Advanced Drug Delivery Reviews 39 (1999) 117–151
`
`L
`
`www.elsevier.com / locate / drugdeliv
`
`Development and optimization of anti-HIV nucleoside analogs
`and prodrugs:
`A review of their cellular pharmacology, structure-activity
`relationships and pharmacokinetics
`Xiaolei Tan, Chung K. Chu, F. Douglas Boudinot
`
`*
`
`Department of Pharmaceutical and Biomedical Sciences, University of Georgia College of Pharmacy, Athens, GA 30602-2352, USA
`
`Received 5 October 1998; received in revised form 22 January 1999; accepted 3 February 1999
`
`Abstract
`
`Significant improvements in antiviral therapy have been realized over the past 10 years. Numerous nucleoside analogs, as
`well as prodrugs of active compounds, have been synthesized and tested for anti-HIV activity. In addition to the five
`nucleoside analogs currently used clinically for the treatment of HIV infection, a broad spectrum of anti-HIV nucleoside
`analogs (including 29,39-dideoxynucleoside analogs, oxathiolanyl 29,39-dideoxynucleoside analogs, dioxolanyl 29,39-dideox-
`ynucleoside analogs, carbocyclic 29,39-dideoxynucleoside analogs and acyclic nucleoside analogs) and their prodrugs
`(including ester prodrugs, phospholipid prodrugs, dihydropyridine prodrugs, pronucleotides and dinucleotide analogs),
`targeted at HIV reverse transcriptase, are reviewed with focus on structure-activity relationships, cellular pharmacology and
`pharmacokinetics. Several of these anti-viral agents show promise in the treatment of AIDS.
`1999 Elsevier Science B.V.
`All rights reserved.
`
`Keywords: Anti-HIV; Reverse transcriptase; 29,39-Dideoxynucleoside analog; Nucleotide analog; Prodrug; Cellular pharmacology;
`Pharmacokinetics
`
`Contents
`
`1. Introduction ............................................................................................................................................................................
`2. 29,39-Dideoxynucleoside analogs ..............................................................................................................................................
`2.1. Structure-activity relationships ..........................................................................................................................................
`2.2. Thymidine (T) analogs .....................................................................................................................................................
`2.2.1. Cellular pharmacology............................................................................................................................................
`2.2.2. Pharmacokinetics ...................................................................................................................................................
`2.3. 29-Deoxyuridine (dU) analogs...........................................................................................................................................
`2.3.1. Cellular pharmacology............................................................................................................................................
`2.3.2. Pharmacokinetics ...................................................................................................................................................
`2.4. 29-Deoxycytidine (dC) analogs..........................................................................................................................................
`
`118
`120
`120
`121
`121
`122
`123
`123
`124
`124
`
`*Corresponding author. Tel.: 1 1-706-542-5335; fax: 1 1-706-542-5252.
`E-mail address: boudinot@rx.uga.edu (F.D. Boudinot)
`
`0169-409X / 99 / $ – see front matter
`P I I : S 0 1 6 9 - 4 0 9 X ( 9 9 ) 0 0 0 2 3 - X
`
`1999 Elsevier Science B.V. All rights reserved.
`
`Columbia Ex. 2088
`Illumina, Inc. v. The Trustees
`of Columbia University in the
`City of New York
`IPR2020-00988, -01065,
`-01177, -01125, -01323
`
`(cid:211)
`(cid:211)
`

