Search

The role of GTP hydrolysis in microtubule dynamics has been reinvestigated using an analogue of GTP, guanylyl-(alpha, beta)-methylene-diphosphonate (GMPCPP). This analogue binds to the tubulin exchangeable nucleotide binding site (E-site) with an affinity four to eightfold lower than GTP and promotes the polymerization of normal microtubules. The polymerization rate of microtubules with GMPCPP-tubulin is very similar to that of GTP-tubulin. However, in contrast to microtubules polymerized with GTP, GMPCPP-microtubules do not depolymerize rapidly after isothermal dilution. The depolymerization rate of GMPCPP-microtubules is 0.1 s-1 compared with 500 s-1 for GDP-microtubules. GMPCPP also completely suppresses dynamic instability. Contrary to previous work, we find that the beta--gamma bond of GMPCPP is hydrolyzed extremely slowly after incorporation into the microtubule lattice, with a rate constant of 4 x 10(-7) s-1. Because GMPCPP hydrolysis is negligible over the course of a polymerization experiment, it can be used to test the role of hydrolysis in microtubule dynamics. Our results provide strong new evidence for the idea that GTP hydrolysis by tubulin is not required for normal polymerization but is essential for depolymerization and thus for dynamic instability. Because GMPCPP strongly promotes spontaneous nucleation of microtubules, we propose that GTP hydrolysis by tubulin also plays the important biological role of inhibiting spontaneous microtubule nucleation.

Escherichia coli DNA has been fractionated into extractable and non-extractable DNA after deproteinization of detergent-lysed cell preparations with chloroform—isoamyl alcohol. The former was extracted with dilute buffered saline whereas the latter remained in the interphase layer associated with residual cellular debris from which almost 40% could be released by incubating with pronase. About 20–25% more amino acid residues were bound to the pronase-released DNA than to the extractable DNA, but the relative distribution of the residues in the two DNA samples was virtually identical. The specific activities and the relative amounts of denser (1.709 g · cm−3) and lighter (on the surface of CsCl gradients) DNA fractions from E. coli, grown in the presence of labeled thymidine, indicated that these two corresponded to extractable and non-extractable DNA, respectively. The relative amounts of the two fractions varied with the growth phase, primarily as a function of the growth rate. Age and metabolic state of cells in the culture or those used as inocula could modify this relative distribution. When growth rate was maximal, the ratio of the two remained at about 1. During lag phase when no appreciable net synthesis of DNA could be detected, there was a rapid and preferential incorporation of labeled thymidine into non-extractable DNA. A disproportionate increase in the fraction of the total DNA, which was extractable, was also observed but only when stationary phase cultures were used as inocula. Complete equilibration of the label in the two DNA fractions was attained only after cultures had reached mid-log phase of growth. Similar results were obtained when prelabeled cells were used. These data have been interpreted as suggesting that the rate of cell growth and DNA synthesis are related to the number or size of sites of attachment of DNA to some cellular structure. Newly synthesized DNA would be attached to different sites on this structure and initiation of DNA replication in lag phase would require reorientation of the two kinds of DNA. Small peptides which are firmly bound to the DNA and which vary quantitatively with the rate of DNA synthesis could perhaps be involved in the attachment to the sites.

Amino acids are found tightly bound to DNA isolated from several strains of Escherichia coli. The amount bound was a function of growth conditions; it varied with changes in the composition of the medium and with the phases of the growth cycle. In DNA extracted from the earliest growth phases an unusual amino acid was found which was eluted from a Stein-Moore column with the aromatic amino acids. This odd substance was not found in the cell protein hydrolysates and appeared to be the same compound previously found in several primary tumors. In one instance, DNA and protein were extracted from cells which had been incubated with labeled amino acids under conditions which permitted protein but not DNA synthesis. The protein was highly labeled, but the amino acids associated with the DNA were not. The amino acids were not dissociated from the DNA by banding in cesium chloride or by passage through molecular exclusion gels in 5 m urea. These data have been interpreted as suggesting a functional role for DNA-associated amino acids.

