DMD Fast Forward. Published on December 23, 2008 as DOI: 10.1124/dmd.108.022269 DMD FasTth iFs aorrtiwclea hrads .n Potu bbeelni scohpeyded iotend aDnde fcoermmatbteed.r T2h3e ,f i2na0l 0ve8r saiosn mdaoyi :d1if0fe.r1 f1ro2m4 t/hdism vedrs.1io0n.8.022269 DMD #22269 5’-Aminocarbonyl phosphonates as new AZT depot forms: antiviral properties, intracellular transformations and pharmacokinetic parameters Anastasia L. Khandazhinskaya, Dmitry V. Yanvarev, Maxim V. Jasko, Alexander V. Shipitsin, Vsevolod A. Khalizev, Stanislav I. Shram, Yuriy S. Skoblov, Elena A. Shirokova and Marina K. Kukhanova Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Moscow, Russia (A.L.K., D.V.Y., M.V.J., A.V.S., V.A.K., E.A.S., M.K.K) Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, Moscow, Russia (Y.S.S) D o w Institute of Molecular Genetics, Moscow, Russia (S.I.S) nlo a d e d fro m d m d .a s p e tjo u rn a ls .o rg a t A S P E T J o u rn a ls o n J a n u a ry 1 3 , 2 0 2 3 1 Copyright 2008 by the American Society for Pharmacology and Experimental Therapeutics. DMD Fast Forward. Published on December 23, 2008 as DOI: 10.1124/dmd.108.022269 This article has not been copyedited and formatted. The final version may differ from this version. DMD #22269 RUNNING TITLE PAGE a) AZT 5’-aminocarbonyl phospho nates as new AZT depot forms b) Anastasia L. Khandazhinskaya, Engelhardt Institute of Molecular Biology RAS, 32, Vavilov str., Moscow, 119991, Russia, telephone (7499)1356065, fax (7499)1351405, lhba- [email protected], [email protected] c) 30 text pages 8 tables 8 figures 16 references 247 words in the Abstract 359 words in the Introduction D o w 1464 words in the Discussion nlo a d d) AIDS- Acquired ImmunoDeficiency Syndrome; AZT - 3'-azido-3'-deoxythymidine; AZT-MP ed fro - 3'-azido-3'-deoxythymidine 5’-monophosphate; HIV – human immunodeficiency virus; DNA - m d m deoxyribonucleic acid; NMR - nuclear magnetic resonance; UV – ultraviolet; TEAB –triethyl d .a s p ammonium bicarbonate; HPLC - high-pressure liquid chromatography; GSIOC - Gilson Serial e tjo u Input/Output Channel; Cmax - maximum plasma concentration; Tmax - time to maximum plasma rna ls concentration; AUC - area under concentration-time curve from 0 to ∞; CL - total clearance; .org Т a MRT - mean residence time; 1/2 –elimination half-life; LD50 – mean lethal dose; CD50 – t A S P cytotoxic dose; ID – inhibitory dose; SI – selectivity index; SD - standard deviation; Thy - E 50 T J o thymine; T - thymidine; DCC – dicyclohexyl carbodiimide; C - cumulation coefficient u c rn a ls o n J a n u a ry 1 3 , 2 0 2 3 2 DMD Fast Forward. Published on December 23, 2008 as DOI: 10.1124/dmd.108.022269 This article has not been copyedited and formatted. The final version may differ from this version. DMD #22269 ABSTRACT The main disadvantages of 3'-azido-3'-deoxythymidine (Zidovudine®, AZT), the most common anti-HIV drug, are toxicity and a short half-life in the organism. The introduction of an H- phosphonate group into the AZT 5’ position resulted in significant improvement of its therapeutical properties and allowed a new anti-HIV drug Nikavir®. In this work we described a new group of AZT derivatives, namely, AZT 5'-aminocarbonylphosphonates. The synthesized compounds displayed antiviral properties in cell cultures infected with HIV-1 and the capacity to release the active nucleoside in animals (rabbits and dogs) in a dose-dependent manner. The compounds were less toxic in MT-4 and HL-60 cell cultures and experimental animals when compared with Zidovudine. Major metabolites found in MT-4 cells after their incubation with D AZT 5’-aminocarbonyl phosphonate 1 were AZT and AZT 5’-phosphate (25 and 55%, o w n respectively). Among the tested compounds, phosphonate 1 was the most effective AZT donor loa d e d and its longest T1/2 and Tmax values in the line phosphonate 1 - Nikavir - Zidovudine imply that fro m compound 1 is an extended depot-form of AZT. Although bioavailability of AZT following oral d m d administration of phosphonate 1 was lower than those