`

`118
`
`X. Tan et al. / Advanced Drug Delivery Reviews 39(1999)117–151
`
`2.4.1. Cellular pharmacology............................................................................................................................................
`2.4.2. Pharmacokinetics ...................................................................................................................................................
`2.5. 29-Deoxyadenosine (dA) and 29-deoxyinosine (dI) analogs..................................................................................................
`2.5.1. Cellular pharmacology............................................................................................................................................
`2.5.2. Pharmacokinetics ...................................................................................................................................................
`3. Oxathiolanyl 29,39-dideoxynucleoside analogs ...........................................................................................................................
`3.1. Structure-activity relationships ..........................................................................................................................................
`3.2. Cellular pharmacology .....................................................................................................................................................
`3.3. Pharmacokinetics .............................................................................................................................................................
`4. Dioxolanyl 29,39-dideoxynucleoside analogs .............................................................................................................................
`4.1. Structure-activity relationships ..........................................................................................................................................
`4.2. Cellular pharmacology .....................................................................................................................................................
`4.3. Pharmacokinetics .............................................................................................................................................................
`5. Carbocyclic 29,39-dideoxynucleoside analogs ............................................................................................................................
`5.1. Structure-activity relationships ..........................................................................................................................................
`5.2. Cellular pharmacology .....................................................................................................................................................
`5.3. Pharmacokinetics .............................................................................................................................................................
`6. Acyclic nucleoside analogs ......................................................................................................................................................
`6.1. Structure-activity relationships ..........................................................................................................................................
`6.2. Cellular pharmacology .....................................................................................................................................................
`6.3. Pharmacokinetics .............................................................................................................................................................
`7. Ester prodrugs of 29,39-dideoxynucleoside analogs ....................................................................................................................
`8. Phospholipid prodrugs of 29,39-dideoxynucleoside analogs .........................................................................................................
`8.1. Rationale for prodrug design .............................................................................................................................................
`8.2. Cellular pharmacology .....................................................................................................................................................
`8.3. Pharmacokinetics .............................................................................................................................................................
`9. Dihydropyridine prodrugs of 29,39-dideoxynucleoside analogs ....................................................................................................
`9.1. Rationale for prodrug design .............................................................................................................................................
`9.2. Cellular pharmacology .....................................................................................................................................................
`9.3. Pharmacokinetics .............................................................................................................................................................
`10. Pronucleotides: phosphatidyl 29,39-dideoxynucleoside analogs ..................................................................................................
`10.1. Rationale for prodrug design ...........................................................................................................................................
`10.2. Structure-activity relationships ........................................................................................................................................
`10.3. Cellular pharmacology....................................................................................................................................................
`10.4. Pharmacokinetics ...........................................................................................................................................................
`11. Dinucleotide analogs .............................................................................................................................................................
`11.1. Rationale for prodrug design ...........................................................................................................................................
`11.2. Structure-activity relationships ........................................................................................................................................
`12. Other pronucleotide approaches ..............................................................................................................................................
`13. Conclusions ..........................................................................................................................................................................
`References ..................................................................................................................................................................................
`
`125
`125
`125
`125
`127
`127
`127
`128
`128
`129
`129
`130
`130
`131
`131
`131
`132
`132
`132
`134
`135
`135
`135
`135
`136
`136
`136
`136
`137
`137
`137
`137
`138
`138
`139
`140
`140
`140
`141
`141
`141
`
`1. Introduction
`
`immunodeficiency
`acquired
`combating
`In
`syndrome (AIDS) and AIDS related complex (ARC),
`the search for therapeutic agents possessing activity
`against human immunodeficiency virus (HIV) has
`yielded a number of compounds demonstrating po-
`tent and selective antiviral activity. Human immuno-
`deficiency virus reverse transcriptase (HIV-RT) re-
`mains a primary target for the treatment of HIV
`infection [1]. As HIV-RT inhibitors, five anti-HIV
`
`nucleoside analogs have been approved by Food and
`Drug Administration (FDA) and are currently used
`clinically. These anti-HIV nucleoside analogs include
`zidovudine (AZT), didanosine (ddI), zalcitabine
`(ddC),
`stavudine (d4T), and lamivudine (3TC).
`Except for 3TC, which has a b-L-configuration, these
`nucleoside analogs have b-D-configurations similar
`to natural nucleosides.
`The active antiretroviral form of a nucleoside
`analog is its triphosphate anabolite. By sharing the
`anabolic pathway of
`the naturally occurring nu-
`
`

