Antiretroviral therapy-induced mitochondrial toxicity: potential mechanisms beyond polymerase-γ inhibition.

We hypothesized that competition between nucleotide reverse-transcriptase inhibitor triphosphate and endogenous deoxyribonucleotide triphosphate (dNTP) may lead to depletion of dNTP pools and mitochondrial dysfunction independent of polymerase-γ (pol-γ) inhibition. We collected peripheral blood mononuclear cells from 75 adults (25 cases: HIV-infected patients with mitochondrial toxicity, 25 HIV-infected positive controls, and 25 HIV-negative controls). We observed statistically significant individual and group differences in ribonucleotide (RN) and deoxyribonucleotide (dRN) pools. The median values for the RN pools were 10,062 (interquartile range (IQR): 7,090-12,590), 4,360 (IQR: 3,058-6,838), and 2,968 (IQR: 2,538-4,436) pmol/10(6) cells for negative controls, positive controls, and cases, respectively. Cases had significantly higher absolute mitochondrial DNA copy number as compared with negative controls (P < 0.05). Moreover, cases had significantly higher expression levels of pol-γ, nucleotide transporters, cellular kinases, and adenosine triphosphate (ATP)-binding cassette (ABC) proteins as compared with controls. Antiretroviral therapy (ART) perturbs RN and dRN pools. Depletion of RN and dRN pools may be associated with ART-induced mitochondrial toxicity independent of pol-γ inhibition.

Direct determination of phosphorylated intracellular anabolites of stavudine (d4T) by liquid chromatography/tandem mass spectrometry.

The objective was to develop and validate a routine assay for active intracellular anabolites of stavudine (d4T), a nucleoside reverse transcriptase inhibitor in human PBMC, applicable to pharmacokinetic studies and treatment monitoring. This was achieved using liquid chromatography coupled to tandem mass spectrometry (LC/MS/MS), which theoretically allies optimum sensitivity, specificity and high sample throughput. After cellular lysis in a Tris/methanol buffer, the extract spiked with 2[H(8)]-ATP (internal standard) is directly injected into the LC/MS/MS system. Phosphorylated metabolites of d4T as well as deoxythymidine-triphosphate, the competitor on the reverse transcriptase, are separated from d4T on a reverse-phase microbore column with ion pairing. The detection is performed in the multiple reaction monitoring (MRM) mode after drug ionisation in negative mode electrospray. The limit of quantitation for d4T-TP was 138 fmol per 7 mL blood (9.8 fmol per 10(6) cells) and CV% for repeatability and intermediate precision were lower than 15%. Stability of compounds was checked before and during the process of isolation of PBMC. Cellular samples from several d4T-treated patients were successfully analysed using this method and d4T-triphosphate and deoxythymidine triphosphate were recovered. In conclusion, we have developed and validated a routine LC/MS/MS method that allows the simultaneous determination of mono-, di- and triphosphorylated anabolites of d4T in PBMC as well as the natural corresponding triphosphate in one analysis. For the first time, the chain terminator ratio (d4T-TP/dT-TP) could be directly measured. This method can be used simply and routinely on more than 35 samples per day. Extension to other nucleoside analogues is under development.

An ion-pairing high-performance liquid chromatographic method for the direct simultaneous determination of nucleotides, deoxynucleotides, nicotinic coenzymes, oxypurines, nucleosides, and bases in perchloric acid cell extracts.

