Physical properties of the Escherichia coli transcription termination factor rho. 2. Quaternary structure of the rho hexamer

Abstract

Under approximately physiological conditions, the transcription termination factor rho from Escherichia coli is a hexamer of planar hexagonal geometry [Geiselmann, J., Yager, T. D., Gill, S. C., Calmettes, P., & von Hippel, P. H. (1992) Biochemistry (preceding paper in this issue)]. Here we describe studies that further define the quaternary structure of this hexamer. We use a combination of chemical cross-linking and treatment with mild denaturants to show that the fundamental unit within the rho hexamer is a dimer stabilized by an isologous (or pseudoisologous) bonding interface. Three identical dimers of rho interact via a second type of isologous bonding interface to yield a hexamer with C3 or D3 symmetry. Cross-linking and denaturation experiments definitely rule out C6 and C2 symmetry for the rho hexamer. Data from fluorescence quenching, lifetime, and energy transfer experiments also argue against C2 symmetry. The simplest symmetry assignment that is not contradicted by any experimental data is D3; thus we conclude that the rho hexamer has D3 symmetry. We also consider the positioning of the binding sites for RNA and ATP relative to the coordinate reference frame of the D3 hexamer. Fluorescence energy transfer data are presented and integrated with data from the literature to arrive at a self-consistent model for the quaternary structure of the rho hexamer.

Functional interactions of ligand cofactors with Escherichia coli transcription termination factor rho. I. Binding of ATP

“…This result is confirmed by several independent determinations of the ATP binding stoichiometry, using different techniques. We have previously shown (Geiselmann et al, 1992b) that the rho hexamer is the functionally important association state of rho and that the hexamer is composed of a trimer of dimers characterized by D3 symmetry Geiselmann et al, 1992a). The existence of two classes of ATP binding sites corroborates this trimer of dimers organization and also demonstrates that each dimer does not possess an exact twofold symmetry axis (as required for a structure characterized by D3 symmetry).…”

“…We know that ATP is not carried along nonspecifically by the detergent because ATP does not comigrate with any of the marker proteins. In addition, no ATP binding was observed when the gel was run in a phosphate (Geiselmann et al, 1992a) or Tris-borate-EDTA buffer (Hammes & Rickwood, 1981). (Phosphate is a competitive inhibitor of ATP binding and EDTA would complex the Mg2+ that is required for ATP binding [Stitt, 19881.)…”

Functional interactions of ligand cofactors with Escherichia coli transcription termination factor rho. II. Binding of RNA

“…The other six (cryptic) RNA binding sites are either occluded because they are located at the hexamer-hexamer interface or because their binding affinities are too weak to be detected under our experimental conditions. Both possibilities, as well as the fact that the six ATP binding sites of rho fall into two affinity classes (Geiselmann & von Hippel, 19921, support the hypothesis that, in the hexameric form of rho, the functional unit is the rho dimer (Geiselmann et al, 1992a;Seifried et al, in prep. ).…”

“…In a hexamer displaying D3 symmetry (Geiselmann et al, 1992a) three subunits have a particular surface pointing up. These alternate with three other subunits that have the same surface pointing down (see Fig.…”

“…Therefore this sedimentation boundary most likely reflects the properties of hexamers carrying a single bound (rC)70 ligand. Hexamers themselves tend t o associate somewhat with one other (see Geiselmann et al, 1992a), and this tendency appears to persist or be enhanced when (rC)70 is bound. This could explain the solution heterogeneity that is observed upon analysis of the sedimentation boundary by the method of van Holde and Weischet (Fig.…”