Pharmaceutical Research, Vol. 5. No. 11, 1988
`
`Report
`
`Synthesis, PhysicochemicalProperties, and Cytotoxicity of a
`Series of 5'-Ester Prodrugs of 5-Iodo-2’-deoxyuridine
`
`Milind M. Narurkarl and Ashim K. Mitralvz
`
`Received January 14, I988; accepted May I], 1988
`
`Five aliphatic 5'-esters of 5-iodo-2'deoxyuridine (IDU) were synthesized via an acid chloride alcohol-
`ysis reaction. The solubility in pH 7.4 phosphate buffer, lipophilicity as determined by partition ex-
`periments in octanol/pH 7.4 buffer. and cytotoxicity of these potential prodrugs were evaluated. The
`esters showed a 43- to 250-fold increase in lipophilicity and a 1.6- to 14-fold decrease in aqueous
`solubility relative to IDU. At a concentration of 50 uM. all esters showed reduced cytotoxicity toward
`uninfected Vero cells relative to IDU.
`
`KEY WORDS: 5—iodo-2'-deoxyuridine; 5'-ester prodrugs; solubility; partition coefficient‘, cytotox-
`icity.
`
`INTRODUCTION
`
`Herpes simplex virus type I infection (HSV-1) is the
`leading cause of corneal scarring and visual impairment in
`the United States (1,2). 5-Iodo-2’-deoxyuridine (IDU) is
`used in the treatment of ocular HSV-1 infections such as
`keratilis. Its therapeutic usefulness is limited, however, by
`frequent administration (3), a high incidence of treatment
`failure unassociated with resistant viral strains, and the in-
`ability to eradicate virus particles from deep ocular tissues
`(4), These problems can be attributed largely to the polar
`nature of IDU resulting in poor permeability of the drug
`across the lipoidal epithelial layer of the corneal membrane.
`Incomplete ocular absorption of drugs is due primarily
`to low corneal permeability and/or rapid drainage from the
`application site (5,6). The prodrug approach has been suc-
`cessfully applied to improve the ocular bioavailability of epi-
`nephrine (7-9), nadolol (10), various prostaglandins (l 1). pi-
`locarpine (12), and timolol (13).
`Five aliphatic 5’-ester derivatives of IDU (II—VI) (Fig.
`1) were synthesized as potential prodrugs and tested for cy-
`totoxicity. It was felt that transient blocking of the 5'-hy-
`droxyl group will enhance the lipophilicity of IDU and also
`help reduce phosphorylation at the S’- postion by thymidine
`kinase. Prodrug-based delivery, while resulting in appre-
`ciable IDU concentrations at the infected ocular sites, may
`also help in minimizing its toxicity, due to reduction in drug
`entrapment by uninfected normal cells.
`
`MATERIALS AND METHODS
`
`Chemicals
`
`IDU and trimethyl acetyl chloride were obtained from
`
`1 Department of Industrial and Physical Pharmacy, School of Phar-
`macy and Pharmacal Sciences, Purdue University, West La-
`fayette, Indiana 47907.
`2 To whom correspondence should be addressed.
`
`0724-874l/8811100-[)734S06.U()I0 © I988 Plenum Publishing Corporation
`
`734
`
`Sigma Chemical Co., M0. All other acid chlorides were ob-
`tained from Aldrich Chemical Co., Wis. Chemicals and sol-
`vents used were of reagent grade and were used as received.
`Distilled, deionized water was used for the preparation of
`buffer solutions as well as mobile phases.
`
`Methods
`
`Melting points were determined on at Thomas Hoover
`Unimelt capillary device and are uncorrected. ‘H-NMR
`spectra were run on a Chemagnetics A—200 spectrometer at
`200 MHZ. Chemical ionization mass spectra were obtained
`from a Finnegan 4000 mass spectrometer. pH measurements
`were made at the temperature of the study using a Corning
`Model 125 pH meter equipped with a combination electrode
`(Corning Science Products, Medfield, Mass).