An ion-pairing high-performance liquid chromatographic method for the direct and simultaneous determination of nucleotides, deoxynucleotides, cAMP, nicotinic coenzymes, oxypurines, nucleosides, and bases in perchloric acid cell extracts is presented. By using an Alltima C-18, 250 x 4.6-mm, 5-microns particle size column, a high resolution of 38 acid-soluble compounds, including ATP, GTP, dTTP, CTP, UTP, ADP, GDP, dTDP, CDP, UDP, dATP, dGTP, dCTP, dUTP, dADP, dGDP, dCDP, dUDP, and cAMP, is obtained. Elution is performed with a step gradient from buffer A (consisting of 10 mM tetrabutylammonium hydroxide, 10 mM KH2PO4, 0.25% methanol, pH 7.00) to buffer B (consisting of 2.8 mM tetrabutylammonium hydroxide, 100 mM KH2PO4, 30% methanol, pH 5.50). Perchloric acid extracts of resting and phytohemagglutinin-stimulated human lymphocytes were analyzed. Data indicate that this chromatographic method offers, for the first time to the best of our knowledge, the possibility of simultaneously determining di- and triphosphate nucleosides and their corresponding deoxynucleosides without any chemical manipulation of samples except for perchloric acid deproteinization. Hence, the present HPLC assay minimizes the risks of modification or loss of metabolite concentration and allows one to obtain, with a single chromatographic run, the complete pattern of those metabolites which are known to be involved in energy metabolism and in DNA and RNA synthesis, resulting therefore of great advantage in cell biology studies.

Low levels of deoxynucleotides in peripheral blood lymphocytes: a strategy to inhibit human immunodeficiency virus type 1 replication.

Human immunodeficiency virus type 1 (HIV-1) viral DNA synthesis in quiescent and activated peripheral blood lymphocytes (PBLs) was studied. Incomplete viral DNA (previously demonstrated to be associated with HIV-1 virions) is carried by HIV-1 virions into quiescent and activated PBLs, contributing to the formation of an early viral DNA pool in these cells. The viral DNA is subsequently completed but only extremely slowly and inefficiently in quiescent PBLs compared to that in stimulated PBLs. We find that this correlates with significantly lower levels of dNTP substrates in quiescent compared to activated PBLs. At these low dNTP concentrations, HIV-1 reverse transcriptase acts in a partially distributive manner. Increasing dNTP concentrations from the levels of quiescent PBLs to the levels of activated PBLs increases the processive action of reverse transcriptase, which in turn stimulates rapid and efficient formation of full-length DNA. Furthermore, hydroxyurea treatment of stimulated PBLs decreases the dNTP levels and the DNA synthesis rate to levels comparable to quiescent PBLs. Our data therefore indicate that low levels of dNTP may explain why HIV-1 DNA is synthesized slowly and inefficiently in quiescent PBLs and suggest that pharmacologic induction of low dNTP levels represents a therapeutic approach for inhibition of HIV-1 replication.

Differential phosphorylation of azidothymidine, dideoxycytidine, and dideoxyinosine in resting and activated peripheral blood mononuclear cells.

The antiviral activity of azidothymidine (AZT), dideoxycytidine (ddC), and dideoxyinosine (ddI) against HIV-1 was comparatively evaluated in PHA-stimulated PBM. The mean drug concentration which yielded 50% p24 Gag negative cultures were substantially different: 0.06, 0.2, and 6 microM for AZT, ddC, and ddI, respectively. We found that AZT was preferentially phosphorylated to its triphosphate (TP) form in PHA-PBM rather than unstimulated, resting PBM (R-PBM), producing 10- to 17-fold higher ratios of AZTTP/dTTP in PHA-PBM than in R-PBM. The phosphorylation of ddC and ddI to their TP forms was, however, much less efficient in PHA-PBM, resulting in approximately 5-fold and approximately 15-fold lower ratios of ddCTP/dCTP and ddATP/dATP, respectively, in PHA-PBM than in R-PBM. The comparative order of PHA-induced increase in cellular enzyme activities examined was: thymidine kinase > uridine kinase > deoxycytidine kinase > adenosine kinase > 5'-nucleotidase. We conclude that AZT, ddC, and ddI exert disproportionate antiviral effects depending on the activation state of the target cells, i.e., ddI and ddC exert antiviral activity more favorably in resting cells than in activated cells, while AZT preferentially protects activated cells against HIV infection. Considering that HIV-1 proviral DNA synthesis in resting lymphocytes is reportedly initiated at levels comparable with those of activated lymphocytes, the current data should have practical relevance in the design of anti-HIV chemotherapy, particularly combination chemotherapy.

Different pathways for deoxyguanosine toxicity in T-lymphocytes of various developmental stages.

