In vitro and in vivo toxicity of 2',3'-dideoxycytidine in mice.

3T3 mouse embryo fibroblast cell growth was inhibited in a concentration dependent manner by 2',3'-dideoxycytidine (ddCyd), a strong inhibitor of human immunodeficiency virus. Cell growth inhibition was associated with an increased incorporation of ddCyd into cell DNA. In contrast SP2/0-Ag14 (a mouse myeloma) cell growth is not inhibited by 100 microM ddCyd both in the presence or absence of hypoxanthine and thymidine. Furthermore, in vitro spleen cell proliferation, upon phytohemagglutinin (PHA) addition, was much more affected by ddCyd in C57BL/6 mice than in Swiss albino mice. That indeed ddCyd affects spleen cell proliferation was confirmed by studies on splenocytes obtained from C57BL/6 mice that received ddCyd for 2 weeks in drinking water. These results suggest that ddCyd toxicity in mice is cell and strain dependent and that the toxicity mechanism is related to the incorporation of the drug in cell DNA.     

In vitro toxicity and metabolism of 2',3'-dideoxycytidine, an inhibitor of human immunodeficiency virus infectivity.

U937 human monoblastoid cell growth was inhibited in a concentration-dependent manner by 2',3'-dideoxycytidine (ddCyd) (an antiretroviral drug) up to 500 microM. Cell growth inhibition was associated with a pronounced increase in cell volume, however this was not due to cell ATP or NAD+ depletion that could effect osmotic balance or DNA repair. This ddCyd toxicity paralleled the accumulation of ddCyd into acid soluble material where 2',3'-dideoxycytidine-5'-triphosphate (ddCTP) was the predominant labelled nucleotide up to an extracellular ddCyd concentration of 150 microM. At higher ddCyd concentrations, the amount of 2',3'-dideoxycytidine-5'-diphosphate (ddCDP) became predominant over ddCTP. This increase of phosphorylated dideoxycytidine in U937 cells was also associated with an increased incorporation of the drug into cell DNA suggesting a possible toxicity mechanism. That ddCyd does indeed become cytotoxic to human cell by incorporation into DNA was shown by incubating human resting and stimulated lymphocytes with ddCyd. While the drug does not affect cell viability in resting cells it strongly affects cell proliferation upon phytohemagglutinin (PHA) addition.     

Both 2',3'-dideoxythymidine and its 2',3'-unsaturated derivative (2',3'-dideoxythymidinene) are potent and selective inhibitors of human immunodeficiency virus replication in vitro

2',3'-Dideoxythymidine (ddThd) and its 2',3'-unsaturated derivative 2',3'-dideoxythymidinene (ddeThd) are potent and selective inhibitors of human immunodeficiency virus (HIV) in vitro. When evaluated for their inhibitory effects on the cytopathogenicity of HIV in MT-4 cells, ddThd and ddeThd completely protected the cells against destruction by the virus at a concentration of 1 microM and 0.04 microM, respectively. In this aspect, ddeThd was about 5 times more potent than 2',3'-dideoxycytidine (ddCyd), one of the most potent and selective anti-HIV compounds now pursued for its therapeutic potential in the treatment of AIDS. ddThd and ddeThd also suppressed HIV antigen expression at 1 microM and 0.04 microM, respectively. Their selectivity indexes, as based on the ratio of the 50% cytotoxic dose to the 50% antiviral effective dose, were 120 (ddeThd) and greater than 625 (ddThd).     

2',3'-dideoxycytidine permeation of the human erythrocyte membrane

The mechanism by which 2,3'-dideoxycytidine, an inhibitor of HIV-I infectivity, permeates the cell membrane was investigated. The influx of ddCyd into human erythrocytes was nonconcentrative. The initial velocity of both ddCyd influx and efflux was, in contrast to compounds that permeate the cell membrane via the nucleoside transporter, a linear function of nucleoside concentration in the 1 microM to 10 mM range and relatively insensitive to temperature. Furthermore, potent inhibitors of nucleoside transporter and other nucleosides were found to inhibit ddCyd influx only partially or not at all suggesting that ddCyd permeates the human erythrocyte membrane predominantly by nonfacilitated diffusion. This unusual characteristic seems to be due to the lack of 3'-hydroxyl moiety of ddCyd which appears to be an important determinant for the nucleoside carrier specificity rather than to lipid solubility itself. As far as permeation of the cell membrane is concerned ddCyd shares these properties with 2',3'-dideoxythymidine and 3'-azido-3'-deoxythymidine.     

