throbber
Pronucleotides:Toward the InVivo
`Delivery of Antiviral and
`Anticancer Nucleotides
`
`CarstonR.Wagner,VidhyaV.Iyer,EdwardJ.McIntee
`
`Department of Medicinal Chemistry, College of Pharmacy, University of Minnesota,
`Minneapolis, MN 55455
`
`!
`
`Abstract: To overcome the many hurdles preventing the use of antiviral and anticancer
`nucleosides as therapeutics, the development of a prodrug methodology (i.e., pronucleotide) for
`the in vivo delivery of nucleotides has been proposed as a solution. The ideal pronucleotide should
`be non-toxic, stable in plasma and blood, capable of being i.v. and/or orally dosed, and
`intracellularly convertible to the corresponding nucleotide. Although this goal has yet to be
`achieved, many clever and imaginative pronucleotide approaches have been developed, which are
`likely to be important pharmacological tools. This review will discuss the major advances and
`future directions of the emerging field of antiviral and anticancer pronucleotide design and
`development. (cid:223) 2000 John Wiley & Sons, Inc. Med Res Rev, 20, No. 6, 417–451, 2000
`
`Key words: prodrug; pronucleotide; nucleotide; antiviral; anticancer
`
`1. INTRODUCTION
`
`Nucleosides and nucleotides have demonstrated wide-spread utility as antiviral and anti-cancer
`therapeutics.1,2 Natural endogenous nucleosides must be phosphorylated to the corresponding 50-
`triphosphates (TP) to be incorporated into the DNA strand being synthesized in the cell. The first
`phosphorylation step leading to the formation of nucleoside 50-monophosphate (MP), is commonly
`catalyzed by a nucleoside kinase encoded by the host cell or the virus infecting the host cell (Fig. 1).3
`Conversion of nucleoside-MPs to the corresponding 50-diphosphates (DP) and triphosphates is
`carried out by nucleoside, nucleotidyl, and nucleoside diphosphate kinases, respectively. Thus,
`cellular kinases and virally-encoded kinases play a vital role in the metabolism and replication of
`cells and viruses.
`Based on the metabolite–antimetabolite approach, nucleoside analogs such as 20,30-
`dideoxynucleosides (ddNs) have been developed as competitors of natural 20-deoxynucleoside
`50-triphosphates (dNTPs). Typically, modifications at the 20 or the 30carbon atoms of the glycone
`(sugar) moiety of nuelcosides are introduced. By virtue of their resemblance to the natural 20-
`deoxynucleosides, ddNs are phosphorylated to the corresponding 50-triphosphates and incorporated
`
`Correspondence to: Carston R.Wagner
`
`(cid:223) 2000 John Wiley & Sons, Inc.
`
`417
`
`IPR2018-00119
`
`Page 1 of 35
`
`I-MAK 1010
`
`

`

`418
`
`* WAGNER ETAL.
`
`Figure1. Intracellular metabolism of nucleosides.3
`
`into a growing DNA strand by a DNA polymerase, resulting in chain termination. Therefore, ddNs
`are in essence prodrugs, since they must be phosphorylated intracellularly in order to be biologically
`active.
`Therapies involving long-term administration of ddNs such as 30-azido-2030-dideoxythymidine
`(AZT) have been reported to lead to decreased activity of the first phosphorylating enzyme,
`thymidine kinase and thus, resistance.4 This type of resistance is observed not only in host tissues of
`the patients undergoing ddN therapy, but also in viruses. This resistance mechanism renders the
`ddNs less effective since their activation is hindered at the first phosphorylation step. In addition, for
`antiviral acyclic nucleosides such as acyclovir and penciclovir, the dependence of the nucleosides
`on activation to the triphosphates by virally encoded thymidine kinase, limits their spectrum of
`antiviral activity to those viruses such as herpes simplex virus (HSV) and Varicella zoster virus
`(VZV) which encode their own thymidine kinase.5 Thus, viruses such as hepatitis B virus (HBV),
`which do not encode their own nucleoside kinase, do not fall within the purview of activity of these
`antiviral nucleosides. Moreover, ddNs such as AZT, are associated with myelosuppressive side-
`effects, such as anemia and neutropenia.6 Toxic side-effects have been widely reported to lead to the
`discontinuation of ddN therapy. Certain other ddNs such as 20,30-dideoxyuridine (ddU), are poor
`substrates for thymidine kinase or other cellular kinases, and are, therefore, not converted to the
`corresponding triphosphates.7,8
`In principle, administration of the 50-phosphates would aid in overcoming the drawbacks of
`ddN therapy posed by resistance mechanisms and inherent biological differences. However, because
`phosphates are strongly acidic and thus negatively charged at physiological pH (pH(cid:136) 7.4), they are
`too hydrophilic to penetrate the lipid-rich cell membranes. In addition, blood and cell surface
`
`IPR2018-00119
`
`Page 2 of 35
`
`I-MAK 1010
`
`

