2'-Fluoro-arabinonucleic Acid (FANA): A Versatile Tool for Probing Biomolecular Interactions.

This Account highlights the structural features that render 2'-deoxy-2'-fluoro-arabinonucleic acid (FANA) an ideal tool for mimicking DNA secondary structures and probing biomolecular interactions relevant to chemical biology.The high binding affinity of FANA to DNA and RNA has had implications in therapeutics. FANA can hybridize to complementary RNA, resulting in a predominant A-form helix stabilized by a network of 2'F-H8(purine) pseudohydrogen bonding interactions. We have shown that FANA/RNA hybrids are substrates of RNase H and Ago2, both implicated in the mechanism of action of antisense oligonucleotides (ASOs) and siRNA, respectvely. This knowledge has helped us study the conformational preferences of ASOs and siRNA as well as crRNA in CRISPR-associated Cas9, thereby revealing structural features crucial to biochemical activity.Additionally, FANA is of particular use in stabilizing noncanonical DNA structures. For instance, we have taken advantage of the anti N-glycosidic bond conformation of FANA monomers to induce a parallel topology in telomeric G-quadruplexes. Subsequent single-molecule FRET studies elucidated the mechanism by which these parallel G-quadruplexes are recognized and extended by telomerase. Similarly, we have utilized FANA to stabilize elusive telomeric i-motifs in the presence of concomitant parallel G-quadruplexes and under physiological conditions, thereby reinforcing their potential relevance to telomere biology. In another study, we adapted microarray technology and used FANA substitutions to enhance the binding affinity of the G-quadruplex thrombin-binding aptamer to its thrombin target.Finally, we discovered that DNA polymerases can synthesize FANA strands from DNA templates. On the basis of this property, other groups demonstrated that FANA, like DNA, can store hereditary information. They did so by engineering polymerases to efficiently transfer genetic information from DNA to FANA and retrieve it back into DNA. Subsequent studies showed that FANA could be evolved to acquire ribozyme-like endonuclease or ligase activity and to form high-affinity aptamers.Overall, the implications of these studies are remarkable because they promise a deeper understanding of human biochemistry for innovative therapeutic avenues. This Account summarizes past achievements and provides an outlook for inspiring the increased use of FANA in biological applications and fostering interdisciplinary collaborations.

Activation of guanine-ß-D-arabinofuranoside and deoxyguanosine to triphosphates by a common pathway blocks T lymphoblasts at different checkpoints.

The deoxyguanosine (GdR) analog guanine-ß-d-arabinofuranoside (araG) has a specific toxicity for T lymphocytes. Also GdR is toxic for T lymphocytes, provided its degradation by purine nucleoside phosphorylase (PNP) is prevented, by genetic loss of PNP or by enzyme inhibitors. The toxicity of both nucleosides requires their phosphorylation to triphosphates, indicating involvement of DNA replication. In cultured cells we found by isotope-flow experiments with labeled araG a rapid accumulation and turnover of araG phosphates regulated by cytosolic and mitochondrial kinases and deoxynucleotidases. At equilibrium their partition between cytosol and mitochondria depended on the substrate saturation kinetics and cellular abundance of the kinases leading to higher araGTP concentrations in mitochondria. dGTP interfered with the allosteric regulation of ribonucleotide reduction, led to highly imbalanced dNTP pools with gradual inhibition of DNA synthesis and cell-cycle arrest at the G1-S boundary. AraGTP had no effect on ribonucleotide reduction. AraG was in minute amounts incorporated into nuclear DNA and stopped DNA synthesis arresting cells in S-phase. Both nucleosides eventually induced caspases and led to apoptosis. We used high, clinically relevant concentrations of araG, toxic for nuclear DNA synthesis. Our experiments do not exclude an effect on mitochondrial DNA at low araG concentrations when phosphorylation occurs mainly in mitochondria.

Interactions between 2-fluoroadenine 9-beta-D-arabinofuranoside and the kinase inhibitor UCN-01 in human leukemia and lymphoma cells.

Interactions between the purine analogue 2-fluoroadenine 9-beta-D-arabinofuranoside (F-ara-A) and the kinase inhibitor UCN-01 have been examined in human leukemia cells (U937 and HL-60) with respect to induction of mitochondrial damage, caspase activation, apoptosis, and loss of clonogenic survival. Simultaneous or subsequent exposure of F-ara-A-treated cells (2 microM) to UCN-01 (100 nM) resulted in a marked potentiation of apoptosis, manifested by loss of mitochondrial membrane potential (delta psi(m)), cleavage/activation of procaspase-9 and procaspase-3, DNA fragmentation, and degradation of poly-ADP(ribosyl) polymerase. Coadministration of UCN-01 with F-ara-A was also associated with diminished phosphorylation of the cdc25 phosphatase. In contrast, exposure of cells to the sequence UCN-01, followed by F-ara-A, resulted in only a modest increase in apoptotic cells. The ability of UCN-01 to potentiate F-ara-A-mediated lethality was not mimicked by the selective PKC inhibitor bisindolylmaleimide, nor did treatment of cells with UCN-01 enhance formation of F-ara-ATP or increase incorporation of [3H]F-ara-A into DNA. Enhanced apoptosis in cells exposed sequentially or simultaneously to F-ara-A and UCN-01 was accompanied by a substantial reduction in colony formation (e.g., to 0.01% of control values). Cotreatment with UCN-01 also increased F-ara-A-mediated apoptosis and loss of delta psi(m) in U937 cells ectopically expressing Bcl-2, although not to the same extent as that observed in empty-vector controls. Finally, simultaneous exposure (24 h) of malignant B lymphocytes from the pleural effusion of a patient with indolent non-Hodgkin's lymphoma to F-ara-A and UCN-01 ex vivo resulted in a striking increase in apoptosis, as determined by terminal deoxynucleotidyltransferase-mediated nick end labeling assay. These findings indicate that UCN-01 increases F-ara-A-induced mitochondrial damage and apoptosis in human leukemia cells in a sequence-dependent manner, and that these events occur in at least some primary human lymphoma cells.

