Discovery of the hemifumarate and (α-L-alanyloxy)methyl ether as prodrugs of an antirheumatic oxindole: Prodrugs for the enolic OH group

Article abstract of DOI:10.1021/jm950575k

Ether, ester, and carbonate derivatives of the antirheumatic oxindole 1 were prepared and screened as potential prodrugs of 1. This effort led to the discovery of the (α-L-alanyloxy)methyl ether and hemifumarate derivatives of 1 which deliver the drug eff

Full text of DOI:10.1021/jm950575k

10
J . Med. Chem. 1996, 39, 10-18
Discover y of th e Hem ifu m a r a te a n d (r-L-Ala n yloxy)m eth yl Eth er a s P r od r u gs of
a n An tir h eu m a tic Oxin d ole: P r od r u gs for th e En olic OH Gr ou p
Ralph P. Robinson,* Lawrence A. Reiter, Wayne E. Barth, Anthony M. Campeta, Kelvin Cooper,
Brian J . Cronin, Rosalina Destito, Kathleen M. Donahue, Fred C. Falkner, Eugene F. Fiese, Diane L. J ohnson,
Alexander V. Kuperman, Theodore E. Liston, Deborah Malloy, J ohn J . Martin, David Y. Mitchell,
Frank W. Rusek, Sheri L. Shamblin, and Charles F. Wright
Central Research Division, Pfizer Inc., Groton, Connecticut 06340
Received August 2, 1995X
Ether, ester, and carbonate derivatives of the antirheumatic oxindole 1 were prepared and
screened as potential prodrugs of 1. This effort led to the discovery of the (R-L-alanyloxy)-
methyl ether and hemifumarate derivatives of 1 which deliver the drug efficiently into the
circulation of test animals, are stable in the solid state, and possess good stability in solution
at low pH as required to ensure gastric stability. Success in achieving acceptable bioavailabilties
of 1 across species (rats, dogs, and monkeys) followed the inclusion of ionizable functionality
within the promoiety to compensate for masking the polar enolic OH group of the free drug.
However, the introduction of ionizable functionality was often associated with decreased
stability, as demonstrated by the hemisuccinate, hemiadipate, hemisuberate, and R-amino ester
derivatives of 1 which could not be isolated. A clear exception was the hemifumarate derivative
of 1 which was not only isolable but actually more stable at neutral pH than the nonionizable
ester analogues. The solution and solid state stability of the hemifumarate, together with its
activity as a prodrug of 1, suggests that hemifumarate be considered as an alternative to
hemisuccinate as a prodrug derivative for alcohols, particularly in situations where solution
state stability is an issue.
In tr od u ction
Antirheumatic oxindoles constitute a new class of
drugs for the treatment of arthritis.1 The lead example,
tenidap, has demonstrated excellent activity in rheu-
matoid arthritis and osteoarthritis clinical trials2 and
is now awaiting regulatory approval for the treatment
of these diseases. Although being a potent inhibitor of
cyclooxygenase (CO),3 tenidap can be clearly distin-
guished from classical nonsteroidal anti-inflammatory
agents (NSAIDs) by its cytokine-modulating properties.4
Like the NSAID piroxicam, tenidap and other anti-
rheumatic oxindoles undergoing evaluation possess an
acidic enolic OH group which is essential for CO
inhibitory activity.5 In the case of piroxicam, the enolic
OH group has been exploited as a “handle” for the
preparation of piroxicam derivatives which are inactive
against CO but converted to piroxicam during or after
absorption, thus minimizing the potential for gastro-
Particular structural and physicochemical properties
of 1 were of special concern in considering the prepara-
tion of prodrugs of this agent. Compound 1 itself is
poorly soluble in water (40 µg/mL at pH 6.2); masking
of the enolic OH group was likely to produce compounds
with even lower aqueous solubility unless the promoiety
intestinal irritation due to topical CO inhibition and
prostaglandin depletion by the free drug.6-9 Several of
these piroxicam prodrugs are now marketed including
ampiroxicam, an ether carbonate derivative of piroxicam
discovered in these laboratories.6,7 Here we report on
our parallel efforts to discover prodrugs of the antirheu-
itself possessed solubilizing functionality. Compound
matic oxindole 1, a close analogue of tenidap with a
1 is also quite acidic (pKa ) 3.5), and consequently the
enolate of 1 is a better leaving group than that of
piroxicam (pKa ) 6.5). Thus, there was concern that
carbonyl-linked derivatives (e.g., esters and carbonates)
similar pharmacologic profile. As in the case of piroxi-
cam, the strategy was to prepare derivatives with a
masked enolic OH group to prevent direct exposure of
the stomach to the free drug. These derivatives would
would be too readily hydrolyzed and that the enolate of
be stable under gastric conditions but sufficiently labile
1 would lack sufficient reactivity for derivatization
under some circumstances. Since 1 is known to exist
predominantly as a mixture of E and Z enol tautomers,
another chemical issue was the likelihood that mixtures
toward hydrolysis to allow release of 1, preferably
during or immediately after absorption from the gas-
trointestinal track.10
of geometric isomers would be formed upon acylation
or alkylation of the exocyclic enol group.
X Abstract published in Advance ACS Abstracts, November 15, 1995.
0022-2623/96/1839-0010$12.00/0 © 1996 American Chemical Society

Prodrugs of an Antirheumatic Oxindole
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 11
Ta ble 1
Rb
aq solubility,
no.