The basis of the selective cellular immunodeficiency which occurs in patients with purine nucleoside phosphorylase (PNP) deficiency still is not completely understood. We studied the mechanism of deoxyguanosine (dGuo) toxicity in proliferating lymphoid T-cells of different maturation stage, i.e. in T-cells of adult peripheral blood and cord blood and in CD3+ and CD3- subfractions of thymocytes. The mitogen-induced proliferation of T-cells from peripheral blood and cord blood and of CD3+ and CD3- subfractions of thymocytes. The mitogen-induced proliferation of T-cells from peripheral blood and cord blood and of CD3+ thymocytes, as well as the spontaneous proliferation of CD3- thymocytes, are inhibited by dGuo. CD3+ and CD3- thymocytes are significantly more sensitive to dGuo than T-cells from peripheral blood or cord blood. Among the thymocyte subfractions CD3- thymocytes appeared to be extremely sensitive. In all cell types studied, inhibition of proliferation is accompanied by intracellular increases in both guanosine triphosphate (GTP) and deoxyguanosine triphosphate (dGTP) concentrations. By use of the PNP inhibitor 8-aminoguanosine, or the metabolites hypoxanthine or deoxycytidine, the metabolism of dGuo could be selectively directed to the formation of GTP or to dGTP. Based on the pattern of rescue from dGuo intoxication under these different metabolic conditions we conclude that in CD3- thymocytes dGuo toxicity is mediated by dGTP. In all other cell types studied GTP mediates dGuo intoxication. Altogether the results show that during the maturation from immature thymocytes to mature peripheral blood T-cells a shift occurs in the pattern of dGuo toxicity since dGuo toxicity in the former is primarily caused via the dCyd kinase pathway, and in the latter mainly the degradation route is involved. Since in PNP deficiency mature T-cells do occur in the peripheral blood, we must conclude that some cells escape the stage of T-cell maturation in the thymus which is extremely sensitive to dGuo. Furthermore, the results imply that as far as T-cell development in the normal thymus is concerned, survival and death of cells might be regulated by local (deoxy) nucleoside availability.

A new case of purine nucleoside phosphorylase deficiency: enzymologic, clinical, and immunologic characteristics.

Deficiency of purine nucleoside phosphorylase (PNP) was detected in a 3-yr-old boy who was admitted for investigation of a behavior disorder and spastic diplegia. The urinary excretion of purines, analyzed by high-performance liquid chromatography, showed the presence of large amounts of (deoxy)inosine and (deoxy)guanosine and low uric acid levels. Analysis of the (deoxy)nucleotide pools of erythrocytes showed elevated levels of deoxyguanine nucleotides and NAD and decreased guanine nucleotides. PNP activity in red blood cells was 0.1-0.5% of normal on two occasions and undetectable on four later measurements. Furthermore no immunoreactive material could be detected in his red cell lysate using an anti-PNP antiserum. PNP activities in the red cells of the patient's parents were 35 and 50% of normal. The presence of (minor) residual PNP activity in the patient enabled the investigation of some enzyme properties after partial purification. No abnormalities could be detected in substrate affinity for inosine, heat stability, and electrophoretic properties. In the heterozygous parents no signs of a mutant enzyme could be found. The molecular specific activities of the parental enzymes were also normal, indicating that no immunoreactive material attributable to inactive-mutant enzyme subunits was present. A striking feature of the patient is the prevailing neurologic abnormalities presumably caused by the metabolic disorder. A severe lymphopenia exists; however, clinical symptoms of an immune deficiency did not become apparent until the age of 4 yr.

Some biochemical mechanisms underlying the impairment of T and B cell immunity in C3HA mice during hepatoma growth.