`The high-performance liquid chromatographic (l-IPLC)
`setup consisted of a Waters model 510 pump equipped with a
`Waters U6K injector, a Waters Lambda Max Model 481
`variable wavelength LC spectrophotometer, and a Fisher
`Recordall series 5000 strip chart recorder. The mobile phase
`for prodrug analyses was 25-40% (v/v) acctonitrite in water
`
`1
`
`R=H
`
`.
`R=—C—CH2—CH3
`E
`331 Rt” ‘C”2‘°“2“3“3
`o
`ll
`
`E! R=—C—CH(CH3l2
`‘R
`R =—C~*(CH2l3CH3
`
`:1
`
`0n
`
`0
`
`u
`
`NH
`N)‘“*o
`
`H
`
`I
`
`|
`
`ROHZC
`H
`
`o
`
`H
`
`H
`
`OH H
`
`Fig. 1. IDU and five aliphatic S’-ester derivatives.
`
`El
`
`Fl =—C—C(CH3l3
`
`Patent Owner, UCB Pharma GmbH — Exhibit 2055 - 0001
`
`

`
`5'-Ester Prodrugs of 5-Iodo-2'-Deoxyuridine
`
`at a flow rate of 1.0 ml/min. Alltech C12 econosphere octa-
`deeyl silane column was used. IDU was analyzed with a
`mobile phase of 25% (v/v) of methanol in water at a flow rate
`of 1.0 ml/min on a Waters p.Bondapak phenyl column. Ultra-
`violet detection was performed at 261 nm. Temperature was
`ambient. Parabens and benzamide were used as internal
`
`standards for prodrugs and IDU analyses, respectively.
`
`General Procedure for Synthesis of 5'-Esters
`
`Direct acylation of IDU was effected by adding slowly a
`10% molar excess of the appropriate aeyl chloride to a
`chilled solution of IDU in a 1:1 mixture of pyridine:N,N-di-
`methyl formamide. The solvent—base combination was
`found to facilitate greatly the selectivity of acylation of the
`primary hydroxyl group over the secondary hydroxyl group
`(14). This could be a consequence of having the acylating
`agent as a charged species (N-acyl pydridinium chloride) in
`an aprotic, polar solvent such as DMF (15,16). The reaction
`was allowed to continue in an ice bath for 2-3 days. At
`completion, the mixture was evaporated to dryness with a
`rotary evaporator, under reduced pressure. The residue was
`dissolved in chloroform and the organic phase was washed
`with water, saturated sodium bicarbonate solution, and
`water. Evaporation of chloroform under reduced pressure
`gave the corresponding 5'-ester derivatives.
`Compounds II—VI were recrystallized from benzene-
`methanol in over a 50% yield. Purity was determined by
`HPLC, elemental analysis, and melting point determina-
`tions. Structural confirmation was made by NMR and
`CI-MS.
`
`‘1-1-NMR (Me2SO-a',,): 81.07 (t, 3,
`11.5’-Prop1’onyt'IDU.
`J = 7 Hz, CH3), 2.18 (m, 2, C2H0, 2.42 (q, 2. J = 7 Hz.
`CH2), 3.97 (m, 1, C411), 4.22 (m, 3, C311 and C5H), 6.09 (t, 1,
`J = 7 Hz, C,H), and 7.97 (s, 1,11,). CI-MS (CH4) m/e 411
`(m-1-1).
`Anal. Calcd. for C,2H,5IN2O,-, (410.1): C, 35.12; H,
`3.66; N, 6.83. Found: C, 35.31; H, 3.41; N, 7.13.
`111. 5'-Butyryl IDU.
`‘H-NMR (CDC13): 80.99 (t, 3, J :
`7 Hz, CH2), 1.73 (m, 2, CH2), 2.45 (m, 4, C2H and CH2),
`4.16-4.52 (m, 4, (:,H, C411 and C,H), 5.24 (1, 1, J = 7 Hz,
`C.I-1), and 7.98 (s, I, H,-,), CI-MS (CH4) mfe 425 (m +1).