`

`*
`
`419
`
`PRONUCLEOTIDES
`phosphohydrolases (acid and alkaline phosphatases, 50-nucleotidases)
`phosphates to the corresponding nucleosides.
`In order to overcome the poor cell penetration of nucleoside 50-phosphates, Montgomery9
`proposed that ‘‘this difficulty might be overcome if one could prepare an ester of a nucleotide which
`could penetrate the cell wall and then be metabolized to the nucleotide itself.’’ Consequently,
`various prodrug or ‘‘pronucleotide’’8 approaches have been devised and investigated. In general, the
`goal of these approaches has been to promote passive diffusion through cell membranes and
`increase the bioavailability of phosphorylated nucleosides. This approach of derivatization has been
`applied using various protecting groups for the phosphate moiety. Ideally, the attempts have been
`designed to achieve stability in the extracellular medium and rapid intracellular hydrolysis to release
`the phosphate. In most cases, to be biologically active, the phosphate has to be activated to the
`diphosphate and triphosphate. This review will discuss what we believe to be the major advances
`and future directions of the emerging field of antiviral and anticancer pronucleotide design and
`development.
`
`rapidly convert
`
`the
`
`2. ALKYL AND ARYL PHOSPHATE
`DERIVATIVES
`
`A. AlkylandArylPhosphodiesters
`
`A simple solution to the delivery of nucleotide monophosphates to cells is the use of alkyl phosphate
`esters. For example, aryl phosphodiesters of 6-mercaptopurine (6-MP) riboside were prepared in
`order to deliver 6-MP to 6-MP resistant neoplasms.9 The attempt was unsuccessful, probably due to
`poor cell penetration.10
`Alkyl and aryl phosphodiesters of b-D-arabinofuranosylcytosine (Ara-C), 20,30-didehydro-30-
`deoxyadenosine (d4A), 20,30-didehydro-30-deoxycytosine (d4C), and ddU have also been
`synthesized (Fig. 2). It had been reported that of a series of lipophilic 50-(alkyl phosphate) esters
`of Ara-C, only the Ara-C 50-(glyceryl phosphate) possessed cytotoxicity comparable to that of Ara-
`C-MP (CC50(cid:136) 0.35 and 0.65 mM, respectively) towards L1210 leukemia cells.11 The glyceryl
`phosphate ester also demonstrated in vivo activity against a kinase-deficient P388 murine
`leukemia.12 These results suggested that the development of b-hydroxyalkyl phosphate esters and
`other hydroxyalkyl (steroids) substitutions would be a worthwhile endeavor.11 Hong and co-
`workers13 reported that alkyl esters of both Ara-C-MP and Ara-C-DP demonstrated lower growth
`inhibitory activity than steroidal esters of Ara-C and Ara-C-DP. Nevertheless, the Ara-C-DP
`steroidal esters were found to possess lower antitumor activity than Ara-C-MP and Ara-C-DP alkyl
`esters against L1210 lymphoid leukemia in mice in vivo.13 A decrease in anti-leukemic activity was
`observed with increasing alkyl chain length towards cultured L1210 leukemic cells. However, Ara-
`C-MP alkyl esters containing alkyl groups of 16–20 carbon atom length were reported to be orally
`active prodrugs of Ara-C.14
`Mullah and co-workers15 reported that 50-phenyl- and 50-methyl-phosphate diesters of d4A and
`d4C demonstrated in vitro anti-HIV activity and cytotoxicity comparable to the corresponding
`parent nucleosides (Fig. 2). The compounds were cleaved under test conditions and released the
`parent nucleosides or the nucleoside 50-monophoshates in medium containing 10% serum.15
`Synthesis of ionophore–nucleotide conjugates of 15 crown ether-AZT and 15 crown ether-ddU 50-
`phosphate diesters has also been reported (Fig. 2).16 After association with a metal cation, these
`conjugates likely form a lipophilic ion-pair which could diffuse through a bilayer membrane. Once
`inside the membrane, the aryl phosphate ester linkage of the conjugate could be cleaved to release
`the nucleoside 50-monophosphate. Although AZT was found to be approximately 14-fold more
`potent than the AZT–ionophore conjugate, the ddU–ionophore conjugate had approximately 11
`
`IPR2018-00119
`
`Page 3 of 35
`
`I-MAK 1010
`
`