Potent and selective anti-HTLV-III/LAV activity of 2',3'-dideoxycytidinene, the 2',3'-unsaturated derivative of 2',3'-dideoxycytidine

2',3'-Dideoxycytidinene (ddeCyd), the 2',3'-unsaturated derivative of 2',3'-dideoxycytidine (ddCyd) is, like ddCyd itself, a potent and selective inhibitor of HTLV-III/LAV in vitro. This conclusion is based on the relatively high ratio of effective antiviral dose (0.3 microM) versus cell growth inhibitory concentration (20-35 microM) and the lack of any appreciable inhibitory activity against a series of non-oncogenic RNA and DNA viruses. Both compounds were considerably more inhibitory to human lymphoid cell lines than human nonlymphoid or murine cell lines. They were highly dependent on prior activation by deoxycytidine kinase to exert their anti-HTLV-III/LAV and cytostatic effects. In contrast with ddCyd, ddeCyd lost part of its anti-retrovirus effect upon prolonged incubation (10 days) with the virus-infected cells in culture.     

Genetic analysis of 2',3'-dideoxycytidine incorporation into cultured human T lymphoblasts

In order to analyze the cellular determinants that mediate the action of 2',3'-dideoxycytidine, the growth inhibitory and cytotoxic effects and the metabolism of the dideoxynucleoside were examined in wild type human CEM T lymphoblasts and in mutant populations of CEM cells that were genetically deficient in either nucleoside transport or deoxycytidine kinase activity. Whereas 2',3'-dideoxycytidine at a concentration of 5 microM inhibited growth of the wild type CEM parental strain by 50%, two nucleoside transport-deficient clones were 4-fold resistant to the pyrimidine analog. The deoxycytidine kinase-deficient cell line was virtually completely resistant to growth inhibition by the dideoxynucleoside at a concentration of 1024 microM. An 80% diminished rate of 2',3'-[5,6-3H]dideoxycytidine influx into the two nucleoside transport-deficient lines could account for their resistance to the dideoxynucleoside, while the resistance of the deoxycytidine kinase-deficient cells to 2',3'-dideoxycytidine toxicity could be explained by a virtually complete failure to incorporate 2',3'-[5,6-3H]dideoxycytidine in situ. Two potent inhibitors of mammalian nucleoside transport, 4-nitrobenzylthioinosine and dipyridamole, mimicked the effects of a genetic deficiency in nucleoside transport with respect to 2',3'-dideoxycytidine toxicity and incorporation. These data indicate that the intracellular metabolism of 2',3'-dideoxycytidine in CEM cells is initiated by the nucleoside transport system and the cellular deoxycytidine kinase activity.     

Human red blood cells as bioreactors for the release of 2',3'-dideoxycytidine, an inhibitor of HIV infectivity

2',3'-Dideoxycytidine (ddCyd) is one of the most potent antiviral nucleosides for killing the human immunodeficiency virus (HIV). ddCyd is currently used in the treatment of severe HIV infections but due to its rapid clearance it must be administered to patients every 4 h reaching concentrations that are toxic. We have synthesized 2',3'-dideoxycytidine-5'-phosphate (ddCMP) as a prodrug, encapsulated it in human erythrocytes and found that it is dephosphorylated by endogenous pyrimidine nucleotidases and subsequently released by the cells as ddCyd. Encapsulated ddCMP does not affect erythrocyte metabolism and was not deaminated by cytidine deaminase. The dephosphorylation reaction has an apparent Km of 6mM, an optimum pH of 6.8 and is not inhibited by ATP or 2,3-bisphosphoglycerate. The efflux of ddCyd from the erythrocyte is a linear function of ddCyd concentration and relatively insensitive to nucleoside transporter inhibitors suggesting that ddCyd permeates the erythrocyte membrane predominantly by nonfacilitated diffusion. Thus, ddCMP-loaded erythrocytes might be used as endogenous bioreactors for ddCyd delivery in the treatment of HIV infection.     

Deoxycytidine deaminase-resistant stereoisomer is the active form of (+/-)-2',3'-dideoxy-3'-thiacytidine in the inhibition of hepatitis B virus replication.