`Anal. Calcd. for C,3H,2IN206 (424.1): C, 36.79; H,
`4.01; N, 6.61. Found: C, 36.93; H, 3.79; N, 6.32.
`IV. 5'—Isobutyryl IDU.
`‘H-NMR (CDCI3): 31.25 (d, 6, J
`= 6.8 Hz, 2 CH3), 2.53 (In, 2, C2H), 2.71(septet, 1, J = 6.8
`1-1z),4.08—4.53 (m,4, C,H, C.,H, and C,H),6.23 (1, 1,1 = 6.3
`Hz, C111’). and 7.96(s, 1, H6) CI-MS (CH ,) m/1» 425 (m + 1).
`
`735
`
`Anal. Calcd. for C,3H1—,-IN2O6 (424.1): C, 36.79; H,
`4.01; N, 6.61. Found: C, 36.99; H, 4.14; N, 6.58.
`V. 5'-Val'ery1'IDU. ‘H-NMR (CDCl3,: 80.97 (t, 3, J = 7
`Hz, CH3), 1.37 (m, 2, CH2), 1.65 (In, 2, CH2), 2.09-2.56 (m,
`4, CH2 and C2H), 4.13-4.49 (m, 4, C2H, C411’, and C5H) 6.25
`(t, 1, J = 6 Hz, CIH), and 7.99 (S, 1, H6). CI-MS (CH4) m/e
`439 (m + 1).
`Anal. Calcd. for C;4H,9lN2O6 (438.1): C, 38.34; H,
`4.34; N, 6.39. Found: C, 38.48; M, 4.46; N, 6.18.
`VI. 5'-Pivaloyl IDU. ‘H-NMR (CDC13): 81.27 (s, 9,
`3CH3), 2.02-2.63 (m, 2, C2H), 4.24-4.47 (tn, 4, C3H, C4H
`and C5H), 6.24 (t, 1, J = 7 Hz, C.H), and 7.9 (s, 1, H6).
`CI-MS (CH4) m/e 439 (m + 1).
`Anal. Calcd. for C14H,9IN2O5- H20 (456.1): C, 36.83;
`H, 4.60; N, 6.14. Found: C, 36.94; H, 4.30; N, 6.46.
`
`Determination of Aqueous Solubility
`
`Excess solid in 0.05 M phosphate buffer (pH 7.4) was
`allowed to equilibrate at 25°C with continous stirring for 72
`hr. The suspensions were then filtered through 0.45—uM
`ny1on—66 filters (Rainin) and the filtrate was analyzed by
`HPLC. Triplicate samples were run and the mean aqueous
`solubility was calculated.
`
`Determination of Partition Coefficients
`
`Apparent partition coefficients were determined at 34°C
`between l-octanol and 0.05 M phosphate buffer at pH 7.4.
`Mutually presaturated aqueous and organic phases were
`used. The compounds were dissolved in the aqueous buffer
`phase and then mixed with an equal volume of l-octanol and
`stirred at 34°C till HPLC analysis ensured equilibrium. The
`aqueous phase was sampled and analyzed by HPLC. Parti-
`tion coefficients were calculated from Eq. (1).
`
`C3,, - C2,,
`au
`Kicj
`
`0)
`
`where C,,,, is the total IDU or prodrug concentration and C2,,
`is the aqueous IDU or prodrug concentration at equilibrium.
`Hydrolysis of prodrugs during the experiment was insignifi-
`cant.