of Nikavir and Zidovudine (8% against .a s p e 14% and 49%), we expect that this reduction would not cause essential fall of antiviral activity tjo u rn but noticeably decrease toxicity due to gradual accumulation of AZT in blood and the absence of a ls .o sharp difference between C and C . Such a combination of properties makes the compounds rg max min a t A of this group promising for further studies as extended-release forms of AZT. S P E T J o u rn a ls o n J a n u a ry 1 3 , 2 0 2 3 3 DMD Fast Forward. Published on December 23, 2008 as DOI: 10.1124/dmd.108.022269 This article has not been copyedited and formatted. The final version may differ from this version. DMD #22269 INTRODUCTION The progress towards the treatment of HIV infections has steadily increased in the past two decades. At the end of the 1980's the life expectancy of HIV-infected patients was only 2 to 5 years, whereas today many patients often survive between 10 to 15 years, however, once the infection progresses to full blown AIDS, the mortality rate is 100%. Currently, more than 20 drugs have been approved for treatment of HIV, these include inhibitors of crucial viral enzymes such as reverse transcriptase, protease, and integrase. Despite significant progress in the design of anti-HIV drugs, many problems remain such as toxicity and side effects, as well as rapid elimination from the body resulting in more frequent dosing. More significantly, the development of drug-resistant strains has increased dramatically. As a result, there is a critical D o w need for more effective and less toxic therapeutics. nlo a d In that regard, HIV reverse transcriptase has proven to be an attractive target. HIV RT ed fro inhibitors are primarily modified nucleosides that are converted by way of an intracellular m d m cascade of phosphorylations to the corresponding triphosphates. The triphosphates are then d .a s p incorporated into the growing DNA chain, which results in termination of DNA synthesis. As the e tjo u efficacy of the triphosphates is low, drug doses must therefore be high, which generally leads to rn a ls significant toxicity. One solution is to design the corresponding depot form, i.e., a prodrug-like .o rg a derivative capable of revealing the active compound in the organism at a controlled rate t A ń S P (Sta czak and Ferra, 2006). Many depot forms of anti-HIV drugs have been developed E T J o (Beaumont et al., 2003; Calogeropoulou et al., 2003) and some have resulted in new drugs u rn a (Kearney et al., 2004; Lyseng-Williamson et al., 2005; De Clercq and Field, 2006). In particular, ls o n the depot form of AZT, Nikavir® (AZT H-phosphonate, phosphazide, AZT 5’- Ja n u a hydrogenphosphonate), proved to be highly effective and has been approved in Russia for the ry 1 3 prevention and treatment of AIDS (Kravchenko, 2004; Skoblov et al., 2004; Kukhanova and , 2 0 2 3 Shirokova, 2005). In this work we describe preliminary biological data for several new AZT depot analogues (phosphonates 1-5, Fig. 1), including their anti-HIV properties in cell systems, cellular uptake, intracellular transformations (for phosphonate 1 as an example) and pharmacokinetic and toxicological data. 4 DMD Fast Forward. Published on December 23, 2008 as DOI: 10.1124/dmd.108.022269 This article has not been copyedited and formatted. The final version may differ from this version. DMD #22269 METHODS Chemistry All reagents and solvents were purchased from Acrus (Belgium). The syntheses of compounds 1-3 and 5 have been described previously (Shirokova et al., 2004; Ias’ko et al., 2006). Compound 4 was obtained from AZT 5'-ethoxycarbonylphosphonate and n-butylamine. AZT was a gift from the “AZT Association” (Moscow, Russia). NMR spectra were obtained on an AMX III-400 spectrometer (Bruker) with the working frequency of 400 MHz for 1H NMR (Me Si as an internal standard for organic solvents and 4 sodium 3-trimethylsilyl-1-propanesulfonate for D O), 162 MHz for 31P NMR (with phosphorus- 2 С proton interaction decoupling, 85% H PO as an external