Thomas A Steitz — Page 3

PubMed2002

A pre-translocational intermediate in protein synthesis observed in crystals of enzymatically active 50S subunits.

The large ribosomal subunit catalyzes peptide bond formation during protein synthesis. Its peptidyl transferase activity has often been studied using a 'fragment assay' that depends on high concentrations of methanol or ethanol. Here we describe a version of this assay that does not require alcohol and use it to show, both crystallographically and biochemically, that crystals of the large ribosomal subunits from Haloarcula marismortui are enzymatically active. Addition of these crystals to solutions containing substrates results in formation of products, which ceases when crystals are removed. When substrates are diffused into large subunit crystals, the subsequent structure shows that products have formed. The CC-puromycin-peptide product is found bound to the A-site and the deacylated CCA is bound to the P-site, with its 3prime prime or minute OH near N3 A2486 (Escherichia coli A2451). Thus, this structure represents a state that occurs after peptide bond formation but before the hybrid state of protein synthesis.

Alcohols↗

PubMed2002

Rational design of 'water-soluble' bacteriorhodopsin variants.

We have explored the interchangeability of soluble and membrane proteins by attempting to render a helical membrane protein 'water soluble' through mutation of its lipid-exposed residues. Using an atomic resolution structure of bacteriorhodopsin (bR), two different strategies were developed to identify lipid-exposed residues for mutation. In the first strategy all residues in trimeric bR with solvent accessibility >35% were marked for replacement. Replacement residues were chosen so as to map an average surface of helical soluble proteins onto the bR surface, resulting in the mutagenesis of 14.9% of surface residues. The second strategy took into account the observation that accessible residues can be categorized as fully or partially accessible. Consequently, three mutants were designed based on monomeric bR, all with their accessible residues changed and with varying extents of mutagenesis of partially accessible residues. 13.5-24.3% of the wild-type surface was altered in these designs. The construct for the first design was cloned into Escherichia coli. Trace amounts of the mutant protein were expressed with the concurrent overexpression of an endogenous prolyl isomerase. In contrast, all three mutant proteins of the second design expressed well and could be purified to homogeneity. Systematic refolding trials were undertaken with limited success at solubilization in aqueous media. We have discussed the feasibility of applying the 'solubilization strategy' outlined here to membrane proteins.

Amino Acid Sequence↗