REF2 encodes an RNA-binding protein directly involved in yeast mRNA 3'-end formation.

The Saccharomyces cerevisiae mutant ref2-1 (REF = RNA end formation) was originally identified by a genetic strategy predicted to detect decreases in the use of a CYC1 poly(A) site interposed within the intron of an ACT1-HIS4 fusion reporter gene. Direct RNA analysis now proves this effect and also demonstrates the trans action of the REF2 gene product on cryptic poly(A) sites located within the coding region of a plasmid-borne ACT1-lacZ gene. Despite impaired growth of ref2 strains, possibly because of a general defect in the efficiency of mRNA 3'-end processing, the steady-state characteristics of a variety of normal cellular mRNAs remain unaffected. Sequencing of the complementing gene predicts the Ref2p product to be a novel, basic protein of 429 amino acids (M(r), 48,000) with a high-level lysine/serine content and some unusual features. Analysis in vitro, with a number of defined RNA substrates, confirms that efficient use of weak poly(A) sites requires Ref2p: endonucleolytic cleavage is carried out accurately but at significantly lower rates in extracts prepared from delta ref2 cells. The addition of purified, epitope-tagged Ref2p (Ref2pF) reestablishes wild-type levels of activity in these extracts, demonstrating direct involvement of this protein in the cleavage step of 3' mRNA processing. Together with the RNA-binding characteristics of Ref2pF in vitro, our results support an important contributing role for the REF2 locus in 3'-end processing. As the first gene genetically identified to participate in mRNA 3'-end maturation prior to the final polyadenylation step, REF2 provides an ideal starting point for identifying related genes in this event.

A quaternary transcription termination complex. Reciprocal stabilization by Rho factor and NusG protein.

The Escherichia coli protein NusG is known to modulate Rho-dependent transcription termination in vivo. We have shown that it can also alter the pattern of Rho-dependent RNA endpoints in vitro, at lower NusG concentrations than can be explained by reported interactions between NusG and Rho or RNA polymerase. Three observations in vitro now suggest a model to account for these effects of NusG on Rho-dependent termination. First, the presence of NusG circumvents the interference with Rho function caused by adding DNA oligonucleotides complementary to particular segments of the Rho binding site. Second, when NusG is added to stalled elongation complexes, the off-rate of Rho from nascent RNA is slowed. Third, NusG associates stably with the elongation complex only when Rho is also present and bound to the nascent RNA. Our observations are consistent with a model in which NusG and Rho participate in an interdependent association with the transcribing RNA polymerase and the nascent RNA to facilitate the recognition and use of termination signals. Common structural and functional features shared with complexes that carry out processive antitermination are discussed.

NusG alters rho-dependent termination of transcription in vitro independent of kinetic coupling.

To complement the recent discovery that rho-dependent termination in E. coli requires nusG protein in vivo, we have tested the effect of purified nusG protein on rho-dependent termination in vitro. With the well-characterized trp t' terminator of E. coli, and no other proteins than E. coli RNA polymerase and rho factor, nusG causes a proximal shift in the terminated RNA endpoints, compared to the endpoints generated by rho alone. The presence of nusG also enhances rho-mediated termination on partially defective mutant trp t' templates. We rule out explanations such as a change in the kinetic coupling between rho and RNA polymerase or a nusG-mediated increase in the affinity of rho for RNA. We also detect no difference in the helicase rate of rho in the presence of nusG. Even assays with completely stalled and isolated ternary complexes indicate that rho is able to effect the release of RNA with the assistance of nusG at points preceding the most proximal release sites observed in the absence of nusG. Our observations support a model in which nusG acts as a component of the transcription complex, possibly interacting with both rho and RNA polymerase as it governs accessibility to the nascent transcript.

Yeast tRNA ligase mutants are nonviable and accumulate tRNA splicing intermediates.

We show here that yeast tRNA ligase protein is essential in the cell and participates in joining together tRNA half-molecules resulting from excision of the intron by the splicing endonuclease. A haploid yeast strain carrying a chromosomal deletion of the ligase gene is viable only if ligase protein can be supplied from a plasmid copy of the gene. When synthesis of the plasmid-borne ligase gene is repressed, cells eventually die and accumulate endonuclease cut but unligated half-molecules and intervening sequences. Half-molecules that accumulate appear to be fully end-processed. Two temperature-sensitive ligase mutant strains have been isolated; these strains accumulate a similar set of unligated half-molecules at the nonpermissive temperature.