`

`X. Tan et al. / Advanced Drug Delivery Reviews 39(1999)117–151
`
`119
`
`cleosides, antiviral nucleoside analogs are phos-
`phorylated to the corresponding mono-, di-, and
`triphosphates. Phosphoryation of nucleosides is cata-
`lyzed by cellular kinases as illustrated in Fig. 1.
`Although the efficiency of the phosphorylation of
`nucleoside analogs is generally lower than that of the
`naturally occurring nucleoside, the potent and selec-
`tive anti-HIV activity of
`the nucleoside analog
`resides in strong inhibition of HIV-RT with relatively
`little effect on cellular polymerases exerted by the
`nucleoside-triphosphate. Once incorporated into the
`growing proviral DNA chain, nucleoside analogs
`terminate the viral DNA elongation owing to a lack
`of the 39-OH. While the mechanism of antiviral
`
`activity of nucleoside analogs may ultimately be the
`same, it should be noted that each compound has its
`own distinct metabolic and pharmacological prop-
`erties. Besides the inhibition of HIV-RT, the effect of
`triphosphates of nucleoside analogs on cellular poly-
`merases (pol a, b, d, g and «) are also important to
`the understanding of the potential mechanisms in-
`volved in in vitro and in vivo activity and toxicity.
`Pol a, which is the main polymerase responsible for
`cellular DNA synthesis, is not inhibited by all FDA-
`approved nucleoside analogs.
`A number of cell lines have been employed to
`investigate the in vitro anti-HIV activity and cytotox-
`icity of the nucleoside analogs,
`including human
`
`Fig. 1. The intracellular phosphorylation pathways of FDA-approved nucleoside analogs.
`
`

