32 control (116), located immediately upstream of lon (140). It is currently unclear whether ClpA responds to σ 32 (30). Sequences homologous to both ClpP and ClpA are widespread, and these proteins may be conserved among all organisms (77, 142).
The ClpAP protease functions as a large assembly. ClpP is a 22-kDa protein whose functional form is a 14-subunit protein consisting of two seven-membered rings (108, 140) ClpA is an 81-kDa protein which binds as a hexamer to ClpP. The functional form of this protein suggests a structural organization similar to that of the 26S proteosome (M. Kessel, M. R. Maurizi, B. Kim, E. Kocsis, B. L. Trus, S. K. Singh, and A. C. Steven, submitted for publication). ATP plays two roles in the activity of ClpAP protease. First, ATP, or a nonhydrolyzable ATP analog, is required for the assembly of the dimeric form of ClpA to a hexameric protein, which is competent to bind ClpP (139). Second, ATP hydrolysis is required for proteolytic activity (76, 96). The intrinsic ATPase activity of ClpA is modulated by binding either ClpP or substrate (239). Very few specific substrates for ClpAP have been found (109). One substrate appears to be ClpA itself, and experiments with ClpA-lacZ fusion proteins indicate that the first 40 aa contain the targeting signal (74). In addition, ClpAP degrades proteins that have a certain amino acid at their N-terminus (202). In this sense, ClpAP may be the prokaryotic counterpart of the eukaryotic ubiquitin-dependent system of protein degradation in which the N-terminal amino acid is an essential component of the degradation signal.
The discovery of ClpAP resulted from a search for proteases that carried out the residual degradation of abnormal proteins in a Δ lon strain. Hence, it was of some interest to determine whether ClpAP actually participates in this process. Cells deleted either for clpA or clpP exhibit very little, if any, reduction in the ATP-dependent degradation of abnormal peptides (109, 140). However, in combination with a Δ lon mutation, more significant effects are seen. A Δ lon Δ clpP double mutant degrades abnormal canavanyl-containing proteins at only half the rate of a Δ lon single mutant. Thus, ClpP is involved in about 50% of the residual degradation of abnormal proteins seen in cells lacking Lon (73). As described below, there is a second regulatory subunit for ClpP called ClpX. Both ClpAP and ClpXP contribute to the residual degradation in the absence of Lon (108).
Sequence comparisons indicate that the ClpA family includes rather divergent members, which at present are distinguished by whether they have one or two ATP-binding motifs and the spacing of these motifs within the protein (77, 187). Whether these differences in structure translate to functional differences remains to be determined. E. coli has three ClpA homologs, as follows.
ClpY (HslU). Sequence similarity identified protein ClpY (HslU) as a member of the ClpA class (73, 76, 223), and it has been shown to increase the proteolytic activity of HslV (D. Missiakas and S. Raina, personal communication).
ClpX. The clpX gene is cotranscribed with ClpP and ClpX and functions as another regulatory subunit for ClpP (73, 228). It was first identified because ClpXP degrades λO protein (228) and has a substrate specificity distinct from that of ClpA (73, 228). ClpXP also degrades the repressor of a virulent Mu derivative and an unstable protein encoded by bacteriophage P1 (73).
ClpB (previously called F84.1). clpB is located immediately upstream of the rrnG operon. It is unclear whether ClpB is a protease or acts as a chaperone (112, 188). ClpB has not been demonstrated to combine with ClpA or to proteolyze any peptides or proteins (76). Moreover, of the E. coli Clp family members, ClpB is most similar to hsp100 from S. cerevisiae, which has been demonstrated to protect cells from thermal killing and to function as a chaperone (166). Cells lacking ClpB grow more slowly at very high temperatures and also die more rapidly at 50°C, possibly suggesting an analogous role for ClpB (112, 188).
In summary, E. coli contains at least three, and possibly more, heat-inducible two-component proteases that degrade a variety of specific proteins and play an auxiliary role in degrading abnormal proteins. The regulatory subunit of these proteases has characteristics in common with chaperones in that it binds to specific proteins. Indeed, it has recently been demonstrated that both ClpA and ClpX alone function as chaperones in vitro (219, 223). ClpA converts the inactive, dimeric RepA protein of bacteriophage P1 to the active monomer form, thus substituting for the DnaK-DnaJ chaperone team. ClpX can both disassemble aggregated λO protein and prevent its heat-induced aggregation in an ATP-dependent manner.
HflB.
HflB is an hsp whose gene is cotranscribed from a σ 32 promoter along with an upstream gene called ftsJ (90). hflB was originally identified because mutations in that gene led to high-frequency lysogenization of bacteriophage λ (64). The lysis-lysogeny decision in λ is controlled by the concentration of a transcriptional activator, λCII, with higher concentrations favoring lysogeny. λCII is an unstable protein which is degraded by host proteases and stabilized by another λ protein, λCIII. hflB strains are defective in degrading λCII, thus elevating the concentration of λCII and enhancing lysogeny (29, 64). σ 32 is also an unstable protein, and certain mutations in hflB have recently been reported to stabilize σ 32 in vivo (90, 203). These phenotypes suggest the possibility that HflB is a protease. Indeed, in vitro experiments indicate that purified HflB is an ATP-dependent protease (203). It degrades a version of σ 32 tagged with six histidines at its C-terminus. In the next section I will consider the regulatory implications of the fact that a σ 32 regulon member is involved in σ 32 degradation.
HflB is a 70-kDa integral membrane protein with a small N-terminal periplasmic domain and a large C-terminal cytoplasmic domain (204) with weak similarity to several eukaryotic proteins (YME1, SEC18, PAS1-CDC48) that are putative ATPases (198, 205). HflB is the only essential protease currently known in E. coli (89). Moreover, mutations in the gene give pleiotropic phenotypes. hflB is identical to ftsH, a gene identified because certain mutations in it result in defects in cell division (89). In addition, other alleles of hflB are defective in protein secretion (1, 2). It is possible that the diverse phenotypes of hflB mutant strains result from alteration in the levels of the various proteins degraded by HflB. Alternatively, HflB may function both as a protease and as a chaperone.
This brief survey reveals that members of the σ 32 regulon constitute a rich diversity of proteases and chaperones devoted to various protein transactions in the cell. It also raises a number of questions about the functions and relationships of these proteins. Why do chaperones and proteases have specific targets, as well as their general role in protein transactions? How much redundancy of function exists in the system? Is there a clear distinction between chaperones and proteases? How do chaperones facilitate proteolysis? Some of these issues are considered below.
It is intriguing that some specific cellular processes are tied to the action of particular chaperones. There are many possible reasons for this connection. First, the involvement of chaperones makes a regulatory connection between that process and the folding state of the cell. If the cell is coping with an unusually high level of unfolded proteins, sufficient amounts of free chaperones may not be available to permit their participation in the process. That connection may underlie the participation of DnaK and cohorts in the regulation of σ 32 (see below). Second, chaperones may be required to alter the stability of a complex. To carry out some processes, stable structures must first be assembled and then disassembled. If a stable structure assembles spontaneously, it cannot effectively disassemble spontaneously without the intervention of additional players. Involvement of the chaperones, which could favor partial denaturation by stabilizing a non-native state, could be one mechanism to facilitate disassembly of structures. That connection may underlie the involvement of chaperones in λ replication (see chapter 61). Finally, the active species in a reaction may be one that is exceptionally susceptible to aggregation, possibly because it involves exposing a hydrophobic surface. In this case, binding to chaperones may be essential either to prevent aggregation or to disassemble other inactive forms of the protein. That connection may underlie the involvement of chaperones in P1 and F replication and in λ head assembly (see chapter 61). Similar kinds of considerations could underlie the connection of the heat shock proteases to the stability of particular substrates.
It is already clear that there are many species of chaperones, probably to allow both functional specialization and redundancy. Functional specialization underlies at least some of the diversity of players. Yet the distinction is not absolute, and considerable redundancy exists. As discussed earlier, mechanistic studies of the DnaK and the GroEL chaperone teams indicate their distinct roles in mediating protein folding, and in vivo studies confirm this: at physiological levels, both the DnaK and the GroEL chaperone teams must be present for protein folding to occur. However, overexpression of either chaperone team also restores some folding capacity to cells lacking the other chaperone team (see discussion on GroEL above); thus, there is some overlap in function. Some additional chaperones also appear to have discrete functions, as evidenced by the preferential association of the small IbpA and IbpB chaperones with inclusion bodies and the unique ability of the HtpG chaperone (hsp90) to revert a secY mutation. There is an obvious need for mechanistic studies to be performed on these chaperones so that we may more clearly assess their role in protein folding. Moreover, there are likely to be many additional chaperones to be discovered, only some of which may be hsps. Many of these chaperones may have somewhat discrete targets, but overexpression or subtle alteration in properties may permit one chaperone to substitute for another. This reservoir of certain types of chaperones may underlie the ability of cells to accumulate suppressor mutations that permit growth in the absence of DnaK. In this regard, it is likely that the GroE-type chaperone is relatively unique, as the cell does not seem to be able to compensate for the absence of this chaperone (51). Overall, it appears that many different chaperones may normally be involved in the optimal maturation of cellular proteins. Our investigations thus far have probably identified only those with the most severe effects on this process.
It is also clear that there are many proteases, and again one wonders why this is so. Bear in mind that many heat shock proteases form large, complex structures. For example, the two-subunit Clp proteases are large assemblies. E. coli has three known Clp-type proteases and may have more. In this case, the substrate specificity of the recognition subunit may account for the diversity of proteases. Unfortunately, it has not yet been possible to demonstrate significant cellular phenotypes for these proteins. One possible reason is that the role of the Clp proteins in general proteolysis is redundant and that all of the players have not yet been simultaneously deleted. Another possibility is that these proteins have a primary role in response to stress—they may exhibit strong phenotypes after recovery from severe stress, rather than during normal growth. Finally, at least the regulatory subunit of the Clp proteases may play an additional role in the cell (see below).
The heat shock proteases and chaperones share many characteristics in common. Both are ATP dependent and use the ATP, in part, to mediate a conformational change. Both recognize substrates specifically, both probably recognize some feature signaling disordered proteins, and both modulate their conformation in response to binding substrate. Finally, both kinds of proteins generally exist as large, rather complex assemblies in the cell. This convergence is clearest when one examines the Clp proteases, where the substrate binding and proteolytic subunits are distinct. Indeed, as discussed above, two of the substrate-binding subunits, ClpA and ClpX, can function as chaperones, at least in vitro, and ClpB may function exclusively as a chaperone. It will be important to establish the extent to which the protein-binding subunits of proteases also serve to increase the chaperone pool of the cell.
It has been evident for a while that chaperones are important participants both in general proteolysis and in the degradation of specific cellular proteins in prokaryotic and eukaryotic cells. Although their exact molecular mechanism of action is not clear, the chaperones could play a direct role in the proteolytic event. For example, the protease could preferentially recognize the chaperone-protein complex. As an alternative view, the chaperone would play its same role in these reactions as in protein folding. Unstable proteins are probably particularly prone to off-pathway reactions like aggregation. By preventing such reactions, the chaperones will keep proteins in a "protease-competent" form. Maurizi, Gottesman, and colleagues have found that decreased proteolysis of the Lon substrate RcsA in a dnaJ mutant strain is correlated with RcsA aggregation, leading them to favor the latter point of view (Jubete et al., submitted). Likewise, an investigation of the role of the hsp70 chaperone team in protein degradation in mitochondria indicates that degradation by the Lon homolog takes place after DnaK release, again favoring the indirect model (215). However, several investigations by Goldberg and coworkers found that chaperones preferentially associate with abnormal proteins targeted for degradation and constitute a rate-limiting step for that process. In some cases, these complexes were also thought to contain proteases. Based upon these experiments, the Goldberg group has proposed a direct role for chaperones, suggesting that the chaperone-protein complex or a specific conformation of the released protein is the target for the proteolysis machinery (35, 106, 183). If one admits to the possibility that the released protein exists in a special conformation, it may be difficult to distinguish models in which chaperone intervention simply prevents aggregation from those in which chaperones play a more direct role in promoting substrate accessibility.