standard) and 100.6 MHz for 13 NMR 3 4 (with carbon-proton interaction decoupling). UV spectra were recorded on a Shimadzu UV- D o w 2401PC spectrophotometer (Japan) and were in line with known standard values for thymidine nlo a d derivatives. ed fro 5'-(Butylamino)carbonylphosphonyl-3'-azido-3'-deoxythymidine, ammonium salt 4. m d m 1H-NMR (D O): 7.65 (1H, s, H6), 6.22 (1H, t, J 6.5, H1'), 4.44 (1H, m, H3'), 4.18 (2H, m, H5'), d 2 и с .as p 4.11 (1H, m, H4'), 3.20 (2H, dd, J 6.9 7.2, CH N (butyl)), 2.45 (2H, m, H2'), 1.88 (3H, , 5- e 2 СН и tjo u MСНe), 1.44 (2H, m, CH2CH2N (butyl)), 1.26 (2H, m, CH2 3 (butyl)), 0.83 (3H, dd, J 6.5 6.9, rnals (butyl)). 31P-NMR (D O): -1.27 s. 13C-NMR (D O): 186.8 (d, 1J 212.1, C(O)P), 169.2 (s, .o 3 2 2 C,P rg a C4), 154.4 (s, C2), 140.1 (s, C6), 114.4 (s, C5), 87.7 (s, C1'), 85.6 (d, 3JC,P 7.2, C4'), 67.7 (d, 2JC,P t A СН СН S P 4.8, C5'), 63.0 (s, C3'), 41.4 (d, 3J 6.4, CH NHC(O)P), 39.1 (s, C2'), 33.3 (s, CH ), E C,P 2 2 2 3 T СН СН СН СН J o 22.2 (s, 2CH2 3), 15.7 (s, 2CH2 3), 14.4 (s, 5-Me). urn a Synthesis of 5'-aminocarbonylphosphonyl-3'-azido-3'-deoxy-[6-3H]thymidine ([6- ls o n 3H]-1). Trimethylsilyl bromide (5 mmol, 650 μL) was added to triethylphosphonoformate (1 Ja n С u a mmol, 190 μL) and the mixture stirred for 20 h at 22° under argon. Anhydrous toluene (1 mL) ry 1 С 3 was added, the solvent was evaporated in vacuum (1 mm Hg, 30° ) and a solution of SOCl (10 , 2 2 0 2 3 mmol, 730 μL) in anhydrous CCl (1 mL) was added to the residue. The mixture was refluxed 4 for 2 h under argon; the solvent evaporated under vacuum and coevaporated with anhydrous benzene (2 mL). The residue was then dissolved in anhydrous benzene (1 mL) and a 10 μL (a.c. 10 μmol) aliquot of the solution was treated with [6-3H]AZT (3 mCi, 14 Ci/mmol) and AZT (0.5 С μmol, 133.5 μg). The mixture was stirred for 2 h at 22° , evaporated under vacuum and 1M С TEAB (1 mL) added to the residue. The solution was then stirred for 30 min at 45° , evaporated under vacuum, and 32% NH /H O (100 μL) added to the residue. The mixture was stirred for 18 3 2 С h at 0° , the solvents removed under vacuum and the residue dissolved in H O (200 μL). The 2 product was isolated by HPLC on a Lichrosorb RP-18 column (5μm, 4×150 mm) using a Gilson 5 DMD Fast Forward. Published on December 23, 2008 as DOI: 10.1124/dmd.108.022269 This article has not been copyedited and formatted. The final version may differ from this version. DMD #22269 chromatograph (France) supplied with a digital GSIOC 506 controller (Gilson) and a Gilson-315 UV detector (254 nm). Solution A: 50 mM TEAB; solution B: 75% methanol. Gradient of B: 0 min at 0%, 10 min at 25%, 25 min at 25%, and 30 min at 100%. Retention times: 14.5 min for 1, 20 min for AZT and 28 min for AZT 5’-ethoxycarbonyl phosphonate. The yield of product [6- 3H]-1 was 2.6 mCi (87%), specific activity 4.2 Ci/mmol. The structure was confirmed using a nonradioactive reference standard by reverse-phase (for the conditions, see above) and ion- М exchange (SynChropak AX300 column, 4.6 × 250 mm; solution A: water; solution B: 0.2 KCl; retention time of 1 was 6.5 min) chromatography. Antiviral activity of the synthesized compounds was studied in MT-4 cell culture infected with HIV-1, strain GKV-4046, using previously published procedures (Shirokova et al., 2004). D o w Cytotoxicity of the synthesized compounds was studied in MT-4 cell culture, using nlo a d previously published procedures (Shirokova et al., 2004). ed fro Cellular uptake m х d m A suspension of MT-4 cells (0.7- 1 106 cells/mL) in the RPMI–1640 medium (50 mL) d .a s containing 2 mM L-Gln and 10% calf serum was incubated with 4 µM [6-3H]AZT (200 µCi) or 4 pe tjo u µM [6-3H]-1 (300 µCi) at 37°C in the atmosphere containing 5% CO2. rna ls Incubation of the tested compounds with HL-60 cells was carried out as described .o rg a previously (Yanvarev et al. 2007). t A S P After 1, 3, 7 and 24 h incubation, two aliquots (2.5 mL) were removed and centrifuged at E T J o 1500 g for 10 min. The supernatants were