`

`120
`
`X. Tan et al. / Advanced Drug Delivery Reviews 39(1999)117–151
`
`T-cell lines such as MT-2, MT-4, H9, CEM, and
`ATH8, human peripheral blood mononuclear (PBM)
`cells, human peripheral blood lymphocytes (PBL),
`and murine C3H cells. The antiviral activity of
`nucleoside analogs depends critically on cellular
`kinases for the metabolic steps to produce the 59-
`triphosphate. It is, therefore, not surprising that a
`compound’s activity may vary to some extent with
`the type of cell line and the phase in the cell cycle,
`since the occurrence and the relative amounts of
`these kinases vary under different circumstances
`[2,3]. Based on the phosphorylation profiles, AZT
`and d4T are considered as cell-activation-dependent
`nucleoside analogs, while ddI, ddC and 3TC are
`classified as cell-activation-independent nucleoside
`analogs [4].
`Myelosuppressive effects such as neutropenia and
`anemia are the main clinical adverse effects associ-
`ated with AZT [5,6]. Nucleoside analogs are evalu-
`ated for hematological toxicity by inhibition of in
`vitro colony formation of human hematopoietic
`progenitors CFU-granulocyte (CFU-G) and burst-
`forming unit-erythroid (CFU-E)
`in bone marrow
`cells. Peripheral neuropathy is the primary adverse
`effect associated with ddI, ddC and d4T [7,8]. The
`decreased mitochondrial DNA (mtDNA) synthesis
`caused by nucleoside analogs is monitored as it is
`proposed to be related to delayed cytotoxicity in
`vitro and in vivo side effects of these compounds.
`Unlike the natural nucleosides entering cells via
`facilitated transporters, anti-HIV nucleoside analogs
`permeate into cells by either simple diffusion or by a
`combination of simple diffusion and nucleoside / nu-
`cleobase carrier-mediated transportation [9,10]. Gen-
`erally,
`in vivo nucleoside analogs are extensively
`distributed throughout
`the body but with limited
`penetration into the central nervous systems (CNS);
`they are rather rapidly cleared from the systemic
`circulation with apparent half-lives usually less than
`3 h. Severe adverse effects and drug resistance are
`frequently observed clinically. Efforts have therefore
`been continuously made to develop and optimize
`nucleoside analogs. In addition to rigorous searches
`for new compounds with increased anti-HIV potency
`and selectivity, novel prodrug approaches have also
`been extensively investigated to improve the thera-
`peutic characteristics of
`the drugs. Prodrugs are
`designed to increase exposure of anti-HIV nucleoside
`
`analogs to the CNS and to target delivery of the
`drugs to the lymphatic system. The lymphatic system
`and CNS serve as reservoirs for HIV.
`This current review is intended to give an over-
`view of the field of nucleoside analogs and their
`prodrugs, with special focuses on the structure-ac-
`tivity relationships, cellular pharmacology and phar-
`macokinetics of the representative compounds.
`
`2. 29,39-Dideoxynucleoside analogs
`
`2.1. Structure-activity relationships
`
`29,39-Dideoxynucleoside analogs (dT, ddC, ddA,
`ddI and ddG) except for ddU have anti-HIV activity
`in ATH8 cell cultures to various extents [11]. In
`ATH8 cells, 39-azido (AZT) or 39-fluoro substitution
`(FLT) or 29,39-didehydro (d4T) modification greatly
`potentiates the anti-HIV activity of dT. On the other
`hand, 39-fluoro substitution (FLC) or 29,39-didehydro
`(Fd4C) modification decreases the anti-HIV activity
`of ddC. Similarly, 39-azido (AZdA) substitution or
`29,39-didehydro (d4A) modification decreases the
`anti-HIV activity of ddA and increases its cytotoxici-
`ty [12]. In MT-4 and CH3 cells, 39-azido (AZdU) or
`39-fluoro (FddU) substitution significantly increases
`the anti-HIV activity of ddU [13]. In MT-4 cells,
`39-azido (AZdG) or 39-fluoro (FddG) substitution
`increases
`the
`anti-HIV activity of ddG. 2,6-
`Diaminopurine
`29,39-dideoxyribose
`(ddDAPR),
`AZddDAPR and FddDAPR have higher anti-HIV
`activity than ddG. However, AZddDAPR is also one
`of the most cytotoxic nucleoside analogs [14]. 29,39-
`Didehydro (d4G) modification of ddG severely in-
`creases the cytotoxicity [15]. In H9 cells, 29-‘up’
`fluoro-substitution abolishes the anti-HIV activity of
`AZT, FLT and d4T, and decreases that of ddC
`[16,17]. 29-Azido-modification of dideoxypyrimidine
`analogs do not possess any anti-HIV activity [18]. In
`MT-4 cells, 59-isocyano-modification annihilates the
`anti-HIV activity of dT, AZT and AZdU [19]. In
`MT-4 cells, 59-O-phosphonomethylation abolishes
`the anti-HIV activity of dT and ddC, and decreases
`that of FLT and AZT [20]. In CEM cells, selected
`5-halo-6-alkoxy (or azido)-derivatives of AZT and
`FLT show equipotency against HIV [21,22].
`2-Chloro-modification substantially decreases the
`
`

`

`X. Tan et al. / Advanced Drug Delivery Reviews 39(1999)117–151
`
`121
`
`anti-HIV activity of ddA and d4A [23]. In MT-4
`cells, 5-chloro-derivative of FddU (FddClU) and
`AZdU (AZdClU) have significantly higher selectivity
`than that of FddU and AZdU [24,25]. 5-Chloro
`derivatives of FddC, AZdC, d4C and ddC result in
`reduced cytotoxicity with slightly reduced anti-HIV
`activity [26]. 6-Halo-29,39-dideoxypurine ribofurano-
`sides have potent anti-HIV activity in various cells
`[27].
`In CEM cells, b-L-ddC and b-L-FddC have potent
`anti-HIV and HBV activity with little inhibition of
`mtDNA synthesis, compared to b-D-ddC (natural
`configuration) [28]. Relative to d4C, b-L-Fd4C has
`more potent anti-HIV activity and less cytotoxicity,
`while b-L-d4C has comparable anti-HIV potency and
`cytotoxicity; b-L-Fd4C and b-L-d4C have little effect
`on mtDNA synthesis [29]. Among the isomeric
`dideoxynucleoside analogs, 4(S)-(6-amino-9H-purin-
`9-yl)tetrahydro-2(S)-furanmethanol
`(isoddA)
`and
`isoddG are potent anti-HIV activity in various cell
`cultures and low cytotoxicity [30].
`
`2.2. Thymidine (T) analogs
`
`thymidine and the antiviral
`The structures of
`thymidine analogs dT, AZT, FLT and d4T are shown
`in Fig. 2. The cellular kinases responsible for the
`phosphorylation of thymidine analogs are thymidine
`kinase (TK), thymidylate kinase (TmpK), and nu-
`cleoside diphosphate (NDP) kinase, which, respec-
`tively, mediate the metabolism to the mono-, di- and
`triphosphate anabolites. In this category, AZT and
`d4T are FDA approved anti-HIV drugs. FLT was
`tested in clinical
`trials without success owing to
`severe toxicity at doses required for anti-HIV effica-
`cy.
`
`2.2.1. Cellular pharmacology
`Zidovudine has potent and selective anti-HIV
`activity in various human cell cultures [31,32]. AZT
`is phosphorylated to its monophosphate (AZTMP),
`diphosphate (AZTDP) and triphosphate (AZTTP)
`anabolites by TK, TmpK and NDP kinase, respec-
`
`Fig. 2. Thymidine (T) and its analogs.
`
`