The transcription of the σ 32-controlled heat shock genes is exquisitely sensitive both to temperature upshift and temperature downshift. When cells are shifted to growth temperatures above 30°C, the rate of transcription of the heat shock genes increases, with larger increases exhibited at higher growth temperatures (123, 230). This increase is both rapid and significant. For example, transcription of the heat shock genes increases about 10-fold within 5 min after temperature upshift from 30 to 42°C (123, 193, 230). Thereafter, the induction phase of the response ends and an adaptation phase ensues during which transcription declines to a new steady-state level. Both the duration of the response and the extent of the decline are determined by the particular temperature (123, 230). The higher the temperature, the longer the duration of maximal transcription and the higher the final rate. At a limit, when cells are shifted to lethal temperatures, transcription of the σ 32 regulon continues at maximal rate for as long as the cells can carry out this process (157).
When cells are shifted to low temperature, the converse response ensues (195, 196). The rate of transcription of the heat shock genes abruptly and dramatically decreases. For example, transcription decreases about 10-fold within 5 min after downshift from 42 to 30°C and continues to decrease. At 20 min, the rate of heat shock gene transcription is only about 5% of that at 42°C. Transcription eventually recovers, but takes one to two cell doublings after downshift.
This abrupt and dramatic change in the transcription rate of the heat shock genes after temperature shift is the mechanism by which the cell achieves a new steady-state level of hsps as rapidly as possible. The steady-state level of hsps in cells growing at 42°C is only twice that at 30°C. The rapid and transient increase in transcription achieves this twofold increase within about 15 min after upshift. Conversely, upon temperature downshift, transcription of hsps is basically shut off until the existing hsps are diluted out by growth to the steady-state level characteristic of 30°C.
A variety of studies indicate that the transient increase in heat shock gene transcription observed after temperature upshift results from a transient increase in the amount of σ 32 (124, 185, 193). During steady-state growth at 30°C, cells contain very little σ 32, on the order of 10 to 30 molecules per cell. Upon temperature upshift, the amount of σ 32 increases rapidly so that by 5 min after upshift, the level of σ 32 is 15-fold greater than it was at 30°C. Following this induction phase, the amount of σ 32 declines and reaches a new steady-state level, somewhat higher than that at low temperature. The kinetics and magnitude of the increase in σ 32 level are sufficient to account for the observed transient increase in heat shock gene transcription.
The transient accumulation of σ 32 results both from changes in σ 32 stability and synthesis (193). During steady-state growth at either low or high temperature, σ 32 is very unstable and is degraded with a half-time of about 1 min. However, immediately following temperature upshift, σ 32 is transiently stabilized for the duration of the induction phase of the heat shock response (about 5 min). Concomitantly, the synthesis of σ 32 transiently increases about 10-fold because of increased translation of rpoH mRNA. Taken together, these two factors account for the buildup of σ 32 in the cell. During the adaptation phase, σ 32 is again rapidly degraded and its rate of synthesis drops, thus lowering its intracellular levels.
In contrast to the situation upon upshift, the initial rapid drop in heat shock gene transcription observed after temperature downshift results from a decrease in σ 32 activity, rather than from a decrease in its concentration (195). During the initial 5 min after downshift, there is a 10-fold decrease in σ 32 transcriptional capacity, but the level of σ 32 drops less than twofold. Although decreased activity of σ 32 is most easily observed after temperature downshift, this is not the only condition that alters its activity. Overexpression of the hsps at constant temperature also reduces σ 32 activity. Moreover, during the adaptation phase of the heat shock response, inactivation of excess σ 32 probably contributes to a rapid return to lower rates of transcription of the heat shock genes.
Expression of the heat shock genes is ultimately controlled by altering σ 32 translation, stability, and activity. Thus, in order to understand how temperature regulates σ 32 gene expression, it is necessary to understand the mechanism for each of these σ 32 transactions.
σ
32Translation.
The evidence that σ 32 synthesis is regulated primarily at the translational rather than transcriptional level came initially from the finding that protein fusions but not operon fusions between rpoH and lacZ exhibited increased synthesis upon temperature upshift (195). That general finding has been confirmed in other ways: temperature regulation of σ 32 synthesis occurs when an exogenous promoter is substituted for the authentic rpoH promoter and is manifest even when rifampin is added to inhibit new initiation of transcription (151). There are two components to the thermal regulation: an initial increase in synthesis and a subsequent decrease (195). These two components could reflect two different stages of a single regulatory event or could result from two independent mechanisms. We do not understand the nature of this translational regulation sufficiently to have a definitive answer to this question. However, investigations to date suggest that these two components have different requirements and may represent distinct regulatory mechanisms (150, 151, 192, 236).
A priori, increased translation upon temperature upshift could be due to either relieving repression or activation. The initial suggestion that temperature upshift relieved translational repression came from the finding that transcription of rpoH was relatively high but translation of its message was poor (195). Support for this idea came from determining the cis-acting sequences in rpoH required for induction after upshift (105, 151). By making a set of successive 3' or 5' deletions of rpoH fused to lacZ, Yura and coworkers identified two regions of rpoH involved in heat induction. The first, located immediately around the translation start point, has sequence homology to the "downstream box," a region found in several prokaryotic mRNAs. Downstream box sequences provide additional homology to 16S rRNA and lead to increased translational capacity. Consistent with this idea, deletion of the downstream box region leads to synthesis of σ 32 that is uninducible and even lower than that seen at 30°C. The second region identified is specific to rpoH and consists of a 100-nucleotide (nt) region internal to the gene. Deletion of this region of rpoH (nt 150–250) results in high, constitutive synthesis of σ 32, indicating that a negative regulatory element had been removed. This finding is consistent with the idea that rpoH translation is usually repressed and that the mechanism of repression involves either the mRNA or the partially translated σ 32 encoded by this region of rpoH.
How might this region act? Secondary structure predictions indicated that portions of the internal region could pair with the downstream box, raising the possibility that mRNA structure per se could be involved in regulation (151). For example, increased temperature or other inducers might function to destabilize this base pairing and allow increased translation. To test this idea, the Yura group made site-directed mutants that would disrupt potential interactions and compensatory changes that would restore base pairing (236). The results of this exercise were mixed. A number of the mutants constructed had significant phenotypes, suggesting that the regions involved in base-pairing interactions were of regulatory importance; however, in general, restoring base pairing did not restore regulation. At the present, it seems most likely that these regions interact with additional factors rather than providing an "RNA sensor" for temperature change. However, to date, no additional factors have been identified.
Repression of translation during the adaptation phase of the heat shock response has both cis and trans requirements that differ from those involved in heat induction. The two RNA segments described above, which mediate thermal induction, are not sufficient for repression. A third segment, located more toward the C-terminus (nt 364–433; aa 122–144), is also necessary for this event (150). A frameshift mutation encompassing this region abolished repression, suggesting that the amino acid sequence rather than nucleotide structure is involved in the regulatory event. Several trans-acting factors are also required for repression. The DnaK-DnaJ-GrpE chaperone team is involved in this event, as mutations in any one of these components reduce or eliminate translational repression without affecting induction of translation (192). Taken together, these studies suggest that translational repression minimally involves chaperone interaction with σ 32. However, no mechanism for repression of translation that involves such protein-protein interactions has been suggested.
σ
32 Stability.
The instability of σ 32 is a key feature of this regulatory system (193). The instability of the protein permits a change either in stability or synthesis to immediately translate into changes in the levels of σ 32. Thus, understanding the mechanism of degradation, and ultimately how the rate of degradation is regulated, is a focus of much attention.
Several factors involved in the degradation of this protein have been identified. The HflB protease, which is a member of the σ 32 regulon, is likely to be the major protease degrading σ 32. In vivo studies indicate that depleting the cell of HflB, or inactivating a mutant HflB by shift to nonpermissive temperature, decreases the rate of degradation of σ 32 about 10-fold (90). Moreover, purified HflB can degrade a σ 32 variant that is tagged with six histidines at its C-terminus (203). The DnaK-DnaJ-GrpE chaperone team is also involved in σ 32 degradation in vivo (192, 199). All three proteins are involved in σ 32 degradation at 30°C, as mutations in each of the corresponding genes decrease the rate of σ 32 degradation 10- to 20-fold. In addition, DnaK and GrpE, but not DnaJ, are required for rapid degradation at 42°C. The particular role played by these chaperones in promoting degradation is not yet established, and it is curious that DnaJ plays no apparent role at 42°C. We do not yet know all of the features of the protein that make it unstable. However, one protein feature contributing to instability has recently been described (150). The frameshift mutation described above that prevented translational repression also stabilized σ 32, indicating that aa 122–144 are essential for the instability of the protein. Both processes have in common the requirement for the DnaK-DnaJ-GrpE chaperone team. This has led to the suggestion that aa 122–144 may be a binding site for either DnaK or DnaJ. One caveat to these conclusions is that the frameshift mutation altering aa 122–144 may induce a global change in protein structure, making it difficult to attribute a particular function to this limited region of σ 32.
σ
32 Activity.
We are furthest advanced in our understanding of how σ 32 is inactivated. In vivo experiments indicate that inactivation of the sigma factor occurs most prominently upon temperature downshift, but also under other conditions where hsps are in excess (192, 196). Inactivation is reversible because σ 32 regains activity after extraction from the cell (195). Moreover, the DnaK-DnaJ-GrpE chaperone team was expected to be involved in this process as mutations in these genes were defective in inactivation (cited in reference 195). These observations led to the proposal that reversible association of these hsps inhibited either the formation or the activity of holoenzyme containing σ 32 (Eσ 32), which resulted in decreased transcriptional activity of Eσ 32 (195).
Recently, elegant in vitro studies, carried out by investigators in the Bukau and Georgopoulos laboratories, are beginning to establish the molecular basis for the inactivation phenomenon. Both DnaK and DnaJ bind independently to σ 32 with comparable affinity (63, 126, 127; B. Bukau, personal communication). Binding of DnaK to σ 32 is dissociated by ATP, whereas binding by DnaJ is not. As is the case for many proteins, a ternary complex consisting of σ 32, DnaK, and DnaJ can be formed and formation of this complex is promoted by ATP (127). It is this ternary complex that is involved in sequestering or inactivating σ 32. There is general agreement that inactivation occurs when σ 32 exists preferentially as a σ 32-DnaK-DnaJ ternary complex rather than as Eσ 32 holoenzyme.
