Rate of synthesis and half-life of globin messenger ribonucleic acid. Rate of synthesis of globin messenger ribonucleic acid calculated from data of cell haemoglobin content

By the use of the favoured models defining mRNA synthesis and half-life from the preceding paper (Hunt, 1974) and the known content of globin in a reticulocyte it is possible to estimate the absolute rate of mRNA and globin synthesis and the mRNA and globin content in each type of erythroid cell. The best model requires an mRNA-synthetic rate of 3000 molecules per h/cell. This rate compares favourably with the estimated chain-extension rate of 43 nucleotides/s in Escherichia coli (Manor et al., 1969) provided that the four alpha- and beta-chain cistrons per cell are transcribed by polymerases spaced 50 nucleotide base pairs apart. Similar calculations can be made for erythropoiesis in the chick embryo, where cell times and relative globin content at each mitosis have been measured (Campbell et al., 1971), but where no reliable estimates of mRNA half-life have been made. In this case it was estimated that a constant rate of mRNA synthesis at 10000 molecules per h/cell through six cell divisions is necessary if the mRNA half-life is 15h; after the sixth mitosis the mRNA synthesis would stop and its half-life would increase to approx. 20h. If an mRNA half-life of 4.5h is used, the synthesis rate through the six mitoses would be 21000 molecules per h/cell, ceasing at the sixth mitosis, when the half-life would need to increase to 25h. The chain-elongation rate for the four alpha- and beta-globin cistrons per cell would be 1-2 times higher than in E. coli and would either require a greater rate, polymerases spaced between 25 and 50 nucleotide base pairs apart on the DNA, or limited gene replication. These possibilities are discussed in the light of the low values found for globin cistron multiplicity in ducks and mice.     

The identification of globin messenger ribonucleic acid in newt erythropoietic cells.

Polyadenylated [poly(A)+]-RNA isolated from newt (Triturus cristatus) erythropoietic cells contained two main species sedimenting at 9S and 25S, and minor amounts of a 15-20S component. The 9S poly(A)+-RNA fraction induced synthesis of newt haemoglobin and globins in frog oocytes and in an mRNA-dependent rabbit reticulocyte lysate, confirming its identity as newt globin mRNA. Translation of 9S globin mRNA in reticulocyte lysate was concentration-dependent, the patterns of globin synthesis suggesting both preferential utilization and unequal amounts of the different globin mRNA subspecies. Globin mRNA activity was also evident in the 25S poly(A)+-RNA fraction whose localization in polyribosomes excluded its function as a nuclear globin mRNA precursor. Denaturation in formamide and estimation of its relative methyl content indicated that the 25S poly(A)+-RNA fraction contained equimolar amounts of 9S globin mRNA and 26S rRNA. Translation of the 25S fraction in reticulocyte lysate was less efficient than that of comparable amounts of 9S globin mRNA and induced a pattern of globin synthesis similar to that obtained with subsaturating amounts of 9S mRNA. The 25S mRNA-rRNA complex was considered to be a non-physiological aggregate generated by extraction of RNA in the presence of buffers of moderate to high ionic strength.     

Analysis of the ratio of alpha- to beta-globin and globin messenger ribonucleic acid content of fractionated rabbit erythroid bone-marrow cells.

The ratio of alpha- to beta-globin mRNA was measured by hybridization of a constant amount of highly purified alpha- or beta-globin cDNA (complementary DNA) with increasing amounts of RNA in the range up to 20% cDNA hybridization, where an essentially linear reaction is obtained. Statistical analysis indicates that the ratio of alpha- to beta-globin can be measured within a maximal error of +/- 0.3 and in most cases is better than +/- 0.15. Under these conditions there is no significant deviation from the ratio of 1.3 in the alpha- to beta-globin mRNA ratio of RNA isolated from erythroid cells rich in pronormoblasts through to reticulocytes. If the ratio of alpha- to beta-globin mRNA exceeded 1.7 or was less than 0.9 in pronormoblasts, it would be detected in these experiments. The overall globin mRNA content increases to a maximal value in the fractions rich in basophilic normoblasts of 30,000--50,000 molecules/cell. However, the accuracy of these determinations is not as great as for the ratio determinations, and no significant deviations were seen except in the cells rich in pronormoblasts, which contained less globin mRNA than the later stages.     

Interaction between polyuridylic acid and rabbit globin messenger ribonucleic acid

Poly(U) binds to globin mRNA in 0.1m-NaCl. Studies with ribonuclease digestion of this complex suggest that there are polyadenylate-rich sequences in the mRNA containing about 30-40 adenylate residues. The sequences appear to be homogenous and of approximately the same length for both alpha- and beta-globin mRNA. They are most likely located at the 3' terminus of the molecule.