R
E,a
%
formulab
mp, °C
recryst solv
µg/mL (pH)
rat
dog
100
1.4
mon
100
1
2
3
4
5
6
7
8
9
H
C14H8ClFN2O3S
C19H16ClFN2O6S
C20H18ClFN2O6S
220-221
176-179 EtOAc
183-184 Et2O
40 (6.2)
0.009 (6.2)
100
26
12
2.3
0
0
CH(Me)OCO2Etc
CH(Me)OCO2i-Prc
CH(Me)OCOMec
CH2OCOt-Bu
>95
>95
94
>95
>95
>95
C18H14ClFN2O5S‚H2O 209
MeCN
0.04 (4.8)
0.34 (6.7)
C20H18ClFN2O5Sd
C22H14ClFN2O5S
C19H16ClFN2O6S
199-200 EtOAc
CH2OCOPh
202-203 EtOAc
CH(Me)OCOCH2OMec
CH2OCOCH2OMe
CH2OCOCH2OCH2Ph
175-176 MeCN
47
62
42
17
36
90
45
93
86
58
46
80
27
81
49
5.2
83e C18H14ClFN2O6S
144-147 hexane/EtOAc
130-131 hexane/EtOAc
176-177 MeCN
>95
C24H18ClFN2O6S
C19H14ClFN2O7S
C22H21ClFN3O7S
C18H14ClFN2O6S
C19H16ClFN2O6S
C20H18ClFN2O6S
C26H22ClFN2O6S
C27H24ClFN2O6S
C24H21ClFN3O8S2
C25H23ClFN3O8S2
C27H27ClFN3O8S2
C28H29ClFN3O8S2
C21H26Cl3FN4O6S
10 CH2OCOCH2OAc
>95
>95
>95
92
<0.02g
11 CH2OCOCH2OCONEt2
12 CH2OCOCH(Me)OHc
13 CH2OCOCH(Me)OMec
14 CH2OCOCH(Et)OMec
15 CH2OCOCH(Et)OCH2Phc
157-158 Et2O
184-187 hexane/EtOAc
149-151 hexane/EtOAc
134-135 hexane/EtOAc
111-113 hexane/EtOAc
121-123 cyclohexane/EtOAc
210-211 EtOH/EtOAc
210-214 MeOH/EtOAc
168-173 EtOH/EtOAc
210-211 EtOH/EtOAc
0.38 (6.5)
22
>95
>95
12
7.5
78
78
16 CH2OCOCH(n-Pr)OCH2Phc >95
17 CH2OCH2NH2‚TsOH
18 CH2OCH(Me)NH2‚TsOHf
19 CH2OCH(i-Pr)NH2‚TsOHf
20 CH2OCH(i-Bu)NH2‚TsOHf
21 CH2OCH[(CH2)4NH2]NH2‚
2HCl‚H2Of
95
>95
>95
>95
88
150 (6.5)
0.06g
78
135
EtOH/EtOAc
22 CH2OCH2NHCO2t-Bu
83
>95
C22H21ClFN3O7S
C23H23ClFN3O7S
171-174 hexane/EtOAc
165-170 hexane/EtOAc
0
20
23 CH2OCH(Me)NHCO2t-Buf
a
1
b
d
Determined by H NMR. Except where indicated, all compounds analyzed within (0.4% for C, H, and N. c Racemic. Anal. Calcd:
C, 53.04; H, 4.01; N, 6.19. Found: C, 52.73; H, 3.53; N, 6.10. e The other 17% is C-alkylated material. f S-Configuration. Unbuffered
g
distilled water; final pH not determined.
Ch em istr y
latter with neat TFA and 1 equiv of p-toluenesulfonic
acid or with HCl in dioxane. The method chosen was
based on pilot experiments to determine whether the
tosylate or hydrochloride salt of the final product was
crystalline. The N-t-BOC precursors to compounds 19-
21 were prepared as described for compounds 22 and
23.
The ether derivatives 2-16, 22, and 23 (Table 1) were
prepared in low to moderate yields by alkylation of a
salt of 1 with a 2-3-fold excess of an R-haloalkyl ester
or carbonate: compound 2 was obtained by the reaction
In all cases, the ethers 2-23 were predominantly to
exclusively the E isomers. This was established by
single-crystal X-ray structure analysis of compounds 12
and 18 and correlation of the 1H NMR spectra with
those of the other ether derivatives. In each case, the
chemical shifts and coupling constants of the aromatic
protons of the major (or exclusive) isomer closely
matched those of 12 and 18. The possibility that the E
and Z isomers could produce closely similar spectra was
discounted by the 1H NMR spectra of the ester and
carbonate derivatives (24-45), where it was clear that
the chemical shifts of the aromatic protons in the E and
Z isomers were significantly and consistently different
(vide infra).
of the triethylammonium salt with 1-chloroethyl ethyl
carbonate in chloroform in the presence of silver nitrate;
compounds 3, 4, and 7 were obtained by reaction of the
corresponding chloroethyl carbonate or ester with the
sodium salt in acetone in the presence of sodium iodide;7
the (acyloxy)methyl ethers 5, 6, 8-16, 22, and 23 were
prepared by treatment of the sodium salt with the
corresponding iodomethyl ester in refluxing acetone. In
nearly all cases, chromatography and recrystallization
were required for isolation of the desired product which
often included separation of the unwanted product
arising from C-alkylation of the enolate of 1.
The [(aminoacyl)oxy]methyl ethers 17-21 were pre-
pared by deprotection of the corresponding N-tert-
butoxycarbonyl (N-t-BOC) amino acid derivatives (e.g.,
22 and 23). This was accomplished by treatment of the

12 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Robinson et al.
Ta ble 2
Rb
aq solubility,
no.
R
E,a
%
formulab
mp, °C
recryst solv
µg/mL (pH)
rat
dog mon
100 100
1
H
C14H8ClFN2O3S
220-221
40 (6.2)
100
14
6.6
17
3.7
0
4
13
24 COMe
25 COEt
26 COn-Bu
27 COt-Bu
28 COPh
29 COCHdCHPh (E)
30 CO2Me
31 CO2Et
40
C
C
C
C
C
C
16H10ClFN2O4S
17H12ClFN2O4S
19H16ClFN2O4S
19H16ClFN2O4S
21H12ClFN2O4S
24H16ClFN2O4S
200-203 acetone/i-PrOH 0.011 (6.2)
182-188 hexane/EtOAc 0.003 (6.2)
181-183 hexane/EtOAc 0.013 (6.2)
196-197 hexane/EtOAc
5
8
7
20
90
216-222 i-PrOH
214-215 acetone/i-PrOH
40 C16H10ClFN2O5S
15 C17H12ClFN2O5S
>95 C18H14ClFN2O5S
24 C18H14ClFN2O5S
>95 C23H24ClFN2O5S
19 C20H14ClFN2O6S
>95 C20H16ClFN2O6S
203-208 i-PrOH
164-166 i-PrOH
197-199 i-PrOH
175-178 i-PrOH
142-145 i-PrOH
165-168 i-PrOH
182-183 MeCN
184-185 i-PrOH
201-203 i-PrOH
219-220 i-PrOH
199-200 MeCN
188-189 MeCN
165-167 toluene
0.023 (6.2)
3.5
32 CO2i-Pr
33 CO2i-Pr
0.6
9.9
5.7
34 CO2n-C8H17
35 COCHdCHCO2Et (E)
36 CO(CH2)2CO2Et
37 CO-2-pyridyl
38 CO-3-pyridyl
39 CO-4-pyridyl
40 COCH2OMe
41 COCH2OAc
42 CO(CH2)2CO2CH2CONEt2 >95 C24H23ClFN3O7S
43
44 CO-3-C6H4CO2H
45 COCHdCHCO2H (E)
<0.02 (6.7)
21
6.4
3.5
1.6
0
2.2
8.9
6.6
16
25
4.2
95
c
c
c
C
C
C
20H11ClFN3O4S
20H11ClFN3O4S
20H11ClFN3O4S
<0.06 (2), <0.06 (7.5)
<0.06 (2), <0.06 (7.5)
0.32 (2), <0.07 (7.5)
28 C17H12ClFN2O5S
C18H12ClFN2O6S
8
C
15H6ClFN2O4S
>260
MeCN
244-248 AcOH
229-232 EtOAcf
17d
>95 C22H12ClFN2O6Se
99 C18H10ClFN2O6S
0.58d
0.08 (2), 2.5 (6.5), 24 (7)
91
73
a
b
Determined by 1H NMR except compound 45 for which HPLC was used. Except where indicated, all compounds analyzed within
(0.4% for C, H, and N. c Stereochemistry not assigned. 37: ratio of isomers, 95:5. 38: ratio of isomers, 90:10. 39: ratio of isomers,
88:12. Unbuffered distilled water; final pH not determined. e Anal. Calcd: C, 54.27; H, 2.48; N, 5.75. Found: C, 53.80; H, 2.18; N, 5.71.
d
f Trituration.