There are several unsolved issues concerning this interaction. The nature of the cooperation between the DnaK and DnaJ chaperones in forming this ternary complex and its effect on σ 32 is currently unclear. Because both DnaK and DnaJ bind independently to σ 32, it would seem that both binding events are important for formation of the complex. However, a recent study from Georgopoulos indicated that a C-terminally truncated DnaJ, which cannot bind substrate, is capable of mediating inactivation in vivo and promoting the tight association of DnaK with σ 32 in vitro (130a, 217). This study suggests that DnaJ may function primarily to modify DnaK. Moreover, it is not clear whether the chaperone team associates only with free σ 32 or can remove σ 32 from holoenzyme. Finally, the in vivo signals that trigger inactivation have yet to be worked out. The interaction of σ 32 with its chaperone regulators is clearly a central feature of the regulatory loop governing σ 32 activity. Moreover, it is a paradigm for the general action of chaperones in in vivo protein folding. From both points of view, new insights should emerge from a careful study of this reaction.
How does the cell perceive the signals that alter expression of the regulon? We have a great deal of information about both ends of this regulatory loop; many inducers have been identified and the responses of σ 32 have been investigated. However, in spite of the many hypotheses, there is little solid information on the signal transduction pathway connecting the two.
One fundamental question is whether a single signal transduction pathway regulates the stability, synthesis, and activity of σ 32 or whether multiple independent pathways carry out these functions. Evidence points to more than one pathway. There is clear evidence that the signal transduction pathway leading to the translational induction of σ 32 is distinct from that stabilizing σ 32. First, the inducing signals are distinct. Temperature upshift, addition of ethanol, and increased production of cytoplasmic unfolded proteins all stabilize σ 32; however, only the first two conditions lead to translational induction (107, 193, 227). Second, the participating players are different. DnaK, DnaJ, and GrpE are involved in σ 32 stability and stabilization, but not in translational induction (192). Finally, RNA interactions are important for translational induction but not for adjusting σ 32 stability (151, 236).
The other σ 32 responses are tied together because each of them (inactivation, stabilization and translational repression) requires the DnaK-DnaJ-GrpE chaperone team. That has led to several different proposals that have in common the idea that these chaperones play a regulatory role in these processes by sensing the signals created by the inducers. It is important to emphasize at the outset that there is currently no solid evidence for this idea. One proposal suggests that DnaK, through its highly temperature-dependent ATPase, senses temperature directly (143). Altering the ATPase (and possibly phosphorylation) of DnaK could affect its properties as a chaperone and lead directly to thermal induction or repression. In this proposal, inducers that work at constant temperature would utilize a different signal transduction pathway. A second proposal sees the chaperone team as the central element of a homeostatic mechanism that senses changes in the folding state of the cell. In this proposal, the "cellular thermometer" is the amount of free DnaK, DnaJ, and GrpE in the cell. The idea is that σ 32 competes with the vast bulk of the unfolded or misfolded proteins in the cell for DnaK, DnaJ, and GrpE. Inducing signals (ethanol, temperature upshift, misfolded proteins) will increase the general substrates for DnaK, DnaJ, and GrpE, thus titrating these chaperones away from σ 32 and relieving their negative regulatory effects on stability and translation. As a consequence, the amount of σ 32 in the cell will rise. Conversely, repressing signals (temperature downshift) will decrease the general substrates for DnaK, DnaJ, and GrpE, thus freeing these chaperones to inactivate σ 32. In each case, the response would be self-limiting because the ensuing overproduction or underproduction of DnaK, DnaJ, and GrpE would act to restore the free pool of these proteins to an appropriate level. Several versions of this model have been proposed (192). The scheme for a homeostatic mechanism is presented diagramatically in Fig. 2.
How likely is it that the homeostatic mechanism described above will prove to be correct? It is very likely that such a mechanism senses the conditions signaling inactivation. In vitro studies indicate that a ternary σ 32-DnaK-DnaJ complex is the active species mediating inactivation. In vivo studies indicate that a twofold overexpression of DnaK and DnaJ at constant temperature can mimic the effect of temperature downshift (N. Kusukawa, P. Rossmeissl, T. Yura, and C. A. Gross, unpublished results). Taken together, these results make it likely that the size of the pool of free DnaK and/or DnaJ determines the extent of σ 32 inactivation.
It is less certain that the free pool of chaperones regulates translational repression or σ 32 stability. So little is known about translational repression that it is hard to evaluate the relationship. It is difficult to imagine, however, that the chaperones alone mediate translational repression. Indeed, the fact that some protein synthesis inhibitors mimic the effects of temperature upshift on hsp synthesis led VanBogelen and Neidhardt to hypothesize that the ribosome itself might be involved in these regulatory loops (212). The involvement of additional players certainly raises the possibility that one of them will respond directly to the signal. Finally, the recent finding that σ 32 is degraded by HflB, another hsp, certainly raises the possibility that the concentration of the protease itself is limiting. In support of this idea, it has been demonstrated that overproduction of HflB increases the rate of degradation of σ 32. Experiments are under way to critically test this possibility.
σ E was initially identified in a study of the transcription of rpoH, encoding σ 32. The rpoH gene is transcribed from four different promoters, three of which are recognized by Eσ 70. The fourth promoter, called rpoHp3, was not recognized by Eσ 70 and showed an interesting pattern of expression. It was a very minor promoter at 30°C (<2% total transcription), but was the major promoter driving transcription at the lethal temperature of 50°C (>90% total transcription) (50). At 50°C, the σ 32 regulon hsps account for most protein synthesis; thus the need for σ 32 production is high (158, 162). Purification of the factor necessary for transcription of rpoHp3 was carried out in two laboratories and identified σ E (also called σ 24) (49, 218).
A study of the signals inducing σ E indicates that activity of σ E, but not that of σ 32, responds to the rate of expression of outer membrane proteins (OMPs) (144). Overexpressing OMPs increases σ E activity, whereas underexpressing OMPs decreases σ E activity. To understand how OMP expression affects σ E activity, the location of the signal generated by OMP overexpression was determined. The signal was not generated in the cytoplasm, since introduction of a secB null allele that results in buildup of cytoplasmic OMP precursors did not induce σ E. On the contrary, such strains showed a decrease in σ E activity. The signal was not generated by titrating translocase, since overproduction of a variety of wild-type and mutant periplasmic proteins (which use the same translocase as OMPs) did not induce σ E. Finally, the signal was generated before insertion in the outer membrane, since overproduced mutant OMPs incapable of outer membrane insertion still induced σ E activity. Taken together, these results suggest that the signal affecting σ E is generated by the extracytoplasmic accumulation of misfolded or immature OMP precursors. For example, a chaperone involved in transporting proteins to the outer membrane could also interact with the signal transduction pathway affecting σ E activity. Change in occupancy of the chaperone by OMPs could generate the signal.
Why does E. coli have two heat-inducible sigma factors? E. coli may have two heat-inducible regulons because they respond to stress in different cellular compartments. Some inducers, like heat and ethanol, affect all compartments and induce both regulons. Other inducers increase the amounts of misfolded or unfolded cytoplasmic proteins and uniquely induce σ 32. Finally, inducers that increase certain misfolded or immature extracytoplasmic proteins uniquely induce σ E. Interestingly, just as the σ 32 response has a close parallel in the eukaryotic heat shock response, theσE response also has a eukaryotic counterpart. Accumulation of unfolded proteins in the endoplasmic reticulum induces a response separate from the heat shock response, called the unfolded protein response (36, 149). It remains to be seen whether the σ E response and the unfolded protein response share either common proteins or mechanisms of induction.
The rpoE gene is located at 55.5 min on the E. coli chromosome (171, 176). Cells lacking σ E are viable at 30°C and form colonies with greatly reduced efficiency at 42 to 43°C (10–3 to 10–5). Δ rpoE strains die more rapidly than wild-type cells after brief exposure to lethal temperature. These phenotypes confirm the importance of the σ E regulon for resistance to thermal stress. However, it should be noted that cells lacking σ 32 do not grow above 20°C. The much more severe phenotype of the σ 32 deletion strains suggests that the major functions for coping with thermal stress are encoded by the σ 32 regulon and that the σ E regulon provides auxiliary functions necessary under more extreme conditions.
Who are the members of the σ E regulon? Overexpression of σ E induced expression of at least 10 proteins, suggesting that the regulon is at least that large (33, 103). However, to date, only three members have been identified. In addition to rpoH, σ E transcribes degP (htrA), which encodes a periplasmic protease, and one of the two promoters (rpoEp2) driving expression of rpoE (171, 176).
The fact that σ E directs transcription of itself raised the possibility that σ E is at least in part transcriptionally regulated. Indeed, there seems to be a good correlation between the steady-state "activity" of σ E as measured by relative expression of an rpoHp3-lacZ reporter construct and the amount of transcription from the rpoEp2 promoter. In contrast, increased transcription of rpoE cannot account for the rapid increase in activity immediately after temperature induction. Moreover, the fact that σ E transcribes its own gene sets up the potential for a positive feedback loop, leading to runaway transcription of rpoE (176). Both situations demand additional proteins regulating σ E activity or expression. Some clue as to what these are comes from homology arguments. σ E is a member of a subgroup of sigmas that is very diverged from most sigma factors. Like σ E, the other members of this family also regulate extracytoplasmic functions, leading Lonetto et al. (133) to call this the ECF sigma group. Interestingly, most ECF sigmas are encoded in operons that include one or more negative regulators. AlgU from Pseudomonas aeruginosa, the ECF sigma most closely related to σ E, is negatively regulated by two proteins called MucA and MucB (43, 137). Sequences about 40% homologous to MucA (but not to MucB) are found immediately downstream of rpoE, suggesting that σ E is also negatively regulated (171, 176). Consistent with this idea, overexpressing rpoE alone leads to high constitutive expression of σ E, whereas expressing rpoE with its downstream sequences does not. These results suggest a model for rapidly inducing σ E activity. An extracytoplasmic signal is transduced to the cytoplasm by a membrane-spanning protein and decreases either the activity or amount of the negative regulator(s) controlling σ E activity. Rapid induction ensues, followed by an adaptation phase as the signal is removed.
Work on this regulon is just beginning. There is no doubt that the next few years will see rapid progress in understanding the function of regulon members, the nature of the signal inducing σ E, and the construction of a regulatory network that links two compartments of the cell.
How general is the heat shock regulatory paradigm presented for E. coli? Although the jury is still out, it is already clear that this paradigm is not universal. The heat shock response has been investigated in several gram-positive organisms, most notably Clostridium acetobutylicum and Bacillus subtilis. So far, there is no evidence that an alternative sigma factor is involved in the primary response to heat. Instead, a highly conserved inverted repeat DNA sequence appears crucial for regulation (152, 153, 178, 222, 240). Interestingly, this DNA structure has so far only been found upstream of those stress genes that are also expressed under normal growth conditions. In contrast, the hsp-18 gene of C. acetobutylicum, which encodes a member of the hsp20 family and is synthesized only after stress, does not contain the inverted repeat (H. Bahl, personal communication). Moreover, in Bacillus subtilis, proteins involved in protection against a wide variety of stresses, including temperature, are synthesized after induction of the alternative sigma factor, SigB (18). These observations suggest that these gram-positive organisms have multiple responses to stress that differ in their regulation from those found in E. coli. Moreover, the E. coli paradigm is unlikely to prove universal even among gram-negative bacteria. The inverted repeat element has been implicated in heat shock control in gram-negative bacteria. Some organisms have both σ 32 and inverted repeat elements (Yura, personal communication) Their relative roles in the heat shock response have not yet been resolved.
