Summary: | Metabolic instability, a hallmark property of all mRNAs, can impart a number of important
consequences on the regulation of gene expression. To further study the process of mRNA
degradation in Escherichia coli, we have developed an in vitro system to determine the requirements
for the complete decay of the rpsTmKNA, encoding ribosomal protein S20. The data show that both
purified RNase II and polynucleotide phosphorylase can catalyze the degradation of the 5'-twothirds
of the rpsT mRNA and that prior oligoadenylation of the 3'-termini of truncated rpsT
mRNA substrates can stimulate significantly the initiation of degradation by the 3'-exonucleases.
The intact rpsT mRNA, however, is insensitive to attack by RNase II or polynucleotide
phosphorylase. Furthermore, the single addition of a poly (A) tail to the 3'-end of the rpsT mRNA
cannot overcome its resistance to either 3'-exonuclease in vitro.
Although previous work has implicated the product of the pcnB gene in the decay of a
number of RNAs from Escherichia coli, poly(A) polymerase I does not promote the initiation of
the decay of the rpsT mRNA in vivo. It does, however, facilitate the degradation of highly folded
degradative intermediates by polynucleotide phosphorylase. As expected, purified degradosomes
generate an authentic 147-residue RNase E cleavage product from the rpsT mRNA in vitro.
However, degradosomes are unable to degrade the 147-residue fragment in the presence of ATP
even when it is oligoadenylated. Rather, both continuous cycles of polyadenylation and
polynucleotide phosphorylase activity are necessary and sufficient for the complete decay of the
147-residue fragment in a process which can be antagonized by the action of RNase II. Moreover,
both ATP and a non-hydrolyzable analog, ATPyS, support the poly (A) polymerase I and polynucleotide phosphorylase-dependent degradation of the 147-residue intermediate implying that
ATPase activity, such as that which may reside in RhlB, a putative RNA helicase, is not
necessarily required. Alternatively, the rpsT mRNA can be degraded in vitro by a second 3'-decay
pathway which is dependent on poly(A) polymerase I, polynucleotide phosphorylase and ATP
alone.
Complete degradation of a fragment of the malE-malF mRNA in vitro exhibits additional
requirements. Degradation of this RNA is dependent on the degradosome, ATP and poly(A)
polymerase I. Unlike the situation for the rpsT mRNA, the non-hydrolyzable analog, ATPyS,
cannot substitute for ATP suggesting that ATP hydrolysis is required for decay of structured
portions of the malE-malF mRNA. Our results demonstrate that a hierarchy of RNA secondary
structures controls access to exonucleolytic attack on 3'-termini. Moreover, decay of a model
mRNA can be reconstituted in vitro by a small number of purified components in a process which
is more dynamic and ATP-dependent than previously imagined. The implications of these findings
are discussed in a number of models of mRNA decay.
|