The ester and carbonate derivatives of 1 (Table 2)
were prepared by the reaction of 1 with acid chlorides
or chloroformates in the presence of a base, Et3N, or
i-Pr2NEt. These conditions invariably gave rise to
mixtures of geometric isomers. The products were
typically isolated and tested as mixtures, although, in
certain instances, isolation of the E isomer was achieved
by fractional crystallization. Assignments of olefin
geometry in the ester and carbonate series are based
on 1H NMR studies with the octyl carbonate isomers,
where a significant NOE (16%) between the oxindole
C4 and thiophene C3′ hydrogens was detected with the
Z isomer and not the E isomer (34). Stereochemical
assignments for other carbonates and esters are based
isomerization took place under the conditions for re-
moval of the protecting group. Thus, while the E/Z ratio
of the 2-(trimethylsilyl)ethyl ester was about 2:1, that
of the product (45) was about 99:1. Attempts to prepare
the corresponding hemisuccinate, hemiadipate, and
hemisuberate derivatives failed during cleavage of the
corresponding mixed 2-(trimethylsilyl)ethyl esters. In
these cases, quantitative regeneration of 1 took place
upon treament with HF/pyridine. For the hemisucci-
nate especially, this was most likely due to intramo-
lecular displacement of 1 (pKa ) 3.5) by the free
ω-carboxy group.12
The relatively high stability of the enolate of 1
undoubtedly played a role in the failure to isolate other
ester derivatives: R-amino acid esters after treatment
of 1 with various activated R-amino acid derivatives
(recovered 1) and 4-(aminomethyl)benzoyl esters13 by
treatment of the 4-(chloromethyl)benzoate derivative
with ammonia or alkylamines (ester aminolysis and
regeneration of 1).
1
on comparisons of the H NMR spectra to those of the
octyl carbonate isomers which allowed characteristic
patterns to be discerned. For the E isomer series in
CDCl3, the thiophene hydrogens are typically well
separated by chemical shift; of these, the C3′ hydrogen
appears furthest downfield. In the Z isomers series, the
thiophene C3′ hydrogen is typically shifted 0.4 ppm
upfield, often overlapping with the signal due to the C5′
hydrogen. This is typically accompanied by a 0.2 ppm
upfield shift of the oxindole C4 proton. Both shifts are
ascribed to mutually interacting anisotropies of the
oxindole and thiophene aromatic rings.
The hemiisophthalate and hemifumarate derivatives
44 and 45 were prepared by selective cleavage of the
corresponding mixed 2,2,2-trichloroethyl and 2-(tri-
methylsilyl)ethyl esters using Zn/AcOH and HF/pyri-
dine, respectively.11 In the latter case, a fortuitous olefin
The cyclic carbamate derivative 43 was prepared as
shown in Scheme 1.14
Resu lts a n d Discu ssion
The prodrug activity of all compounds was first
assessed in the rat. Compounds were administered at
a dose equivalent to 3 mg of 1/kg by oral gavage in 0.1%
methyl cellulose. Relative bioavailabilities were deter-
mined by measurement of plasma AUCs of 1 and
comparison with plasma AUCs of 1 achieved following

Prodrugs of an Antirheumatic Oxindole
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 13
Sch em e 1
(R-L-valinyloxy)methyl ether tosylate 19) to excellent
(81% for the (R-L-leucinyloxy)methyl ether tosylate 20).
Fortunately, in this series, the goal of achieving good
bioavailability in dogs and monkeys was attained; the
bioavailability of 1 for the (glycinyloxy)methyl ether
tosylate 17 was 78% in monkeys, and that for the (R-
L-alanyloxy)methyl ether tosylate 18 was 78% in both
dogs and monkeys. Not unexpectedly, the [[(N-t-BOC-
amino)acyl]oxy]methyl ethers 22 and 23, the immediate
nonionizable precursors of compounds 17 and 18, re-
spectively, exhibited poor prodrug activity in the rat.
oral dosing of the free drug. Select compounds were
advanced to dogs and/or monkeys for determination of
relative bioavailabilities in these species. The relative
bioavailability (Rb) data are included in Table 1 for
ether derivatives and in Table 2 for esters and carbon-
ates.
Relative to the ether carbonate 2, the (R-L-alanyloxy)-
methyl ether tosylate 18 was quite unstable in solution,
another factor possibly contributing to the high bio-
availabilities of 1 in the three test species. At pH 7.4,
18 was highly unstable, with a time for 10% decomposi-
tion (t10%) to 1 of only 0.9 min. At lower pH, stability
improved; at pH 6.0, t10% increased to 9.1 min, and at
pH 2, t10% was 90 min.16 Thus a reasonable assumption
was that 18 would remain mostly unchanged during
passage through the stomach, as desired, but would
hydrolyze to 1 to some extent prior to absorption upon
encountering the higher pH within the small intestine.
The stability of 18 in the solid state, in contrast to that
in solution, was excellent with no decomposition occur-
ring during 4 weeks at 50 °C under 75% relative
humidity.
Ether-linked derivatives of 1 were prepared to avoid
the potential for hydrolytic instability anticipated for
the carbonyl-linked compounds (esters and carbonates)
due to the good leaving group ability of the enolate of
1. Based on its structural relationship to ampiroxicam,
the ether carbonate 2 was of particular interest as a
potential prodrug of 1. However, this compound and
the other nonpolar ether-linked derivatives 3-6 were
only weakly active as prodrugs of 1 in the rat (Table 1).
