Certain phosphoramidate derivatives of dideoxy uridine (ddU) are active against HIV and successfully by-pass thymidine kinase FEBS Letters 351 (1994) II-14 FEBS 14389 Christopher McGuigan”,*, Patrice Bellevergue b, Hendrika Sheekab, Naheed Mahmood”, Alan J. Hayd
`
`a Weish School of Pharmacy, University of Wales Card$ Redwood Building, King Edward VII Avenue, Cardiff; CFI 3XE UK
`‘Department of Chemistry, University of Southampton, Highfield, Southampton, SO9 5NH, UK
`‘Medical Research Council Collaborative Cenrre. 1-3 Burtonhole Lnne, Mill Hill, London, N W7 IAD, UK
`‘National Institute for Medical Research, The Ridgeway, London, NW7 IAA, UK
`
`Received 27 June 1994 Abstract As part of our effort to deliver masked phosphates inside living cells we have discovered that certain phosphate triester derivatives of the inactive nucleoside analogue, dideoxy uridine (ddU) are inhibitors of HIV replication at PM levels. Moreover, we note that certain phosphoramidate derivatives retain their activity in thymidine kinase-deficient cells, which indicates that they do indeed act by intracellular release of the free nucleotide, and that they successfully by-pass the nucleoside kinase. The increased structural freedom in drug design which this allows may have implications for dealing with the emergence of resistance and may stimulate the discovery of improved therapeutic agents. Key words: Nucleoside; Nucleotide; Anti-HIV
`
`1. Introduction
`
`1,
`Although nucleoside analogues, such as 3’-azidothy- midine (AZT,
`
`Fig. 1) continue to dominate anti-HIV drug therapy they have a number of major limitations, such as their inherent toxicity, a dependence on kinase- mediated activation to generate the bio-active (tri)phosphate forms, and the emergence of resistance [l-2]. We [3-61 and others [7-91 have pursued a masked phosphate approach in an attempt to improve on the therapeutic potential of the parent nucleoside analogues. In this approach, inactive phosphate derivatives of the nucleoside analogue are designed to penetrate the cell membrane and liberate the bio-active nucleotides intra- cellularly. Masking of the phosphate group is necessary on account of the extremely poor membrane penetration by the polar (charged) free nucleotide. One mechanism by which masked phosphates may lead to enhanced selectivity of action arises from what we have termed ‘kinase bypass’ [3]. Thus, the complete dependence of administered (anti-HIV) nucleoside analogues on host nucleoside kinase-mediated activation places constraints upon the structures of nucleoside analogues which might be active. Nucleoside analogues which fall outside these strict constraints will be inactive, even if their 5’-tri- phosphates (the bio-active form) are potent and selective inhibitors of a viral target, such as reverse transcriptase (RT). Several such cases are known. Dideoxythymidine, and 3’-O-methylthymidine are examples of nucleoside *Corresponding author. Fax: (44) (222) 874537 or (44) (222) 874180. analogues which are inactive against HIV, whilst their triphosphates are exceptionally potent inhibitors of HIV RT [lo]; the inactivity of the nucleoside being attributed to poor phosphorylation by host kinases. If the masked phosphate strategy were able to deliver nucleotides in- tracellularly, the nucleoside kinase would be by-passed and the structural constraints such host enzymes impose would be obviated. In this way, wider structural varia- tion of the nucleoside analogue would be permitted, and more specific (less toxic) inhibitors of viral function may arise. We have recently reported on the success of this ‘kinase by-pass’ strategy with several highly modified 3’-substituted nucleosides [ 1 l-l 21. We now report the success of this approach with the simple nucleoside analogue dideoxy uridine (ddU, 2). This is essentially inactive against HIV, but judicious phosphorylation leads to the introduction of a signifi- cant, selective antiviral effect. Moreover, this effect is retained in thymidine kinase-deficient cells, indicating a successful by-pass of this enzyme, and strongly support- ing the suggested intracellular delivery of free nucleotides by this strategy. Whilst other researchers have recently reported the failure of the by-pass approach with phosphoramidates derived from ddu [13], we herein clearly demonstrate the success of this strategy with our previously reported aryloxy phosphoramidates [6] 2.
`
`Materials and methods
`
`2.1. Chemistry
`
`General synthetic procedures were similar to those we have described [S]. All nucleotides were pure by high-field multi-nuclear NMR and reverse phase High Performance Liquid Chromatography (HPLC: OOl4-5793/94/$7.00 0 1994 Federation of European Biochemical Societies. All rights reserved.
