Elizabeth Burgi

Sedimentation Rate as a Measure of Molecular Weight of DNA

Zone centrifugation of mixtures of two labeled DNAs at low concentrations in density gradients of sucrose permits accurate measurement of relative sedimentation rates. The individual rates are constant during the run. Measurements with DNAs from phages T2, T5, and lambda conform to the relation D(2)/D(1) = (M(2)/M(1))(0.35), where D and M refer to distances sedimented and molecular weights of the DNA pair. The results show that high molecular weight DNAs sediment artificially fast in the optical centrifuge, owing to a hitherto unknown effect of molecular interactions. The molecular weight of lambda DNA is 31 million, measured either from sedimentation rate or from tests of fragility under shear.

A relative molecular weight series derived from the nucleic acid of bacteriophage T2.

Sedimentation velocity was measured for the following materials: (1) T2 DNA isolated without mechanical breakage and purified by ion-exchange chromatography; (2) the first breakage product produced by stirring T2 DNA; (3) subfractions of (2) separated by chromatography, including putative half molecules; and (4) the quarter-length analogues to (2) and (3) produced by stirring half molecules. Some of these materials were also studied by capillary viscometry and by density gradient centrifugation. Results are summarized in Table 1 and Fig. 6. The results establish a relation between relative molecular weight, sedimenta- tion coefficient, and viscosity over a fourfold range higher than that of DNA preparations studied previously. Quarter length fragments of T2 DNA are similar in physical properties to preparations of bacterial DNA to which a molecular weight of 16 million has been assigned. This and other facts set limits to the molecular weight of T2 DNA, which must lie between 50 and 120 million but cannot be further specified at this time.

Molecular Homogeneity of the Deoxyribonucleic Acid of Phage T2

Chromatographic analysis by means of a column of basic protein is used to measure the breakage of T2 DNA by stirring. Broken molecules elute from the column at lower salt concentrations than do the original molecules. At a critical low speed of stirring, one can produce single breaks near the centers of the molecules, as shown by the all-or-none character of the initial change in chromatographic behavior, and by the survival of unbroken molecules according to an exponential function of time of stirring. In an analogous way, one can produce fragments resulting from three breaks per molecule, and in general stirring for a long time at a given speed produces a moderately homogeneous collection of fragments whose mean size is smaller, the higher the speed of stirring. The sensitivity of the DNA to breakage by stirring is strongly dependent on the concentration of DNA, being greater the lower the concentration. This self-protective action is greatly reduced when the DNA is broken by stirring. By chromatographic refractionation of the fragments produced by initial breaks, it can be shown that the fragments are not identical in chromatographic properties, presumably because of a continuous distribution of lengths centered about the mean half-length. Unbroken DNA, on the contrary, is homogeneous in chromatographic properties, therefore presumably homogeneous in molecular length. We conclude that DNA can be extracted from T2 by phenol without manipulative breakage, and that it exists in the phage particle in the form of one or more molecules of identical length.

Segmental distribution of nucleotides in the DNA of bacteriophage lambda.

DNA molecules isolated from phage λ were fragmented by shear and fractionated in various ways, mainly with respect to buoyant density of mercury complexes in Cs2SO4. Six segments of different composition were found, ranging in G + C content from 37 to 57 mole %. The lengths and left-to-right sequence of segments were determined. The segments are reasonably homogeneous internally and the boundaries between segments rather sharp.

Genetic significance of the transfer of nucleic acid from parental to offspring phage

Putnam and Kozloff (1950) first demonstrated that labeled atoms derived from a parental generation of phage particles were partly conserved among the viral offspring. Subsequent work showed that the conserved atoms were derived from and transferred to deoxyribonucleic acid (DNA), chiefly or exclusively (Hershey and Chase, 1952; Kozloff, 1953; French, 1954). What can this phenomenon tell us about genetic replication and recombination? The selective transfer of DNA is not itself very informative: most of the viral protein fails to enter the cell at the time of infection and is unavailable for transfer (Hershey and Chase, 1952). Moreover, the same selectivity is observed when one examines the utilization of labeled bacterial constituents for viral growth: only the DNA is utilized efficiently (Kozloff et al., 1951; Hershey et al., 1954).

Local denaturation of DNA by shearing forces and by heat

The DNA of phage T5 is partially denatured by stirring at 25°C in 0·1 M - or 0·6 M -NaCl at speeds and DNA concentrations similar to those required to initiate breakage of the molecules. The denaturation can be recognized by abnormal retention of the DNA on fractionating columns and by limited susceptibility to enzymes specific for denatured DNA, but not by alteration of absorbancy—temperature relationships. These measurements show that a small degree of denaturation affects a large fraction of the molecules, which must be denatured in restricted regions along their length. The lesions cannot be healed by thermal treatment. After partial breakage of a DNA sample by stirring, broken and unbroken molecules prove to be denatured about equally. Denaturation but not breakage is suppressed when DNA is stirred at 5°C or in 2·6 M -NaCl. Conversely, denaturation without breakage can be produced at 45°C and diminished stirring speeds. Fast stirring, sufficient to break the DNA quickly, reduces the amount of denaturation. Thus breakage and denaturation are independent processes in that they do not show any marked tendency to occur simultaneously in a single molecule, nor does one predispose to the other. Stirring first at 5°C at the speed producing half-length fragments, then restirring at a higher temperature, causes less denaturation than stirring at the higher temperature only. This means that denaturation, like breakage, is produced by forces acting near the centers of the molecules, forces whose effectiveness depends both on the conditions of stirring and on the length of the molecules. Local denaturation can be produced also by bringing T5 DNA to temperatures near the midpoint of the hyperchromic transition and then cooling either rapidly or slowly. Probably heat and shear act in part through similar mechanisms. The DNA of phage T2 is also subject to denaturation under shear, but is appreciably less susceptible than T5 DNA.

Sedimentation Coefficient and Fragility under Hydrodynamic Shear as Measures of Molecular Weight of the DNA of Phage T5.

T5 DNA molecules resemble fragments of T2 DNA of molecular weight 84 x 10(6) with respect to sedimentation coefficient and susceptibility to breakage under hydrodynamic shear. The sedimentation coefficient falls by the same factor when either T2 or T5 DNA is broken at its characteristic critical shear rate. At a given high rate of shear, both DNAs are broken into fragments exhibiting the same sedimentation coefficient. It follows that 84 x 10(6) is a proper estimate of the molecular weight of T5 DNA, and that particles of phage T5, like those of T2, contain a single DNA molecule.

Preferred breakage points in T5 DNA molecules subjected to shear

Abstract T5 DNA molecules subjected to critical rates of shear break into two pieces measuring 0.6 and 0.4 of the original molecular length. The fragments of length 0.6, isolated and subjected to a higher rate of shear, break into two pieces measuring 0.4 and 0.2 of the original molecular length. The molecules must contain two to four sites of preferred breakage at prescribed locations. Preferred breakage points have not been detected and may be absent in DNAs of some other species.