The study of the heat shock response in E. coli and other organisms has opened a new window on the life of the cell. It has led to the realization that multiple protein factors are involved in the normal intracellular folding pathway and that various protein transactions, including both refolding and degradation, are crucial to cellular survival after a wide variety of stresses. The study of the in vivo role of chaperones and their protease counterparts is in its infancy. Moreover, we have just become aware that distinct pathways may be necessary to specifically protect the extracytoplasmic compartments of the cell. Finally, the mechanism of action of these fascinating proteins has just now become a popular topic for investigation. This field has become a melting pot for physiologists, biochemists, and biophysicists. The view from this window is bound to change over the next few years.
I thank Mike Lonetto for performing the promoter alignment in Fig. 1; Hubert Bahl, Bernd Bukau, Elizabeth Craig, Costa Georgopoulos, Susan Gottesman, Michael Maurizi, Fred Neidhardt, Kenn Rudd, and Takashi Yura for sharing unpublished information and providing many useful suggestions concerning presentation of material; Alexander Johnson for his editorial comments; and Pam Blittersdorf for editing the manuscript and doing the innumerable tasks necessary to put this manuscript in final form.
References
1. Akiyama, Y., T. Ogura, and K. Ito. 1994. Involvement of FtsH in protein assembly into and through the membrane. I. Mutations that reduce retention efficiency of a cytoplasmic reporter. J. Biol. Chem. 269:5218–5224.
2. Akiyama, Y., Y. Shirai, and K. Ito. 1994. Involvement of FtsH in protein assembly into and through the membrane. II. Dominant mutations affecting FtsH functions. J. Biol. Chem. 269:5225–5229.
3. Alix, J.-H., and M.-F. Guérin. 1993. Mutant DnaK chaperones cause ribosome assembly defects in Escherichia coli. Proc. Natl. Acad. Sci. USA 90:9725–9729.
4. Allen, S. P., J. O. Polassi, J. K. Gierse, and A. M. Easton. 1992. Two novel heat shock genes encoding proteins produced in response to heterologous protein expression in Escherichia coli. J. Bacteriol. 174:6938–6947.
5. Anfinsen, C. B. 1973. Principles that govern the folding of protein chains. Science 181:223–230.
6. Ang, D., and C. Georgopoulos. 1989. The heat-shock-regulated grpE gene of Escherichia coli is required for bacterial growth at all temperatures but is dispensable in certain mutant backgrounds. J. Bacteriol. 171:2748–2755.
7. Baker, T. A., A. D. Grossman, and C. A. Gross. 1984. A gene regulating the heat shock response in Escherichia coli also affects proteolysis. Proc. Natl. Acad. Sci. USA 81:6679–6683.
8. Bardwell, J. C., K. Tilly, E. Craig, J. King, M. Zylicz, and C. Georgopoulos. 1986. The nucleotide sequence of the Escherichia coli K12 dnaJ+ gene. A gene that encodes a heat shock protein. J. Biol. Chem. 261:1782–1785.
9. Bardwell, J. C. A., and E. A. Craig. 1984. Major heat shock gene of Drosophila and the Escherichia coli heat inducible dnaK gene are homologous. Proc. Natl. Acad. Sci. USA 81:848–852.
10. Bardwell, J. C. A., and E. A. Craig. 1987. Eukaryotic M r 83,000 heat shock protein has a homologue in Escherichia coli. Proc. Natl. Acad. Sci. USA 84:5177–5181.
11. Bardwell, J. C. A., and E. A. Craig. 1988. Ancient heat shock gene is dispensable. J. Bacteriol. 170:2977–2983.
12. Bi, E., and J. Lutkenhaus. 1990. FtsZ ring structure associated with division in Escherichia coli. Nature (London) 354:161–164.
13. Blond-Elguindi, S., S. E. Cwirla, W. J. Dower, R. J. Lipshutz, S. R. Sprang, J. F. Sambrook, and M.-J. Gething. 1993. Affinity panning of a library of peptides displayed on bacteriophages reveals the binding specificity of BiP. Cell 75:717–728.
14. Bochkareva, E., N. Lissin, and A. Girshovich. 1988. Transient association of newly synthesized unfolded proteins with the heat-shock GroEL protein. Nature (London) 336:254–257.
15. Bollinger, L., O. Deloche, B. S. Glick, C. Georgopoulos, P. Jeno, N. Kronidou, M. Horst, N. Morishima, and G. Schatz. 1994. A mitochondrial homolog of bacterial GrpE interacts with mitochondrial hsp70 and is essential for viability. EMBO J. 13:1998–2006.
16. Borodovsky, M., E. V. Koonin, and K. E. Rudd. 1994. New genes in old sequence: a strategy for finding genes in the bacterial genome. Trends Biochem. Sci. 19:309–313.
17. Borodovsky, M., K. E. Rudd, and E. V. Koonin. 1994. Intrinsic and extrinsic approaches for detecting genes in a bacterial genome. Nucleic Acids Res. 22:4756–4767.
18. Boylan, S. A., A. R. Redfield, M. S. Brody, and C. W. Price. 1993. Stress-induced activation of the sigma B transcription factor of Bacillus subtilis. J. Bacteriol. 175:7931–7937.
19. Brissette, J. L., M. Russel, L. Weiner, and P. Model. 1990. Phage shock protein, a stress protein of Escherichia coli. Proc. Natl. Acad. Sci. USA 87:862–866.
20. Brown, J. H., T. S. Jardetzky, J. C. Groga, L. J. Stern, R. G. Urban, J. L. Strominger, and D. C. Wiley. 1993. Three-dimensional structure of the human class II histocompatibility antigen HLA-Dr1. Nature (London) 364:33–39.
21. Buchberger, A., H. Schröder, M. Büttner, A. Valencia, and B. Bukau. 1994. A conserved loop in the ATPase domain of the dnaK chaperone is essential for stable binding of grpE. Struct. Biol. 1:95–101.
22. Buchberger, A., A. Valencia, R. McMacken, C. Sander, and B. Bukau. 1994. The chaperone function of DnaK requires the coupling of ATPase activity with substrate binding through residue E171. EMBO J. 13:1687–1695.
23. Bukau, B. 1993. Regulation of the Escherichia coli heat-shock response. Mol. Microbiol. 9:671–680.
24. Bukau, B., and G. C. Walker. 1989. ΔdnaK52 mutants of Escherichia coli have defects in chromosome segregation and plasmid maintenance at normal growth temperatures. J. Bacteriol. 171:6030–6038.
25. Bukhari, A. L., and D. Zipser. 1973. Mutants of Escherichia coli with a defect in the degradation of nonsense fragments. Nature (London) 243:238–241.
26. Burton, Z., R. R. Burgess, J. Lin, D. Moore, S. Holder, and C. A. Gross. 1981. The nucleotide sequence of the cloned rpoD gene for the RNA polymerase sigma subunit from E. coli K12. Nucleic Acids Res. 9:2889–2903.
27. Chappell, T. G., B. B. Konfoirti, S. L. Schmid, and J. E. Rothman. 1987. The ATPase core of a clatherin uncoating protein. J. Biol. Chem. 262:746–751.
28. Charette, M., G. W. Henderson, and A. Markovitz. 1981. ATP hydrolysis-dependent activity of the lon (capR) protein of E. coli K12. Proc. Natl. Acad. Sci. USA 78:4928–4932.
28a. Charpentier, B., and C. Branlant. 1994. The Escherichia coli gapA gene is transcribed by the vegetative RNA polymerase holoenzyme Eσ70 and by the heat shock RNA polymerase Eσ32. J. Bacteriol. 176:830–839.
29. Cheng, H. H., P. J. Muhlrad, M. A. Hoyt, and H. Echols. 1988. Cleavage of the cII protein of phage λ by purified flA protease: control of the switch between lysis and lysogeny. Proc. Natl. Acad. Sci. USA 85:7882–7886.
30. Chuang, S.-E., and F. R. Blattner. 1993. Characterization of twenty-six new heat shock genes of Escherichia coli. J. Bacteriol. 175:5242–5252.
31. Chuang, S. E., V. Burland, G. R. Plunkett, D. L. Daniels, and F. R. Blattner. 1993. Sequence analysis of four new heat-shock genes constituting the hslTS/ibpAB and hslVU operons in Escherichia coli. Gene 134:1–6.
32. Chuang, S. E., D. L. Daniels, and F. R. Blattner. 1993. Global regulation of gene expression in Escherichia coli. J. Bacteriol. 175:2026–2036.
33. Chung, C. H., and A. L. Goldberg. 1981. The product of the lon (capR) gene in Escherichia coli is the ATP-dependent protease, protease La. Proc. Natl. Acad. Sci. USA 78:4931–4935.
34. Cooper, S., and R. Ruettinger. 1975. A temperature sensitive nonsense mutation affecting the synthesis of a major protein of Escherichia coli K-12. Mol. Gen. Genet. 139:167–176.
35. Cowing, D. W., J. C. Bardwell, E. A. Craig, C. Woolford, R. W. Hendrix, and C. A. Gross. 1985. Consensus sequence for Escherichia coli heat-shock gene promoters. Proc. Natl. Acad. Sci. USA 82:2679–2683.
36. Cox, J. S., C. E. Shamu, and P. Walter. 1993. Transcriptional induction of genes encoding endoplasmic reticulum resident proteins requires a transmembrane protein kinase. Cell 73:1197–1206.
37. Craig, A. E., and C. A. Gross. 1991. Hsp70—the cellular thermometer? Trends Biochem. Sci. 16:135–140.
38. Craig, E. A. 1990. Regulation and function of the HSP70 multigene family of Saccharomyces cerevisiae, p. 301–321. In R. I. Morimoto, A. Tissières, and C. Georgopoulos (ed.), Stress Proteins in Biology and Medicine. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
39. Craig, E. A. 1990. Role of hsp70 in translocation of proteins across membranes, p. 279–286. In R. I. Morimoto, A. Tissières, and C. Georgopoulos (ed.), Stress Proteins in Biology and Medicine. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
40. Craig, E. A., B. D. Gambill, and R. J. Nelson. 1993. Heat shock proteins: molecular chaperones of protein biogenesis. Microbiol. Rev. 57:402–414.
41. Cyr, D. M., T. Langer, and M. G. Douglas. 1994. DnaJ-like proteins: molecular chaperones and specific regulators of Hsp70. Trends Biochem. Sci. 19:176–181.
42. Derbyshire, C., M. Kramer, and N. D. F. Grindley. 1990. Role of instability in the cis action of the insertion sequence IS903 transposase. Proc. Natl. Acad. Sci. USA 87:4048–4052.
43. Deretic, V., M. J. Schurr, J. C. Boucher, and D. W. Martin. 1994. Conversion of Pseudomonas aeruginosa to mucoidy in cystic fibrosis: environmental stress and regulation of bacterial virulence by alternative sigma factors. J. Bacteriol. 176:2773–2780.
44. Dodson, M., H. Echols, S. Wickner, C. Alfano, K. Mensa-Wilmont, B. Gomes, I. LeBowitz, and R. McMacken. 1986. Specialized nucleoprotein structures at the origin of bacteriophage lambda: localized unwinding of DNA by a six-protein reaction. Proc. Natl. Acad. Sci. USA 83:7638–7642.
45. Dodson, M., R. McMacken, and H. Echols. 1989. Specialized nucleoprotein structures at the origin of replication of bacteriophage λ. Protein association and disassociation reactions responsible for localized initiation of replication. J. Biol. Chem. 264:10719–10725.