`
`SSDI
`
`0014-5793(94)00776-4
`
`IPR2018-00126
`
`Page 1 of 4
`
`I-MAK 1009
`
`

`

`250
`
`4.6
`
`C. McGuigan et nl.IFEBS Letters 351 (1994) 11-14
`
`mm, Spherisorb 0DS2 5 p column, gradient elution using 5% acetonitrile in-water (A), and 5% water in acetonitrile (B). with 20% B for O-10 min. then a linear gradient to 80% \ II B at 30 min, with a flow rate of 1 ml/min). 2.1.1.
`
`2’,3’-dideoxy uridine (ddlJ, 2).
`(a) 5’-Trityl uridine.
`
`2’,3’-dideoxy uridine 5’-(phenyl methoxyalaninyl)pho.sphate
`
`(3~).
`
`Virology
`2.2. I. Antiviral assays.
`
`This was prepared by a procedure which was entirely analogues to (3a) above, except that the reaction was stirred for 17 h, and 4% methanol in chloroform was used as chromatographic eluant. Thus, from 92 mg of ddU was isolated 178 mg (90%) of (3~). 6, (CDCI,) 1.32,1.33(3H, d, Ala-Me, J = 7 Hz), 1.8-2.1(3H, m, H2’,H3’), 2.3(IH, m, H2’), 3.66,3.67(3H, s, OMe), 4.0(lH, m, Ala-CH), 4.34.4(4H, m, H4’, HS’, Ala-NH), 5.60,5.65(1H, d, H5, J = 9 Hz), 6.0(lH. m, Hl’), 7.25(5H. m. Ph). 7.60.7.61(18. d. H6, J = 8 Hz). lO.O(lH, bs. NH): FAB MS m/e 454.1397 (M-H’, 13% C&H,,N,O,P requires 454.1379) 342(35), 307(14), ZOO(l5); 154(100),136(79); HPLC RT = 34.4, 34.6 min. 2.2.
`The anti-HIV-1 activities and toxicities of compounds were assessed in two cell lines 1141. C8166 (a normal T-cell transformed by co-cultivation with leukaemialymphocytes harbouring HTLV-I) were infected with the III-B strain of HIV-l. Secondly. JM, a semi-mature T-cell line derived from a patient with lymphoblastic leukaemia, were infected with HIV-I strains GB8 or BIB. JM cells are relatively resistant to the antiviral effects of AZT and a number of its derivatives. Cells were grown in RPM1 1640 with 10% calf serum. 4 x IO4 cells per microtiter plate well were mixed with 5-fold dilutions of compound prior to addition of IO CCID,,, units of virus and incubated for 557 days. Formation of syncytia was examined from 2 days post-infection. Culture fluid was collected at 557 davs and an120 antigen production measured by ELISA. Cell viability ofinfectezcells andcytotoxicity to uninfected cell controls were assessed by the MTT-Formazan method
`added to the wells and incubated at 4°C for 12-16 h. Bound antigen was detected using human anti-HIV-I antibodies, and anti-human IgG antibodies coupled to horseradish peroxidase. 3. Results and discussion Dideoxy uridine (ddU, 2) was prepared via three inde- pendent routes, all involving a final hydrogenation of the didehydro dideoxy compound d4U. Thus, 5’-silylation of uridine, followed by thiocarbonate formation, elimina- tion with triethyl phosphite, and deprotection with fluo- ride, gave 2’,3’-dideoxy 2’,3’-didehydro uridine (d4U) [17] in an overall yield of 50%. Alternatively, a 5’-trityl protecting group could be used in this synthesis to give d4U, again via the 2’,3’-thiocarbonate, in an overall yield of 58%. Alternatively, by the procedures of Horwitz [18] 2’-deoxyuridine could be converted to its 3’,5’-dimesyl- ate, which gave the 3’,5’-oxetane, and which could be converted to d4U on treatment with strong base. The
`
`1151.
`2.2.2. gpl20 antigen assay.
