On the mutagenic action of adenine.

The deamination of cytosine and adenine is mutagenic; the deamination of guanine is not. The deamination of cytosine leads to G=C-->A=T point mutation and to G-->A and C-->T transition in the DNA molecule; the deamination of adenine leads to the opposite A-->G and T-->C transition. It is shown that adenine lack could be as mutagenic as adenine deamination and it is also shown schematically that adenine lack through defective adenine synthesis could give rise to a population of genetically abnormal cells incapable of any degree of differentiation, a state perhaps reminiscent of the most acute of leukaemias and the most anaplastic of cancers.

Mutagenic mechanisms in leukemia and cancer: a new concept Cytosine lack could be as mutagenic as cytosine deamination.

The molecular mechanisms involved in mutagenesis caused by nitrous acid, ultraviolet light and cytosine methylation are reviewed. All three mutagens lead to a G=C --> A=T point mutation in the DNA molecule and the biochemical basis of these mutations in each case is the deamination of cytosine to uracil. It is shown that the lack of cytosine could be as mutagenic as the deamination of cytosine and it is shown schematically how cytosine lack could give rise to a population of genetically abnormal cells which are completely incapable of any degree of differentiation; a state perhaps reminiscent of the most acute of leukemias and the most anaplastic of cancers. A metabolic block in the amination of uracil to cytosine is suggested as a possible cause of the cytosine lack, a block which is unique in that it incorporates a mutagenic mechanism.

On the pathogenesis of erythroleukaemia (H0493).

In erythroleukaemia megaloblastic changes can co-exist with leukaemic changes in the marrow. The cause of the disease must therefore be such as can cause megaloblastosis and at the same time be mutagenic. Failure of the thymidylate synthelase reaction, the commonest cause of megaloblastic anaemia, can be eliminated in erythroleukaemia because (a) the dU suppression test is normal in the disease and (b) failure of the thymidylate synthelase reaction is not mutagenic. The deamination of both cytosine and adenine is mutagenic but the deamination of cytosine alone is apparent and the nucleotide of cytosine is the prime mutagenic nucleotide in leukaemia and cancer. Megaloblastic changes can result from an inadequate supply of any one of the four nucleotides that enter into the composition of DNA and it is suggested that an inadequate supply of the mutagenic nucleotide of cytosine, possibly through impaired synthesis, could cause both the megaloblastic and leukaemic changes in erythroleukaemia.

The non-random distribution of point mutations in leukaemia and myelodysplasia--a possible pointer to their aetiology.

A conventional and a computer search of the literature yielded 627 sequenced point mutations in the ras and p53 genes in 575 patients with leukaemia and myelodysplasia (MDS) out of a total of 4214 investigated. ras Mutations predominated in myeloid leukaemia and were more common in the disease in relapse than at presentation. There was no clinical, or haematological difference or difference in survival between ras positive and ras negative patients with acute myeloid leukaemia (AML) in adults or children, but ras mutations carried a poorer prognosis in childhood acute lymphocytic leukaemia and an increased risk of leukaemia in MDS. p53 mutations predominated in lymphoid leukaemia and were several fold more frequent in leukaemia in relapse than in the de novo disease, were associated with loss of the normal p53 allele (monosomy 17) in > 50% of cases and carried a poor prognosis in AML, MDS and chronic lymphatic leukaemia and a 3.8-fold increase risk of death in T cell acute lymphocytic leukaemia. There were 163 transitions for every 100 transversions, the expected number being ca 50. Consideration of the molecular mechanisms by which nitrous acid produces transitions allows transitions resulting from the deamination of cytosine to be distinguished from those resulting from the deamination of adenine. The former constitute 84.67% and the latter 15.33% of the 372 transitions present. Again purine-->pyrimidine and pyrimidine-->purine transversions form 80.35 and 19.65%, respectively, of the 228 transversions present. The possible bearing of this highly non-random distribution on the aetiology of point mutations in leukaemia and myelodysplasia is discussed.

Folate responsive neuropathy.

The literature on folate related neuropathy has been reviewed. Twenty patients fulfilled the following criteria (a) they presented with neurological findings for which no other cause could be found (b) the serum or red cell and/or the CSF folate was low (c) the serum vitamin B12 or vitamin B12 absorption was normal and (d) they showed a significant response to folic acid. Ten presented with a peripheral neuropathy, eight with subacute combined degeneration of the cord and two with a myelopathy. In two patients the neuropathy occurred when treatment for congenital malabsorption of folate--an isolated lesion affecting folate alone--lapsed. Two patients with subacute combined degeneration died and posterio-lateral sclerosis of the cord was confirmed at autopsy. Three patients were mentally retarded and nine showed mental changes which also responded to folate in addition to the neurological disorder. A single biochemical reaction, the methionine synthetase reaction, is suggested as the basis for the neurological as well as the haematological consequences of both vitamin B12 and folate deficiency. The pitfalls in diagnosis are discussed and a greater awareness of the condition urged.