`

`122
`
`X. Tan et al. / Advanced Drug Delivery Reviews 39(1999)117–151
`
`tively. The rate-limiting step for the conversion of
`AZT to AZTTP is the diphosphorylation, which
`results in the intracellular accumulation of AZTMP
`while levels of AZTTP are two orders of magnitude
`lower. AZTTP is a competitive inhibitor of HIV-RT,
`and has much less affinity to cellular pol a. The
`intracellular half-life of AZTTP is about 2 h [33]. In
`the presence of 2 mM AZTTP,
`the activities of
`HIV-RT and cellular pol g are inhibited by more than
`80 and 90%, respectively [34]. The incorporation of
`AZT–MP into viral DNA causes the chain termina-
`tion [35]. The hematopoietic toxicity of AZT is
`generally attributable to the high intracellular levels
`of AZTMP which are found to be incorporated into
`DNA of human bone marrow cells [36]. One
`paradigm proposes that high intracellular level of
`AZTMP may inhibit the cellular 39-exonuclease(s)
`and thus trigger the toxicity of AZT by impairing the
`repair of AZTMP-terminated DNA [37]. AMTMP
`may also participate in the inhibition of this type of
`repairing enzymes [38]. The other paradigm pro-
`poses the myelosuppresive toxicities of AZT are due
`to the inhibition of normal cellular protein and / or
`lipid glycosylation by AZTMP [39,40]. The recent
`X-ray crystallography structure of (TK) reveals that,
`when the 39-OH of thymidine is replaced by an azido
`˚
`group, the bulkier –N group causes a 0.5 A shift of
`3
`the p-loop (the binding site for phosphoryl donor).
`This increases the activation energy for phosphoryl
`transfer step and thus is attributing to 200-fold
`reduced phosphorylation rate of AZTMP in com-
`parison to TMP [41].
`FLT exhibits very strong anti-HIV activity in
`various cell lines [13,42]. FLTTP is one of the most
`potent inhibitors of HIV-RT [43]. In contrast to AZT,
`FLT is phosphorylated intracellularly to FLTTP
`without the high accumulation of FLTMP [44]. In
`cultures of normal human hematopoietic progenitor
`cells, FLT produces toxicity similar to AZT [45].
`The intracellular half-life of FLTTP is about 1 h
`[46]. Unlike the non-facilitated membrane transport
`of AZT, FLT permeates the cell membrane by a
`carrier-mediated mechanism as well as by simple
`diffusion [46].
`In different human cell lines, d4T has potent and
`selective anti-HIV activity [47–49]. The phosphoryl-
`ation of the parent drug to its monophosphate is the
`rate-limiting step of the sequential conversion of d4T
`
`to d4TTP which inhibits HIV-RT equipotently as
`AZTTP. The intracellular half-life of d4TTP is about
`3.5 h [50,51]. d4TTP inhibits pol g [34]. d4TTP is
`preferentially incorporated into the elongating viral
`DNA and terminates DNA synthesis at the incorpo-
`ration site. The 39,59-exonuclease pol e can not
`remove d4TMP from the 39-end DNA once it
`is
`incorporated into cellular DNA, while in the case of
`AZTTP,
`the enzyme remains about 20% of
`its
`normal deoxynucleotide excision [52]. Similar to
`AZT, d4T enters cells by non-facilitated diffusion
`[53]. d4T shows 10-fold lesser toxicity to human
`hematopoietic progenitor cells compared to AZT
`[54]. After exposure of human bone marrow cells to
`similar extracellular levels of parent drugs, steady-
`state levels of d4TMP incorporated into cellular
`DNA are 10- to 50-fold less that of AZTMP [55]. In
`CEM cells, d4T decreases mtDNA synthesis with
`potency higher than AZT [7].
`
`2.2.2. Pharmacokinetics
`The pharmacokinetics of AZT have been reviewed
`[56,57]. Briefly, in patients with AIDS and ARC,
`AZT is absorbed quickly following oral administra-
`tion; it distributes extensively throughout the body,
`and appears in cerebrospinal fluid (CSF). The plasma
`protein binding of AZT is less than 25% and it is
`cleared rapidly from the body. Renal clearance
`(CL ) of the unchanged drug accounts for 20% of
`R
`total body clearance (CL ) [58,59] (Table 1). The
`T
`main metabolite of AZT is the 59-O-glucuronide
`(GAZT) [60]. Minor amounts of the toxic 39-amino-
`39-deoxythymidine (AMT) and its 59-O-glucuronide
`(GAMT) are also formed [61]. Bone marrow sup-
`pression is observed with AZT therapy, and more
`frequently the clinical adverse effects are anemia and
`neutropenia [62].
`Despite the promising in vitro anti-HIV profiles of
`FLT, clinical trials with this nucleoside analog failed
`because of its severe hemotological
`toxicity [63].
`The pharmacokinetics of FLT has been studied in
`animal models such as rat [64] and monkey [47,65].
`The pharmacokinetics of d4T have been reviewed
`[66,67]. Briefly, in patients with AIDS and ARC,
`d4T is absorbed readily after oral administration. The
`compound distributes moderately throughout
`the
`body, and enters CSF. d4T is cleared moderately
`from the body and about 40% of the administered
`
`