Compound 2 did have the highest activity among these
nonpolar ether derivatives (27%) and was therefore
advanced into dogs; however, in this species the bio-
availability of 1 was even lower (1.4%). These results
were attributed to a lack of absorption since in no
instance were circulating levels of these potential pro-
drugs detected in the plasma of test animals and since,
as expected, masking the enolic OH group led to
compounds having very low aqueous solubility (e.g., 2:
0.009 µg/mL at pH 6.2). Stability measurements on the
ester carbonate 2 were encouraging, however, being
within acceptable limits: In solution (1:2:1 H2O/MeCN/
MeOH) times for 10% decomposition (t10%) of 3.6 and
28 h were determined at pH 2 and 7.4, respectively, and
in the solid state, 93% of 2 remained unchanged after 4
weeks at 50°C under 75% relative humidity.
Also promising was the moderate to excellent bio-
availability of 1 achieved in the rat for the more polar
ethers (7-16) derived from 1 and the chloromethyl
esters of various glycolic, lactic, 2-hydroxybutyric, and
2-hydroxyvaleric acid derivatives. Unfortunately, the
four compounds among these selected for testing in dogs
or monkeys failed to deliver 1 with even moderate
bioavailability in these higher species. What appeared
to be a lack of correlation between the improved rat
bioavailability of 1 and aqueous solubility (Table 1)
suggested a variable other than solubility to be deter-
minant of activity in the rat. This was not necessarily
true for the other species, and we thus hypothesized that
although significant increases in solubility had been
achieved in some cases (e.g., 12: 0.38 µg/mL), the
aqueous solubilities were still too low on an absolute
scale to allow good absorption in dogs and monkeys.
The [(aminoacyl)oxy]methyl ether derivatives 17-21
were prepared in the hope that the ionizable amino
functionality would further increase aqueous solubil-
ity.15 This appears to have been the case based on the
>300-fold increase in solubility of the (R-L-alanyloxy)-
methyl ether tosylate 18 (150 µg/mL) over the lactate
ether 12. Like the earlier series (7-16) derived from
R-hydroxy acids, the rat bioavailabilities of 1 for the
amino acid ethers ranged from moderate (27% for the
Owing to their ease of synthesis, ester derivatives of
1 were also explored as potential prodrugs despite the
concerns over stability. The simple nonpolar esters 24-
29 and carbonates 30-34 uniformly exhibited poor
activities as prodrugs of 1 in the rat (Table 2). With
the exception of the valerate 26, bioavailabilities of 1
were <10%. Not surprisingly, these ester and carbonate
derivatives exhibited very low aqueous solubility (<0.02
µg/mL), and as anticipated, the solution state stabilities
were poor at pH 7.4 (e.g., 27, t10% ) 0.6 min; 31, t10%
)
5.1 min). However, since solution state stabilities
increased markedly at lower pH (e.g., at pH 2: 27, t10%
) 4.9 h; 31, t10% ) 5.9 h) and since the solid state
stabilities were good, the ester series was further
pursued to improve aqueous solubility and bioavailabil-
ity of 1.
Introduction of nonionizable or poorly ionizable polar
functionality, as in the various mixed diester, pyridine
carboxylates, and glycolates (35-41), was not sufficient
to improve bioavailability of 1. The mixed succinate
glycolamide 42, a potential prodrug of the unstable
hemisuccinate derivative (a pro-prodrug of 1), likewise
displayed poor activity.17,18 These results were again
attributed to poor absorption resulting from low aqueous
solubility as confirmed for several of these compounds
(Table 2). Despite having comparatively good aqueous
solubility (17 µg/mL), the cyclic carbamate 43 (analogous
to the piroxicam prodrug droxicam8) delivered 1 with
only slightly improved bioavailability (25% in the rat).
In parallel with the ethers of 1, ionizable ester
derivatives were next examined. Unfortunately, all
attempts to prepare amino-functionalized esters (R-
amino esters and 4-(aminomethyl)benzoates13) were
unsuccessful as were efforts to prepare the acidic
hemisuccinate, hemiadipate, and hemisuberate deriva-
tives (Chemistry). In the latter cases, the results were
attributed to susceptibility of the hemiesters to in-
tramolecular attack by the free acid group. Thus, acidic

14 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Robinson et al.
compounds were sought in which intramolecular attack
would be precluded by geometric restriction of the tether
between the ester and the ω-acid functionalities. In the
event, both the hemiisophthalate 44 and the hemi-
fumarate 45 proved to be stable and isolable.11 The rat
bioavailability of 1 for 44 was very low (4.2%), probably
a reflection of its low aqueous solubility (0.58 µg/mL).
On the other hand, the prodrug activity of the hemi-
fumarate 45 proved to be excellent, not only in the rat
(95%) but in dogs (91%) and monkeys (73%) as well.
Interestingly, the aqueous solubility of this compound
was found to increase markedly from pH 2 (0.08 µg/mL)
to pH 7.0 (24 µg/mL), a finding in keeping with the high
bioavailability of 1 since, in the small intestine, 45 was
likely to have a solubility toward the higher end of this
range.
In sharp contrast to the nonionizable ester and
carbonate derivatives as well as the (R-L-alanyloxy)-
methyl ether 18, 45 was quite stable in solution at pH
7.4 (t10% ) 3 h). The stability at pH 2, although
relatively low (t10% ) 70 min), was considered to be
sufficient enough to allow passage of 45 through the
stomach without significant hydrolysis. This assump-
tion was supported by studies in vitro which demon-
strated 45 to be stable (no detectable decomposition over
30 min) in gastric fluids obtained from rats and mon-
keys. Like (R-L-alanyloxy)methyl ether 18, 45 exhibited
excellent stability in the solid state with 98% remaining
unchanged after 4 weeks at 50 °C under 75% relative
humidity.
Although both the (R-L-alanyloxy)methyl ether 18 and
the hemifumarate 45 fulfilled the initial bioavailability
and stability criteria as prodrugs of 1, 45 was selected
for further profiling based largely on its better overall
solution state stability profile. In all species (rats, dogs,
and monkeys), circulating levels of 45 were below the
assay limit of quantitation (<0.02 µg/mL) following oral
administration of the prodrug, indicating that, as
desired, systemic exposure to 45 was extremely low.
This was confirmed by the observation that, after oral
administration of 45 to hepatic portal vein-cannulated
dogs, concentrations of intact prodrug within the portal
and systemic circulation were undetectable and sys-
temic concentrations of 1 were similar to systemic
concentrations achieved following oral administration
of 1 to noncannulated dogs.