2',3'-Dideoxy-3'-thiacytidine (+/-)-SddC) was found to have potent activity against human hepatitis B virus as well as human immunodeficiency viruses in culture. The (-)form ((-)-SddC) which is resistant to deoxycytidine deaminase was found to be the more active antiviral stereoisomer than the (+)-form ((+)-SddC). The (+)-SddC is susceptible to deamination by deoxycytidine deaminase and is 25- and 12-fold more toxic than (-)-SddC in CEM cells in terms of anti-cell growth and anti-mitochondrial DNA synthesis, respectively. Similar results were obtained using a mixture of their 5-fluoro analogs ((+/-)-FSddC). Unlike 2',3'-dideoxycytidine, which is a potent inhibitor of mitochondrial DNA synthesis and results in such delayed toxicity as peripheral neuropathy with long term usage, (-)-SddC does not affect mitochondrial DNA synthesis. The (-)form is phosphorylated to (-)-SddCMP and is subsequently converted to (-)-SddCDP and (-)-SddCTP. One additional major metabolite which has been tentatively assigned the name "(-)-SddCMP sialate" was also identified. No significant difference in terms of the profiles of the metabolites was found between 4 and 24 h. There is an appreciable amount of (-)-SddCTP detectable 24 h after removal of the drug. (-)-SddCTP was also found to be approximately 3-fold more potent than (+)-SddCTP in inhibiting human hepatitis B virus DNA polymerase. This is the first nucleoside analog with the unnatural sugar configuration demonstrated to have antiviral activity.     

Diverging substrate specificity of pure human thymidine kinases 1 and 2 against antiviral dideoxynucleosides

The two thymidine (dThd) kinases in human cells, the cytosolic, S-phase-specific TK1 and the mitochondrial, constitutively expressed TK2 were purified to homogeneity as judged from sodium dodecyl sulfate-gel electrophoresis. The substrate specificity of TK1 and TK2 toward natural substrates and important nucleoside analogues was compared. With TK1, the Km values for 5-fluorodeoxyuridine (FdUrd), 3'-azido-2',3'-dideoxythymidine (AZT), and 3'-fluoro-2',3'-dideoxythymidine (FLT) were 2.2, 0.6, and 2.1 microM as compared to 0.5 microM for dThd and 9 microM for deoxyuridine (dUrd). With TK2, dUrd, deoxycytidine (dCyd), and 5-fluorodeoxyuridine (FdUrd) were efficiently phosphorylated, but with distinctly different kinetics: Michaelis-Menten kinetics with dCyd, dUrd, and FdUrd; negative cooperativity with dThd. Negative cooperativity was also observed with AZT, although this drug was a very poor substrate for TK2 with a Vmax of 5-6% of that with dThd. FLT, 2',3'-dideoxycytidine (ddCyd), and arabinofuranosylcytosine (araC) were not substrates for TK2, and 2',3'-didehydrodideoxy-thymidine (D4T) was not a substrate for TK1 or TK2. On the other hand, AZT, FLT, and D4T were competitive inhibitors with Ki values of 0.6, 6, and 2073 microM for TK1, and 2, 10, and 78 microM for TK2, respectively. The much lower tolerance for modifications of the deoxyribose moiety of TK2 as compared to TK1 is important for the design of new antiviral nucleoside analogues intended for use in cells with different expression of TK1 and TK2.     

Metabolism of 4'-azidothymidine. A compound with potent and selective activity against the human immunodeficiency virus.

4'-Azidothymidine (ADRT) is a novel nucleoside analog, that selectively inhibits human immunodeficiency virus replication in human lymphocytes. Unlike the dideoxyribonucleoside analogs and 3'-azido-2',3'-dideoxythymidine (AZT), ADRT retains the 3'-hydroxy group. The pathways of ADRT metabolism were elucidated by determining: (i) the kinetics of the interactions of ADRT and its metabolites with enzymes of thymidine metabolic pathways, (ii) the pool sizes of phosphorylated metabolites, and (iii) the nature of ADRT incorporation into human DNA. ADRT is not a substrate for thymidine phosphorylase, but is metabolized by kinases. Thymidine kinase phosphorylates ADRT to ADRT monophosphate (ADRT-MP). For this enzyme, ADRT has a Ki value of 5.2 microM, in comparison to a Km value of 0.7 microM for thymidine. The Km value of ADRT toward thymidine kinase is 8.3 microM and the rate of ADRT phosphorylation is 1.4% that of thymidine phosphorylation. ADRT-MP has a low affinity toward thymidylate kinase (a Ki value of 28.9 microM versus a Km value of 0.56 microM for thymidylate), and toward thymidylate synthase (a Ki value of 180 microM versus a Km value of 8 microM for deoxyuridylate). The results suggest that ADRT can be activated effectively by cellular kinases without significant interference of normal thymidine metabolism. In cultured human lymphocytes (A3.01, H9, and U937 cells), ADRT was phosphorylated efficiently to ADRT 5'-triphosphate (ADRT-TP), which is the major metabolite of ADRT. The intracellular concentrations of ADRT-TP ranged from 1 to 3.3 microM after 24 h of incubation with 2 microM of ADRT and the half-life of ADRT-TP varied from 3 to 6 h. Although ADRT-TP is a poor competitive inhibitor against dTTP toward DNA polymerases alpha and beta with Ki values of 62.5 and 150 microM, respectively. ADRT-MP was found to be internally incorporated into cellular DNA. The extent of ADRT-MP substitution for dTMP in DNA was 1 in 6979 for A3.01 cells incubated with 2.9 microM ADRT for 24 h. Internal incorporation of ADRT-MP contrasts with the mechanism of other 2',3'-dideoxynucleoside analogs (i.e. AZT, ddC, ddI, d4T...), which are DNA chain terminators. This finding indicates that a 3'-deoxy structure in a nucleoside analog is not a prerequisite for anti-human immunodeficiency virus activity.     