`

`X. Tan et al. / Advanced Drug Delivery Reviews 39(1999)117–151
`
`123
`
`Table 1
`Pharmacokinetic parameters of FDA-approved nucleoside analogs in patients with AIDS and ARC
`
`AZT
`[58,59]
`
`5 (iv)
`10 (po)
`63
`1.4
`
`1.3
`2
`(0.8 l / m / min)
`1.1
`19
`15–135
`
`D4T
`[68]
`
`0.67–4
`
`. 80
`0.88–1.06
`
`0.4–0.5
`
`1–1.6
`39–42
`1
`
`ddC
`[91,92]
`
`0.25–0.5
`
`88
`0.54
`
`2
`0.2 l / m / min
`
`1.2
`75
`0.14 (2 h)
`
`3TC
`[131]
`
`0.25–8
`
`82
`1.3
`
`2.4
`
`2.5
`70
`0.06 (2 h)
`
`ddI
`[106,109]
`
`0.2–6.4
`
`38
`1.01
`
`1
`
`0.6
`36
`0.21 (1 h)
`
`Dose
`(mg / kg)
`F (%)
`V
`SS
`(l / kg)
`CL
`T
`(l / kg / h)
`(h)
`t
`1 / 2,b
`(%)
`f
`SF:plasma
`
`eC
`
`dose is excreted unchanged in the urine. No metabo-
`lites are found in plasma or urine [68] (Table 1).
`Small amounts of d4T 59-glucuronide are detected in
`the urine of monkeys [65]. The primary dose-depen-
`dent adverse effect associated with d4T therapy is
`sensory peripheral neuropathy [69].
`
`2.3. 29-Deoxyuridine (dU) analogs
`
`AZdU (CS-85) and 39-fluoro-5-chloro-29,39-deox-
`yuridine (FddClU) are two 29-deoxyuridine analogs
`demonstrating clinical potential in AIDS chemother-
`apy (Fig. 3). The intracellular phosphorylation of
`29-deoxyuridine analogs is mediated by the same
`enzymes involved in the intracellular anabolism of
`thymidine analogs as described above.
`
`2.3.1. Cellular pharmacology
`AZdU is phosphorylated to AZdUTP that acts as
`both the HIV-RT inhibitor [70] and proviral DNA
`chain terminator [71]. AZdU inhibits HIV replication
`with less potency than AZT in human PBM cells but
`exhibits 20-fold less toxicity toward human marrow
`cells [72]. In human PBM cells, AZdUMP is the
`predominant
`intracellular metabolite. Levels of
`AZdUMP are two orders of magnitude greater than
`AZdUTP. dUMP, the substrate of thymidylate synth-
`ase, is readily converted into TMP intracellularly.
`AZdUMP, however, is not converted to AZT-MP;
`rather uniquely, besides phosphorylation to AZdUDP
`and AZdUTP, AZdU is metabolized to its 59-O-
`diphosphohexose
`and
`59-O-diphospho-N-acetyl-
`glucosamine, which may be accounted for the lower
`
`Fig. 3. 29-Deoxyuridine (dU) and its analogs.
`
`