46. Donnelly, C. E., and G. C. Walker. 1989. GroE mutants of Escherichia coli are defective in umuDC-dependent UV mutagenesis. J. Bacteriol. 171:6117–6125.
47. Duchene, A. M., H. M. Kieser, D. A. Hopwood, C. J. Thompson, and P. Mazodier. 1994. Characterization of two groEL genes in Streptomyces coelicolor A3 (2). Gene 144:97–101.
48. Ellis, R. J., and S. M. van der Vies. 1991. Molecular chaperones. Annu. Rev. Biochem. 60:321–347.
49. Erickson, J. W., and C. A. Gross. 1989. Identification of the σE subunit of Escherichia coli RNA polymerase: a second alternate sigma factor involved in high-temperature gene expression. Genes Dev. 3:1462–1471.
50. Erickson, J. W., V. Vaughn, W. A. Walter, F. C. Neidhart, and C. A. Gross. 1987. Regulation of the promoters and transcripts of rpoH, the Escherichia coli heat-shock regulatory gene. Genes Dev. 1:419–432.
51. Fayet, O., T. Ziegelhoffer, and C. Georgopoulos. 1989. The groES and groEL heat shock gene products of Escherichia coli are essential for bacterial growth at all temperatures. J. Bacteriol. 171:1379–1385.
52. Fischer, H. M., M. Babst, T. Kaspar, G. Acuna, F. Arigoni, and H. Hennecke. 1993. One member of a gro-ESL-like chaperonin multigene family in Bradyrhizobium japonicum is co-regulated with symbiotic nitrogen fiaxation genes. EMBO J. 12:2901–2912.
53. Flaherty, K., D. B. McKay, W. Kabsch, and K. C. Holmes. 1991. Similarity of the three-dimensional structures of actin and the ATPase fragment of a 70-kDa heat shock cognate protein. Proc. Natl. Acad. Sci. USA 88:5041–5045.
54. Flaherty, K. M., C. DeLuca-Flaherty, and D. B. McKay. 1990. Three-dimensional structure of the ATP-ase fragment of a 70K heat-shock cognate protein. Nature (London) 346:623–628.
55. Flajnik, M. F., C. Canel, J. Kramer, and M. Kasahara. 1991. Which came first, MHC class I or class II? Immunogenetics 33:295–300.
56. Flynn, G. C., C. J. Beckers, W. A. Baase, and F. W. Dahlquist. 1993. Individual subunits of bacterial luciferase are molten globules and interact with molecular chaperones. Proc. Natl. Acad. Sci. USA 90:10826–10830.
57. Flynn, G. C., T. G. Chappell, and J. E. Rothman. 1989. Peptide binding and release by proteins implicated as catalysts of protein assembly. Science 245:385–390.
58. Flynn, G. C., J. Pohl, M. T. Flocco, and J. E. Rothman. 1991. Peptide-binding specificity of the molecular chaperone BiP. Nature (London) 353:726–730.
59. Fremont, D. H., M. Matsumura, E. A. Stura, P. A. Peterson, and I. A. Wilson. 1992. Crystal structure of two viral peptides in complex with murine MHC class I H-2Kb. Science 257:919–927.
60. Friedman, D. E., E. R. Olson, C. Georgopoulos, K. Tilly, I. Herskowitz, and F. Banuett. 1984. Interactions of bacteriophage and host macromolecules in the growth of bacteriophage λ. Microbiol. Rev. 48:299–325.
61. Fujita, N., and A. Ishihama. 1987. Heat-shock induction of RNA polymerase sigma-32 synthesis in Escherichia coli: transcriptional control and a multiple promoter system. Mol. Gen. Genet. 210:10–15.
62. Gaitanaris, G. A., A. G. Papavassiliou, P. Rubock, S. J. Silverstein, and M. E. Gottesman. 1990. Renaturation of denatured λ repressor requires heat shock proteins. Cell 61:1013–1020.
63. Gamer, J., H. Bujard, and B. Bukau. 1992. Physical interaction between heat shock proteins DnaK, DnaJ, and GrpE and the bacterial heat shock transcription factor σ32. Cell 69:833–842.
64. Gautsch, J. W., and D. L. Wulff. 1974. Fine structure mapping, complementation, and physiology of Escherichia coli hfl mutants. Genetics 77:435–448.
65. Gayda, R. C., P. E. Stephens, R. Hewick, J. M. Schoemaker, W. J. Dreyer, and A. Markovitz. 1985. Regulatory region of the heat shock-inducible capR (lon) gene: DNA and protein sequences. J. Bacteriol. 162:271–275.
66. Georgopoulos, C., D. Ang, K. Liberek, and M. Zylicz. 1990. Properties of the Escherichia coli heat shock proteins and their role in bacteriophage λ growth, p. 191–222. In R. I. Morimoto, A. Tissières, and C. Georgopoulos (ed.), Stress Proteins in Biology and Medicine. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
67. Georgopoulos, C., R. Hendrix, A. Kaiser, and W. Wood. 1972. Role of the host cell in bacteriophage morphogenesis: effects of a bacterial mutation on T4 head assembly. Nature (London) New Biol. 239:38–41.
68. Georgopoulos, C., K. Tilly, D. Drahos, and R. Hendrix. 1982. The B66.0 protein of Escherichia coli is the product of the dnaK + gene. J. Bacteriol. 149:1175–1177.
69. Georgopoulos, C. G., R. W. Hendrix, S. R. Casjens, and A. D. Kaiser. 1973. Host participation in bacteriophage lambda head assembly. J. Mol. Biol. 76:45–60.
70. Goff, S. A., L. P. Casson, and A. L. Goldberg. 1984. Heat shock regulatory gene htpR influences rates of protein degradation and expression of the lon gene in Escherichia coli. Proc. Natl. Acad. Sci. USA 81:6647–6651.
71. Goldberg, A. L., and A. C. St. John. 1976. Intracellular protein degradation in mammalian and bacterial cells: part 2. Annu. Rev. Biochem. 45:747–803.
72. Goloubinoff, P., J. T. Christeller, A. A. Gatenby, and G. H. Lorimer. 1989. Reconstitution of active dimeric ribulose biphosphate carboxylase from an unfolded state depends on two chaperonin proteins and Mg-ATP. Nature (London) 342:884–889.
73. Gottesman, S., W. P. Clark, V. de Crecy-Lagard, and M. R. Maurizi. 1993. ClpX, an alternative subunit for the ATP-dependent Clp protease of Escherichia coli. J. Biol. Chem. 268:22618–22626.
74. Gottesman, S., W. P. Clark, and M. R. Maurizi. 1990. The ATP-dependent Clp protease of Escherichia coli: sequence of clpA and identification of a clp-specific substrate. J. Biol. Chem. 265:7886–7893.
75. Gottesman, S., M. E. Gottesman, J. E. Shaw, and M. L. Pearson. 1981. Protein degradation in E. coli: the lon mutation and bacteriophage lambda N and cII protein stability. Cell 24:225–233.
76. Gottesman, S., and M. R. Maurizi. 1992. Regulation by proteolysis: energy-dependent proteases and their targets. Microbiol. Rev. 56:592–621.
77. Gottesman, S., C. Squires, E. Pichersky, M. Carrington, M. Hobbs, J. S. Mattick, B. Dalrymple, H. Kuramitsu, T. Shiroza, T. Foster, Clark, W. P., B. Ross, C. L. Squires, and M. R. Maurizi. 1990. Conservation of the regulatory subunit for the Clp ATP-dependent protease in prokaryotes and eukaryotes. Proc. Natl. Acad. Sci. USA 87:3513–3517.
78. Gottesman, S., and D. Zipser. 1978. Deg phenotype of Escherichia coli lon mutants. J. Bacteriol. 133:844–851.
79. Gragerov, A., and M. E. Gottesman. 1994. Different peptide binding specificities of hsp70 family members. J. Mol. Biol. 241:133–135.
80. Gragerov, A., E. Nudler, N. Komissarova, G. A. Gaitanaris, M. E. Gottesman, and V. Nikiforov. 1992. Cooperation of GroEL/GroES and DnaK/DnaJ heat shock proteins in preventing protein misfolding in Escherichia coli. Proc. Natl. Acad. Sci. USA 89:10341–10344.
81. Gragerov, A., L. Zeng, X. Zhao, W. Burkholder, and M. E. Gottesman. 1994. Specificity of DnaK-peptide binding. J. Mol. Biol. 235:848–854.
82. Gross, C. A., D. B. Straus, J. W. Erickson, and T. Yura. 1990. The function and regulation of heat shock proteins in Escherichia coli, p. 167–190. In R. I. Morimoto, A. Tissières, and C. Georgopoulos (ed.), Stress Proteins in Biology and Medicine. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
83. Grossman, A. D., J. W. Erickson, and C. A. Gross. 1984. The htpR gene product of E. coli is a sigma factor for heat-shock promoters. Cell 38:383–390.
84. Hayer-Hartl, M. K., J. J. Ewbank, T. E. Creighton, and F. U. Hartl. 1994. Conformational specificity of the chaperonin GroEL for the compact folding intermediates of α-lactalbumin. EMBO J. 13:3192–3202.
85. Hemmingsen, S., C. Woolford, S. M. van der Vies, K. Tilly, D. T. Dennis, C. Georgopoulos, R. W. Hendrix, and J. R. Ellis. 1988. Homologous plant and bacterial proteins chaperone oligomeric protein assembly. Nature (London) 333:330–334.
86. Hendrick, J. P., and F.-U. Hartl. 1993. Molecular chaperone functions of heat-shock proteins. Annu. Rev. Biochem. 62:349–384.
87. Hendrick, J. P., T. Langer, T. A. Davis, F. U. Hartl, and M. Wiedmann. 1993. Control of folding and membrane translocation by binding of the chaperone DnaJ to nascent polypeptides. Proc. Natl. Acad. Sci. USA 90:10216–10220.
88. Herendeen, S. L., R. A. VanBogelen, and F. C. Neidhardt. 1979. Levels of major proteins of Escherichia coli during growth at different temperatures. J. Bacteriol. 139:185–194.
89. Herman, C., T. Ogura, T. Tomoyasu, S. Hiraga, Y. Akiyama, K. Ito, R. Thomas, R. D’Ari, and P. Bouloc. 1993. Cell growth and lambda phage development controlled by the same essential Escherichia coli gene, ftsH/hflB. Proc. Natl. Acad. Sci. USA 90:10861–10865.
90. Herman, C., D. Thévenet, R. D’Ari, and P. Bouloc. 1995. Degradation of σ32, the heat shock regulator in Escherichia coli, is governed by HflB. Proc. Natl. Acad. Sci. USA 92:3516–3520.
91. Hoffman, H. J., S. K. Lyman, C. Lu, M.-A. Petit, and H. Echols. 1992. Activity of the Hsp70 chaperone complex—DnaK, DnaJ, and GrpE—in initiating phage λ DNA replication by sequestering and releasing λ P protein. Proc. Natl. Acad. Sci. USA 89:12108–12111.
92. Horwich, A. L., K. B. Low, W. A. Fenton, I. N. Hirshfield, and K. Furtak. 1993. Folding in vivo of bacterial cytoplasmic proteins: role of GroEL. Cell 74:909–917.
93. Horwitz, J. 1992. Alpha-crystallin can function as a molecular chaperone. Proc. Natl. Acad. Sci. USA 89:10449–10453.
94. Hutchinson, K. A., K. D. Dittmar, and W. B. Pratt. 1994. All of the factors required for assembly of the glucocorticoid receptor into a functional heterocomplex with heat shock protein 90 are preassociated in a self-sufficient protein folding structure, a foldosome. J. Biol. Chem. 269:27894–27899.
