The antiproliferative effect of beta-carotene requires p21waf1/cip1 in normal human fibroblasts.

In normal human fibroblasts, beta-carotene induces a cell-cycle delay in the G1 phase independent of its provitamin A activity via a mechanism not yet elucidated. In this study we provide biochemical evidence showing that delayed progression through the G1 phase occurs concomitantly with: an increase in both nuclear-bound and total p21waf1/cip1 protein levels; an increase in the amount of p21waf1/cip1 associated with cdk4; the inhibition of cyclin D1-associated cdk4 kinase activity; and a reduction in the levels of hyperphosphorylated forms of retinoblastoma protein, and particularly, in phosphorylated Ser780. The role of p21waf1/cip1 in the antiproliferative effect of the carotenoid was further supported by genetic evidence that neither changes in cell-cycle progression nor in the phosphorylation status of retinoblastoma protein were observed in p21waf1/cip1-deficient human fibroblasts treated with beta-carotene. These results clearly demonstrate that p21waf1/cip1 is involved directly in the molecular pathway by which beta-carotene inhibits cell-cycle progression.

The replication factory targeting sequence/PCNA-binding site is required in G(1) to control the phosphorylation status of DNA ligase I.

The recruitment of DNA ligase I to replication foci in S phase depends on a replication factory targeting sequence that also mediates the interaction with proliferating cell nuclear antigen (PCNA) in vitro. By exploiting a monoclonal antibody directed at a phospho-epitope, we demonstrate that Ser66 of DNA ligase I, which is part of a strong CKII consensus site, is phosphorylated in a cell cycle-dependent manner. After dephosphorylation in early G(1), the level of Ser66 phosphorylation is minimal in G(1), increases progressively in S and peaks in G(2)/M phase. The analysis of epitope-tagged DNA ligase I mutants demonstrates that dephosphorylation of Ser66 requires both the nuclear localization and the PCNA-binding site of the enzyme. Finally, we show that DNA ligase I and PCNA interact in vivo in G(1) and S phase but not in G(2)/M. We propose that dephosphorylation of Ser66 is part of a novel control mechanism to establish the pre-replicative form of DNA ligase I.

Preferential perinuclear localization of poly(ADP-ribose) glycohydrolase.

The transient nature of poly(ADP-ribosyl)ation, a posttranslational modification of nuclear proteins, is achieved by the enzyme poly(ADP-ribose) glycohydrolase (PARG) which hydrolyzes the poly(ADP-ribose) polymer into free ADP-ribose residues. To investigate the molecular size and localization of PARG, we developed a specific polyclonal antibody directed against the bovine PARG carboxy-terminal region. We found that PARG purified from bovine thymus was recognized as a 59-kDa protein, while Western blot analysis of total cell extracts revealed the presence of a unique 110-kDa protein. This 110-kDa PARG was mostly found in postnuclear extracts, whereas it was barely detectable in the nuclear fractions of COS7 cells. Further analysis by immunofluorescence revealed a cytoplasmic perinuclear distribution of PARG in COS7 cells overexpressing the bovine PARG cDNA. These results provide direct evidence that PARG is primarily a cytoplasmic enzyme and suggest that a very low amount of intranuclear PARG is required for poly(ADP-ribose) turnover.

Congenital disorders sharing oxidative stress and cancer proneness as phenotypic hallmarks: prospects for joint research in pharmacology.

In spite of very distinct genotypic assets, a number of congenital conditions include oxidative stress as a phenotypic hallmark. These disorders include Fanconi's anaemia, ataxia telangiectasia, xeroderma pigmentosum and Bloom's syndrome, as well as two frequent congenital conditions: Down's syndrome and cystic fibrosis. Cancer proneness is a clinical feature shared by these disorders, while other manifestations include early ageing, neurological symptoms or congenital malformations. The onset of oxidative stress has been related to excess formation, or defective detoxification, of reactive oxygen species (ROS). This can arise from either the abnormal expression or inducibility of ROS-detoxifying enzymes, or by defective absorption of nutrient antioxidants. Resulting oxidative injury has been characterized through: (i) DNA, protein or lipid oxidative damage; (ii) excess ROS formation (in vitro and ex vivo); (iii) sensitivity to oxygen-related toxicity; (iv) improvement of cellular defects by either hypoxia or antioxidants; and (v) circumstantial evidence for in vivo oxidative stress (as e.g. clastogenic factors). Investigations conducted so far have been confined to individual disorders. Comparative studies of selected indicators for oxidative stress could provide further insights into the pathogenesis of each individual condition. Such a unified approach may have wide-ranging consequences for studies of ageing and cancer.

Structural requirements for inhibitors of poly(ADP-ribose) polymerase.

The purpose of this study was to examine the structure/activity relationships of a series of substituted benzamides as poly(ADP-ribose) polymerase inhibitors. The experimental approach has involved the use of in vitro and in vivo assays in order to gather information either on the intrinsic activity of the benzamides or on the effect of various pharmacodynamic parameters on the activity in vivo. Although some discrepancies between the data obtained in vivo and in vitro were found in this study, results seem to indicate that most powerful inhibitors were characterized by acylation of the -NH2 function in the 3 position or by substitution in this same position with hydroxy or methoxy groups. The best inhibitors were not cytotoxic under these experimental conditions. Computed calculations of molecular electrostatic potential of these molecules were also performed and a good correlation was found between the similarity index and the experimental inhibitory activity.

Prognostic significance of adenosine deaminase determinations in subjects with the lymphoadenopathy syndrome.