`

`124
`
`X. Tan et al. / Advanced Drug Delivery Reviews 39(1999)117–151
`
`toxicity of AZdU [73]. AZdU is cross resistant with
`AZT-resistant HIV virus [74].
`FddClU shows antiviral activity against HIV and
`remarkably low cytotoxicity in human leukemic cells
`and bone marrow progenitor cells [75,76]. FddClU
`hardly induces any resistance of HIV-1, moreover, it
`is sensitive to strains of HIV which are, respectively,
`resistant to AZT, ddI and ddC, 3TC and FTC, as
`well as many non-nucleosides [76]. FddClU is
`metabolized to its mono-, di- and triphosphate in
`human cells, and the monophosphate is the pre-
`dominant metabolite;
`its
`triphosphate selectively
`inhibits HIV-RT and DNA pol g while has little
`effects on DNA pol a and b [76].
`
`2.3.2. Pharmacokinetics
`AZdU is currently being developed as an anti-HIV
`agent. Pharmacokinetic profiles of AZdU are com-
`parable to those of AZT. These two nucleosides have
`similar oral bioavailabilities, steady-state volumes of
`distribution (V ),
`terminal half-lives,
`total body
`SS
`clearance values, metabolic pathways and excretion
`
`routes, as has been demonstrated in rat and monkey
`models [77,78].
`of
`studies
`pharmacokinetic
`The
`preliminary
`FddClU have been performed in mice and monkeys.
`FddClU is well absorbed with oral bioavailability of
`86% in mice and 60% in monkeys. FddClU is
`cleared from the systemic circulation at about 2- to
`3-fold lower rate than AZT in mice and monkeys. It
`is metabolized similarly as AZT in mice, while in
`monkeys the percent of the dose excreted in urine as
`glucuronide conjugate is much lower for FddClU
`(4%) than AZT (45%) [76].
`
`2.4. 29-Deoxycytidine (dC) analogs
`
`The cellular enzymes mediating the mono-, di-
`and triphosphorylation of dC analogs are dC kinase,
`CMP/ dCMP kinase and NDP kinase, respectively,
`the same as those for dC. The 29-deoxycytidine
`analogs
`discussed
`here
`include
`ddC,
`FddC,
`AZddMeC (Fig. 4).
`
`Fig. 4. 29-Deoxycytidine (dC) and its analogs.
`
`