The effects of the administration of single injections of sheep red blood cells (SRBC) and of 6-mercaptopurine (6-MP) on the synthesis of nucleic acids and proteins were studied. Under the experimental conditions, extensive antibody production was induced by the SRBC and this synthesis was blocked by the dose of 6-MP used. One day after the various treatments, inhibition of nucleic acid synthesis was seen in spleen, kidney, and liver of those animals which had received 6-MP. No extensive reduction in protein synthesis was seen. No consistent reduction was seen in the synthesis de novo of purines. Eight days after initiation of treatment, at which time antibody production was extensive, there was still a reduced incorporation of formate into DNA in the spleen of those animals which had received 6-MP or-6-MP plus SRBC. The incorporation of thymidine into DNA, on the other hand, was reduced in the spleen of animals receiving 6-MP but in those animals which had received 6-MP plus SRBC there was much less inhibition. Eight days after treatment there was still considerable inhibition of thymidine incorporation into the liver of animals which had received 6-MP alone, no change in the animals which had received SRBC, and considerable increase in those animals which had received both 6-MP and SRBC. The findings suggest a role of DNA synthesis in response to antigens. The data also demonstrate that the effects of antimetabolites persist long after they have been catabolized per se , and in view of the biological effects seen concurrently suggest general significance to these long-lived effects.

WE have characterised two forms of deoxythymidine kinase (TK) using antisera prepared against the enzyme partially purified from human colonic mucosa. One of these forms is associated with rapidly proliferating lower colonic crypt cells and the other, with non-dividing surface cells. Antibody to the latter cross reacted with several tumours and placental extracts but not with phytohaemagglutinin (PHA)-stimulated or leukaemic lymphocytes. This cross reactivity of surface cells and colonic tumours supports the concept that surface cells rather than crypt cells are the site of origin of colonic cancer1,2.

The immunological properties of thymidine kinase from a variety of human tumors suggest that the form of the tumor enzyme resembles that found in the placenta and in the nondividing colonic flat mucosa. To examine the placenta-like characteristics of tumor thymidine kinase, the jejunum and colon from rats ranging in age from fetal to old and from animals treated with dimethylhydrazine (DMH), an intestinal carcinogen, have been studied. In normal jejunum, thymidine kinase activity decreased rapidly with age. Both the activity and the response to phospholipase C and to mercaptans in DMH-induced tumors resembled that of fetal gut, while those in abnormal appearing DMH-treated jejunum were intermediate between normal control of the same age and tumor. Similar but less pronounced changes were seen in the colon. In the jejunum, the level of another enzyme normally associated with rapid cell division, ornithine decarboxylase, was found to be over 100 times that of the liver, colon, and stomach. Treatment of the animals with acetylaminofluorene and with DMH resulted in elevated levels of the enzyme in liver and in colon, respectively, but had little effect on this enzyme in other tissues. The data presented indicate that there were premalignant changes in the levels of both of these enzymes in target tissues of animals treated with carcinogens.

DNA was isolated from liver, kidney, spleen and small intestines of rats treated with high doses of actinomycin D, proflavin, nitrogen mustard (HN2) and mitomycin C. The DNA-associated amino acids differed profoundly in relative and absolute amounts. Treatment in vitro of liver homogenate with one of these drugs (proflavin) did not produce this or any change in the amino acids bound to DNA, which suggests a complex interaction. The changes are not random, since they: (1) are qualitatively and quantitatively reproducible; (2) include both additions and deletions of amino acids; and (3) include an extra ninhydrin-positive component associated with the aromatic amino acids but not found in the residual host proteins. The new amino acid is apparently the same one previously found in certain tumor DNA's. Despite the differences in molecular structure, reported pharmacologic effects, and nature of their interactions with DNA of the four compounds studied, they all produced profound changes in the amount and kind of amino acids bound to DNA so as to suggest that this may be a significant factor in the mechanisms of their action.

A small but persistent amount (0.1–0.6 %) of amino acids is found bound to the DNA of all tissues examined. The amount is constant and reproducible for the DNA of a given tissue taken from animals of the same species, sex and age. These observations suggest that the residues have a role. There is a difference in absolute and relative amounts of the amino acids bound to DNA from different tissues of the same species. There are also differences among the amino acids bound to the DNA of the liver of several species.

SEVERAL investigators have found amino-acids among the hydrolysis products of DNA's from many sources and some have suggested that small peptides may play a part in DNA structure and/or function. These have been reviewed by Bendich and Rosenkranz1. We have considered the possibility that such small peptides are concerned with the regulation of enzyme synthesis. For this reason we have examined the amino-acid residues found in samples of DNA from Escherichia coli grown in ‘minimal’ and in ‘complete’ media. Bacteria grown in minimal media must synthesize large numbers of compounds not made by those grown in complete media. Therefore, more enzymes should be formed and more DNA should be functional in bacteria grown in minimal media. If control of DNA function is exerted through firmly bound amino-acids and peptides, then DNA from bacteria grown under these two sets of conditions should contain different amounts of amino-acids.