`
`A microtiter antigen capture ELISA was developed [16] using a lectin (GNA) from Galanthus nivalis (Vector Labs., Peterborough, UK) and anti-HIV antibodies in human serum. The plates were coated with lectin (0.5pg), and after blocking with 10% calf serum, dilutions of virus supernatants in 0.25% Empigen solution (Albright and Wilson Ltd., Whitehaven, UK)
`
`were
`
`2:3’-Dideoxy
`
`uridine 5’-bis[2,2,2-trichloroethyl]
`
`phosphate
`
`2’.3’-Dideoxy
`
`uridine 5’-(ethyl methoxyalaninyl)phosphate
`
`(3b). DdU (2) (0.05 g, 0.23 mmol) and N-methyl imidazole (0.09 g, 1.18 mmol) were mixed in THF (IO ml) and ethyl methoxyalaninyl phosphorochloridate (0.11 g, 0.47 mmol) was added slowly at ambient temperature. After 4h the solvent was evaporated and the crude prod- uct purified entirely as described for (3a) above, except that a second chromatographic column was necessary, using an eluant of 15% meth- anol in diethyl ether, in order to obtain pure (3b) (0.051 g, 54%). &(CDCI,) 1.3(6H, m, Ala-Me, POCC), 2.0(3H, m, H2’, H3’). 2.4(lH,m, H2’), 3.70,3.71(3H, s, OMe), 3.84.25(7H,m, Ala-CH, Ala- NH, H4’, H5’, POC), 5.7(lH, d, H5, J = 8 Hz), 6.05(lH, m, Hl’), 7.68,7.73(lH,d,H6,J=8Hz),9.5(lH,bs,NH);FABMSm/e406.1392 (MH’, IO%, CSH,,N,O,P requires 406.1379) 294( IOO), 94(7); HPLC RT = 29.9,30.3 min. 2.1.4.
`
`5’-Trityl-2’.3’-thiocarbonyl uridine.
`
`2’,3’-Didehydro-2’,3’-dideoxy
`
`uridine (d4l.f).
`
`(d) 2’,3’-dideoxy uridine (2,
`
`Trityl chloride (4.1 g, 14.7 mmol) was added to a solution of uridine (3 g, 12.3 mmol) in pyridine (50 ml) and the reaction mixture heated at 50°C for 24 h. The solvent was removed under high vacuum and the residue purified by chromatography on silica using 5% methanol in chloroform as eluant. Pooling and evapora- tion of appropriate fractions, followed by recrystallisation from ethanol gave the product as a white solid (5.32 g, 89%). G,(d,-DMSO) 3.3(28, dd, H5’, J = 10.8, 4.5 Hz), 4.0(1H, m, H3’), 4.15(28, m, H2’, H4’), 5.2(IH, d, OH, J = 4.8 Hz), 5.35(1H, d, H5,J = 8.1 Hz), 5.6(1H, d, OH,J = 3.7 Hz), 5.8(1H, d, HI’, 3 = 3.1 Hz), 7.2-7.5(15H, m, Ar), 7.8(lH, d, H6, J = 8.1 Hz), 11.2(lH, s,NH). &(d,- DMSO) 63.24(CS), 69.56(C3’), 73.44(C2’), 82.36(C4’), 86,45(Cl’), 88.97(CPh,), 101.47(C5); 127.18, 127.00,128.32(Ar), 140.62(C6), 143,43(Ar), 150,52(C4), 163.07(C2). (h)
`A solution of 5’-trityl uridine (1 g, 2.05 mmol) and thiocarbonyldiimidazole (400 mg, 2.24 mmol) in THF (20 ml) was stirred at ambient temperature for 17 h. After evap- oration of the solvent under reduced pressure the residue was purified by chromatography on silica using 5% methanol in chloroform as eluant. Pooling and evaporation of appropriate fractions gave the prod- uct as a white solid (990 mg, 91%). 6, (CDCI,) 3.5(2H, m, HS), 4.5(1 H, m, H4’), 5.5(lH, m, H2’); 5.65.9(2H,m, H3’, H5), 7.2-7.6(16H,m, Ar, H6), 9.5(lH, s, NH). (c)
`A solution of thio- carbonate (850 mg, 1.61 mmol) in triethylphosphite (10 ml) was heated at 150°C for 30 min. Excess triethylphosphite was evaporated under high vacuum and the residue purified by chromatography on silica using 5% methanol in chloroform as eluant. Pooling and evaporation of appropriate fractions gave the product and triethylphosphite (ratio = 60/40 by NMR). This mixture was dissolved in acetic acid and left for 17 h with stirring at ambient temperature. The solid obtained after evaporation of the solvent was purified by chromatography on silica using 8% methanol in chloroform as eluant. Pooling and evapora- tion of appropriate fractions gave the product (240 mg, 71%). 6, (CDCI,. CD,OD) 3.7(28, dd, H5’. J = 12.3, 3.0 Hz’), 4:85(lH, m, H4’);5.6(lH, d, H5, J = 8.1 Hz), 5.8(1H, m, H2’), 6.3(IH, m. H3’). 6.9(1H. m. HI’). 7.7(1H. d.H6. J = 8.1 Hz). 6,(CDCI,. CD,ODj 62.&l(&), ‘87.4i(C4’),‘89.86(Cl’j, 101.95(C5): 12?81(C2’j; 134.71(C3’), 141.29(C6), 15l.Ol(C4), 164.25(C2).