Serum valine, methionine and isoleucine levels in patients anaesthetized with and without nitrous oxide.

Serum methionine and valine have been assayed microbiologically in 31 patients anaesthetized with and in 13 controls anaesthetized without nitrous oxide. Isoleucine has been similarly assayed in five cases and three controls. The preinduction levels of both methionine and valine were considerably below those of semifasting healthy adults. Serum methionine fell significantly in both groups as did valine and isoleucine in the controls. These findings are attributed to dietary restriction before and during the operation. Serum valine was significantly elevated in patients anaesthetized with nitrous oxide but only in those anaesthetized for 3 h or longer. Isoleucine also was elevated in two of the five cases exposed to nitrous oxide. Reasons are given for attributing these elevated levels to the direct action of nitrous oxide. Valine is entirely and isoleucine is partly catabolized along the adenosylcobalamin dependent propionyl-CoA-succinyl-CoA pathway. It is concluded that their elevated levels are caused by the inactivation of adenosylcobalamin by nitrous oxide thereby inducing the methylmalonyl mutase block in this pathway. It is suggested also that the relief of this block by adenosylcobalamin in pernicious anaemia could provide a possible metabolic basis for the well recognized early subjective improvement encountered in this disease following vitamin B12 therapy.

A new case of congenital dyserythropoietic anaemia, type III: studies of the cell cycle distribution and ultrastructure of erythroblasts and of nucleic acid synthesis in marrow cells.

The clinical and laboratory findings in an asymptomatic 19-year-old Welshman with congenital dyserythropoietic anaemia (CDA) type III are described. The blood film showed macrocytosis and red cell fragmentation and there was biochemical evidence of intravascular haemolysis. The bone marrow showed erythroid hyperplasia, megaloblastic erythropoiesis and several giant multinucleate erythroblasts. Some mononucleate erythroblasts were large and had relative DNA contents of 4-8c and the bi- and multinucleate erythroblasts had total DNA contents of 2-16c. Some of the multinucleate erythroblasts displayed a variety of ultrastructural abnormalities, including marked differences in the appearances of the individual nuclei within the same cell. The marrow cells gave a normal deoxyuridine-suppressed value indicating that the megaloblastic changes were not caused by an impairment of the methylation of deoxyuridylate. The rates of incorporation of 14C-glycine and 14C-adenine into both the DNA and RNA of bone marrow cells were within the normal range. Furthermore, the average rate of elongation of newly-synthesised, 3H-thymidine-labelled daughter DNA strands, assessed by hydroxyapatite chromatography of alkali-denatured DNA was found to be normal. The results suggest that there is no impairment of DNA replication in the majority of the erythroblasts and that the abnormality of erythropoiesis resulted from disturbances during mitosis and the G2 phase.

Leukaemia classification: a study of the accuracy of diagnosis in 456 patients.

A panel of five haematologists has examined, without consultation or prior knowledge of the diagnosis, blood films and bone marrow smears from 456 patients with a diagnosis of leukaemia. A diagnostic classification which recognized various subtypes of acute myelogenous leukaemia was used but no attempt was made to subdivide acute lymphoblastic leukaemia. Complete agreement with the initial diagnosis was low (56.4%) and was particularly poor (45.7%) when the patient had one of the forms of acute leukaemia. However, disagreements which would have involved the patient in a change of treatment were unusual (2.0%). We conclude that a high degree of diagnostic agreement for patients with leukaemia is unlikely from morphological classifications alone.

Electron microscope autoradiographic studies of the erythroblasts of a case of congenital dyserythropoietic anaemia, type II.

The bone marrow cells of a patient with congenital dyserythropoietic anaemia, type II, were incubated with 3H-thymidine, 3H-uridine or 3H-leucine for 1 h and studied using the technique of electron microscope autoradiography. Several of the erythroblasts which either displayed the characteristic subsurface double membranes or showed various non-specific abnormalities of the nuclear membrane were found to be actively engaged in DNA, RNA and protein synthesis. Both members of some pairs of erythroblasts which were joined together by a spindle bridge were found to be engaged in DNA synthesis, indicating that some spindle bridges persist for a period longer than the duration of the G1 phase. A small proportion of mononucleate and binucleate late (non-dividing) erythroblasts showed a marked depression or arrest of protein synthesis and some or all of such cells were presumably destined to be phagocytosed by the bone marrow macrophages.