Nucleobase transporter-mediated permeation of 2',3'-dideoxyguanosine in human erythrocytes and human T-lymphoblastoid CCRF-CEM cells.

Several 2',3'-dideoxynucleosides (ddNs), agents that inhibit the replication of human immunodeficiency virus and hepatitis B virus, enter mammalian cells by simple diffusion. In this report, we show that the membrane permeation of 2',3'-dideoxyguanosine (ddG) in human erythrocytes and CCRF-CEM cells, in contrast with that of other ddNs, is transporter-mediated. Inward fluxes of ddG in both cell types were inhibited by adenine, hypoxanthine, and acyclovir, but not by inhibitors of nucleoside transport (nitrobenzylthioinosine, dipyridamole, dilazep). Fluxes of ddG in human erythrocytes were attributable to a single, rate-saturable process (Km, 380 +/- 90 microM and Vmax, 7.9 +/- 0.8 pmol/s/microliter cell water) that was competitively inhibited by adenine (Ki, 16 microM). These results showed that ddG entered human erythrocytes and CCRF-CEM cells by a transporter-mediated process that was also the basis for entry of purine nucleobases. In contrast, inward fluxes of 2,6-diaminopurine-2',3'-dideoxyriboside (ddDAPR), a prodrug of ddG, were not affected by purine nucleobases or nucleoside transport inhibitors in either cell type. Thus, the permeation properties of ddDAPR resembled those of 2',3'-dideoxyadenosine, a diffusional permeant (cell uptake is transporter-independent), and contrasted with those of ddG, the deamination product of ddDAPR. This study demonstrated that the nucleobase moiety of ddNs is an important determinant of membrane permeation.     

Unusual structural features of 2',3'-dideoxycytidine, an inhibitor of the HIV (AIDS) virus

The structure and conformation of 2',3'-dideoxycytidine, a potent inhibitor of the human immunodeficiency virus, was determined by X-ray crystallography. The nucleoside crystallizes in the tetragonal space group P4(1)2(1)2 with cell dimensions a = b = 8.698(4) and c = 26.155(9) A. Atomic parameters were refined by full-matrix least squares to a final value of R = 0.037 for 1926 observed reflections. The conformation of the furanose ring corresponds to the unusual C3'exo/C4'endo (3T4) pucker, similar to that found in one of the molecules of 3'-azidothymidine (AZT). The glycosidic torsion angle is also smaller than expected. The relevance of these unusual structural features to anti-AIDS activity is assessed.     

Cross-resistance to anti-HIV nucleoside analogs in multidrug-resistant human cells

Human multidrug-resistant K562/ADM cells showed 12-fold and 31-fold resistance to AZT (3'-azido-2', 3'dideoxythymidine) and DDC (2', 3'-dideoxycytidine), respectively. Other multidrug-resistant human cells CEM/VLB100 and AdrRMCF-7 also showed resistance to these nucleoside analogs. However, verapamil (10 microM) failed to reverse the resistance to the nucleoside analogs. Accumulation of [3H]AZT in human multidrug-resistant K562/ADM, CEM/VLB100 and AdrRMCF-7 cells decreased by 23, 35, and 42% respectively, as compared to their parental cells. These results suggest that anti-HIV nucleoside analogs including AZT, DDC could be transported by outward drug-transport system in the multidrug-resistant cells.     