`

`420
`
`* WAGNER ETAL.
`
`Figure2. Alkyl and aryl phosphodiesters.9,11,15,16,18
`
`times the activity of ddU against HIV-1 infected CEM-SS cells, suggesting nucleoside 50-
`monophosphate delivery.16
`
`B. SteroidPhosphodiesters
`
`Based on the synergistic effect observed between certain drugs and the steroid prednisolone against
`human lymphoid leukemias and lymphomas, nucleotide–steroid conjugates were developed and
`evaluated for
`their anti-cancer effects. Ara-C-MP-prednisolone and Ara-C-MP-prednisone
`conjugates had greater anti-cancer effects against L1210 lymphoid leukemia in female mice than
`Ara-C alone or in combination with the individual steroid components.17 Ara-C-MP conjugates of
`corticosterone, cortexolone, 6a-methylprednisolone, and 11-deoxycorticosterone were also found to
`increase the life-span of mice with L1210 leukemia to a greater extent than Ara-C18 (Fig. 2) The
`conjugates were also found to be resistant to human liver cytidine deaminase and alkaline
`phosphatase, but sensitive to phosphodiesterase I, 50-nucleotidase, and acid phosphatase. They were
`shown to hydrolyze slowly in blood (t1/2(cid:136) 24–48 hr) and were able to cross the blood–brain barrier.
`These nucleotide–steriod conjugates did not deliver Ara-C-MP to the cells, as was evident from
`their lack of activity against a kinase-deficient L1210 leukemia resistant to Ara-C. However, many
`of
`the conjugates did show superior potency, compared to Ara-C, upon intraperitoneal
`administration against L1210 leukemia implanted intracerebrally.19,20 Other examples of steroid
`conjugates include the 3-fluorodeoxyuridine (FUdR)-MP-7b-hydroxycholestrol conjugate, which
`was found to possess in vitro cytotoxic activity against EL-4 murine leukemia cells and in vivo
`antitumor activity in mice with Krebs II ascitic carcinoma.21 The AZT-MP-3b(7b-hydroxychole-
`sterol) conjugate also demonstrated in vitro anti-HIV activity.22
`
`IPR2018-00119
`
`Page 4 of 35
`
`I-MAK 1010
`
`

`

`PRONUCLEOTIDES
`
`*
`
`421
`
`C. AlkylandArylPhosphotriesters
`
`In general, alkyl and aryl phosphodiesters have proven to be unsuitable for the delivery of nucleo-
`tides, presumably due to their inherent polarity and sensitivity to phosphodiesterase. Consequently,
`simple alkyl and aryl phosphotriesters of nucleosides have been developed in order to increase the
`lipophilicity of the phosphate by the neutralization of negative charges of the phosphate moiety.9 The
`increased lipophilicity was postulated to overcome the inactivity of phosphodiesters by promoting
`passive diffusion of the phosphotriesters into the cells (Fig. 3). Literature reports have documented
`the synthesis and biological evaluation of alkyl and aryl phosphotriesters of AZT, Ara-C, dideoxy-
`cytidine (ddC), b-D-arabinofuranosyladenine (Ara-A), 20,30-didehydrodideoxythymidine (d4T), 30-
`acylthymidine and netivudine (1-(b-D-arabinofuranosy)-5-prop-1-ynyluracil) and 6-MP. In general,
`the phosphotriesters were found to be less potent than the corresponding parent nucleosides.9,10,23–34
`Ara-A and Ara-C are inactivated by the deamination catalyzed by adenosine deaminase and
`cytidine deaminase, respectively. However, the simple alkyl phosphotriesters of Ara-A and Ara-C
`were reported to display resistance to the action of deaminases, phosphodiesterase I, and lipase
`(Fig. 3).24,25 These phosphotriesters were found to be hydrolyzed slowly by alkaline phosphatase
`
`Figure3. Nucleoside alkyl and aryl phosphotriesters.24^26,28,29,33,34,38
`
`IPR2018-00119
`
`Page 5 of 35
`
`I-MAK 1010
`
`