95. Hwang, B. J., K. M. Woo, C. H. Chung, and A. L. Goldberg. 1987. Escherichia coli contains a soluble ATP-dependent protease (Ti) distinct from protease La. Proc. Natl. Acad. Sci. USA 84:4490–4494.
96. Hwang, B. J., K. M. Woo, A. L. Goldberg, and C. H. Chung. 1988. Protease Ti, a new ATP-dependent protease in Escherichia coli, contains protein-activated ATPase and proteolytic function in distinct subunits. J. Biol. Chem. 263:8727–8734.
97. Ikeda, E., S. Yoshida, H. Mitsuzawa, I. Uno, and A. Toh-e. 1994. YGE1 is a yeast homologue of Escherichia coli grpE and is required for maintenance of mitochondrial functions. FEBS Lett. 339:265–268.
98. Ishiai, M., C. Wada, Y. Kawasaki, and T. Yura. 1994. Replication initiator protein RepE of mini-F plasmid: functional differentiation between monomers (initiator) and dimers (autogenous repressor). Proc. Natl. Acad. Sci. USA 91:3839–3843.
99. Jaenicke, R. 1987. Folding and association of proteins. Prog. Biophys. Mol. Biol. 49:117–237.
100. Jakob, U., and J. Buchner. 1994. Assisting spontaneity: the role of hsp90 and small hsps as molecular chaperones. Trends Biochem. Sci. 19:205–211.
101. Jakob, U., M. Gaestel, K. Engel, and J. Buchner. 1993. Small heat shock proteins are molecular chaperones. J. Biol. Chem. 268:1517–1520.
102. Johnson, C., G. N. Chandrasekhar, and C. Georgopoulos. 1989. Escherichia coli DnaK and GrpE heat shock proteins interact both in vivo and in vitro. J. Bacteriol. 171:1590–1596.
103. Jones, C. A., and I. B. Holland. 1984. Inactivation of essential genes, ftsA, ftsZ, suppresses mutations at sfiB, a locus mediating division inhibition during the SOS response in E. coli. EMBO J. 3:1181–1186.
104. Jordan, R., and R. McMacken. 1995. Modulation of the ATPase activity of the molecular chaperone DnaK by peptides and the DnaJ and GrpE heat shock proteins. J. Biol. Chem. 270:4563–4569.
105. Kamath-Loeb, A. S., and C. A. Gross. 1991. Translational regulation of σ32 synthesis: requirement for an internal control element. J. Bacteriol. 173:3904–3906.
106. Kandror, O., L. Busconi, M. Sherman, and A. L. Goldberg. 1994. Rapid degradation of an abnormal protein in Escherichia coli involves the chaperones GroEL and GroES. J. Biol. Chem. 269:23575–23582.
107. Kanemori, M., H. Mori, and T. Yura. 1994. Induction of heat shock proteins by abnormal proteins results from stabilization and not increased synthesis of σ32 in Escherichia coli. J. Bacteriol. 176:5648–5653.
108. Katayama, Y., S. Gottesman, J. Pumphrey, S. Rudikoff, W. P. Clark, and M. R. Maurizi. 1988. The two-component, ATP-dependent Clp protease of Escherichia coli. Purification, cloning, and mutational analysis of the ATP-binding component. J. Biol. Chem. 263:15226–15236.
109. Katayama-Fujimura, Y., S. Gottesman, and M. R. Maurizi. 1987. A multiple-component ATP-dependent protease from Escherichia coli. J. Biol. Chem. 262:4477–4485.
110. Kawula, T. H., and M. J. Lelivelt. 1994. Mutations in a gene encoding a new hsp70 suppress rapid DNA inversion and bgl activation, but not proU derepression, in hns-1 mutant Escherichia coli. J. Bacteriol. 176:610–619.
111. Keller, J. A., and L. D. Simon. 1988. Divergent effects of a dnaK mutation on abnormal protein degradation in Escherichia coli. Mol. Microbiol. 2:31–41.
112. Kitagawa, M., C. Wada, S. Yoshioka, and T. Yura. 1991. Expression of ClpB, an analog of the ATP-dependent protease regulatory subunit in Escherichia coli, is controlled by a heat shock σ factor (σ32). J. Bacteriol. 173:4247–4253.
113. Kohara, Y., K. Akiyama, and K. Isono. 1987. The physical map of the whole E. coli chromosome: application of a new strategy for rapid analysis and sorting of a large genomic library. Cell 50:495–508.
114. Koonin, E. V., and S. M. van der Vies. 1995. Conserved sequence motifs in bacterial and bacteriophage chaperonins. Trends Biochem. Sci. 20:14–15.
115. Kornitzer, D., D. Teff, S. Altuvia, and A. B. Oppenheim. 1991. Isolation, characterization, and sequence of an Escherichia coli heat shock gene, htpX. J. Bacteriol. 173:2944–2953.
116. Kroh, H. E., and L. E. Simon. 1990. The ClpP component of Clp protease is the σ32-dependent heat shock protein F21.5. J. Bacteriol. 172:6026–6034.
117. Kusukawa, N., and T. Yura. 1988. Heat shock protein GroE of Escherichia coli: key protective roles against thermal stress. Genes Dev. 2:874–882.
118. Kusukawa, N., T. Yura, C. Ueguchi, Y. Akiyama, and K. Ito. 1989. Effects of mutations in heat-shock genes groES and groEL on protein export in Escherichia coli. EMBO J. 8:3517–3521.
119. Laloraya, S., B. D. Gambill, and E. A. Craig. 1994. A role for a eukaryotic GrpE-related protein, Mge1p, in protein translocation. Proc. Natl. Acad. Sci. USA 91:6481–6485.
120. Landick, R., V. Vaughn, E. T. Lau, R. A. VanBogelen, J. W. Erickson, and F. C. Neidhart. 1984. Nucleotide sequence of the heat-shock regulatory gene of E. coli suggests its product may be a transcription factor. Cell 38:175–182.
121. Landry, S. L., R. Jordan, R. McMacken, and L. M. Gierasch. 1992. Different conformations for the same polypeptide bound to chaperones DnaK and GroEL. Nature (London) 355:455–457.
122. Langer, T., C. Lu, H. Echols, J. Flanagan, M. K. Hayer, and F.-U. Hartl. 1992. Successive action of DnaK, DnaJ and GroEL along the pathway of chaperone-mediated protein folding. Nature (London) 356:683–689.
123. Lemaux, P. G., S. L. Herendeen, P. Bloch, and F. C. Neidhardt. 1978. Transient rates of synthesis of individual polypeptides in E. coli following temperature shifts. Cell 13:427–434.
124. Lesley, S. A., N. E. Thompson, and R. R. Burgess. 1987. Studies of the role of the Escherichia coli heat shock regulatory protein σ32 by the use of monoclonal antibodies. J. Biol. Chem. 262:5404–5407.
125. Léveque, F., P. Plateau, P. Dessen, and S. Blanquet. 1990. Homology of lysS and lysU, the two Escherichia coli genes encoding distinct lysyl-tRNA synthetase species. Nucleic Acids Res. 18:305–312.
126. Liberek, K., T. P. Galitski, M. Zylicz, and C. Georgopoulos. 1992. The DnaK chaperone modulates the heat shock response of Escherichia coli by binding to the σ32 transcription factor. Proc. Natl. Acad. Sci. USA 89:3516–3520.
127. Liberek, K., and C. Georgopoulos. 1993. Autoregulation of the Escherichia coli heat shock response by the DnaK and DnaJ heat shock proteins. Proc. Natl. Acad. Sci. USA 90:11019–11023.
128. Liberek, K., C. Georgopoulos, and M. Zylicz. 1988. Role of the Escherichia coli DnaK and DnaJ heat shock proteins in the initiation of bacteriophage λ DNA replication. Proc. Natl. Acad. Sci. USA 85:6632–6636.
129. Liberek, K., J. Marzlalek, D. Ang, and C. Georgopoulos. 1991. Escherichia coli DnaJ and GrpE heat shock proteins jointly stimulate ATPase activity of DnaK. Proc. Natl. Acad. Sci. USA 88:2874–2878.
130. Liberek, K., D. Skowyra, M. Zylicz, C. Johnson, and C. Georgopoulos. 1991. The Escherichia coli DnaK chaperone, the 70-kDa heat shock protein eukaryotic equivalent, changes conformation upon ATP hydrolysis, thus triggering its dissociation from a bound target protein. J. Biol. Chem. 266:14491–14496.
130a. Liberek, K., D. Wall, and C. Georgopoulos. 1995. The DnaJ chaperone catalytically activates the DnaK chaperone to preferentially bind the σ32 heat shock transcriptional regulator. Proc. Natl. Acad. Sci. USA 92:6224–6228.
131. Lipinska, B., J. King, D. Ang, and C. Georgopoulos. 1988. Sequence analysis and transcriptional regulation of the Escherichia coli grpE gene, encoding a heat shock protein. Nucleic Acids Res. 16:7545–7562.
132. Lipinska, B., S. Sharma, and C. Georgopoulos. 1988. Sequence analysis and regulation of the htrA gene of Escherichia coli: a sigma 32-independent mechanism of heat-inducible transcription. Nucleic Acids Res. 16:10053–10067.
133. Lonetto, M., K. L. Brown, K. E. Rudd, and M. J. Buttner. 1994. Analysis of the Streptomyces coelicolor sigmaE gene reveals the existence of a subfamily of eubacterial RNA polymerase sigma factors involved in the regulation of extracytoplasmic functions. Proc. Natl. Acad. Sci. USA 91:7573–7577.
134. Lutkenhaus, J. 1983. Coupling of DNA replication and cell division: sulB is an allele of ftsZ. J. Bacteriol. 154:1339–1346.
135. Lutkenhaus, J., B. Sanjanwala, and M. Lowe. 1983. Overproduction of FtsZ suppresses sensitivity of lon mutants to division inhibition. J. Bacteriol. 166:733–738.
136. Madden, D. R., J. C. Gorga, J. L. Strominger, and D. C. Wiley. 1992. The three-dimensional structure of HLA-B27 at 2.1Å resolution suggests a general mechanism for tight peptide binding to MHC. Cell 70:1035–1048.
137. Martin, D. W., M. J. Schurr, M. H. Mudd, J. R. Govan, B. W. Holloway, and V. Deretic. 1993. Mechanism of conversion to mucoidy in Pseudomonas aeruginosa infecting cystic fibrosis patients. Proc. Natl. Acad. Sci. USA 90:8377–8381.
138. Maurizi, M. R. 1987. Degradation in vitro of bacteriophage lambda N protein by Lon protease from Escherichia coli. J. Biol. Chem. 262:2696–2703.
139. Maurizi, M. R. 1991. ATP-promoted interaction between ClpA and ClpP in activation of Clp protease from Escherichia coli. Biochem. Soc. Trans. 19:719–723.
140. Maurizi, M. R., W. P. Clark, Y. Katayama, S. Rudikoff, J. Pumphrey, B. Bowers, and S. Gottesman. 1990. Sequence and structure of ClpP, the proteolytic component of the ATP-dependent Clp protease of Escherichia coli. J. Biol. Chem. 265:12536–12545.