used for HPLC analysis of the extracellular hydrolysis u rn a of phosphonate 1. The precipitates were resuspended in PBS and washed three times (3 x 3 mL) ls o n with PBS. The final pellets were resuspended in water (200 μL), subjected to cryolysis, and Ja n С u a methanol (400 μL) was added. The samples were kept at -18° . Prior to HPLC analysis the ry 1 3 samples were centrifuged at 1500 g for 10 min, the supernatant was evaporated under vacuum, , 2 0 2 3 the residue was dissolved in water (100 μL), and the standards (thymine, thymidine, AZT-MP, phosphonate 1 and AZT 5'-H-phosphonate, 5 μg each) were added. The HPLC analysis was performed on a Nucleosil 100 C-18 column (5 μm, 4 x 150 mm) with UV detection at 254 nm using a Gilson chromatograph (see above); the flow rate 0.5 mL/min. System A: 50 mM ammonium bicarbonate; system B: 75% aqueous ethanol. Gradient: 0 min at 0%B, 10 min at 10%B, 50 min at 20%B, 55 min at 100%B. Retention times: 37 min for AZT; 25 min for 1, 22 min for AZT-MP, 19 min for thymidine,12 min for thymine. The aliquots were taken out every 0.5 min (0.25 mL), scintillation liquid (Beckman ready value, 5 mL) was added, and radioactivity was measured using an LS-counter SL-4000 Intertechnique (France). 6 DMD Fast Forward. Published on December 23, 2008 as DOI: 10.1124/dmd.108.022269 This article has not been copyedited and formatted. The final version may differ from this version. DMD #22269 Based on the value of specific activity, the concentration of the radioactive compounds was calculated. Chemical stability Chemical stabilities of compounds 1-5 at the concentrations of 0.5 mM were studied in 0.1 M sodium phosphate buffer at 37°C and pH in the range of 5.5 - 8.5. The aliquots (10 μL) were taken out after certain intervals, frozen in liquid nitrogen, and the products analyzed on a liquid Gynkotek chromatograph (Germany) supplied with a UV detector (Model 320), at a wavelength of 265 nm. Chromatographic conditions: an Ultrasphere ODS - IP column, 5μm, 4.6 х 150 mm; mobile phase: acetonitrile-0,1% phosphoric acid containing 0,2% of triethylamine, pH 2.7-2.9. The ratios of the mobile phase components were individual for each compound D (Table 1). o w n Stability in whole blood lo a d e Water solutions of the tested compounds (10-20 μL, 1 μg/μL) were placed into tubes with Na- d fro m heparin, fresh dog whole blood (1 mL) was added into each tube and the tubes were kept at d оС m 37 . The aliquots were taken out every 1 h during 6 h and centrifuged for 10 min at 1500 g to d.a s p e separate plasma. Methanol (0.5 mL) was added to the supernatant (0.25 mL) and the mixture was tjo u shaken for 30 s and again centrifuged for 10 min at 1500 g. The supernatant was evaporated at rna С ls.o 40° . The dry residue was dissolved in water and analyzed as described in section “Chemical rg a stability” t A S P E Animal experiments were carried out according to the protocols of Council Directive T J o u 86/609/EEC of November 24, 1986, on the protection of animals used for experimental rn a ls purposes. o n J a Pharmacokinetic parameters were calculated using the Thermo Kinetika 4.4.1 program n u a (Thermo electron corporation, USA). Pharmacokinetics following oral administration were ry 1 3 studied by the extravascular noncompartmental model of the Thermo Kinetica program. For , 20 2 3 intravenous administration, the noncompartmental IV Infusion model was used. Pharmacokinetics in dogs The studies were performed following standard protocols (Kurlyandskii, 2000), with outbred dogs (males and females, 13.6 ± 2.6 kg body weight). The tested compounds were administered orally in the fasting state; the animals were fed 3 h after dosing (50 or 20 mg/kg body weight). Intravenous administration (1-5 mL of aqueous solution, 5 mg/kg) was performed in the antebrahial vein of anterior foot (v. cephalica antebrachii) for 2-5 min. Ten blood samples (not less than 3 mL) were taken out from the femoral vein 0-24 h after administration and placed into tubes containing heparin (5 μL, 5000 U/mL) and