Further in vitro studies were undertaken in order to
determine the sites and kinetics of the conversion of the
hemifumarate 45 to 1. The prodrug was rapidly hy-
drolyzed to the parent drug in monkey intestinal ho-
mogenates and in rat, dog, monkey, and human plasma
with in vitro half-lives of <1 min. Together with the in
vitro studies showing stability of 45 in rat and monkey
gastric fluid, these data support the “ideal” scenario
wherein 45 is stable within the upper gastrointestinal
tract but rapidly hydrolyzed to 1 as it is absorbed and
circulated.
sufficient stability in solution at low pH to ensure
gastric stability. Success in achieving acceptable bio-
availabilties of 1 across species followed the inclusion
of ionizable functionality within the promoiety to com-
pensate for masking the polar enolic OH group of the
free drug and the accompanying decrease in aqueous
solubility observed with nonionizable derivatives. As
demonstrated by the low stability of 18 at neutral pH
and our inability to isolate the hemisuccinate, hemi-
adipate, hemisuberate, and R-amino ester derivatives
of 1, the introduction of ionizable functionality was often
associated with decreased stability. A clear exception
to this observation was the hemifumarate 45 which was
not only isolable but actually more stable at neutral pH
than its nonionizable analogues. We ascribe the stabil-
ity of 45 relative to the hemisuccinate and other diacid
hemiesters to the presence of the trans olefinic double
bond which prevents intramolecular attack on the ester
by the free carboxylate. On the other hand, the stability
of 45 relative to the nonionizable ester derivatives at
neutral pH may be due to the presence of the ionized,
negatively charged carboxylate group (pH ) 4.5) which
may electronically or electrostatically hinder hydrolysis
of the adjacent ester linkage.
Despite limited solution phase stability, hemisucci-
nates have been commonly used to increase the aqueous
solubility of compounds containing free hydroxyl groups
(particularly corticosteroids).10 On the other hand, to
our knowledge, hemifumarates have never been studied
as prodrug derivatives of alcohols or enolic OH com-
pounds. Together with the activity of 45 as a prodrug
of 1, the stability of 45 relative to the corresponding
hemisuccinate (which could not be isolated) and its
stability in solution relative to the nonionizable esters
suggest that hemifumarate be considered as an alterna-
tive to hemisuccinate as a prodrug derivative for alco-
hols, particularly in situations where solution state
stability is an issue.
Exp er im en ta l Section
The 1H NMR spectra were recorded at 300 MHz (Bruker
AC 300 or Varian XL-300 spectrometer) or at 250 MHz (Bruker
AC 250 spectrometer) as indicated. Flash chromatography
was carried out with use of silica gel (J . T. Baker Inc.; 40 µm
flash chromatography packing). All reactions requiring an-
hydrous conditions were run in flame-dried glassware under
nitrogen. Melting points are uncorrected.
1-Chloroethyl ethyl carbonate (used for preparing 2) was
used as purchased from Fluka. 1-Chloroethyl 2-propyl carbon-
ate (used for preparing 3) was prepared from 1-chloroethyl
chloroformate.19 1-Chloroethyl acetate (used for preparing 4)
and 1-chloroethyl methoxyacetate (used for preparing 7) were
prepared by the reaction of acetyl chloride and methoxyacetyl
chloride, respectively, with acetaldehyde in the presence of
catalytic zinc chloride.20 The iodomethyl esters used in the
preparation of 5, 6, and 8-23 were prepared by reacting the
tetrabutylammonium or cesium salt of the appropriate car-
boxylic acid with
a large excess of bromochloromethane
followed by a Finkelstein reaction.21 This method gave
reasonable yields of the desired esters with only minimal gem-
diester formation.15
Con clu sion
Except for the cases of compounds 35, 42, 44, and 45, the
acid chlorides and chloroformates used for preparing the ester
and carbonate derivatives of 1, respectively, were obtained
from commercial sources. Fumaric acid monoethyl ester
monochloride (used in the preparation of 35) was prepared
according to the published procedure for the corresponding
methyl ester.22
We have described here the discovery of the (R-L-
alanyloxy)methyl ether 18 and the hemifumarate 45 as
prodrugs of the antirheumatic oxindole 1. These com-
pounds deliver 1 efficiently into the circulation of test
animals, are stable in the solid state, and possess

Prodrugs of an Antirheumatic Oxindole
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 15
Car bon ic Acid 1-[[1-(Am in ocar bon yl)-6-ch lor o-5-flu or o-
1,2-dih ydr o-2-oxo-3H-in dol-3-yliden e]-2-th ien ylm eth oxy]-
eth yl Eth yl Ester (2). To a solution of 1 (49.8 g, 147 mmol)
and triethylamine (22.5 mL, 161 mmol) in chloroform (1 L)
were added 1-chloroethyl ethyl carbonate (67.3 g, 441 mmol)
and silver nitrate (25 g, 147 mmol). The resulting mixture
was stirred at room temperature for 36 h and then cooled in
an ice bath, and the reaction was quenched by addition of
concentrated HCl (14 mL). After filtration to remove silver
salts, the filtrate was washed with aqueous 1 N HCl solution,
saturated aqueous NaHCO3, and brine, dried (MgSO4), and
concentrated. The residue was chromatographed twice on
silica gel eluting with 5% MeOH/CHCl3, and the appropriate
fractions were concentrated. The solid residue was recrystal-
lized from EtOAc to provide 2 as a yellow solid, 7.4 g (11%).
Mp: 176-179 °C. 1H NMR (250 MHz, DMSO-d6): δ 8.25 (d,
J ) 7.1 Hz, 1 H), 8.05 (dd, J ) 1.0, 4.9 Hz, 1 H), 7.95 (br s, 1
H), 7.81 (br s, 1H), 7.77 (d, J ) 9.7 Hz, 1 H, overlapped), 7.67
(dd, J ) 1.0, 3.5 Hz, 1 H), 7.28 (dd, J ) 3.5, 4.9 Hz, 1 H), 6.14
(q, J ) 5.4 Hz, 1 H), 4.01 (q, J ) 7.1 Hz, 2 H), 1.70 (d, J ) 5.4
Hz, 3 H), 1.03 (t, J ) 7.1 Hz, 3 H). Anal. (C19H16ClFN2O6S)
C, H, N.
2-Hyd r oxyp r op a n oic Acid [[1-(Am in oca r bon yl)-6-ch lo-
r o-5-flu or o-1,2-d ih yd r o-2-oxo-3H-in d ol-3-ylid en e]-2-t h i-
en ylm eth oxy]m eth yl Ester (12). (a ) To a solution of benzyl
lactate (21.26 g, 0.118 mol) and diisopropylethylamine (51 mL,
0.293 mol) in CH2Cl2 (60 mL) was added [2-(trimethylsilyl)-
ethoxy]methyl chloride (25 mL, 0.141 mol). The mixture was
stirred at room temperature for 2 h and then at reflux for 1 h.