In thymocytes of C3HA mice carrying the transplantable and ortoaminoazotoluene induced hepatomas at the time of their intense growth a drastic decrease in adenosine deaminase activity set in and 3-4-fold augmentation of intracellular concentration of dATP and dGTP, potential inhibitors of ribonucleoside diphosphate reductase was observed, leading to the reduction of the DNA synthesis. The latter event was evidenced by a suppressed 14C-thymidine incorporation into thymocytes DNA in vitro, decreased thymidine kinase activity, intracellular dTTP and depletion of dCTP pools. Only in the terminal period of hepatocarcinogenesis (12 months) a 4-fold increase in the corticosterone serum concentration was observed. As for the mice carrying transplantable 22a hepatoma, serum hormone levels augmented 4-fold as early as 24 h after tumor implantation and thereafter kept increased two fold. An elevated activity of terminal deoxynucleotidyl transferase in mouse thymocytes has been shown to be characteristic of the late periods of tumor growth reflecting the arrest of the immature cortical thymocyte differentiation. By the time hepatomas emerged in the course of hepatocarcinogenesis in spleen T and B lymphocytes a significant drop in the activity of adenosine deaminase (3-4-fold) and purine nucleoside phosphorylase (2-8-fold) was noted--the events directly correlated with the weakening of cell immune functions. The disorders described were accompanied by the accumulation of dGTP in spleen T lymphocytes, dATP in B lymphocytes and inhibition of DNA synthesis, predominantly in T lymphocytes. In the latter instance the pool of dCTP was found to be depleted. In spleen T and B lymphocytes of mice carrying solid 22a hepatoma when the peak of its growth was reached (day 5) the rate of DNA synthesis dropped. Later on (from day 8 to the animal death), however, in spite of the suppression of immune function and the decrease in adenosine deaminase activity a drastic stimulation of DNA synthesis in spleen T and B lymphocytes was observed. The increase in spleen T suppressor activity in the course of intense growth of the both types of hepatomas coincided in the time with the stimulation of the CTP-dependent thymidine kinase isoenzyme activity in total T lymphocyte population of the same organ.

The mechanism of inhibition and "reversal" of mitogen-induced lymphocyte activation in a model of purine-nucleoside phosphorylase deficiency.

Purine-nucleoside phosphorylase (PNP) is a purine degradative enzyme that catalyzes the phosphorolysis of (deoxy) inosine or (deoxy) guanosine to their respective bases and (deoxy) ribose 1-phosphate. A severe T-cell immune deficiency syndrome with hypouricemia is associated with impaired PNP function. To study the biochemical basis for this syndrome we created an in vitro model of PNP deficiency in mitogen (phytohemagglutinin)-stimulated normal human peripheral blood lymphocytes using guanosine to competitively inhibit deoxyguanosine phosphorolysis. Guanosine-induced guanine toxicity was reversed by adenine. Under these conditions, deoxyguanosine (5-45 microM) diminished mitogen stimulation to 30% of control while increasing the deoxyguanosine triphosphate pool (dGTP) by over 20-fold. Deoxycytidine reversed deoxyguanosine toxicity with a diminution of dGTP accumulation, but no significant change in the deoxycytidine triphosphate pool. Thymidine reversed the deoxyguanosine toxicity, repleted the thymidine triphosphate (dTTP) pool, and caused an even further increase in the accumulation of dGTP. These data support a model of lymphotoxicity in PNP deficiency based on dGTP accumulation with inhibition of ribonucleotide reductase and depletion of the thymidine triphosphate pool. Thymidine triphosphate depletion is reversed by either deoxycytidine or thymidine; however, the former diminishes dGTP accumulation (probably by competition for phosphorylation) and the latter potentiates dGTP accumulation (probably through feedback augmentation of guanosine diphosphate (GDP) reduction by ribonucleotide reductase secondary to an increased dTTP pool).

The effect of deoxyguanosine on human lymphocyte function. I. Analysis of the interference with lymphocyte proliferation in vitro.