`
`Protocol for Cytotoxicity Study
`Vero cells in MEM-Earl‘s salts with 10% hcat-inactl-
`
`vated fetal calf serum (FCS), 100 Ufml penicillin. and 100
`pg/ml streptomycin were plated in fifty-four 25-cmz flasks at
`a density of 220,000 cells/2.5 ml/flask. The cells were al-
`
`Table 1. Physlcochemical Properties of1DU and 11s 5’-Ester Prodrugs
`
`Solubility“ in pH 7.4
`phosphate buffer, 25°C
`K“ : SD
`
`
`
`
` Compound m.p. (°C) [M/1.. : SD (X 103)] (octanol/water)
`
`
`
`I
`11
`111
`IV
`V
`V1
`"N = 3
`
`168-171 {dec)
`167-168
`145-146
`144-145
`142-143
`106-107
`
`5.65 (0.5)
`3.48 (0.3)
`1.45 (0.1)
`1.75 (0.3)
`0.40 (0.2)
`0.44 (0.1)
`
`0.11 (0.02)
`4.77 (0.1)
`7.50 (0.3)
`6.92 (0.8)
`27.54 (2.0)
`22.10 (1.5)
`
`Patent Owner, UCB Pharma GmbH — Exhibit 2055 - 0002
`
`

`
`736
`
`Narurkar and Mitra
`
`Table II. Effect of IDU and its 5‘-Ester
`Prodrugs on the Replication of Unin-
`fected Vero Cells In Vitro“
`
`Percentage inhibition
`of Vero cells
`
`Compound
`
`50 p.M
`
`400 p.M
`
`I
`II
`III
`IV
`V
`V1
`
`25.9
`6.0
`2].!
`18.8
`12.9
`9.3
`
`58
`36
`50
`33
`41
`72
`
`“ Assays were carried out in triplicate
`with appropriate controls.
`
`lowed to attach for 24 hr. The cytotoxicity of compounds
`(I—VI) on uninfected Vero cells was determined at 50 and
`400 u.M.
`On day 1, test compounds were added to the medium to
`yield a concentration of 50 and 400 p.M. Immediately after
`adding test samples to flasks, two flasks from each group
`were counted. The flasks were washed with phosphate—defi—
`cient buffered saline and then trypsinized with 1 ml of
`trypsin—Versene, 0.05% trypsin, and 0.02% EDTA, until the
`cells were dislodged. The number of cells was counted using
`a hemocytometer after adding 1 ml of trypan blue solution.
`Each day for the next 2 days, two of the remaining
`flasks were harvested in the aforementioned manner for de-
`
`termination of cell number. Assays were carried out in tripli-
`cate with appropriate controls.
`
`RESULTS AND DISCUSSION
`
`Melting points, aqueous solubilities, and octanol/water
`partition coefficients of compounds are shown in Table I. All
`prodrugs have lower melting points than that of IDU, which
`may be due to the loss of tightly bound crystal structure of
`IDU resulting from the weakening of intramolecular hy-
`
`O
`ll
`
`i
`
`{NHAN
`
`\0
`
`drogen bonding. The aqueous solubility measurements
`showed the expected decrease, whereas octanolfwater parti-
`tion coefficients exhibited an increase with an increase in
`the number of carbon atoms in the promoiety. For the same
`number of carbons, branched-chain esters were more sol-
`uble and had a lower partition coefficient than their normal-
`chain analogues.
`The effect of IDU and 5 '—ester prodrugs on the replica-
`tion of uninfected host Vero cells is shown in Table II. All
`
`5'-ester derivatives (II—VI) appear to be less cytotoxic than
`IDU to Vero cells at a concentration of 50 p.M. Similar re-
`sults are obtained at a concentration of 400 p.M for com-
`pounds II—V. Compound VI, however, appears to be more
`toxic to Vero cells than the parent drug at a concentration of
`400 p.M. As postulated in Scheme 1, IDU is sequentially
`phosphorylated at the 5'-position to the triphosphate by
`both viral and cellular thymidine kinase. Viral thymidine ki-
`nase, however, possesses a higher affinity for [DU and is
`able to trap greater amounts of IDU than the cellular en-
`zyme. IDU triphosphate is then incorporated into viral DNA
`instead of deoxythymidine triphosphate by viral DNA poly-
`merase, which is the basis for antiviral activity of IDU.