After cooling, the mixture was washed with aqueous 1 N HCl
(3 × 50 mL) and concentrated to afford an oil which was
chromatographed on silica gel eluting with 1% and then 10%
EtOAc/hexane as eluant. Concentration of the appropriate
fractions gave 2-[[2-(trimethylsilyl)ethoxy]methoxy]propionic
acid benzyl ester as an oil, 27.1 g (74%). 1H NMR (250 MHz,
CDCl3): δ 7.32 (s, 5 H), 5.20 (s, 2 H), 4.75 (s, 2 H), 4.31 (q, J
) 6.8 Hz, 1 H), 3.69-3.60 (m, 2 H), 1.44 (d, J ) 6.8 Hz, 3 H),
0.92-0.87 (m, 2 H), 0.00 (s, 9 H).
(b ) A solution of [[2-(2-trimethylsilyl)ethoxy]methoxy]-
propionic acid benzyl ester (26.7 g, 86 mmol) in methanol was
hydrogenated over 10% Pd on charcoal (2.65 g) in a Parr
shaker for 1 h at 2 atm of pressure. The catalyst was removed
by filtration. Evaporation of the solvent provided 2-[[2-
(trimethylsilyl)ethoxy]methoxy]propionic acid as an oil, 9.97
g (58%). 1H NMR (250 MHz, CDCl3): δ 4.76 (s, 2 H), 4.28 (q,
J ) 7.1 Hz, 1 H), 3.71-3.63 (m, 2 H), 1.47 (d, J ) 7.1 Hz, 3
H), 0.95-0.88 (m, 2 H), 0.00 (s, 9 H).
(c) A mixture of 2-[[2-(trimethylsilyl)ethoxy]methoxy]-
propionic acid (8.95 g, 40.6 mmol), CsHCO3 (9.47 g, 48.8 mmol),
BrCH2Cl (27 mL, 0.415 mol), and dioxane was heated at 80
°C for 3 h. The reaction mixture was taken up in EtOAc,
washed twice with water, dried (MgSO4), and concentrated to
afford an oil (a mixture of the chloromethyl and bromomethyl
esters, 6.36 g) which was combined with sodium iodide (14.2
g, 94.7 mmol) in acetone (100 mL) and refluxed for 18 h. The
mixture was filtered, and the filtrate was concentrated to leave
an oil. This was dissolved in EtOAc and washed with 10%
aqueous sodium thiosulfate, water, and brine. Drying (MgSO4)
and concentration yielded crude 2-hydroxypropionic acid
iodomethyl ester as an oil, 2.49 g (46%). 1H NMR (300 MHz,
CDCl3): δ 5.98 (ABd, J ) 6.5 Hz, 1 H), 5.93 (ABd, J ) 6.5 Hz,
1 H), 4.32 (q, J ) 7 Hz, 1 H), 1.40 (d, J ) 7 Hz, 3 H).
s, 1 H), 7.65 (dd, J ) 1.4, 3.5 Hz, 1 H), 7.29 (dd, J ) 3.5, 4.9
Hz, 1 H), 5.80 (ABd, J ) 6.2 Hz, 1 H), 5.72 (ABd, J ) 6.2 Hz,
1 H), 5.56 (d, J ) 6.0 Hz, 1 H), 4.18-4.09 (m, 1 H), 1.18 (d, J
) 6.9 Hz, 3 H). Anal. (C18H14ClFN2O6S) C, H, N. The E
stereochemistry was assigned by X-ray crystallographic analy-
sis.
Gen er a l Meth od for th e P r ep a r a tion of (Acyloxy)-
m eth yl Eth er Der iva tives of 1. N-[(1,1-Dim eth yleth oxy)-
car bon yl]-L-alan in e [[1-(Am in ocar bon yl)-6-ch lor o-5-flu or o-
1 ,2 -d i h y d r o -2 -o x o -3H -i n d o l -3 -y l i d e n e ]-2 -t h i e n y l -
m eth oxy]m eth yl Ester (23). (a ) A solution of N-t-BOC-L-
alanine (9.46 g, 50 mmol) in ethanol (100 mL) was neutralized
with a solution of 40% aqueous tetrabutylammonium hydrox-
ide using phenolphthalein as indicator. The solution was
concentrated under vacuum leaving the tetrabutylammonium
salt which was dissolved in BrCH2Cl (200 mL) and stirred for
48 h in the dark. After removal of the excess BrCH2Cl under
vacuum, the mixture was taken up in CHCl3 and washed with
water and brine. Drying (MgSO4) and concentration gave an
oil. This was chromatographed on silica gel eluting with
CH2Cl2 to afford N-t-BOC-L-alanine chloromethyl ester as an
oil, 6.25 g (53%) (containing a trace of the corresponding
bromomethyl ester). 1H NMR (250 MHz, CDCl3): δ 5.83 (ABd,
J ) 5.9 Hz, 1 H), 5.63 (ABd, J ) 5.9 Hz, 1 H), 5.10-4.90 (br
m, 1 H), 4.40-4.28 (m, 1 H), 1.43 (s, 9 H), 1.40 (d, J ) 7.3 Hz,
3 H).
(b) N-t-BOC-L-alanine chloromethyl ester (6.20 g, 26.1
mmol) and sodium iodide (19.5 g, 130 mmol) were combined
in acetone (150 mL) and stirred for 18 h. The mixture was
then filtered, and the filtrate was concentrated to leave an oil.
This was dissolved in EtOAc and washed with 10% aqueous
sodium thiosulfate, water, and brine. Drying (MgSO4) and
concentration yielded crude N-t-BOC-L-alanine iodomethyl
ester as an oil, 7.40 g (86%).
(c) Crude N-t-BOC-L-alanine iodomethyl ester (7.30 g, 22.2
mmol) and the sodium salt of 1 (2.67 g, 7.4 mmol) were
combined in acetone (75 mL) and refluxed for 6 h. The reaction
mixture was concentrated, and the residue, after preadsorption
onto silica gel, was chromatographed eluting with CH2Cl2 and
then 5% MeOH/CH2Cl2. Combination and concentration of the
appropriate fractions gave a yellow solid which was recrystal-
lized from EtOAc/hexane to yield 23, 1.16 g (29%). Mp: 165-
170 °C. 1H NMR (250 MHz, DMSO-d6): δ 8.25 (d, J ) 7.0 Hz,
1 H), 8.07 (d, J ) 4.9 Hz, 1 H), 7.96 (br s, 1 H), 7.81 (d, J ) 9.7
Hz, 1 H, overlapped), 7.79 (br s, 1 H), 7.63 (d, J ) 3.5 Hz, 1
H), 7.39 (d, J ) 7.0 Hz, 1 H), 7.31-7.28 (m, 1 H), 5.78 (ABd,
J ) 6.5 Hz, 1 H), 5.71 (ABd, J ) 6.5 Hz, 1 H), 4.09-3.94 (m,
1 H), 1.27 (s, 9 H), 1.15 (d, J ) 7.4 Hz, 3 H). Anal. (C23H23
ClFN3O7S) C, H, N.