`ddll). D4U (240 mg, 1.14 mmol) was dissolved in methanol (10 ml) and ethanol (10 ml) and 10% palladium on charcoal (200 mg) was added. This mixture was shaken under a hydrogen atmosphere for 4 h.The catalyst was removed by filtration through celite and the residue purified by chromatography on silica using 8% methanol in chloroform as eluant. Pooling and evaporation of appropriate fractions gave the product as a white solid (200 mg, 83%). 6, tCDCI,,CD,OD) l.9-2.1(3H, m, H2’,H3’), 2.4(1H, m, H2’), 3,8(2H,‘hd, H5’, J =-12.2; 2.7 Hz); 4.15(1H, m, H4’), 5.7(lH, d, H5, J = 8.1 Hz).6.0(1H.Hl’.dd.J =6.6.3.3Hz)7.9(lH.d.H6.J =8.l Hz): FAB MS ‘m/e ‘213,0886(MH’, 72% C H N-0 requires 213.0875), 9 I? 2 4 149(32), 137(93), 129(40), 113(68), lOl(1OO). 2.1.2.
`(3a). DdU (2) (0.08 g, 0.38 mmol) and N-methyl imidazole (0.155 g, 1.89 mmol) were mixed in THF (10 ml ) and bis(2,2,2-trichloroethyl) phosphorochloridate (0.186 g, 0.49 mmol) was added slowly at ambient temperature. After I h the solvent was evaporated and the residue dissolved in CHCI, (30 ml) washed with HCI (1 M; 10 ml), sodium bicarbonate solution (15 ml) and water (15 ml). The organic phase was dried (MgSO,) and evaporated under reduced pressure. The crude product was purified by chromatography on silica eluting with 4% methanol in chloroform. Pooling and evaporation of appropriate frac- tions under reduced pressure gave the product (0.20 g, 95%). 6,, (CDCI,) 2.1(3H, m, H2’, H3’), 2.45(18, m, H2’), 4.4(28, m, HS), 4.55(18, m, H4’), 4.7(4H, m, CHZOP), 5.75(1H, d, H5, J = 8 Hz), 6.l(lH, m, HI’), 7.6(1H, d, H6, J = 8 Hz), 9.4(lH, s, NH); FAB MS m/e 552,8792(MH’, lo%, C,,H,,N20,PCI, requires 552.8826); HPLC RT = 32.6 min. 2.1.3.
`
`IPR2018-00126
`
`Page 2 of 4
`
`I-MAK 1009
`
`12
`ACS system, 50 +
`mm x
`

`

`C. McGuigan et al. IFEBS Letrers 351 (1994)
`
`II-14
`
`13
`
`I N, 1
`
`0
`
`2
`
`3a
`
`R oHlN ’ 3 Et-o-P-o AH Me-_CH I9 c=o 6 MB 3b Fig. 1. The structures of potential anti-HIV nucleoside and nucleotide analogues. overall yield by this route was 29%. Thus, in terms of yield the best route to d4U was via the 5’-trityl protected 2’,3’- thiocarbonate. The didehydro compound d4U generated by either of these routes was hydrogenated to give ddU (2) in 83% yield, and the structure and purity of this compound was confirmed by ‘H, 13C NMR, mass spectrometry and HPLC. The nucleoside analogue was then subjected to 5’- phosphorylation with a range of phosphates. A labile phosphate blocking group is clearly a pre-requisite for the masked phosphate approach, in order to facilitate (intracellular) release of the free nucleotides. Thus, sim- ple dialkyl phosphate derivatives of AZT are extremely resistant to phosphate hydrolysis, and they display no antiviral effect [3]. On the other hand bis(trihaloethy1) phosphate derivatives of zidovudine and of 2’,3’-dide- oxycytidine (ddC) show significant lability towards hy- drolysis [ 191, and are potent inhibitors of viral prolifera- tion [5]. We have also noted that one of these phosphate masking groups, the bis(2,2,2-trichloroethyl) phosphate moiety is successful in the kinase by-pass activation of certain inactive 3’-modified nucleosides [l l-121. We therefore chose this phosphate as the first blocking