`

`422
`* WAGNER ETAL.
`and were stable at 37(cid:14)C, pH 7.4 for up to one week.25 In addition, because of their greater
`lipophilicity, they were shown to penetrate cell membranes. Unfortunately, they displayed no
`antiviral activity or cytotoxicity up to 100 mg/ml, although they were found to inhibit [3H]thymidine
`incorporation into the DNA of the mammalian epithelial cell line, CNCM 1221.24,25
`In an attempt to overcome the stability of simple alkyl phosphotriesters and induce the
`decomposition of these phosphotriesters, bis(2,2,2-trichloroalkyl) phosphates of thymidine and ddC
`were synthesized (Fig. 3).26 While the thymidine derivatives demonstrated modest anti-HIV activity,
`none of the analogs demonstrated antiviral potency comparable to that of the parent nucleosides.26
`Mixed alkyl monohaloalkyl phosphates of Ara-C were found to be more potent than simple alkyl
`phosphotriesters at inhibiting [3H]thymidine incorporation into DNA, with increasing potency
`correlating with increased lipophilicity (Fig. 3). Similarly, bis(haloalkyl) phosphotriesters of Ara-C
`were more potent than alkyl monohaloalkyl phosphotriesters of Ara-C.28
`Mixed aryl haloalkyl and diaryl phosphotriesters of AZT were synthesized and tested in human
`T-cell C8166 and the thymidine kinase-deficient JM cells for anti-HIV-1 activity. Except for the
`di(p-nitrophenyl) phosphate, none of the phosphotriesters were found to have greater anti-HIV-1
`potency than the nucleoside and were only marginally more potent than AZT in JM cells.29 The
`diaryl phosphates of AZT were found to have comparable or greater anti-HIV-2 activity compared to
`AZT in CEM/0, MT4, and the thymidine kinase-deficient CEM/TK(cid:255) cell lines, thus demonstrating
`their ability to not only penetrate the cell membranes, but also deliver
`the nucleoside
`monophosphates into the cells.32 With the exception of 30-acetyl, 30-ethyl, and 30-mesyl thymidine,
`bis(2,2,2-trihaloalkyl), simple alkyl, and mixed alkyl haloalkyl phosphates of thymidine analog
`antiviral nucleosides (i.e., d4T, and AZT) were found to be less potent against HIV-1 in C8166 cells
`than the parent nucleosides.27,30,31,35,36 Glycolate and lactate analogs of d4T were also examined,
`but were not found to be particularly active against HIV-1 or HIV-2 in the human T-lymphocyte
`CEM/0 and CEM/TK(cid:255) cell lines (Fig. 3).34 Similarly, alkyl, haloakyl, and aryl phosphotriesters
`of netivudine were synthesized and evaluated for anti-HCMV, HSV-2, VZV, influenza A, and
`HIV-1 activities (Fig. 3).34 All
`the phosphotriesters were found to be inactive at noncyto-
`toxic concentrations except for the 50-(4-nitrophenyl)phosphate which inhibited influenza A viral
`replication (EC50(cid:136) 2.3 mM).33
`
`D. CyclicAlkylPhosphotriesters
`
`Literature reports have demonstrated that while 5-membered cyclic phosphates are highly acid and
`alkali labile, 6-membered and 7-membered cyclic phosphates are stable under these conditions.37 A
`series of cyclic phosphates of FUdR and thymidine were synthesized and evaluated for cytotoxicity
`towards murine leukemia L1210 cells.38 A 50 0,50 0-difluoro-10 0,30 0-dioxa-20 0-phosphacyclohex-20 0-
`yl-20 0-oxide-50-cyclophosphate derivative of FUdR was found to be nearly as cytotoxic as FUdR
`toward L1210 cells (ID50(cid:136) 0.003 g/ml and 0.001 mg/ml, respectively) (Fig. 3). However, despite its
`
`lack of inhibitory activity towards cell-free thymidylate synthetase, the cytotoxicity of the cyclic
`phosphate of FUdR toward L1210 cells was reversed by the addition of exogenous thymidine
`(ID50(cid:136) 5 mg/ml). The comparable activity of the cyclic phosphate in the thymidine kinase-deficient
`L1210/BdUrd cell line to that of FUdR and FUdR-MP (ID50(cid:136) 2.42(cid:6) 0.4 mg/ml, 1.75(cid:6) 0.4 mg/ml,
`and 2.43(cid:6) 0.43 mg/ml, respectively) suggested that the compound was hydrolyzed extracellularly to
`FUdR, before cellular uptake.38
`
`E. Cyclosaligenyl(CycloSal)Phosphotriesters
`
`The cyclosaligenyl (cycloSal) phosphotriester approach has been investigated as a potential vehicle
`for the delivery of monophosphates of d4T, FUdR, acyclovir (ACV), penciclovir (PCV),
`T-penciclovir (T-PCV), dideoxyadenosine (ddA), d4A, dideoxyinosine (ddI), didehydro-dideox-
`
`IPR2018-00119
`
`Page 6 of 35
`
`I-MAK 1010
`
`