`

`X. Tan et al. / Advanced Drug Delivery Reviews 39(1999)117–151
`
`125
`
`2.4.1. Cellular pharmacology
`ddC shows potent and selective anti-HIV activity
`in human cells [11]. In various cell lines ddC is
`phosphorylated intracellularly to its mono-, di- and
`triphosphates, and their concentrations are at
`the
`same order of magnitude. ddC is not subjected to
`deamination [79]. The affinity of ddCTP to DNA pol
`a is very poor, while intermediate to pol b and high
`to pol g [80]. At concentrations lower than the
`plasma concentrations of clinical
`relevance, ddC
`exerts delayed cytotoxicity and reduces the cellular
`content of mtDNA [81]. Compared to the other four
`FDA-approved nucleoside analogs, ddC has the
`highest potency of inhibiting mtDNA synthesis [7].
`At higher concentrations, ddC causes a delayed
`distortion of mitochondrial ultrastructure [82]. ddC
`enters slowly into cells via passive diffusion and
`nucleoside carrier mediated pathways [10].
`b-L-ddC and its derivative b-L-29,39-dideoxy-5-
`fluorocytidine (b-L-FddC) show strong anti-HIV and
`hepatitis B (HBV) activity in CEM cells and no
`inhibition against mtDNA synthesis at a concen-
`tration up to 100 mM [83]. In different human cell
`cultures, the anti-HIV activity of b-L-FddC is about
`10-fold more potent
`than b-L-ddC and the same
`order of potency as ddC. b-L-FddC has the lowest
`toxicity compared to AZT, ddC, b-L-ddC and FddC.
`b-L-ddC and b-L-FddC are sensitive to AZT-resistant
`viruses, while cross resistant to 3TC and (2)-FTC-
`resistant viruses [84]. b-L-ddCTP and b-L-FddCTP
`exhibit potent inhibition against HIV-RT similar to
`their D-enantiomers, but not human pol a and b.
`They can be incorporated into viral DNA and act as
`chain terminators [85]. Furthermore, b-L-ddCTP and
`b-L-FddCTP are 100-fold less potent than their D-
`counterparts in inhibiting pol g. Not all L- and
`D-enantiomers are substrates for pol d and « [86].
`The combinations of b-L-FddC with other nucleoside
`analogs act synergistically (AZT, d4T) or additively
`(ddI, ddC) to inhibit the replication of HIV in vitro.
`b-L-FddC reduces mitochondrial
`toxicity of these
`analogs [87].
`AZddMeC demonstrates potent anti-HIV activity
`in human PBM cells and macrophages. This nu-
`cleoside analog has 40- to 60-fold less toxicity than
`AZT in human bone marrow cells. AZddMeCTP
`efficiently inhibits HIV-RT competing with dCTP
`while binds human DNA pol a with much lower
`affinity ( , 6000-fold) [88].
`
`2.4.2. Pharmacokinetics
`The pharmacokinetics of ddC has been reviewed
`[89,90]. Briefly, in patients with AIDS and ARC,
`ddC is well absorbed after oral administration. The
`drug distributes to the extent of the volume of total
`body water and penetrates the CSF, although to a
`lesser degree than other nucleosides such as AZT.
`ddC is moderately cleared from the body, mainly via
`urine excretion, no ddU and other metabolite is
`detected [91,92] (Table 1). A minor metabolite ddU
`is found in the urine samples of monkey [93].
`Peripheral neuropathy is the major dose-limiting
`adverse effect associated with ddC treatment [94].
`Pharmacokinetic studies of AZddMeC have been
`carried out
`in rats and monkeys [88,95,96].
`In
`monkeys, AZddMeC is absorbed with an oral bio-
`availability of 26%. It distributes extravascularly
`with a steady-state volume of distribution of 0.9
`l / kg;
`its penetration into CSF is negligible.
`In
`monkeys, approximately 32% of the administrated
`dose is deaminated to AZT. No glucuronide metabo-
`lite of AZddMeC is found, however AZT-glucuro-
`nide
`is
`detected
`following
`administration
`of
`AzddMeC to monkeys. Unchanged AZddMeC is
`excreted in urine. The total clearance of the nu-
`cleoside analog (2.0 l