-
L-Ala n in e [[1-(Am in oca r bon yl)-6-ch lor o-5-flu or o-1,2-
d ih yd r o-2-oxo-3H -in d ol-3-ylid en e]-2-t h ien ylm et h oxy]-
m eth yl Ester Tosyla te Sa lt (18). To cold (0°C) TFA (12 mL)
was added 23 (998 mg, 1.85 mmol). The resulting solution
was allowed to stand for 1 h at 0 °C, and then TsOH‚H2O (352
mg, 1.85 mmol) was added. The TFA was removed under
vacuum chasing with toluene, and the residual solid was
crystallized from MeOH/EtOAc to afford 18 as a yellow solid,
352 mg (31%). Mp: 210-214 °C. 1H NMR (250 MHz, DMSO-
d6): δ 8.31 (br s, 3 H, overlapped), 8.26 (d, J ) 7.1 Hz, 1 H),
8.09 (dd, J ) 1.1, 5.0 Hz, 1 H), 7.93 (br s, 1 H), 7.83 (d, J ) 9.7
Hz, 1 H, overlapped), 7.82 (br s, 1 H), 7.67 (dd, J ) 1.1, 3.7
Hz, 1 H), 7.48 (d, J ) 8.1 Hz, 2 H), 7.34 (dd, J ) 3.7, 5.0 Hz,
1 H), 7.12 (d, J ) 8.1 Hz, 2 H), 5.90 (ABd, J ) 6.5 Hz, 1 H),
5.82 (ABd, J ) 6.5 Hz, 1 H), 4.27-4.13 (m, 1 H), 2.30 (s, 3 H),
1.34 (d, J ) 7.2 Hz, 3 H). Anal. (C25H23ClFN3O8S2) C, H, N.
The E stereochemistry was assigned by X-ray crystallographic
analysis.
Gen er a l Meth od for th e P r ep a r a tion of Ester a n d
Ca r bon a te Der iva tives of 1. Ca r bon ic Acid [1-(Am in o-
ca r bon yl)-6-ch lor o-5-flu or o-1,2-d ih yd r o-2-oxo-3H-in d ol-
3-ylid en e]-2-th ien ylm eth yl 2-P r op yl Ester (32 a n d 33).
To a solution of 1 (4.06 g, 12.0 mmol) and triethylamine (1.8
mL, 12.9 mmol) in chloroform (120 mL) was added 1 M
isopropyl chloroformate in toluene (24 mL, 24 mmol). The
mixture was allowed to stir overnight, concentrated, and then
taken up in chloroform. The solution was washed successively
(d ) Crude 2-hydroxypropionic acid iodomethyl ester (2.40
g, 10.4 mmol) and the sodium salt of 1 (1.62 g, 4.5 mmol) were
combined in acetone (50 mL) and refluxed for 4 h. After
concentration of the reaction mixture, the residue was dis-
solved in EtOAc. The solution was washed with water, dried
(MgSO4), and concentrated. The residue was taken up in
CH2Cl2, filtered to remove unreacted 1, and concentrated. The
residue was chromatographed on silica gel three times eluting
with CH2Cl2 and then 1% MeOH/CH2Cl2. Combination and
concentration of the appropriate fractions gave a yellow solid
which was recrystallized from EtOAc/hexane to yield 12, 382
mg (19%). Mp: 184-187 °C. 1H NMR (300 MHz, DMSO-d6):
δ 8.25 (d, J ) 6.8 Hz, 1 H), 8.06 (dd, J ) 1.4, 4.9 Hz, 1 H),
7.94 (br s, 1 H), 7.79 (d, J ) 9.5 Hz, 1 H, overlapped), 7.79 (br

16 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Robinson et al.
with aqueous 1 N HCl, water, saturated aqueous NaHCO3,
and brine and then dried (MgSO4) and concentrated. The
residue was chromatographed on silica gel eluting with 5%
MeOH/CHCl3. Clean fractions containing the desired product
were combined, concentrated, and recrystallized from 2-pro-
panol to afford 32 (E isomer) as a yellow solid, 1.9 g (37%).
Mp: 197-199 °C dec. 1H NMR (300 MHz, CDCl3): δ 8.52 (br
s, 1 H, NH), 8.44 (d, J ) 7.0 Hz, 1 H, C7 H), 8.19 (dd, J ) 1.5,
3.9 Hz, 1 H, C3′ H), 7.74 (dd, J ) 1.5, 5.0 Hz, 1 H, C5′ H), 7.47
(d, J ) 9.7 Hz, 1 H, C4 H), 7.21 (dd, J ) 3.9, 5.0 Hz, 1 H, C4′
H), 5.36 (br s, 1 H, NH), 4.98 (septet, J ) 6.3 Hz, 1 H, isopropyl
1,3-Ben zen ed ica r b oxylic Acid Mon o[[1-(a m in oca r -
b on yl)-6-ch lor o-5-flu or o-1,2-d ih yd r o-2-oxo-3H -in d ol-3-
ylid en e]-2-th ien ylm eth yl] Ester (44). (a ) A solution of
isophthaloyl dichloride (40.6 g, 200 mmol) and 2,2,2-trichloro-
ethanol (30.9 g, 207 mmol) in dry toluene was heated at reflux
for 24 h. The solvent was evaporated, and the remaining
material was distilled under vacuum. The desired product
distilled over along with some of the bis(2,2,2-trichloroethyl)
ester between 125 and 170 °C at 0.2 mmHg. On standing,
the unwanted side product separated out as a solid. The
desired product, 3-[[(2,2,2-trichloroethyl)oxy]carbonyl]benzoyl
chloride (19.0 g, 30%), was decanted off for use in the next
step. 1H NMR (300 MHz, CDCl3): δ 8.82 (s, 1 H), 8.41-8.32
(m, 2 H), 7.69-7.63 (m, 1 H), 5.03 (s, 2 H).
(b) To a solution of 1 (5.0 g, 14.7 mmol) and diisopropyl-
ethylamine (5.1 mL, 29.3 mmol) in CH2Cl2 (100 mL) at room
temperature was added 3-[[(2,2,2-trichloroethyl)oxy]carbonyl]-
benzoyl chloride (9.3 g, 29.4 mmol). The mixture was stirred
overnight, diluted with CH2Cl2, and then poured into water.
The organic layer was washed with water and brine and then
filtered to remove the suspended solid. The filtrate was
concentrated to a yellow residue which, after trituration with
CHCl3, afforded the 2,2,2-trichloroethyl isophthalate mixed
ester of 1 (E isomer), (1.4 g, 15%). 1H NMR (300 MHz,
CDCl3): δ 8.98 (t, J ) 1.6 Hz, 1 H, isophthalate C2 H), 8.55-
8.43 (m, 4 H, NH, C7 H, isophthalate C4 and C6 H), 8.21 (dd,
CH), 1.37 (d, J ) 6.3 Hz, 6 H, 2 × CH3). Anal. (C18H14
-
ClFN2O5S) C, H, N.