group for ddU. Thus, ddU was allowed to react with bis(2,2,2-trichlo- roethyl) phosphorochloridate in tetrahydrofuran con- taining N-methyl imidazole at room temperature to give (3a) in good yield. This was fully characterised by heter- onuclear NMR, FAB mass spectrometry, and HPLC, all data being consistent with its structure and purity. We have also noted that certain phosphoramidate deriva- tives of AZT are potent and selective inhibitors of HIV [4]. Thus, ethyl methoxyalaninyl phosphorochloridate was prepared by methods we have reported [4] and was allowed to react with ddU to give the target compound (3b) in moderate yield. This compound was recently re- ported by another group, following our earlier phosphor- amidate strategy [13]. However, we have found that aryloxy phosphoramidates are especially potent phos- phate blocking groups for AZT, and appear to release the free nucleotides within cells, on the basis of data in thymidine kinase-deficient cells [6,20]. Thus, phenyl methoxyalaninyl phosphorochloridate was similarly pre- pared [20] and was allowed to react with ddU to give (3~) in good yield. Full carbon-13 and (where appropriate) phosphorus-3 1 NMR data for the nucleoside (2) and the phosphates (3a-c) are given in Table 1. The parent nucleoside (2) and the corresponding masked phosphates (3a-c) were tested for their ability to inhibit the replication of HIV-l in C8166 cells, and in thymidine kinase-deficient [JM] cells, data being pre- sented in Table 2. As expected, the parent nucleoside (2) is active only at the highest concentrations tested, and is essentially inactive in JM cells. The bis[trichloroethyl] phosphate (3a) is approximately 5-10 times more active in each assay. On the other hand, the simple phospho- ramidate (3b) is devoid of antiviral activity in this assay, as has been recently noted by other researchers in a different assay [ 131. However, the aryloxy phosphorami- date (3~) is a potent agent, being approximately 50-times more active than the parent nucleoside analogue. This confirms the importance of data derived from assays in kinase deficient cells for the interpretation of the activi- ties of blocked nucleotides. As we have noted [20] only Table 1 Carbon-13 and phosphorus-31 NMR data for compounds (2) and (3ac) 2 3a 3b 3c Base C2 c4 c5 C6 Sugar Cl’ C2’ C3’ c4 C5’ P-OR POC POCC P-OAr Cl” - C2” C3” C4” P-NHR PNC c=o OMe 164.3 150.5 101.4 140.8 89.3 24.6 32.5 81.8 62.6 _ CHMe - - SP _ 163.4 150.6 102.6 139.7 86.6 25.5 32.2 78.66 69.6 77.3” 94.7’O -3.6 163.9 163.7 150.7 150.4 102.1 101.9, 102.0 140.0 139.7 86.5, 86.3 85.9, 86.2 25.3, 25.4 25.0, 25.2 32.5, 32.6 32.0, 32.1 79.46 78.86 66.5, 66.8 66.9, 67.0 66.0 _ 16.2, 16.3 - _ 150.5 119.8, 1 19.95 _ 129.6, 129.7 _ 125.0 50.0, 50.1 50.0, 50.2 174.4, 174.3” 173.8, 174.0’ 52.6 52.4 21.1, 21.05 20.6, 20.7 7.2(d) 3.2 “Multiplet. All spectra were recorded in CDCI,, except for (2) which were recorded in CDCI, plus CD,OD. Data are presented as 6 in ppm. All spectra were recorded using proton decoupling. In the case of carbon data phosphorus coupling constants in Hz are superscripted. Many carbon peaks for (3b-c) display diastereomeric splitting arising from mixed stereochemistry at the phosphate centre.
