The spontaneous refolding of chemically denatured dihydrofolate reductase (DHFR) is completely arrested by chaperonin 60 (GroEL). This inhibition presumably results from the formation of a stable complex between chaperonin 60 and one or more intermediates in the folding pathway. While sequestered on chaperonin 60, DHFR is considerably more sensitive to proteolysis, suggesting a nonnative structure. Bound DHFR can be released from chaperonin 60 with ATP, and although chaperonin 10 (GroES) is not obligatory, it does potentiate the maximum effect of ATP. Hydrolysis of ATP is also not required for DHFR release since certain nonhydrolyzable analogues are capable of partial discharge. "Native" DHFR can also form a stable complex with chaperonin 60. However, in this case, complex formation is not instantaneous and can be prevented by the presence of DHFR substrates. This suggests that native DHFR exists in equilibrium with at least one conformer which is recognizable by chaperonin 60. Binding studies with S-35-labeled DHFR support these conclusions and further demonstrate that DHFR competes for a common saturable site with another protein (ribulose-1,5-bisphosphate carboxylase) known to interact with chaperonin 60.

Mitochondria contain a polypeptide that is functionally equivalent to Escherichia coli chaperonin 10 (cpn10; also known as groES). This mitochondrial cpn10 has been identified in beef and rat liver and is able to replace bacterial cpn10 in the chaperonin-dependent reconstitution of chemically denatured ribulose-1,5-bisphosphate carboxylase. Thus, like the bacterial homologue, mitochondrial cpn10 facilitates a K(+)- and Mg.ATP-dependent discharge of unfolded (or partially folded) ribulose bisphosphate carboxylase from bacterial chaperonin 60 (cpn60; also known as groEL). Instrumental to its identification, mitochondrial cpn10 and bacterial cpn60 form a stable complex in the presence of Mg.ATP. Bacterial and mitochondrial cpn10 compete for a common saturable site on bacterial cpn60. As a result of complex formation, with either mitochondrial or bacterial cpn10, the "uncoupled ATPase" activity of bacterial cpn60 is virtually abolished. The most likely candidate for mitochondrial cpn10 is an approximately 45-kDa oligomer composed of approximately 9-kDa subunits. We propose that, like the protein-folding machinery of prokaryotes, mitochondrial cpn60 requires a cochaperonin for full biological function.

Both the chaperonin- and MgATP-dependent reconstitution of unfolded ribulosebisphosphate carboxylase (Rubisco) and the uncoupled ATPase activity of chaperonin 60 (groEL) require ionic potassium. The spontaneous, chaperonin-independent reconstitution of Rubisco, observed at 15 but not at 25 degrees C, requires no K+ and is actually inhibited by chaperonin 60, with which the unfolded or partly folded Rubisco forms a stable binary complex. The chaperonin-dependent reconstitution of Rubisco involves the formation of a complex between chaperonin 60 and chaperonin 10 (groES). Formation of this complex almost completely inhibits the uncoupled ATPase activity of chaperonin 60. Furthermore, although the formation of the chaperonin 60-chaperonin 10 complex requires the presence of MgATP, hydrolysis of ATP may not be required, since complex formation occurs in the absence of K+. The interaction of chaperonin 60 with unfolded or partly folded Rubisco does not require MgATP, K+, or chaperonin 10. However, discharge of the complex of chaperonin 60-Rubisco, which leads to the formation of active Rubisco dimers, requires chaperonin 10 and a coupled, K(+)-dependent hydrolysis of ATP. We propose that a role of chaperonin 10 is to couple the K(+)-dependent hydrolysis of ATP to the release of the folded monomers of the target protein from chaperonin 60.

Nine different proteins were imported into isolated pea chloroplasts in vitro. For seven of these [the large and small subunits of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco), beta-subunit of ATP synthase, glutamine synthetase, the light-harvesting chlorophyll a/b binding protein, chloramphenicol acetyltransferase, and pre-beta-lactamase], a fraction was found to migrate as a stable high-molecular-weight complex during nondenaturing gel electrophoresis. This complex contained the mature forms of the imported proteins and the groEL-related chloroplast chaperonin 60 (previously known as Rubisco subunit binding protein). Thus, the stable association of imported proteins with this molecular chaperone is widespread and not necessarily restricted to Rubisco subunits or to chloroplast proteins. With two of the imported proteins (ferredoxin and superoxide dismutase), such complexes were not observed. It seems likely that, in addition to its proposed role in assembly of Rubisco, the chloroplast chaperonin 60 is involved in the assembly or folding of a wide range of proteins in chloroplasts.

The import of cytoplasmically synthesized proteins into chloroplasts involves an interaction between at least two components; the precursor protein, and the import apparatus in the chloroplast envelope membrane. This review summarizes the information available about each of these components. Precursor proteins consist of an amino terminal transit peptide attached to a passenger protein. Transit peptides from various precurosrs are diverse with respect to length and amino acid sequence; analysis of their sequences has not revealed insight into their mode of action. A variety of foreign passenger proteins can be imported into chloroplasts when a transit peptide is present at the amino terminus. However, foreign passenger proteins are not imported as efficiently as natural passenger proteins, and some chimeric precursor proteins are not imported into chloroplasts at all. Therefore, the passenger protein, as well as the transit peptide, influences the import process. Import begins by binding of the precursor to the chloroplast surface. It has been suggested that this binding is mediated by a receptor, but evidence to support this hypothesis remains incomplete and a receptor protein has not yet been characterized. Protein translocation requires energy derived from ATP hydrolysis, although there are conflicting reports as to where hydrolysis occurs and it is unclear how this energy is utilized. The mechanism(s) whereby proteins are translocated across either the two envelope membranes or the thylakoid membrane is not known.

