Specific inhibition of chromatin-associated poly(A) synthesis in vitro by cordycepin 5'-triphosphate.

It has been established that many eukaryotic mRNAs contain poly(adenylic acid) tracts at their 3'-termini. The polyadenylation of mRNA occurs post-transcriptionally in the nucleus as a rapid, initial addition of 100-200 adenylate residues to the pre-mRNA (ref. 1). Subsequently, a slower chain extension (6-8 bases) of the poly(A) tail seems to occur both in the nucleus and in the cytoplasm. The initial polyadenylation reaction can be specifically inhibited by the drug cordycepin (3'-deoxyadenosine) in cell culture, presumably by its conversion to the triphosphate analogue which acts as a competitive inhibitor of poly(A) polymerase. Cordycepin, however, has little effect on the slower poly(A) extension reaction or on the formation of mRNA precursor molecules; but it can inhibit rRNA synthesis. Contrary to the in vitro observations, cordycepin 5'-triphosphate (3'dATP) is not a specific inhibitor of poly(A) synthesis in vivo, relative to RNA synthesis, and RNA polymerase I (which synthesises rRNA) is actually less sensitive to inhibition by 3'dATP than RNA polymerase II (ref. 10) (which is presumed to be involved in the synthesis of mRNA). Since nuclear poly(A) polymerase occurs in two functional states as 'free' and 'chromatin-bound' forms, we reasoned that if the chromatin-associated poly(A) polymerase were involved in the initial polyadenylation of mRNA, it might be selectively inhibited by 3'dATP. The present studies, designed to test such an idea, demonstrate that, as in vivo, the initial polyadenylation reaction can be selectively inhibited in vitro by low levels of 3'dATP. These data also show that higher levels of 3'dATP can inhibit RNA synthesis, 'chromatin-bound' RNA polymerase I activity being significantly more sensitive than the 'bound' RNA polymerase II activity.