The association between human immunodeficiency virus type I (HIV-I) infection and high levels of erythrocyte adenosine deaminase (ADA) has been suggested by Cowan et al [1986]. We have analyzed the specific activities of the same enzyme during different stages of acquired immunodeficiency syndrome (AIDS), including asymptomatic subjects at high risk and patients with lymphoadenopathy syndrome (LAS), AIDS-related complex (ARC), full-blown AIDS, and AIDS encephalopathy (AIDS enc). The ADA activities were significantly higher (P less than .05) in asymptomatic HIV-I serum-positive individuals (13.1 U +/- 1.1) and in different groups of patients (LAS = 23.6 U +/- 10.2; ARC = 23.7 +/- 4.1) than those found in controls (9.5 U +/- 1.8) and in HIV-I serum-negative subjects (10.4 +/- 1.5). In patients with AIDS the mean ADA activity was of 32.3 U +/- 7.1, whereas in two cases with AIDS enc it was of 10 U. A tendency to increase in median ADA values with the progression of the disease was observed. In LAS patients the ADA values presented two distinct subsets falling below and above the cut-off line of 15 U/10(9) erythrocytes, respectively. A specific correlation to drug addition and its duration was observed: LAS subjects who discontinued drug abuse (median addiction time: 3 years) presented ADA values (median = 13 U) that are lower than for addicts (median = 27.2 U; median addiction time = 7 years) and are close to those observed for asymptomatic HIV-I serum-positive group. Evidence was also obtained for a progressive increase of ADA values of LAS patients with disappearance of the product of gag gene. These results suggest that LAS subjects with elevated ADA activities present a longer history of HIV-I infection and a higher probability of developing AIDS.

Clinical utility of terminal deoxynucleotidyl transferase and adenosine deaminase determinations in adult leukemia with a lymphoid phenotype.

A prospective study was done to assess the clinical utility of terminal deoxynucleotidyl transferase (TdT) and adenosine deaminase (ADA) assays in adult leukemia with a lymphoid phenotype. The study population consisted of 58 patients with adult lymphoblastic leukemia (ALL) at onset, 12 with lymphoblastic lymphoma, 15 with acute unclassifiable leukemia (AUL), and 30 with lymphoid or mixed acute phase of Philadelphia chromosome-positive (Ph' +) chronic myelogenous leukemia (CML). TdT was present in all cases of T-ALL, in 90% of non-T, non-B ALL, and absent in B-ALL; the ADA activity was significantly higher (P less than .01) in T-ALL. TdT was found in 75% of lymphoblastic lymphomas, in 78% of lymphoid, and in 50% of mixed CML transformations; higher ADA activity correlated with TdT positivity in AUL and CML blastic transformations (P less than .001). TdT-positive ALL had a better chance of response to therapy than TdT-negative ALL (P less than 0.01), but survival was not statistically different. TdT was undetectable in the peripheral blood of patients with ALL in complete remission and within the normal range in bone marrow (0.1%-8% of nucleated cells); median ADA activity was as in control subjects. Relapsing ALL patients had TdT and ADA enzymatic activities as before therapy; TdT immunofluorescence test (IF) was positive in 69% of bone marrow and in 100% of CNS relapses. Twenty percent of TdT-positive ALL at onset became TdT-negative in bone marrow at relapse. TdT IF test was instrumental in detecting meningeal leukemia but neither TdT nor ADA could be used as indicators of complete remission or impending relapse because TdT-positive cells were present in normal marrows and wide fluctuations of TdT IF values and of ADA activity were observed in remission.

Sequence analysis of heteropolymeric DNA synthesized in vitro by the enzyme terminal deoxynucleotidyl transferase and cloned in Escherichia coli.

We have synthesized in vitro single strands of DNA in reaction mixtures containing the terminal deoxynucleotidyl transferase (Bollum's enzyme), an oligo-dG as primer, the four common deoxynucleoside triphosphates and both Mg and Co ions. The resulting heteropolymers have been converted into double strands, tailed with oligo-dC sequences, annealed with an oligo-dG tailed plasmid vector and cloned in E. coli. Six recombinant plasmids have been isolated and characterized. Two of them have been sequenced. The heteropolymeric chains produced by the terminal transferase, ranging in size between 200 and 400 nucleotides, are richer in purines than in pyrimidines, except in the last portions. Open reading frames for 7-20 amino acids with repeated, in phase translational stop codons are present in these sequences and in their complements.

DNA topoisomerase I from mitochondria of Xenopus laevis oocytes.

Activity of DNA topoisomerase I has been characterized in extracts of mitochondria purified from Xenopus laevis oocytes. Several lines of evidence have been obtained for the intramitochondrial localization of the enzyme. The mitochondria-associated of DNA topoisomerase I represents 1% of the activity recovered from a total ovary population of oocytes. The enzymes has been purified by DEAE-cellulose, phosphocellulose and double-stranded DNA cellulose chromatography and its properties compared to those of its nuclear-cytosolic counterpart purified by the same purification steps. Both enzymes appear to possess very similar if not identical physico-chemical and catalytic properties. Neither enzyme shows DNA gyrase activity and they are both equally sensitive towards trypanocide drugs such as ethidium bromide and berenil. The mitochondrial enzyme is able to remove positive superhelical turns and possibly acts as a swivelase during replication of mitochondrial DNA. These results confirm the pioneering work of Fairfield et al. (1979, J Biol. Chem. 254, 9354) on the presence of a DNA topoisomerase I in mitochondria. However our results support the conclusion that in Xenopus laevis oocytes the mitochondrial and nuclear cytosolic DNA topoisomerase I are in all likelihood one and the same enzyme.