`

`PRONUCLEOTIDES
`
`*
`
`423
`
`Figure4. Nucleoside CycloSal phosphotriesters.39^44,46
`
`yinosine (d4I), 20-fluoro-ara-20,30-dideoxyadenosine (F-ara-ddA), 20-fluoro-ara-20,30-dideoxya-
`denosine (F-ara-ddI), 20-fluoro-ribo-20,30dideoxyinosine (F-ribo-ddA), and 20-fluoro-ribo-20,30-
`dideoxyinosine (F-ribo-ddI) derivatives.39 –43 The cycloSal approach was designed to release the
`monophosphate selectively by a controlled, chemical tandem hydrolysis mechanism. The cycloSal-
`nucleotide conjugate incorporates unique phosphate ester bonds into the same molecule. The
`mechanism of generation of the nucleoside 50-monophosphate from the nucleoside cycloSal
`phosphotriester is described in Fig. 4 and is initiated by selective hydrolysis of the phenolic ester
`bond.39 Subsequently, spontaneous cleavage of the benzylic phosphate ester bond releases the
`nucleoside 50-monophosphate and the salicyl alcohol.39 Studies of the hydrolysis of the 5-chloro
`18O water
`cycloSal-d4T-MP and 5-methoxy cycloSal-d4T-MP conducted in the presence of H2
`revealed that 18O was incorporated into both the salicyl alcohol and monophosphate products,
`confirming chemical release of the monophosphate. However, the results of these experiments were
`unable to determine if the reaction proceeds through a benzyl carbocation or quinone methide
`intermediate.44 Although quinone methide has been shown to preferentially alkylate guanines in
`duplex DNA, the ability of the cycloSal derivatives to alkylate DNA has not been examined.45
`Because these compounds are prepared and evaluated as a diastereomeric mixture, the potential
`for the disastereomers to exhibit different levels of antiviral and anticancer activity has been
`evaluated for pronucleotides of d4T and ddA. Interestingly, Meier and co-workers42,44 have
`observed a difference of 3- to 80-fold in antiviral potency between the isolated diastereomers. An
`explanation for this preference has not been reported.
`The stabilities of cycloSal conjugates of FUdR were found to be greater in Tris buffer, pH 6.9
`(t1/2(cid:136) 6.8–59.2 hr) than in phosphate buffer, pH 7.29 (t1/2(cid:136) 0.9–10.6 hr) or in borate buffer, pH 8.9
`
`IPR2018-00119
`
`Page 7 of 35
`
`I-MAK 1010
`
`

`

`424
`
`* WAGNER ETAL.
`
`(t1/2(cid:136) 0.4–2.1 hr).39 Besides decreased stability associated with increasing buffer pH,
`the
`conjugates also showed comparable or greater stability (t1/2(cid:136) 1.5–11.1 hr) in RPMI medium, pH
`7.4 than in RPMI containing 10% fetal calf serum (t1/2(cid:136) 2.2–10.7 hr).39 Increased stability was
`found to be associated with the decreased potency of the cycloSal phosphotriesters of FUdR.39
`In general, the cycloSal phosphotriesters were found to be more potent against HIV-1 in CEM/
`TK(cid:255) cells than the corresponding parent nucleosides. The cycloSal phosphotriesters of T-PCV, ddA,
`ddI, d4A, F-ribo-ddI, F-ribo-ddA, F-ara-ddI, and F-ara-ddA in CEM/0 cells demonstrated greater
`anti-HIV-1 potency, while those of d4T, ACV, PCV, and d4I had comparable or lower anti-HIV-1
`potency than the corresponding parent nucleosides.40 –43 However, the cycloSal phosphotriesters of
`FUdR displayed lower cytotoxicity than FUdR towards FM3A/0 and FM3A/TK(cid:255) cells (murine
`mammary carcinoma cells).39 CycloSal phosphotriesters have been used as mechanistic tools for
`understanding the effect of nucleoside structure on biological activity. For example, of the two
`20-fluoro analogs of ddA, F-ara-ddA and F-ribo-ddA, which were developed to overcome the
`susceptibility of ddA to deamination by adenosine deaminase (ADA) as well as acid-catalyzed
`glycosyl cleavage, F-ara-ddA proved to be more potent against HIV-1 and HIV-2 than F-ribo-
`ddA.43 It has been postulated that the conformations of the sugar rings and thus the activity of the
`two F-ddA isomers are strongly influenced by the orientation of 20-F.46,47 The cycloSal
`pronucleotide approach was applied to F-ara-ddA and F-ribo-ddA to determine if metabolic
`blockade at the monophosphorylation step is responsible for the difference in potency.43 The
`cycloSal approach did indeed appear to improve the anti-HIV activity of F-ribo-ddA by
`circumventing the metabolic barrier of hindered phosphorylation to the monophosphate due to
`the C-30-endo-conformation of the tetrahydrofuran ring.43
`Studies of the in vivo antiviral or anticancer potency of cycloSal phosphotriesters, as well as,
`toxicity have not been reported. In vivo toxicity studies have also not been reported. Salicyl alcohol
`(2-hydroxybenzyl alcohol) has been shown to be mild anesthetic.48 It is not known if similar in vivo
`biological activity is observed for substituted salicyl alcohols.
`
`3. BIOLABILE PROTECTING GROUPS FOR
`NUCLEOSIDE PHOSPHOTRIESTERS
`
`Although neutral lipophilic alkyl and aryl phosphotriesters are easily taken up by cells, in general,
`they were poorly hydrolyzed due to their stability. While phosphodiesters formed from
`phosphotriesters can be enzymatically converted to the nucleoside monophosphates, an endogenous
`eukaryotic phosphotriesterase activity has yet to be reported. In order to retain the advantages of
`nucleoside phosphotriesters with regard to their improved cellular uptake and still induce the
`hydrolysis of the phosphotriesters, a variety of biolabile moieties have been evaluated as potential
`protecting goups of the nucleoside monophosphate.
`
`A. S-Acyl-2-Thioethyl(SATE) Approach
`
`The SATE approach was first described by Imbach and co-workers and has been applied to the
`development of bis(SATE) phosphotriester analogs of ddU, ddA, ddI, AZT, 20,30-dideoxy-30-
`oxyadenosine (isoddA), d4T, and ACV.8,49 –52 The design and synthesis of closely related (iso)SATE
`analogs of D4T and ddA, monoSATE aryl phosphotriesters of AZT, S-pivaloyl-4-thiobutyl
`(tBuSATB) analogs of AZT and ddA and S-glycopyranosidyl (SGTE) analogs of AZT have also
`been reported (Fig. 5).53–56 The general decomposition pathway of these biolabile phosphotriesters
`to yield the nucleoside 50-monophosphates via the formation of a phosphodiester is described in
`Fig. 6. The hydrolysis of the bis(SATE) phosphotriester is initiated by the carboxyesterase-mediated
`hydrolysis of
`the thioester moiety of one of
`the SATE groups,
`to form the unstable
`
`IPR2018-00119
`
`Page 8 of 35
`
`I-MAK 1010
`
`