The effect of deoxyguanosine on mitogen- and antigen-induced proliferation of peripheral blood lymphocytes from healthy donors was studied. Deoxyguanosine was found to inhibit the proliferative response to mitogens and antigens. Concentrations of deoxyguanosine causing 50% inhibition of the proliferation proved to be dependent on the activity of catabolic enzymes, such as purine nucleoside phosphorylase (PNP), in sera used in the culture media. The inhibitory effect of deoxyguanosine on phytohemagglutinin (PHA)-induced cell proliferation was prevented by deoxycytidine as well as by hypoxanthine. These findings were analyzed further by determination of intracellular (deoxy)-nucleotide levels. Stimulation of lymphocytes by PHA in the presence of deoxyguanosine leads to intracellular accumulation of dGTP. The presence of hypoxanthine in addition to deoxyguanosine abolished the inhibitory effect but did not prevent dGTP accumulation. On the other hand, the addition of deoxycytidine in combination with deoxyguanosine did not lead to intracellular accumulation of detectable amounts of dGTP, but only gave partial protection against the toxic effect. Furthermore, guanosine inhibited mitogen-induced cell proliferation to the same extent as did deoxyguanosine provided that the culture media were supplemented with pretreated fetal calf serum. Peripheral blood lymphocytes of a PNP-deficient or a HGPRT-deficient patient in cultures stimulated with PHA or pokeweed mitogen were resistant to the inhibitory effects of guanosine and were less sensitive to deoxyguanosine than cells of normal donors. The present results clearly show the involvement of two pathways contributing to deoxyguanosine-mediated inhibition of the proliferation of normal lymphocytes, i.e., on the one hand degradation of deoxyguanosine by PNP, salvage of guanine by HGPRT, and (possibly) phosphorylation of GMP eventually leading to GTP, and on the other hand formation of dGTP by direct phosphorylation of deoxyguanosine by deoxycytidine kinase.

[Possibility of biological indication of radiation injury from the blood levels of deoxynucleosides and deoxynucleotides in combined radiation-mechanical injuries]. / Vozmozhnost' biologicheskoi indikatsii luchevogo porazheniia po soderzhaniiu dezoksinukleozidov i dezoksinukleotidov v krovi pri radiatsionno-mekhanicheskom porazhenii.

A study was made of the dependence of a total content of deoxynucleosides and deoxynucleotides in blood serum and blood clot leukocytes upon radiation dose and time lapsed after radiation- and radiation-mechanical affection. It was established that after radiation-mechanical affection this dependence was different from that observed after the effect of radiation alone. From the analysis of the data obtained it was inferred that deoxynucleosides and deoxynucleotides could be used for biological indication of radiation damage in conditions of radiation-mechanical affection.

Concentration of nucleotides and deoxynucleotides in peripheral and phytohemagglutinin-stimulated mammalian lymphocytes. Effects of adenosine and deoxyadenosine.

Concentrations of purine and pyrimidine ribonucleotides were measured with HPLC in lymphocytes of man, horse, pig and sheep and in rat thymocytes. The ATP concentration was highest in lymphocytes of all species and about 850 pmol/10(6) cells in human and equine lymphocytes, higher in porcine and lower in ovine lymphocytes and rat thymocytes. The GTP concentration was comparable in human, equine and porcine lymphocytes, but lower in ovine lymphocytes. ATP concentration was also measured in lymphocytes of man, horse and pig with a luciferin-luciferase assay. During culturing with or without phytohemagglutinin the ATP concentrations decreased in these lymphocytes. The concentrations of TTP and dATP were measured with a DNA polymerase assay. Phytohemagglutinin-stimulation increased the TTP concentration in lymphocytes of all three species, the dATP concentration only in human lymphocytes. ATP, TTP and dATP concentrations and thymidine incorporation were measured in phytohemagglutinin-stimulated lymphocytes after 24 and 48 h culturing in the presence of adenosine or deoxyadenosine. Adenosine increased the ATP concentration in porcine and equine, but not in human lymphocytes. Deoxyadenosine and adenosine did not affect the TTP concentration. Deoxyadenosine decreased the ATP concentration only in the presence of EHNA in human lymphocytes, but increased it in other conditions and in equine and porcine lymphocytes. Deoxyadenosine in the presence of EHNA increased the dATP concentration in human, equine and porcine lymphocytes 3-, 10-, and 9-fold, respectively, and decreased considerably thymidine incorporation. Deoxyadenosine without EHNA increased the dATP concentration 2-5-fold, decreased the thymidine incorporation in lymphocytes of man and horse, but stimulated incorporation in porcine lymphocytes about 5-fold. The latter results indicate that accumulation of dATP is not always associated with inhibition of cell proliferation.