The large subunit of ribulose bisphosphate carboxylase from Anacystis nidulans 6301, and the β subunit of chloroplast ATP synthase from maize, were fused to the transit peptide of the small subunit of ribulose bisphosphate carboxylase from soybean. These proteins were assayed for post‐translational import into isolated pea chloroplasts. Both proteins were imported into chloroplasts. Imported large subunits were associated with two distinct macromolecular structures. The smaller of these structures was a hybrid ribulose bisphosphate carboxylase holoenzyme, and the larger was the binding protein oligomer. Time‐course experiments following import of the large subunit revealed that the amount of large subunit associated with the binding protein oligomer decreased over time, and that the amount of large subunit present in the assembled holoenzyme increased. We also observed that imported small subunits of ribulose bisphosphate carboxylase, although predominantly present in the holoenzyme, were also found associated with the binding protein oligomer. In contrast, the imported β subunit of chloroplast ATP synthase did not assemble into a thylakoid‐bound coupling factor complex.

Protein targeting in eukaryotic cells is determined by several topogenic signals. Among these are stop-transfer regions, which halt translocation of proteins across the endoplasmic reticulum membrane. Two different stop-transfer regions were incorporated into precursors for a chloroplast protein, the small subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase. Both chimeric proteins were imported into chloroplasts and did not accumulate in the envelope membranes. Thus, the stop-transfer signals did not function during chloroplast protein import. These observations support the hypothesis that the mechanism for translocation of proteins across the chloroplast envelope is significantly different from that for translocation across the endoplasmic reticulum membrane.

In order to further our understanding of the targeting of nuclear-encoded proteins into intracellular organelles, we have investigated the import of chimeric precursor proteins into pea chloroplasts. Two different chimeric precursor proteins were produced by in vitro expression of chimeric genes. One chimeric precursor contained the transit peptide of the small subunit of soybean ribulose 1,5-bisphosphate carboxylase and the mature peptide of the same protein from pea. The second contained the same transit peptide plus 13 amino acids of the pea mature peptide fused to a cytosolic heat shock protein. The extent of import and binding of the two chimeric proteins was examined by using quantitative assays and was compared to the import of pea small subunit precursor. Both precursor proteins imported well into pea chloroplasts, although the extent of import observed with the chimeric small-subunit-heat shock precursor was less than that observed with the soybean-pea small subunit precursor. The heat shock protein alone did not import into nor bind to chloroplasts. The binding of both the chimeric small-subunit-heat shock protein and the soybean-pea small subunit precursor to chloroplasts was physiologically significant, as shown by the fact that when chloroplasts with bound precursors were isolated, these bound precursors could subsequently be imported.

Precursor forms of chloroplast proteins synthesized in cell-free translation systems can be imported posttranslationally into isolated, intact chloroplasts. Radiochemically pure precursors to the small subunit of ribulose-1,5-bisphosphate carboxylase and to the light-harvesting chlorophyll a/b protein have been prepared by in vitro translation of hybrid-selected mRNA and used to study this import process. If chloroplasts are pretreated with the uncoupler nigericin, import does not occur, but the precursors bind to the chloroplast surface. Reincubation of the precursor-chloroplast complex in the presence of ATP results in import of bound precursors. The binding appears to be mediated by proteins of the outer chloroplast envelope membrane because pretreatment of chloroplasts with protease inhibits their ability to bind as well as to import precursors. These results indicate that at least a portion of the observed binding is to functional receptor proteins involved in the import process.

A cyclic nucleotide-independent protein kinase, protease-activated kinase II, which incorporates up to four phosphates into 40 S ribosomal protein S6, has been purified from the postribosomal supernatant of rabbit reticulocytes. Protease-activated kinase II was purified as an inactive proenzyme by chromatography on DEAE-cellulose, phosphocellulose, Sephadex G-150, and hydroxylapatite. The enzyme was activated in vitro by limited digestion with trypsin or chymotrypsin. No other mode of activation for protease-activated kinase II in vitro was identified. The proenzyme had a molecular weight of 80,000 as measured by gel filtration; following tryptic digestion, the molecular weight of the activated protein kinase was 45,000-55,000. Protease-activated kinase II required Mg2+ for activity but was inhibited by other divalent cations, monovalent cations, and fluoride ion. ATP was the phosphoryl donor in the phosphorylation reaction; GTP had no effect. In vitro, multiple phosphorylation of S6 was observed with some phosphate incorporated into S10. Phosphorylation of S6 by protease-activated kinase II has been shown to be stimulated in serum-starved 3T3-L1 cells by insulin (Perisic, O., and Traugh, J. A. (1983) J. Biol. Chem. 258, 9589-9592) and in reticulocytes by altering the pH of the incubation medium (Perisic, O., and Traugh, J. A. (1983) J. Biol. Chem. 258, 13998-14002.

Casein kinase II, a cyclic nucleotide-independent protein kinase from rabbit reticulocytes, was shown to be inhibited by heparin. Heparin specifically inhibited the enzyme and had no effect on other protein kinases, including casein kinase I, the type I and II cAMP-dependent protein kinases, protease-activated kinase I, and the hemin-controlled repressor. Heparan sulfate was found to be 40-fold less effective than heparin towards casein kinase II; other acid mucopolysaccharides had little or no effect on the enzymatic activity. Steady state studies revealed that heparin acted as a competitive inhibitor with respect to the substrate, casein. A value of 20 ng/ml or about 1.4 nM was obtained for the apparent Ki. The inhibition was not reversed by ATP and varying the ATP and heparin concentrations in the assay only altered the maximum velocity.