`

`PRONUCLEOTIDES
`
`*
`
`425
`
`Figure5. Pronucleotide approaches closely related to the SATE approach.8,49,50,51,53^56
`
`Figure6. Mechanism of decomposition of SATE pronucleotides.49
`
`IPR2018-00119
`
`Page 9 of 35
`
`I-MAK 1010
`
`

`

`426
`
`* WAGNER ETAL.
`
`O-2-mercaptoethylphosphotriester.49 The thiol generated, which is a soft nucleophile, attacks the
`soft electrophilic methylene carbon atom,
`releasing ethylene sulfide and the monoSATE
`phosphodiester.49 This phosphodiester
`is likely subject
`to hydrolysis by an intracellular
`phosphodiesterase to generate the nucleoside 50-monophosphate and S-acyl-thioethanol. Alter-
`natively, a second carboxyesterase-mediated thioester hydrolysis/ethylene sulfide step ensues to
`generate the nucleoside 50-monophosphate.49
`Another possible mechanism of hydrolysis of the bis(SATE) phosphotriester has also been
`proposed to occur via the direct nucleophilic attack of water on the phosphorus atom to release the
`S-acyl-thioethanol.49 The variation of the half-lives of the bis(SATE) phosphotriesters of ACV
`ranged from 2 min in mouse serum, to 5.5 hr in CEM cell extract, to 14 hr in human serum,
`suggesting that an esterase-mediated activation was more plausible than chemical hydrolysis.57
`Although not mutagenic by the Ames test, the potential for electrophilic ethylene sulfide or its
`metabolites released from SATE pronucleotides to react with nucleophilic groups on biological
`macromolecules has not been investigated.58
`The enhanced anti-HIV-1 activity of most of the SATE nucleoside phosphotriester analogs over
`that of the parent nucleosides, has been widely documented in CEM-SS, MT-4 (human T-
`lymphocytes), peripheral blood monocytes (PBMs), macrophages and CEM/TK(cid:255) cells, demon-
`strating that nucleoside 50-monophosphate
`can be
`successfully delivered into these
`cells.8,50,52,54,56,57,59 –61 Bis(SATE) d4T phosphotriesters demonstrated higher anti-HIV-1 potency
`than d4T in MT-4, CEM-SS, PBM, and CEM/TK(cid:255) cells.62 Only the AZT bis(SATE) phosphotriester
`analogs were found to be less potent than AZT in CEM-SS and MT-4 cells against HIV-1, but had
`superior anti-HIV-1 potency in CEM/TK(cid:255) cells, suggesting intracellular delivery of AZT-MP.8,49,61
`The closely related analogs—tBu(iso)SATE (O-pivaloyl-2-oxyethyl) thio-phosphotriesters of
`ddA and d4T, SGTE phosphotriester of AZT, the monoSATE aryl phosphotriesters of AZT, and
`tBuSATB phosphotriesters of AZT and ddA—were found to possess greater anti-HIV activity than
`the corresponding parent nucleosides, but were less active than the corresponding bis(SATE) pho-
`sphotriesters.53 –56 Some of the additional advantages of the bis(SATE) phosphotriester approach
`included not only the delivery of ddA-MP into the cells by the ddA bis(SATE) phosphotriester, but
`also resistance to adenosine deaminase (ADA). Deamination by ADA has been found to severely
`reduce the efficacy of ddA.8,51,63 In addition, unlike ACV, ACV bis(SATE) phosphotriesters were
`found to possess a broader antiviral spectrum of activity than ACV as was evident from their anti-
`hepatitis B viral activity.8,57,64,65 Moreover, AZT bis(SATE) phosphotriesters were able to
`overcome cellular resistance of MOLT4/8rAZR250 cells to AZT, because AZT-MP was delivered.59
`In vivo studies with duck HBV (DHBV) found that ACV bis(SATE) was somewhat more efficient
`than ACV when given over a short period of time orally or intraperitoneally.64 Thus, the bis(SATE)
`phosphotriester approach has proven to be a potentially effective way for both the in vitro and in vivo
`intracellular delivery of nucleoside 50-monophosphates.
`
`B. Dithioethyl(DTE)Approach
`
`The synthesis and biological evaluation of bis(DTE) (dithioethyl) phosphotriesters has been
`reported for AZT and ddU.49,50,59,61 The decomposition of the bis(DTE) analogs is described in
`Fig. 7 and is postulated to be initiated by a reductase-mediated reductive cleavage of the disulfide
`bond.8 As a result of
`the reductive cleavage,
`the unstable O-2-mercaptoethyl monoDTE
`phosphotriester
`is formed, followed by the release of ethylene sulfide or thioethanol by
`intramolecular nucleophilic displacement.8 The monoDTE phosphodiester can undergo hydrolysis
`mediated by phosphodiesterase or undergo a second reductase-mediated disulfide cleavage, to
`generate the nucleoside 50-monophosphate.8 As noted for the SATE approach, in vivo toxicity
`associated with the generation of ethylene sulfide from the DTE derivatives has not been
`reported.
`
`IPR2018-00119
`
`Page 10 of 35
`
`I-MAK 1010
`
`