centrifuged for 10 min at 1500 g. The 7 DMD Fast Forward. Published on December 23, 2008 as DOI: 10.1124/dmd.108.022269 This article has not been copyedited and formatted. The final version may differ from this version. DMD #22269 С plasma samples were kept at –24° . The samples were treated as described in section “Stability in blood” and analyzed under conditions given in section “Chemical stability”. Bioavailability (F) in dogs was calculated by the following formula: F = (AUC × D )/(AUC × D ) × 100% p.os. i.v. i.v. p.os. where D is an intravenous dose in dogs; i.v. D is an oral dose in dogs; p.os. AUC is the area under the time-concentration curve for the intravenous dose i.v. AUC is the area under the time-concentration curve for the oral dose p.os. Pharmacokinetics in rabbits D o w Chinchilla rabbits (males, 3 ± 0.5 kg body weight) were kept in separated cages with free nlo a d e access to water and food and 12 h lighting. The animals did not receive food 14 h before the d fro experiment and were fed 6 h after dosing of the tested compounds. In the case of single m d m intragastric administration the animals were anaesthetized with a 10:1 ether-halotane mixture and d .a s p a polyurethane gastrointestinal tube was introduced 15 cm deep. The tested compounds were etjo u administered as water solutions (12 mL) at concentrations of 7 or 70 mg/kg body weight. rn a ls In the case of intravenous administration (10 mg/kg body weight), the compounds were .org a injected into the Auricularis marginalis vein in physiological solution (1 mL) for 1 min; the t A S P injection time was 1 min. At each day of the experiment, a single heparin dose (0.5 mL of water E T J o solution, 5000 U/mL) was injected to each rabbit prior to taking blood samples. The blood u rn a samples (1mL on average) were taken from the Auricularis marginalis vein into microtubes ls o n containing 5 μL of heparin (5000 U/mL). The tubes were shaken and aliquots of 0.5 mL were Ja n С ua taken out and mixed with methanol (1 mL), rapidly cooled in liquid nitrogen and stored at -24° . ry 1 3 The control blood samples were taken before administration of the tested compounds. , 2 0 2 3 The water-methanol mixture was treated as described in section "Cellular uptake". The samples were analyzed by HPLC on a Gilson chromatograph (see section “Cellular uptake”) on a Nucleosil 100 C-18 column (4 x 150 mm). The chromatographic conditions were as follows: solution A, 50 mM triethylammonium acetate; solution B, 75% methanol; gradient of B: 0 min at 0%B; 12 min at 10%B; 50 min at 20%B; 55 min at 100%B. The flow rate was 0.5 mL/min. Retention times: 32 min for AZT H-phosphonate; 37 min for compound 1; 51 min for AZT. Acute toxicity 8 DMD Fast Forward. Published on December 23, 2008 as DOI: 10.1124/dmd.108.022269 This article has not been copyedited and formatted. The final version may differ from this version. DMD #22269 LD , LD and LD values were obtained according to the standard procedure 16 50 84 (Khabriev, 2005). BALB/c mice (males and females, 19 ± 1 g body weight) were administered single intraperitoneal doses of the tested compounds dissolved in a sterile isotonic solution of sodium chloride. The control animals received the corresponding volume of the isotonic solution of sodium chloride. The animals were examined for 14 days. The BALB/c mice were received from the laboratory animal hatchery of the Scientific Center of Biomedical Technologies of Russian Academy of Medical Sciences. The animals were kept in T-3 cages at artificial illumination (light and darkness, 12 h of С each), mandatory 12-fold per hour ventilation, at 20-22° and 50-65% relative humidity, on mats from wood chippings sterilized in an oven. The animals had a free access to water D and food. ow n lo a Blood samples of mice died after dosing of compound 1 at a dose of 6 g/kg were d e d immediately treated and analyzed by HPLC as described in section "Pharmacokinetics in fro m d rabbits". m d .a s Cumulative effect pe tjo u The experiments on the cumulative