`
`IPR2018-00126
`
`Page 3 of 4
`
`I-MAK 1009
`
`

`

`14 C. McGuigan et al. IFEBS Letters 351 (1994) II-14 Table 2 Anti-HIV1 activity of nucleoside and nucleotide analogues Compound C8166 JM ED,, CCS” ED,,, CC,” 2 200 > 1,000 1,000 > 1,000 3a 40 500 80 400 3b > 1,000 > 1,000 400 400 3c 4 400 20 500 The antiviral activity and cytotoxicity of test compounds in two differ- ent cell lines. ED,, represents the concentration of compound (in ,uM) that decreases viral antigen production in infected cells to 50% of control. CC,, represents the concentration of compound (in ,uM) which causes 50% cytotoxicity to uninfected cells. such data will allow a clear understanding of the likely mechanism of action of blocked nucleotides, and dis- criminate between the release of nucleosides and nucleo- tides. In conclusion, we report the antiviral activity of cer- tain masked phosphate derivatives of the inactive nucle- oside ddU. We note that aryloxy phosphoramidates are particularly efficacious, and attribute the introduction of activity to kinase by-pass. Finally we stress the impor- tance of data in kinase-deficient cells for the clear inter- pretation of results on blocked phosphates. Acknowledgements: We thank the AIDS Directed Programme of the Medical Research Council and the Biomedical research Programme of the European Community for financial support. References [I] Furman, P.A., Fyfe, J.A., St. Clair, M.H. et al. (1986) Proc. Natl. Acad. Sci. USA 83, 8333-8337. [2] Larder, B.A. and Kemp, S.D. (1989) Science 246, 115551158. [3] McGuigan, C., Nicholls, S.R., O’Connor, T.J. and Kinchington, D. (1990) Antiviral Chem. Chemother. 1, 25-33. [4] McGuigan, C., Devine, K.G., O’Connor, T.J. and Kinchington D. (1991) Antiviral Res. 15, 255-263. [5] McGuigan, C., O’Connor, T.J. , Nicholls, S.R., Nickson, C. and Kinchington, D. (1990) Antiviral Chem. Chemother. 1, 355-360. [6] McGuigan, C., Pathirana, R.N., Mahmood, N., Devine, K.G. and Hay, A.J. (1992) Antiviral Res. 17, 31 I-321. [7] Farrow, S.N., Jones, A.S., Kumar, A., Walker, R.T., Balzarini, J. and De Clercq, E. (1990) J. Med. Chem. 33, 1400-1406. [8] Gouyette, C., Neumann, J.M., Fauve, R. and Huynh-Dinh, T. (1989) Tetrahedron Lett. 30, 6019-6022. [9] Henin, Y., Gouyette, C., Schwartz, O., Debouzy, J.C., Neumann, J.M. and Huynh Dinh, T. (1991) J. Med. Chem. 34, 1830-1837. [lo] Herdewijn, P., Balzarini, J., De Clercq, E., Pauwels, R., Masanori. B., Broder, S. and Van der Haeghe, H. (1987) J. Med. Chem. 30, 1270-1278. [I I] McGuigan, C., Kinchington, D., Wang, M.F., Nicholls, S.R., Nickson, C., Galpin, S., Jeffries, D.J. and O’Connor, T.J. (1993) FEBS Lett. 322, 249-252. [12] McGuigan, C., Kinchington, D., Nicholls, S.R., Nickson, C. and O’Connor, T.J. (1993) BioMed. Chem. Lett. 3, 1207-1210. [13] Gosselin, G. and Imbach, J.-L. (1993) Int. Antiviral News I, 100- 102. [ 141 Betbeder, D., Hutchinson, D.W., Richards, A.O’L., Mahmood, N. and Kinchington, D. (I 990) Antiviral Chem. Chemother. 1, 241~ 247. [15] Pauwels, R., Balzarini, J., Baba, M., Snoeck, R., Schols, D., Herdewijn, P., Desmeyter, J. and De Clercq, E. (1988) J. Virol. Methods 20, 309-321. [16] Mahmood, N. and Hay, A.J. (1992) J. Immunol. Methods 151, 9-13. 1171 Chu, C.K., Bhadti, V.S., Doboszewski, B., Gu, Z.P., Kosugi, Y., Pullaiah, K.C. and Van Roey, P. (1989) J. Org. Chem. 54, 2217- 2225. [18] Horwitz, J.P., Chua, J., Da Rooge, M.A., Noel, M. and Klundt, I.L. (1966) J. Org. Chem. 31, 205-211. 1191 see for example: McGuigan, C., Jones, B.C.N.M., Tollerlield, S.M. and Riley, P.A. (1992) Antiviral Chem. Chemother. 3,79-84. [20] McGuigan, C., Pathirana, R.N., Balzarini, J. and De Clercq, E. (1993) J. Med. Chem. 36, 104881052.
`
`IPR2018-00126
`
`Page 4 of 4
`
`I-MAK 1009
`
`