141. Maurizi, M. R., W. P. Clark, S. H. Kim, and S. Gottesman. 1985. Insertional mutagenesis of the lon gene in Escherichia coli: lon is dispensable. J. Bacteriol. 164:1124–1135.
142. Maurizi, M. R., W. P. Clark, S.-H. Kim, and S. Gottesman. 1990. ClpP represents a unique family of serine proteases. J. Biol. Chem. 265:12546–12552.
143. McCarty, J. S., and G. C. Walker. 1991. DnaK as a thermometer: threonine-199 is site of autophosphorylation and is critical for ATPase activity. Proc. Natl. Acad. Sci. USA 88:9513–9517.
144. Mecsas, J., P. E. Rouviere, J. W. Erickson, T. J. Donohue, and C. A. Gross. 1993. The activity of σE, an Escherichia coli heat-inducible sigma factor, is modulated by expression of outer membrane proteins. Genes Dev. 7:2618–2628.
145. Menon, A. S., L. Waxman, and A. L. Goldberg. 1987. The energy utilized in protein breakdown by the ATP-dependent protease La from Escherichia coli. J. Biol. Chem. 262:722–726.
146. Milarski, K. L., and R. J. Morimoto. 1989. Mutational analysis of the human HSP70 protein: distinct domains for nucleolar localization and adenosine triphosphate binding. J. Cell. Biol. 109:1947–1962.
147. Missiakas, D., C. Georgopoulos, and S. Raina. 1993. The Escherichia coli heat shock gene htpY: mutational analysis, cloning, sequencing, and transcriptional regulation. J. Bacteriol. 175:2613–2624.
148. Mizusawa, S., and S. Gottesman. 1983. Protein degradation in Escherichia coli: the lon gene controls the stability of the SulA protein. Proc. Natl. Acad. Sci. USA 80:358–362.
149. Mori, K., M. Wenzhen, M. J. Gething, and J. Sambrook. 1993. A transmembrane protein with a cdc2+/CDC28-related kinase activity is required for signaling from the ER to the nucleus. Cell 74:743–756.
150. Nagai, H., H. Yuzawa, M. Kanemori, and T. Yura. 1994. A distinct segment of the sigma 32 polypeptide is involved in DnaK-mediated negative control of the heat shock response in Escherichia coli. Proc. Natl. Acad. Sci. USA 91:10280–10284.
151. Nagai, H., H. Yuzawa, and T. Yura. 1991. Interplay of two cis-acting mRNA regions in translational control of σ32 synthesis during the heat shock response of Escherichia coli. Proc. Natl. Acad. Sci. USA 88:10515–10519.
152. Narberhaus, F., and H. Bahl. 1992. Cloning, sequencing, and molecular analysis of the groESL operon of Clostridium acetobutylicum. J. Bacteriol. 174:3282–3289.
153. Narberhaus, F., K. Giebeler, and H. Bahl. 1992. Molecular characterization of the dnaK gene region of Clostridum acetobutylicum, including grpE, dnaJ, and a new heat shock gene. J. Bacteriol. 174:3290–3299.
154. Nashimoto, H. 1993. The Translational Apparatus. Plenum Press, New York.
155. Neidhardt, F. C., T. A. Phillips, R. A. VanBogelen, and M. W. Smith. 1981. Identity of the B56.5 protein, the A protein, and the groE gene product of Escherichia coli. J. Bacteriol. 145:513–520.
156. Neidhardt, F. C., and R. A. VanBogelen. 1981. Positive regulatory gene for temperature-controlled proteins in Escherichia coli. Biochem. Biophys. Res. Commun. 100:894–900.
157. Neidhardt, F. C., and R. A. VanBogelen. 1987. Heat shock response, p. 1334–1345. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. American Society for Microbiology, Washington, D.C.
158. Neidhardt, F. C., R. A. VanBogelen, and V. A. Vaughn. 1984. The genetics and regulation of heat-shock proteins. Annu. Rev. Genet. 18:295–329.
159. Nelson, R. J., T. Ziegelhoffer, C. Nicolet, M. Werner-Washburne, and E. A. Craig. 1992. The translation machinery and 70 kd heat shock protein cooperate in protein synthesis. Cell 71:97–105.
160. Newlands, J. T., T. Gaal, J. Mecsas, and R. L. Gourse. 1993. Transcription of the Escherichia coli rrnB P1 promoter by the heat shock RNA polymerase (Eσ32) in vitro. J. Bacteriol. 175:661–668.
161. Osipiuk, J., C. Georgopoulos, and M. Zylicz. 1993. Initiation of λ DNA replication: the Escherichia coli small heat shock proteins, DnaJ and GrpE, increase DnaK’s affinity for the λP protein. J. Biol. Chem. 268:4821–4827.
162. Pack, K. H., and G. C. Walker. 1986. Defect in expression of heat-shock proteins at high temperature in xthA mutants. J. Bacteriol. 165:763–770.
163. Palleros, D. R., K. L. Reid, L. Shi, and A. L. Fink. 1993. DnaK ATPase activity revisited. FEBS Lett. 336:124–128.
164. Palleros, D. R., K. L. Reid, L. Shi, W. J. Welch, and A. L. Fink. 1993. ATP-induced protein-hsp70 complex dissociation requires K+ but not ATP hydrolysis. Nature (London) 365:664–666.
165. Palleros, D. R., W. J. Welch, and A. L. Fink. 1991. Interaction of hsp70 with unfolded proteins: effects of temperature and nucleotides on the kinetics of binding. Proc. Natl. Acad. Sci. USA 88:5719–5723.
166. Parsell, D. A., A. S. Kowal, and S. Lindquist. 1994. Saccharomyces cerevisiae Hsp104 protein. Purification and characterization of ATP-induced structural changes. J. Biol. Chem. 269:4480–4487.
167. Petit, M. A., W. Bedale, J. Osipiuk, C. Lu, M. Rajagopalan, P. McInerny, M. F. Goodman, and H. Echols. 1994. Sequential folding of UmuC by the Hsp70 and Hsp60 chaperone complexes of Escherichia coli. J. Biol. Chem. 269:23824–23829.
168. Phillips, G. J., and T. J. Silhavy. 1992. The E. coli ffh gene is necessary for viability and efficient protein export. Nature (London) 359:744–746.
169. Raina, S., and C. Georgopoulos. 1990. A new Escherichia coli heat shock gene, htrC, whose product is essential for viabiliy only at high temperatures. J. Bacteriol. 172:3417–3426.
170. Raina, S., and C. Georgopoulos. 1991. The htrM gene, whose product is essential for Escherichia coli viability only at elevated temperatures, is identical to the rfaD gene. Nucleic Acids Res. 19:3811–3819.
171. Raina, S., D. Missiakas, and C. Georgopoulos. 1995. The rpoE gene encoding the σE (σ24) heat shock sigma factor of Escherichia coli. EMBO J. 14:1043–1055.
172. Rippmann, F., W. R. Taylor, J. B. Rothbard, and N. M. Green. 1991. A hypothetical model for the peptide binding domain of hsp70 based on the peptide binding domain of HLA. EMBO J. 10:1053–1059.
173. Ritossa, F. M. 1962. A new puffing pattern induced by a temperature shock and DNP in Drosophila. Experientia 18:571–573.
174. Ritossa, F. M. 1963. New puffs induced by temperature shock. DNP and salicylate in salivary chromosomes of Drosophila melanogaster. Drosophila Inf. Service 37:122–123.
175. Ritossa, F. M. 1964. Specific loci in polytene chromosomes of Drosophila. Exp. Cell. Res. 35:601–607.
176. Rouviere, P. E., A. de las Peñas, J. Mecsas, C. Z. Lu, K. E. Rudd, and C. A. Gross. 1995. rpoE, the gene encoding the second heat-shock sigma factor, σE, in Escherichia coli. EMBO J. 14:1032–1042.
177. Schmid, D., A. Baici, H. Gehring, and P. Christen. 1994. Kinetics of molecular chaperone action. Science 263:971–973.
178. Schmidt, A., M. Schiesswohl, U. Völker, M. Hecker, and W. Schumann. 1992. Cloning, sequencing, mapping, and transcriptional analysis of the groESL operon from Bacillus subtilis. J. Bacteriol. 174:3993–3999.
179. Schoemaker, J. M., R. C. Gayda, and A. Markovitz. 1984. Regulation of cell division in Escherichia coli: SOS induction and cellular location of the SulA protein, a key to lon-associated filamentation and death. J. Bacteriol. 158:551–561.
180. Schröder, H., T. Langer, F.-U. Hartl, and B. Bukau. 1993. DnaK, DnaJ, and GrpE form a cellular chaperone machinery capable of repairing heat-induced protein damage. EMBO J. 12:4137–4144.
181. Scidmore, M. A., H. H. Okamura, and M. D. Rose. 1993. Genetic interactions between kar2 and sec63, encoding eukaryotic homologues of DnaK and DnaJ in the endoplasmic reticulum. Mol. Biol. Cell 4:1145–1159.
182. Seaton, B. L., and L. E. Vickery. 1994. A gene encoding a DnaK/hsp70 homolog in Escherichia coli. Proc. Natl. Acad. Sci. USA 91:2066–2070.
183. Sherman, M. Y., and A. L. Goldberg. 1992. Involvement of the chaperonin dnaK in the rapid degradation of a mutant protein in Escherichia coli. EMBO J. 11:71–77.
184. Shi, W., Y. Zhou, J. Wild, J. Adler, and C. A. Gross. 1992. DnaK, DnaJ, and GrpE are required for flagellum synthesis in Escherichia coli. J. Bacteriol. 174:6256–6263.
185. Skelly, S., T. Coleman, C. F. Fu, N. Brot, and H. Weissbach. 1987. Correlation between the 32-kDa σ factor and in vitro expression of Escherichia coli heat shock genes. Proc. Natl. Acad. Sci. USA 84:8365–8369.
186. Spence, J., A. Cegielska, and C. Georgopoulos. 1990. Role of Escherichia coli heat shock proteins DnaK and HtpG (C.62.5) in response to nutritional deprivation. J. Bacteriol. 172:7157–7166.
187. Squires, C., and C. L. Squires. 1992. The Clp proteins: proteolysis regulators or molecular chaperones? J. Bacteriol. 174:1081–1085.
188. Squires, C. L., S. Pedersen, B. M. Ross, and C. Squires. 1991. ClpB is the Escherichia coli heat shock protein F84.1. J. Bacteriol. 173:4254–4262.
189. Sternberg, N. 1973. Properties of a mutant of Escherichia coli defective in bacteriophage λ formation (groE). I. Initial characterization. J. Mol. Biol. 67:1–24.
190. Stout, V., A. Torres-Cabass, M. R. Maurizi, D. Gutnick, and S. Gottesman. 1991. RcsA, an unstable positive regulator of capsular polysaccharide synthesis. J. Bacteriol. 173:1738–1747.
191. Strauch, K. L., K. Johnson, and J. Beckwith. 1989. Characterization of degP, a gene required for proteolysis in the cell envelope and essential for growth of Escherichia coli at high temperatures. J. Bacteriol. 171:2689–2696.
192. Straus, D., W. Walter, and C. A. Gross. 1990. DnaK, DnaJ, and GrpE heat shock proteins negatively regulate heat shock gene expression by controlling the synthesis of σ32. Genes Dev. 4:2202–2209.
193. Straus, D. B., W. A. Walter, and C. A. Gross. 1987. The heat shock response of E. coli is regulated by changes in the concentration of σ32. Nature (London) 329:348–351.