effect of phosphonate 1 were carried out on BALB/c rn a ls mice (males, 17 ± 1 g body weight) using a standard scheme (Table 2) (Khabriev, 2005). The .org a effect was estimated based on the cumulating coefficient C , which was calculated as follows: t A c S P E C = LD / LD , where LD is a mean lethal dose following n-fold administration, T c 50n 50 1 50n J o LD50 1 is a mean lethal dose following single administration (see section “Acute urn a ls toxicity”). o n J Cc >1, habituation; Cc <1, cumulation. anu a Statistical Analysis ry 1 3 Statistical significance of comparisons between treatment groups was assessed by a two- , 2 0 2 3 tailed Student's t test. The values are expressed as mean ± SD (standard deviation). Each pharmacokinetic experiment was repeated at least thrice. Pharmacokinetic parameters were determined using the Thermo Kinetika 4.4.1 ("Thermo Electron Corporation") program. The lower limits of detection of AZT and amide 1 were 10 nmol for unlabelled compounds (UV ε detection, 9700) and 10 pmol for labeled compounds (radioactive detection, a specific 267 activity of 4.2 Ci/mmol). 9 DMD Fast Forward. Published on December 23, 2008 as DOI: 10.1124/dmd.108.022269 This article has not been copyedited and formatted. The final version may differ from this version. DMD #22269 RESULTS Antiviral properties of the synthesized phosphonates were studied in MT-4 cells infected with HIV-1. The results given in Table 3 indicate that the compounds inhibited virus replication similarly to AZT H-phosphonate and by one order of magnitude lower than parent AZT. At the same time their toxicity (except phosphonate 2) was considerably lower than those of AZT H-phosphonate and parent AZT. Compound 2 was excluded from further studies because of comparatively high toxicity. A higher CD of the phosphonates under study allowed 50 better selectivity indexes SI (Table 3). Cellular uptake was studied with radiolabelled phosphonate 1 ([6-3H]-1) in two human lymphoid cell lines MT-4 and HL-60. As is seen in Fig.2, the efficacy of penetration of phosphonate 1 into HL-60 cells was tenfold higher than that for MT-4 cells at the same D o w extracellular concentration of 4 μM. nlo a d e Earlier we showed that penetration of phosphonate 1 into HL-60 cells was about tenfold d fro less effective than that of AZT (Yanvarev et al., 2007). In a similar experiment with MT-4 cells m d m the AZT penetration rate dropped only by 25-30%, if compared with its penetration into HL-60 d .a s p cells, whereas in the case of phosphonate 1, the reduction was considerably higher (about one etjo u order of magnitude). Thus the difference between penetration rates of phosphonate 1 and AZT rn a ls into MT-4 cells exceeded 100 times. .org a The analysis of metabolic products in MT-4 cells after their incubation with phosphonate t A S P 1 showed the presence of AZT (25±10%), AZT-MP (55±10%) and starting 1 (12±4%) (Fig. 3). E T J o In addition, thymine (8±5%) was found. No other metabolites were detected in the mixture. u rn a Compositions of intracellular metabolites were similar for MT-4 and HL-60; variations did not ls o n J exceed 10%. It is noteworthy that a degree of extracellular hydrolysis was only 5% after 24 h a n u a incubation. ry 1 3 Stability , 2 0 2 3 All the synthesized compounds were chemically stable with half-lives in buffer рН solutions ( 5.5-8.5) considerably greater than 48 h. They also demonstrated high оС Т stability in 100% human blood serum at 37 ( >> 6 h) (data not given). None of the 1/2 Т tested phosphonates were hydrolyzed in dog whole blood ( >> 6 h) (data not given). 1/2 Pharmacokinetic parameters Some pharmacokinetic parameters were evaluated in dogs following oral administration of 480-560 mg of phosphonates 1, 3, 4, and 5 (50 mg/kg). In the case of phosphonate 1, the analysis of dog blood serum demonstrated the presence of both AZT and 1 (Fig. 4). The maximum concentration of released AZT was observed 4 h after dosing. The concentration of phosphonate 1 was approaching zero at this time point. 10