Regulation of 1-beta-D-arabinofuranosylcytosine 5'-triphosphate accumulation in human leukemia cells by deoxycytidine 5'-triphosphate.

Cell cycle-specific fluctuations in the ability of human leukemic cells to phosphorylate 1-beta-D-arabinofuranosylcytosine (ara-C) to the toxic metabolite 1-beta-D-arabinofuranosylcytosine 5'-triphosphate (ara-CTP) was investigated in whole cells and in cell extracts. Exponentially growing CCRF-CEM cells were fractionated into populations enriched for G1 phase cells and S phase cells by centrifugal elutriation. The accumulation of ara-CTP by S phase-enriched cells was 50% greater than in G1-enriched cells. However, the ability of extracts of S phase-enriched cells to phosphorylate ara-C was twice that of G1 phase-enriched cell extracts. As cells passed from G1 to S phase, this disproportionality was significant. As demonstrated in other cell types, deoxycytidine 5'-triphosphate (dCTP) also potently inhibited ara-C phosphorylation in CCRF-CEM cell extracts (Ki = 5.9 microM). Deoxynucleotide pool levels determined by high pressure liquid chromatography showed a 5 microM dCTP concentration in G1-enriched cells, whereas S phase-enriched cells contained 15 microM dCTP. These findings suggest that the lack of proportionality between the accumulation of ara-CTP in whole cells and the increase of ara-C phosphorylation in extracts during the G1 to S phase transition may be caused by more stringent regulation of ara-C phosphorylation in whole cells by the concomitant increase in cellular dCTP concentrations. Because such regulation is unlikely to be observed in cell extracts, these results indicated that assays of ara-C phosphorylating activity in cell extracts represent upper limits for that function in whole cells. Such determinations may not reflect the regulated nature of the metabolic pathway.

Deoxyguanosine enhancement of cytarabine nucleotide accumulation in human leukemia cells.

We studied the effect of deoxyguanosine (dGuo) on cellular cytarabine (ara-C) nucleotide accumulation of the human leukemia cell line K562 and of bone marrow blast cells derived from patients with acute nonlymphocytic leukemia. Exposure of cells in culture to dGuo increased ara-C nucleotide accumulation measured in cell lysate, with an average increase of 386% (range, 242%-537%) of control in the presence of 500 microM dGuo. Maximal elevation occurred after 8 hours of exposure and remained constant through 48 hours. dGuo also enhanced deoxycytidine nucleotide accumulation, but dGuo enhancement favored accumulation of ara-C nucleotides over dCyd nucleotides. In cell cycle kinetic studies using flow cytometry, dGuo slowed accumulation of cells with apparent S-phase DNA content in a concentration-dependent fashion. However, neither the rate nor the magnitude of this effect correlated with the increase in ara-C nucleotide accumulation. Since the increase in ara-C nucleotide accumulation caused by dGuo could be prevented by 5 micrograms/ml of cycloheximide, this process appears to require new protein synthesis. Although these data suggest that the elevation of ara-C nucleotide accumulation caused by dGuo may represent induction of enzyme synthesis, other possibilities are discussed. Exposure of bone marrow blast cells obtained from patients with acute nonlymphocytic leukemia to dGuo for 16 hours in liquid culture also increased ara-C nucleotide accumulation. In six of seven studies, exposure to dGuo in concentrations from 50 to 500 microM increased ara-C nucleotide accumulation from 160% to 3400%. These data suggest that dGuo may alter ara-C metabolism in a clinically useful fashion.

Enhanced production of intracellular ara-CTP after sequential use of methotrexate and 1-beta-D-arabinofuranosylcytosine.

Synergistic antiproliferative effect has been proven in vitro when mouse leukemic cells were sequentially treated with MTX and ara-C. The mechanism of this combination effect not well elucidated but the intracellular uptake of ara-C was higher when cells were pre-exposed to MTX. In this experiment, the intracellular ara-CTP was measured by HPLC after MTX and ara-C were sequentially administered to BDF1 mice bearing L1210 leukemic cell, either being sensitive or resistant to MTX. When MTX at the dose of 12 mg/kg was preceded 6 h and 3 h to ara-C at the dose of 25 mg/kg, the intracellular levels of ara-CTP were found to be significantly higher as compared with those of ara-C alone as control group. At 1 h after ara-C, ara-CTP was measured about 165 and 130% of the control levels, respectively, and at 12 h, ara-CTP was over 4 times higher of control level with group of mice to which MTX was preceded 6 h prior to MTX. On the contrary, the enhancement of ara-CTP production was definitely diminished with MTX-resistant cells in the same administrative model. From our present experiment, the time sequential modulation of intracellular ara-CTP production by MTX was reconfirmed in vivo, and this modulation might depend upon the sensitivity of MTX of leukemic cells.