`

`PRONUCLEOTIDES
`
`*
`
`427
`
`Figure7. Mechanism of decomposition of dethioethyl (DTE) pronucleotides.8,51
`
`Like the bis(SATE) analogs, bis(DTE) phosphotriester analogs of AZT were found to possess
`lower anti-HIV-1 potency than AZT in CEM-SS, MT-4, and PBM cells, but did demonstrate greater
`anti-HIV-1 potency than AZT in CEM/TK(cid:255) cells.8,49,61 However, the bis(DTE) phosphotriesters of
`ddU were consistently more active against HIV-1 in CEM-SS, PBM, and CEM/TK(cid:255) cells than the
`parent nucleoside.50,61
`In general, bis(SATE) phosphotriesters of anti-HIV-1 nucleosides
`consistently displayed greater potency than the corresponding bis(DTE) phosphotriesters.8,49,50,61
`
`C. Pivaloyloxymethyl(POM)Approach
`
`The pivaloyloxymethyl (POM) phosphoester approach has been utilized for the development of
`bis(POM) phosphotriesters of ddU, FUdR, AZT and other nucleoside analogs7,8,49,61,66 –68 (Fig. 8).
`Like bis(DTE) phosphotriesters, the bis(POM) phosphotriesters of AZT were found to be less potent
`than AZT towards HIV-1 in CEM-SS, MT-4, and PBMs, but more active in CEM/TK(cid:255) cells.8,49,61,69
`Farquhar and co-workers7,69 demonstrated that unlike ddU or AZT, the incubation of CEM/TK(cid:255)
`cells with bis(POM) esters of ddU and AZT resulted in increased intracellular levels of ddU and
`AZT monophosphate, diphosphate, and triphosphate. The bis(POM) phosphotriesters displayed
`varying degrees of antiviral and cytotoxic activities relative to the corresponding bis(SATE) and
`bis(DTE) derivatives.8,49,61
`The mechanism of decomposition of bis(POM) derivatives to the nucleoside 50-monophosphate
`is described in Fig. 8, and like the bis(SATE) approach, relies on carboxyesterase-mediated
`hydrolysis to initiate release of the nucleoside 50-monophosphate.70 The unstable O-2-hydroxyethyl
`phosphotriester forms, then undergoes intramolecular nucleophilic displacement to generate the
`monoPOM phosphodiester and formaldehyde. Similar to the bis(SATE) approach, conversion of the
`phosphodiester to the nucleoside 50-monophosphate is likely rate limiting and carried out by a
`phosphodiesterase or esterase.70 The generation of electrophilic formaldehyde, as a result of the
`hydrolysis of the bis(POM) phosphotriester, could be a source of long term cellular toxicity arising
`from the interaction of formaldehyde with cellular nucleophiles such as proteins, DNA or RNA.
`The POM derivative of a cyclic phosphate of FUdR was moderately stable in aqueous buffers
`over a pH range of 1–7.4 (t1/2 > 30 hr), but was degraded to FUdR-MP in a concentration-dependent
`manner by carboxylate esterase (Fig. 8).68 The POM ester of FUdR cyclic phosphate was also found
`
`IPR2018-00119
`
`Page 11 of 35
`
`I-MAK 1010
`
`