A novel action of human apurinic/apyrimidinic endonuclease: excision of L-configuration deoxyribonucleoside analogs from the 3' termini of DNA

beta-l-Dioxolane-cytidine (l-OddC, BCH-4556, Troxacitabine) is a novel unnatural stereochemical nucleoside analog that is under phase II clinical study for cancer treatment. This nucleoside analog could be phosphorylated and subsequently incorporated into the 3' terminus of DNA. The cytotoxicity of l-OddC was correlated with the amount of l-OddCMP in DNA, which depends on the incorporation by DNA polymerases and the removal by exonucleases. Here we reported the purification and identification of the major enzyme that could preferentially remove l-OddCMP compared with dCMP from the 3' termini of DNA in human cells. Surprisingly, this enzyme was found to be apurinic/apyrimidinic endonuclease (APE1) (), a well characterized DNA base excision repair protein. APE1 preferred to remove l- over d-configuration nucleosides from 3' termini of DNA. The efficiency of removal of these deoxycytidine analogs were as follows: l-OddC > beta-l-2',3'-dideoxy-2', 3'-didehydro-5-fluorocytidine > beta-l-2',3'-dideoxycytidine > beta-l-2',3'-dideoxy-3'-thiocytidine > beta-d-2',3'-dideoxycytidine > beta-d-2',2'-difluorodeoxycytidine > beta-d-2'-deoxycytidine >/= beta-d-arabinofuranosylcytosine. This report is the first demonstration that an exonuclease can preferentially excise l-configuration nucleoside analogs. This discovery suggests that APE1 could be critical for the activity of l-OddC or other l-nucleoside analogs and may play additional important roles in cells that were not previously known.     

Transport of antiviral 3'-deoxy-nucleoside drugs by recombinant human and rat equilibrative, nitrobenzylthioinosine (NBMPR)-insensitive (ENT2) nucleoside transporter proteins produced in Xenopus oocytes.

In the present study, one has determined the relative role of plasma membrane equilibrative (Na+-independent) ENT nucleoside transport proteins (particularly ENT2) in the uptake of antiviral nucleoside analogues for comparison with the previously reported drug transport properties of concentrative (Na+-dependent) CNT nucleoside transport proteins. The human and rat nucleoside transport proteins hENT1, rENT1, hENT2 and rENT2 were produced in Xenopus oocytes and investigated for their ability to transport three 3'-deoxy-nucleoside analogues, ddC (2'3'-dideoxycytidine), AZT (3'-azido-3'-deoxythymidine) and ddI (2'3'-dideoxyinosine), used in human immunodeficiency virus (HIV) therapy. The results show, for the first time, that the ENT2 transporter isoform represents a mechanism for cellular uptake of these clinically important nucleoside drugs. Recombinant h/rENT2 transported ddC, ddI and AZT, whilst h/rENT1 transported only ddC and ddI. Relative to uridine, h/rENT2 mediated substantially larger fluxes of ddC and ddI than h/rENT1. Transplanting the amino-terminal half of rENT2 into rENT1 rendered rENT1 transport-positive for AZT and enhanced the uptake of both ddC and ddI, identifying this region as a major site of 3'-deoxy-nucleoside drug interaction.     

Changes in purine nucleoside phosphorylase activity during thymosin-induced human null cell differentiation

Purine nucleoside phosphorylase (PNP) is a purine salvage pathway enzyme which we have found to be 8-10 times more active (per cell) in human peripheral blood null lymphocytes than in T lymphocytes. To test the hypothesis that null cells are, in part, pre-T lymphocytes we have defined an in vitro system for null cell differentiation into T cells and examined PNP activity during this differentiation process. We found that about 10% of human null cells could be driven to differentiate into T cells using thymosin fraction 5 (TF5) an extract of bovine thymus glands. The response to TF5 was dose related to up to 250 micrograms/ml with a maximum response occurring by 42-46 hr incubation. Exposure to TF5 was necessary for more than 4 hr but no more than 8 hr in order to obtain a maximum response. Both OKT4 and OKT8 positive cells were present in the newly differentiated T cell population but OKT8 positive cells appeared to predominate (OKT4/OKT8 = 0.698 +/- 0.30, mean +/- 1 SD). The differentiation process did not involve DNA synthesis but was inhibited at 4 degrees C. In the newly differentiated T cells PNP activity per cell was 8- to 10-fold lower (36 +/- 23 nm/hr/106 cells) than in null cells (311 +/- 136), and was at a level similar to mature T cells (56 +/- 7). Thus, human peripheral blood null cells can be induced to differentiate into T lymphocytes which can be characterized by both surface markers and biochemical parameters. Future studies will look at the function of TF5-induced T cells and the regulation of PNP activity during the differentiation process.