Search

(1975). Recent Advances in DNA Sequence Analysi. CRC Critical Reviews in Biochemistry: Vol. 2, No. 4, pp. 455-512.

Measurements of the sedimentation velocity under completely denaturing conditions in dimethylsulfoxide were used to obtain the molecular weights of some T4-specific mRNAs including the biologically active lysozyme mRNA. The amount of folding was determined by comparing their sedimentation in sucrose gradients in the presence and absence of this reagent. T4-specific mRNAs sediment much more slowly in the presence of dimethylsulfoxide than in its absence. This difference is linearly dependent on molecular weight; the dependence was similar to that observed for Escherichia coli stable RNA controls and very different from that of random coils (E. coli stable RNA treated with formaldehyde so that no base-pairing is possible). We conclude that the T4-specific mRNAs are not random coils but resemble much more closely the highly base-paired E. coli stable RNAs.

DNA synthesized in vitro from a phage M13 template has been cleaved at either guanine, adenine, or cytosine residues by ribosubstitution techniques. Fingerprints of the fragments obtained suggest that DNA sequencing will be possible with this technique.

Rabbit globin complementary DNA made with RNA-dependent DNA polymerase (reverse transcriptase) was used as template for in vitro synthesis of 32 P-labeled RNA. The sequences of the nucleotides in most of the fragments resulting from combined ribonuclease T 1 and alkaline phosphatase digestion have been determined. Several fragments were long enough to fit uniquely with the α or β globin amino-acid sequences. These data demonstrate that the cDNA was copied from globin mRNA and contained no detectable contaminants.

The existence of non-random base pairing in T4-specific mRNA is supported by the nuclease digestion patterns of the messenger for phage T4 lysozyme. Under conditions which denature RNA, the nuclease nicked lysozyme mRNA recovers its activity. This behaviour can be explained by a model in which secondary structure controls initiation at existing ribosomal attachment sites.

Deoxynucleotides have been incorporated into RNA synthesized in vitro by RNA polymerase with either double-stranded or single-stranded DNA as a template. By use of this technique to block or promote cleavage at a particular phosphodiester bond, a variety of specific cleavages may be obtained with the available ribonucleases and deoxyribonuclease I. These methods should greatly increase the ease and rapidity of nucleotide sequence determinations.

Summary Three techniques (optical melting curves, sedimentation in the presence and absence of dimethylsulfoxide, and resistance to RNase Tl) are used to investigate different aspects of the structure of random RNA sequences. The hypochromicity of melting curves indicate that these sequences have about 60% base-pairing, but the low T m and large σ T suggest that the base-pairing is much less stable than that observed in rRNA. Both the small hydrodynamic radius revealed in sedimentation studies and the relatively low resistance to RNase are interpreted as confirming the presence of base-pairing but base-pairing which is disorderly .

Publisher Summary Recent developments indicate that in vitro synthesis of ribosubstituted DNA may allow the relatively rapid determination of DNA sequences of small to moderate complexity. In vitro synthesis provides an advantage as one can synthesize the same ribosubstituted DNA with the 32 P labeling a different one of the four bases each time, so that nearest-neighbor labeling information will greatly simplify the subsequent analysis. The ready availability of small molecules such as the tRNAs and 5 S ribosomal RNA was extremely important in permitting the rapid development of RNA sequencing methods. In order to obtain very simple sequences for working out DNA sequencing techniques we have turned to the highly iterated satellite DNA fractions of higher organisms. This chapter describes the methods for synthesis and cleavage of ribosubstituted DNA and discusses the analysis of the sequences of individual fragments and the use of micrococcal nuclease to determine the sequence of larger fragments.

The sequence of the highly repetitive satellite HS-β DNA fraction from kangaroo rat Dipodomys ordii was determined independently by RNA and DNA sequencing techniques. A basic iterated sequence of 10 nucleotides with several mutational variations was found. Base-composition data are consistent with the proposed sequence and revealed a high content of 5-methylcytosine. DNA and RNA sequencing techniques used gave identical results, showing that the fidelity of synthesis of riboguanidine-substituted DNA under our conditions is adequate for nucleotide sequence studies.

PROTEIN synthesis requires initiation of translation at the correct site on mRNA. Two initiation factors, A and C (ref. 1), or F1 and F2 (ref. 2), promote the ApUpG-dependent attachment of formylmethionyl-tRNA (fmet-tRNAf) to Escherichia coli ribosomes2–4, and account for initiation AUG triplets5,6. A mechanism must exist, however, for the discrimination between internal and initial methionine codons. In order to study the possible role of the third initiation factor, B, in this selection mechanism, we have examined the role of the three factors in the synthesis of lysozyme promoted by T4 mRNA7. We assumed that polypeptide synthesis as measured by amino-acid incorporation should not necessarily require correct initiation, while active enzyme formation should be highly dependent on this process. We have used the ratio of enzyme formed to valine incorporated as a measure of the proportion of polypeptide molecules correctly initiated.

A new technique for measuring the rate of turnover and quantity of unstable RNA in exponentially growing cells has been developed which escapes the interpretive difficulties of earlier methods. It depends upon the fact that an unstable RNA fraction has easily measurable effects upon the kinetics of labelling of GTP; before the GTP pool can become fully labelled, unlabelled nucleotides present as GTP or in unstable RNA must be washed into stable RNA. (There is no equilibration with guanine in the medium.) Consequently, if the uptake of label into GTP is rapid, with only the delay predicted from the experimentally measured GTP pool size, then it is possible to conclude that there is little or no unstable RNA. In Bacillus subtilis, however, the GTP pool is labelled very slowly which, with other data, indicates a large amount of unstable RNA (9% of the guanine residues of the cell in unstable RNA). Similar measurements of the rapidity of GTP labelling in Escherichia coli indicate that there is much less unstable RNA (about 3%) in this organism. The curve expressing GTP specific activity as a function of time is shown to be the sum of two exponential terms with very different time constants. From the relative sizes of the two exponential terms, it is possible to estimate the rate of turnover of the unstable RNA. The result is weighted most heavily for the slow components of the decaying messenger, and thus it sets an upper limit to the true average decay time for the unstable RNA. This decay time (upper limit) was three minutes for the unstable RNA in B. subtilis grown at 37 °C with a doubling time of 115 minutes and six minutes when the cells were grown at 30 °C with a doubling time of 225 minutes. For E. coli (growing at 30 °C with a doubling time of 66 minutes) the decay time measured in this way was about four minutes. Thus the decay times as measured in these experiments are the same order of magnitude as have been observed by measuring the decay of pulse-labelled RNA and the decay of protein synthetic capacity after the addition of actinomycin. The various estimates of the decay rate of messenger RNA in bacteria are discussed in terms of the possible theoretical ambiguities in the interpretation of each.