Variants of lambda repressor and cytochrome b562 translated from messenger RNAs without stop codons were modified by carboxyl terminal addition of an ssrA-encoded peptide tag and subsequently degraded by carboxyl terminal-specific proteases present in both the cytoplasm and periplasm of Escherichia coli. The tag appears to be added to the carboxyl terminus of the nascent polypeptide chain by cotranslational switching of the ribosome from the damaged messenger RNA to ssrA RNA.

No abstract

Interruption of translation in Escherichia coli can lead to the addition of an 11-residue carboxy-terminal peptide tail to the nascent chain. This modification is mediated by SsrA RNA (also called 10Sa RNA and tmRNA) and marks the tagged polypeptide for proteolysis. Degradation in vivo of repressor amino-terminal domain variants bearing this carboxy-terminal SsrA peptide tag is shown here to depend on the cytoplasmic proteases ClpXP and ClpAP. Degradation in vitro of SsrA-tagged substrates was reproduced with purified components and required a substrate with a wild-type SsrA tail, the presence of both ClpP and either ClpA or ClpX, and ATP. Clp-dependent proteolysis accounts for most degradation of SsrA-tagged amino-domain substrates at 32°C, but additional proteases contribute to the degradation of some of these SsrA-tagged substrates at 39°C. The existence of multiple cytoplasmic proteases that function in SsrA quality-control surveillance suggests that the SsrA tag is designed to serve as a relatively promiscuous signal for proteolysis. Having diverse degradation systems able to recognize this tag may increase degradation capacity, permit degradation of a wide variety of different tagged proteins, or allow SsrA-tagged proteins to be degraded under different growth conditions.

Several general principles emerge from the studies of Cro, lambda repressor, and CAP. The DNA-binding sites are recognized in a form similar to B-DNA. They do not form cruciforms or other novel DNA structures. There seem to be proteins that bind left-handed Z-DNA (87) and DNA in other conformations, but it remains to be seen how these structures are recognized or how proteins recognize specific sequences in single-stranded DNA. Cro, repressor, and CAP use symmetrically related subunits to interact with two-fold related sites in the operator sequences. Many other DNA-binding proteins are dimers or tetramers and their operator sequences have approximate two-fold symmetry. It seems likely that these proteins will, like Cro, repressor, and CAP, form symmetric complexes. However, there is no requirement for symmetry in protein-DNA interactions. Some sequence-specific DNA-binding proteins, like RNA polymerase, do not have symmetrically related subunits and do not bind to symmetric recognition sequences. Cro, repressor, and CAP use alpha-helices for many of the contacts between side chains and bases in the major groove. An adjacent alpha-helical region contacts the DNA backbone and may help to orient the "recognition" helices. This use of alpha-helical regions for DNA binding appears to be a common mode of recognition. Most of the contacts made by Cro, repressor, and CAP occur on one side of the double helix. However, lambda repressor contacts both sides of the double helix by using a flexible region of protein to wrap around the DNA. Recognition of specific base sequences involves hydrogen bonds and van der Waals interactions between side chains and the edges of base pairs. These specific interactions, together with backbone interactions and electrostatic interactions, stabilize the protein-DNA complexes. The current models for the complexes of Cro, repressor, and CAP with operator DNA are probably fundamentally correct, but it should be emphasized that model building alone, even when coupled with genetic and biochemical studies, cannot be expected to provide a completely reliable "high-resolution" view of the protein-DNA complex. For example, the use of standard B-DNA geometry for the operator is clearly an approximation.(ABSTRACT TRUNCATED AT 400 WORDS)

A genetic system was developed in Escherichia coli to study leucine zippers with the amino-terminal domain of bacteriophage lambda repressor as a reporter for dimerization. This system was used to analyze the importance of the amino acid side chains at eight positions that form the hydrophobic interface of the leucine zipper dimer from the yeast transcriptional activator, GCN4. When single amino acid substitutions were analyzed, most functional variants contained hydrophobic residues at the dimer interface, while most nonfunctional sequence variants contained strongly polar or helix-breaking residues. In multiple randomization experiments, however, many combinations of hydrophobic residues were found to be nonfunctional, and leucines in the heptad repeat were shown to have a special function in leucine zipper dimerization.