`

`428
`
`* WAGNER ETAL.
`
`Figure8. Pivaloyloxymethyl (POM) pronucleotide approach.68,70,71
`
`to be as potent a growth inhibitor as 5-fluorouracil toward Chinese hamster ovary (CHO) cells
`(IC50(cid:136) 5(cid:2) 10(cid:255)6 M). In addition, FUdR monophosphate pronucleotide was found to increase the
`life spans of P388 leukemia bearing mice upon intraperitoneal administration for five consecutive
`days, and was active against a CHO variant cell line and a P388 mutant cell line, both of which were
`resistant to 5-fluorouracil.68 Taken together, this data suggested that the POM phosphate diester
`intracellularly delivered FUdR-MP.
`Like the AZT-bis(POM) phosphotriester, the bis(POM) FUdR phosphotriester was stable at
`pH 1–4 (t1/2 > 100 hr) and at pH 7.4 (t1/2(cid:136) 40.2 hr) but was hydrolyzed quickly in 0.05 M NaOH
`(t1/2 < 2 min).71 FUdR was found to be only 5-fold more cytotoxic (IC50(cid:136) 1(cid:2) 10(cid:255)6 M) towards
`CHO cells than bis(POM) 30-acetyl FUdR (IC50(cid:136) 5(cid:2) 10(cid:255)6 M), which had been synthesized as a
`member of a series of bis(acyloxymethyl) phosphotriesters of 30-acetyl thymidine.71 The bis(POM)
`phosphotriester of ddU was also found to generate the ddU-TP from ddU-MP, which had been
`delivered to MT-4 cells.69 The utility of the approach was evident from the successful decrease in
`the cytopathic effects and virus production of HIV-1 by CEM/TK(cid:255) cells.7
`
`D. Para-Acyloxybenzyl(PAOB)Approach
`
`In an extension of the p-acyloxymethyl approach, the p-acyloxybenzyl approach has also been used
`to develop phosphotriesters of antiviral agents. Among the antiviral agents, bis(PAOB)
`phosphotriesters of AZT, bis(AZT) PAOB phosphotriester homodimers of AZT, and PAOB
`phosphotriester heterodimers of AZT and ddI have been studied (Fig. 9).72 –74 The advantage of the
`p-acyloxybenzyl approach is that upon ester hydrolysis, formation of the quinone methide results in
`cleavage of the C–O bond, thus liberating a phosphate hydroxyl group.46 Similar to the SATE or
`POM approaches, a phosphodiester is generated as an intermediate.
`The mechanism of decomposition of the p-acyloxybenzyl phosphotriesters is described in
`Fig. 9.72,73 Analogous to the bis(POM) phosphotriester approach, the PAOB approach relies on
`esterase-mediated hydrolysis to initiate the activation of the phosphotriesters. Upon formation of the
`
`IPR2018-00119
`
`Page 12 of 35
`
`I-MAK 1010
`
`

`

`PRONUCLEOTIDES
`
`*
`
`429
`
`Figure9. Bis(PAOB) pronucleotide approach.74
`
`unstable p-hydroxybenzyl phosphotriester, generation of the p-hydroxybenzyl carbonium ion is
`followed by the formation of the mono(PAOB) phosphodiester. The phosphodiester in turn,
`undergoes esterase-mediated hydrolysis to release the nucleoside 50-monophosphate, an acylate
`anion and a p-hydroxybenzyl carbonium ion. The disadvantage of this approach was postulated to
`be the reactivity of the p-hydroxybenzyl carbonium ion, which may react with either DNA, protein,
`g

This document is available on Docket Alarm but you must sign up to view it.


Or .

Accessing this document will incur an additional charge of $.

After purchase, you can access this document again without charge.

Accept $ Charge
throbber

Still Working On It

This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.

Give it another minute or two to complete, and then try the refresh button.

throbber

A few More Minutes ... Still Working

It can take up to 5 minutes for us to download a document if the court servers are running slowly.

Thank you for your continued patience.

This document could not be displayed.

We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.

Your account does not support viewing this document.

You need a Paid Account to view this document. Click here to change your account type.

Your account does not support viewing this document.

Set your membership status to view this document.

With a Docket Alarm membership, you'll get a whole lot more, including:

  • Up-to-date information for this case.
  • Email alerts whenever there is an update.
  • Full text search for other cases.
  • Get email alerts whenever a new case matches your search.

Become a Member

One Moment Please

The filing “” is large (MB) and is being downloaded.

Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.

Your document is on its way!

If you do not receive the document in five minutes, contact support at support@docketalarm.com.

Sealed Document

We are unable to display this document, it may be under a court ordered seal.

If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.


Access Government Site

We are redirecting you
to a mobile optimized page.





Document Unreadable or Corrupt

Refresh this Document
Go to the Docket

We are unable to display this document.

Refresh this Document
Go to the Docket