194. Straus, D. B., W. A. Walter, and C. A. Gross. 1988. Escherichia coli heat shock gene mutants are defective in proteolysis. Genes Dev. 2:1851–1858.
195. Straus, D. B., W. A. Walter, and C. A. Gross. 1989. The activity of σ32 is reduced under conditions of excess heat shock protein production in Escherichia coli. Genes Dev. 3:2003–2010.
196. Taura, T., N. Kusukawa, T. Yura, and K. Ito. 1989. Transient shut-off of Escherichia coli heat shock protein synthesis upon temperature shift down. Biochem. Biophys. Res. Commun. 163:438–443.
197. Taylor, W. E., D. B. Straus, A. D. Grossman, Z. F. Burton, C. A. Gross, and R. R. Burgess. 1984. Transcription from a heat-inducible promoter causes heat shock regulation of the sigma subunit of E. coli RNA polymerase. Cell 38:371–381.
198. Thorsness, P. E., K. H. White, and T. D. Fox. 1993. Inactivation of YME1, a member of the ftsH-PAS1-CDC48 family of putative ATPase-encoding genes, causes increased escape of DNA from mitochondria in Saccharomyces cerevisiae. Mol. Cell. Biol. 13:5418–5426.
199. Tilly, K., J. Spence, and C. Georgopoulos. 1989. Modulation of stability of the Escherichia coli heat shock regulatory factor sigma. J. Bacteriol. 171:1585–1589.
200. Tilly, K., R. A. VanBogelen, C. Georgopoulos, and F. C. Neidhardt. 1983. Identification of the heat-inducible protein C15.4 as the groES gene product in Escherichia coli. J. Bacteriol. 154:1505–1507.
201. Tissieres, A., H. K. Mitchell, and V. M. Tracy. 1974. Protein synthesis in salivary glands of Drosophila melanogaster: relation to chromosome puffs. J. Mol. Biol. 84:389–398.
202. Tobias, J. W., T. E. Shrader, G. Rocap, and A. Varshavsky. 1991. The N-end rule in bacteria. Science 254:1374–1377.
203. Tomoyasu, T., J. Gamer, B. Bukau, M. Kanemori, H. Mori, A. J. Rutman, A. B. Oppenheim, T. Yura, K. Yamanaka, H. Niki, S. Hiraga, and T. Ogura. 1995. Escherichia coli Fts is a membrane-bound, ATP-dependent protease which degrades the heat-shock transcription factor σ32. EMBO J. 14:2551–2560.
204. Tomoyasu, T., K. Yamanaka, K. Murata, T. Suzaki, P. Bouloc, A. Kato, H. Niki, S. Hiraga, and T. Ogura. 1993. Topology and subcellular localization of FtsH protein in Escherichia coli. J. Bacteriol. 175:1352–1357.
205. Tomoyasu, T., T. Yuki, S. Morimura, H. Mori, K. Yamanaka, H. Niki, S. Hiraga, and T. Ogura. 1993. The Escherichia coli FtsH protein is a prokaryotic member of a protein family of putative ATPases involved in membrane functions, cell cycle control, and gene expression. J. Bacteriol. 175:1344–1351.
206. Torres-Cabassa, A. S., and S. Gottesman. 1987. Capsule synthesis in Escherichia coli K-12 is regulated by proteolysis. J. Bacteriol. 169:981–989.
207. Tsuchido, T., R. A. VanBogelen, and F. C. Neidhardt. 1986. Heat shock response in Escherichia coli influences cell division. Proc. Natl. Acad. Sci. USA 83:6959–6963.
208. Ueguchi, C., and K. Ito. 1992. Multicopy suppression: an approach to understanding intracellular functioning of the protein export system. J. Bacteriol. 174:1454–1461.
209. Ueguchi, C., M. Kakeda, H. Yamada, and T. Mizuno. 1994. An analogue of the DnaJ molecular chaperone in Escherichia coli. Proc. Natl. Acad. Sci. USA 91:1054–1058.
210. van der Vies, S. M., A. A. Gatenby, and C. Georgopoulos. 1994. Bacteriophage T4 encodes a co-chaperonin that can substitute for Escherichia coli GroES in protein folding. Nature (London) 368:654–656.
211. Van Dyk, T. K., A. A. Gatenby, and R. A. LaRossa. 1989. Demonstration by genetic suppression of interaction of GroE products with many proteins. Nature (London) 342:451–453.
212. VanBogelen, R. A., and F. C. Neidhardt. 1990. Ribosomes as sensors of heat and cold shock in Escherichia coli. Proc. Natl. Acad. Sci. USA 87:5589–5593.
213. Voos, W., B. D. Gambill, S. Laloraya, D. Ang, E. A. Craig, and R. N. Pfanne. 1994. Mitochondrial GrpE is present in a complex with hsp70 and preproteins in transit across membranes. Mol. Cell. Biol. 14:6627–6634.
214. Wada, C., M. Imai, and T. Yura. 1987. Host control of plasmid replication: requirement for the σ factor σ32 in transcription of mini-F replication initiator. Proc. Natl. Acad. Sci. USA 84:8849–8853.
215. Wagner, I., H. Arlt, L. VanDyck, T. Langer, and W. Neupert. 1994. Molecular chaperones cooperate with PIM1 protease in the degradation of misfolded proteins in mitochondria. EMBO J. 13:5135–5145.
216. Wall, D., M. Zylicz, and C. Georgopoulos. 1994. The NH2-terminal 108 amino acids of the Escherichia coli DnaJ protein stimulate the ATPase activity of DnaK and are sufficient for λ replication. J. Biol. Chem. 269:5446–5451.
217. Wall, D., M. Zylicz, and C. Georgopoulos. The conserved G/F motif of the DnaJ chaperone is necessary for the activation of the substrate binding properties of the DnaK chaperone. J. Biol. Chem. 270:2139–2144.
218. Wang, Q. P., and J. M. Kaguni. 1989. A novel sigma factor is involved in expression of the rpoH gene of Escherichia coli. J. Bacteriol. 171:4248–4253.
219. Wawrzynow, A., D. Wojtkowiak, J. Marszalck, B. Banecki, M. Jonsen, B. Graves, C. Georgopoulos, and M. Zylicz. The ClpX heat shock protein of Escherichia coli, the ATP-dependent substrate specificity component of the ClpP-ClpX protease, is a novel molecular chaperone. EMBO J. 14:1867–1877.
220. Waxman, L., and A. L. Goldberg. 1982. Protease La from Escherichia coli hydrolyzes ATP and proteins in a linked fashion. Proc. Natl. Acad. Sci. USA 79:4883–4887.
221. Waxman, L., and A. L. Goldberg. 1986. Protease La, the lon gene product, cleaves specific fluorogenic peptides in an ATP-dependent reaction. J. Biol. Chem. 260:12022–12028.
222. Wetzstein, M., U. Völker, J. Dedio, S. Löbau, U. Zuber, M. Schiesswohl, C. Herget, M. Hecker, and W. Schumann. 1992. Cloning, sequencing, and molecular analysis of the dnaK locus from Bacillus subtilis. J. Bacteriol. 174:3300–3310.
223. Wickner, S., S. Gottesman, D. Skowyra, J. Hoskins, and K. McKenney. 1994. A molecular chaperone, ClpA, functions like DnaK and DnaJ. Proc. Natl. Acad. Sci. USA 91:12218–12222.
224. Wickner, S., J. Hoskins, and K. McKenney. 1991. Monomerization of RepA dimers by heat shock proteins activates binding to DNA replication origin. Proc. Natl. Acad. Sci. USA 88:7903–7907.
225. Wickner, S. H. 1990. Three Escherichia coli heat shock proteins are required for P1. Proc. Natl. Acad. Sci. USA 87:2690–2694.
226. Wild, J., E. Altman, T. Yura, and C. A. Gross. 1992. The DnaK and DnaJ heat shock proteins participate in protein export in Escherichia coli. Genes Dev. 6:1165–1172.
227. Wild, J., W. A. Walter, C. A. Gross, and E. Altman. 1993. Accumulation of secretory protein precursors in Escherichia coli induces the heat shock response. J. Bacteriol. 175:3992–3997.
228. Wojkowiak, D., C. Georgopoulos, and M. Zylicz. 1993. Isolation and characterization of ClpX, a new ATP-dependent specificity component of the Clp protease of Escherichia coli. J. Biol. Chem. 268:22609–22617.
229. Wu, B., D. Ang, M. Snavely, and C. Georgopoulos. 1994. Isolation and characterization of point mutations in the Escherichia coli grpE heat shock gene. J. Bacteriol. 176:6965–6973.
230. Yamamori, T., K. Ito, Y. Nakamura, and T. Yura. 1978. Transient regulation of protein synthesis in Escherichia coli upon shift-up of growth temperature. J. Bacteriol. 134:1133–1140.
231. Yamamori, T., and T. Yura. 1980. Temperature-induced synthesis of specific proteins in Escherichia coli: evidence for transcriptional control. J. Bacteriol. 142:843–851.
232. Escherichia coli K12. Proc. Natl. Acad. Sci. USA 79:860–864.
233. Yura, T., H. Mori, H. Nagai, T. Nagata, A. Ishihama, N. Fujita, K. Isono, K. Mizobuchi, and A. Nakata. 1992. Systematic sequencing of the Escherichia coli genome: analysis of the 0–2.4 min region. Nucleic Acids Res. 20:3305–3308.
234. Yura, T., H. Nagai, and H. Mori. 1993. Regulation of the heat-shock response in bacteria. Annu. Rev. Microbiol. 47:321–350.
235. Yura, T., T. Tobe, K. Ito, and T. Osawa. 1984. Heat shock regulatory gene (htpR) of Escherichia coli is required for growth at high temperature but is dispensable at low temperature. Proc. Natl. Acad. Sci. USA 81:6803–6807.
236. Yuzawa, H., H. Nagai, H. Mori, and T. Yura. 1993. Heat induction of sigma 32 synthesis mediated by mRNA secondary structure: a primary step of the heat shock response in Escherichia coli. Nucleic Acids Res. 21:5449–5455.
237. J. Biol. Chem. 268:7489–7493.
238. Zhou, Y. N., N. Kusukawa, J. W. Erickson, C. A. Gross, and T. Yura. 1988. Isolation and characterization of Escherichia coli mutants that lack the heat shock sigma factor σ32. J. Bacteriol. 170:3640–3649.
239. Ziemienowicz, A., D. Skowyra, J. Zeilstra-Ryalls, O. Fayet, C. Georgopoulos, and M. Zylicz. 1993. Both the Escherichia coli chaperone systems, GroEl/GrosES and DnaK/DnaJ/GrpE, can reactivate heat-treated RNA polymerase. J. Biol. Chem. 268:25425–25431.
240. Zuber, U., and W. Schumann. 1994. CIRCE, a novel heat shock element involved in regulation of heat shock operon dnaK of Bacillus subtilis. J. Bacteriol. 176:1359–1363.
241. Zylicz, M., D. Ang, and C. Georgopoulos. 1984. The grpE protein of Escherichia coli: purification and properties. J. Biol. Chem. 262:17437–17442.
242. Zylicz, M., J. H. LeBowitz, R. McMacken, and C. Georgopoulos. 1983. The dnaK protein of Escherichia coli posseses an ATPase and autophosphorylating activity and is essential in an in vitro DNA replication system. Proc. Natl. Acad. Sci. USA 80:6431–6435
Chapter88