`Esterification of the 5'-hydroxyl group probably results
`in reduction of the molecule’s ability to serve as a substrate
`for thymidine kinase. Consequently, the ester is not phos-
`phorylated and does not get incorporated into viral and cel-
`lular DNA. This may explain the reduced cytotoxicity to
`Vero cells observed with almost all esters. Studies on aden-
`
`osine and related compounds have shown that a free hy-
`droxyl group at the 5' position was required to serve as a
`good substrate for adenosine deaminase (17,18).
`The purpose of this study is to identify IDU prodrugs
`for improved ocular delivery. Enzymatic hydrolysis in oc-
`ular tissue homogenates and in vitro corneal transport of
`prodrugs will be discussed in a future communication.
`
`ACKNOWLEDGMENTS
`
`This investigation was supported in part by a grant from
`
`“W25 o
`
`0“
`
`
`
`-0!-1 c
`2
`
`°
`
`DNA
`—-IDUTPT-DNA
`polymerase
`
`on
`no incorporation into DNA
`
`Thy—KIn<Ise
`
`.m,Tp&_.,,,.,.
`polymerase
`
`Patent Owner, UCB Pharma GmbH — Exhibit 2055 - 0003
`
`

`
`5’-Ester Prodrugs of 5-Iodo-2'~Deoxyuridine
`
`the National Eye Institute (EY 05863) and in part by a new
`investigator grant (A.K.M.) from the American Association
`
`of Colleges of Pharmacy.
`
`REFERENCES
`
`1. C. R. Dawson and B. Togni. Surv. Opthalmoi. 21:121—13()
`(1976).
`2. News from the National Institutes of Health. J. Infect. Dis.
`127:117—l23 (1973).
`3. Physicians’ Desk Reference for Ophlhalmolog , Medical Ec0-
`nomics, N.J., 1987.
`4. V. R. Colemen, E. Tsu, and E. Jawetz. Proc. Soc. Exp. Biol.
`Med. 129:761~765 (1968).
`5. S. S. Chrai, T. F. Patton, A. Mehta, and J. R. Robinson. J.
`Pharm.Sci.62:1112—1121{1973).
`6. S. S. Chrai, M. C. Makoid, S. P. Eriksen, and J. R. Robinson.
`J. Pharm. Sci. 632333-3138 (1974).
`7. D. A. McClure. In T. Higuchi and V.
`
`.1. Stella (eds.), Pradrugs
`
`737
`
`as Novel Drug Delivery Systems, ACS, Washington, D.C.,
`1974, pp. 224-235.
`8. A. Hussain and J. E. Truelove. J. Pharm. Sci. 65:1510—1512
`(19-I-6)_
`9. N. Bodor and G. Visor. Exp. Eye Res. 38:621—626 (1984).
`10. E. Duzman, C. C. Chen, .1. Anderson, M. Blumenthal, and H.
`Twizer. Arch. Ophthalmol. 100:l9l6—1919(1982).
`11. L. 2. Bito. Exp. Eye. Res. 38:18l—194(19S6).
`12. H. Bundgaard, E. Falch, C. Larsen. G. L. Mosher, and T. J.
`Mikkelson. J. Med. Chem. 28979-981 (1985).
`13. H. Bundgaard, A. Buur. S. Chang, and V. H. L. Lee. Int. J.
`Pharm. 33:15-25 (1986).
`14. D. C. Baker, T. H. Haskell. and S. R. Putt. J. Med. Chem.
`21:1218—1221 (1978).
`15. J. M. Knoblich, J. M. Sugihara, and T. Yamazaki. J. Org.
`Chem. 363407-3411 (1971).
`16. A. H. Haines. Adv. Carbohydr. Chem. Biochem. 33:11—109
`(1976).
`17. G. A. LePage. Adv. Enzyme Regal. 8:323—332 (1970).
`18. A. Block. M. J. Robins. and J. R. McCarthy. .1’. Med. Chem.
`10:908—911 (1967).
`
`Patent Owner, UCB Pharma GmbH — Exhibit 2055 - 0004