2'-deoxycoformycin (DCF) and 9-beta-D-arabinofuranosyladenine (Ara-A) in the treatment of refractory acute myelocytic leukemia.

The combination of 2'-deoxycoformycin (DCF), a potent adenosine deaminase (ADA) inhibitor, and 9-beta-D-arabinofuranosyladenine (Ara-A) was used in a patient with acute nonlymphocytic leukemia refractory to all conventional modes of therapy. DCF was given by periodic iv injections to ablate ADA activity. Ara-A was given by continuous iv infusion at an initial dose of 1.5 mg/kg/day, with progressive increases to 6 mg/kg/day. With adequate ADA suppression (less than 2 x 10(-2) mumols of inosine/hr/10(6)h cells), the Ara-A decreased the absolute peripheral blood myeloblast count from 36,332 to 780/microliter. The patient experienced no renal, hepatic, or neurologic complications during therapy.

Biochemical and functional abnormalities in lymphocytes from an adenosine deaminase-deficient patient during enzyme replacement therapy.

Biochemical and immunological properties of lymphocytes were measured repetitively over a period of 40 mo during enzyme replacement by transfusion in a child with adenosine deaminase (ADA) deficiency and severe combined immunodeficiency disease. Catalytically defective ADA protein is present in the child's cells. ADA activity in his lymphocytes is 7 nmol/min per 10(8) cells with 51 ng of ADA protein/10(8) cells by radioimmunoassay. ADA activities in normal cord and adult lymphocytes average 193 and 92 nmol/min per 10(8) cells, respectively, with 429 and 223 ng of ADA protein/10(8) cells. Deoxy(d)ATP accumulates in the patient's erythrocytes and lymphocytes. Transfusion of irradiated packed erythrocytes partially corrects the metabolic defects. Frank metabolic relapse occurs if transfusions are discontinued for several months. The amounts of dATP in erythrocytes and lymphocytes averaged 13 and 2 times normal, respectively, during periods when transfusions were administered every 2-4 wk. Deoxyguanosine triphosphate and deoxycytidine triphosphate in lymphocytes were normal on 11 occasions, but deoxyribosylthymine triphosphate was ninefold increased. On 11 occasions dATP was measured in lymphocytes and erythrocytes isolated simultaneously. There was a positive, but statistically insignificant, correlation between amounts of dATP in the two types of cells (r = 0.25,P > 0.1). The absolute peripheral lymphocyte count was correlated with the activity of ADA in circulating erythrocytes and with the response of lymphocytes to phytohemagglutinin (r = 0.64, P < 0.01; r = 0.49, P < 0.05). Response of lymphocytes to stimulation by phytohemagglutinin in vitro and absolute peripheral lymphocyte counts were not significantly correlated with levels of dATP in the erythrocyte or lymphocyte during periods of intensive therapy. Although there was objective improvement during enzyme replacement, the child remained immunodeficient and biochemically abnormal.

Purinogenic immunodeficiency diseases: clinical features and molecular mechanisms.

Deficiencies of two enzymes that catalyze sequential reactions in the purine catabolic pathway have been causally associated with immunodeficiency states. Adenosine deaminase (ADA) deficiency results in severe combined immunodeficiency disease, while purine nucleoside phosphorylase (PNP) deficiency results in an isolated T-cell defect. Recent work in this area has provided major new insights into the molecular pathology of these syndromes. Deoxyadenosine and deoxyguanosine, substrates that accumulate in ADA and deoxyguanosine, substrates that accumulate in ADA and PNP deficiency, respectively, appear to be selectively phosphorylated by lymphoid cells to the corresponding deoxynucleoside triphosphate, resulting in inhibition of DNA synthesis in these cells. Both deoxynucleosides are far more toxic to cultured T lymphoblasts than to B lymphoblasts. Adenosine and deoxyadenosine may have additional lymphotoxic effects mediated by inhibition of essential methylation reactions. These observations help to explain the immunologic manifestations of ADA and PNP deficiency. Perhaps more important, they lay the foundation for the use of deoxynucleosides or enzyme inhibitors, or both, as selective immunosuppressive and chemotherapeutic agents.