ABSTRACT
This paper represents a first attempt to follow a specific eukaryotic messenger RNA from the moment of its transcription from the nuclear DNA to the site of its expression by protein synthesis in polyribosomes.
A direct demonstration that the globin mRNA sequence is transcribed from DNA in form of a giant precursor RNA (pre-mRNA) is presented: The anti-messenger DNA produced by transcription of purified globin mRNA with the RNA-dependent DNA-polymerase hybridizes to nuclear RNA of more than 2.5 × 106 MW (isolated on DMSO gradients) equally well as to its template.
This proves that the nascent pre-mRNA is an informative precursor to mRNA. However, there exists no direct evidence demonstrating a physical precursor-product relationship between nascent pre-mRNA and mRNA. Nascent pre-mRNA molecules of 5–20 × 106 MW are cleaved with a halftime of about 30 min into intermediate size (1–5 × 106 MW) pre-mRNA molecules which are considerably more stable. This first processing step is accompanied by the solubilisation of about 50 % of the pre-mRNA. Small size nuclear pre-mRNA 1–10 × 105 MW) is metabolically very stable. However, no peak of nuclear 9S globin mRNA can be observed on polyacrylamide gels. Thus the globin mRNA sequence, processed either by direct cleavage of pre-mRNA or by a process of retranscription, is exported very rapidly to the cytoplasm.
Ribosome-free ribonucleoprotein particles containing translatable globin mRNA can be isolated in the cytoplasm. Newly synthesized 9S mRNA rapidly fills up this free pool and the globin messenger is transferred to polyribosomes. The pool of free mRNP particles contains a larger spectrum of messenger molecules than polyribosomes where the 9S mRNA is predominant; some of the other mRNA-types are possibly never translated. Thus we conclude that a pre-translational control must operate, reducing quantitatively and qualitatively the mRNA spectrum in free RNP particles to that translated in polyribosomes.
The globin mRNA in polyribosomes is doubly heterogeneous: newly synthesized molecules have molecular weights of about 230 000 whereas the steady-state spectrum of mRNA molecules is represented by a heterogeneous population of molecules with an average molecular weight of about 180 000. On this basis, we assert that processing of mRNA continues right throughout the translation process.
Nascent globin pre-mRNA is associated with specific proteins; their major component cannot be observed among the proteins associated with cytoplasmic mRNA. The proteins associated with globin mRNA in the free mRNP particles are different from those associated with the translated globin message. Both protein populations include some phosphorylated species which contain phosphoserine. Thus it is possible that exchange of messenger-associated proteins and their phosphorylation reflect pre-translational and translational controls of globin synthesis.
Compared to mRNA from other species, duck globin mRNA is poorly translated in all protein-synthesizing lysate systems tested. However, in a purified and reconstituted ribosome system, it is translated equally well as rabbit 9S mRNA provided that the system is supplemented with a high amount of a particular fraction of heterologous initiation factors. This requirement may be explained by a particular secondary structure in the programming sequence of the duck globin message.
It is proposed that the transcriptional unit of gene regulation reflected in the nascent pre-mRNA molecule contains, in addition to one or several messenger sequences and possible no-sense sequences, programming sequences; interacting with specific proteins, such sequences may permit post-transcriptional controls involved in pre-mRNA processing and in messenger formation and translation. A general discussion of the regulation of the highly complex system of mRNA formation and translation in eukaryotic cells is presented with particular reference to the Cascade Regulation Hypothesis.
Quantitative principles of cis-translational control by general mRNA sequence features in eukaryotes