The mother liquor from the recrystallization was concen-
trated and recrystallized again from 2-propanol to afford a 3:1
mixture of the Z/ E isomers (33), 320 mg (6%). Mp: 175-178
°C. 1H NMR (Z isomer) (300 MHz, CDCl3): δ 8.48 (br s, 1 H,
NH), 8.41 (d, J ) 7.0 Hz, 1 H, C7 H), 7.75-7.73 (m, 2 H, C3′
H, C5′ H), 7.27 (d, J ) 9.7 Hz, 1 H, C4 H), 7.22 (dd, J ) 4.0,
5.0 Hz, 1 H, C4′ H), 5.40 (br s, 1 H, NH), 4.98 (septet, J ) 6.3
Hz, 1 H, isopropyl CH), 1.38 (d, J ) 6.3 Hz, 6 H, 2 × CH3).
Anal. (C18H14ClFN2O5S) C, H, N.
Bu tan edioic Acid [1-(Am in ocar bon yl)-6-ch lor o-5-flu or o-
1,2-d ih yd r o-2-oxo-3H -in d ol-3-ylid en e]-2-t h ien ylm et h yl
2-(Dieth yla m in o)-2-oxoeth yl Ester (42). (a ) A mixture of
2-chloro-N,N-diethylacetamide (11.9 g, 79.5 mmol), succinic
acid monobenzyl ester23 (15.0 g, 72.0 mmol), sodium iodide (1.1
g, 7.4 mmol), and triethylamine (11.2 mL, 80.4 mmol) in EtOAc
(280 mL) was heated at reflux for 3 h. After cooling, the
mixture was filtered, and the filtrate was washed successively
with aqueous 1 N HCl, water, saturated aqueous NaHCO3,
and brine. The solution was then dried (MgSO4) and concen-
trated to afford an oil which was chromatographed on silica
gel eluting with 50% EtOAc/hexane as eluant. Combination
and concentration of the appropriate fractions provided suc-
cinic acid benzyl 2-(diethylamino)-2-oxoethyl ester as a clear
oil, 14.6 g (63%). 1H NMR (250 MHz, CDCl3): δ 7.34 (s, 5 H),
5.13 (s, 2 H), 4.70 (s, 2 H), 3.38 (q, J ) 7.1 Hz, 2 H), 3.22 (q,
J ) 7.1 Hz, 2 H), 2.86-2.71 (m, 4 H), 1.21 (t, J ) 7.1 Hz, 3 H),
1.12 (t, J ) 7.1 Hz, 3 H).
(b) A solution of succinic acid benzyl 2-(diethylamino)-2-
oxoethyl ester (14.6 g, 45.4 mmol) in EtOH (250 mL) was
hydrogenated at 4 atm over 10% Pd on carbon (1.0 g) in a Parr
shaker for 12 h. The catalyst was removed by filtration
through Celite, and the filtrate was concentrated to leave
succinic acid mono[2-(diethylamino)-2-oxoethyl] ester as a
white solid, 10.5 g (100%). 1H NMR (250 MHz, CDCl3): δ 8.33
(br s, 1 H), 4.73 (s, 2 H), 3.37 (q, J ) 7.1 Hz, 2 H), 3.24 (q, J
) 7.1 Hz, 2 H), 2.80-2.64 (m, 4 H), 1.21 (t, J ) 7.1 Hz, 3 H),
1.11 (t, J ) 7.1 Hz, 3 H).
J
) 1.5, 3.9 Hz, 1 H, C3′ H), 7.78 (t, J ) 7.9 Hz, 1 H,
isophthalate C5 H), 7.69 (dd, J ) 1.5, 4.9 Hz, 1 H, C5′ H),
7.24-7.17 (m, 2 H, C4′ H, C4 H), 5.35 (br s, 1 H, NH), 5.02 (s,
2 H, CH2O).
(c) A mixture of the 2,2,2-trichloroethyl isophthalate mixed
ester of 1 (1.0 g, 1.62 mmol) and zinc dust (1.0 g, 15.3 mmol)
in glacial AcOH was stirred at 50 °C overnight. The warm
solution was filtered and then poured into water (250 mL).
The yellow precipitate was collected by filtration and recrys-
tallized from AcOH to afford 44 as a yellow solid (130 mg,
16%). Mp: 244-248 °C. 1H NMR (300 MHz, DMSO-d6): δ
8.71 (s, 1 H), 8.52 (d, J ) 6.5 Hz, 1 H), 8.40 (d, J ) 7.8 Hz, 1
H), 8.30-8.26 (m, 2 H), 8.11 (d, J ) 4.9 Hz, 1 H), 8.05 (br s, 1
H), 8.01 (br s, 1 H), 7.88 (t, J ) 7.8 Hz, 1 H), 7.31-7.24 (m, 2
H). Anal. Calcd for C22H12ClFN2O6S: C, 54.27; H, 2.48; N,
5.75. Found: C, 53.80; H, 2.18; N, 5.71.
Bu ten ed ioic Acid Mon o[[1-(a m in oca r bon yl)-6-ch lor o-
5-flu or o-1,2-d ih yd r o-2-oxo-3H-in d ol-3-ylid en e]-2-th ien yl-
m eth yl] Ester (45). (a ) A solution of 2-(trimethylsilyl)ethanol
(50.0 g, 0.428 mol) in dry benzene (500 mL) was added slowly
with stirring to fumaryl chloride (71.2 g, 0.466 mol) at room
temperature. After stirring overnight, the mixture was filtered
(to remove the precipitated fumaric acid bis[2-(trimethylsilyl)-
ethyl] ester), and the solvent was evaporated to leave a yellow
oil. This was distilled at 18 mmHg and then at 0.8 mmHg to
remove volatiles including unreacted fumaryl chloride. Re-
maining in the pot was an oil (35.7 g) containing the desired
product, fumaric acid mono[2-(trimethylsilyl)ethyl] ester
monochloride, and a small amount of the unwanted diester.
This was added slowly (rinsing with 50 mL of CH2Cl2) to a
solution of 1 (46.8 g, 0.138 mol) and diisopropylethylamine
(26.5 mL, 0.152 mol) in CH2Cl2 (1525 mL) at room tempera-
ture. The mixture was stirred overnight and then worked up
by washing with aqueous 1 N HCl (1 L), saturated aqueous
NaHCO3 (1 L), and brine (1 L). After drying (MgSO4), the
solvent was evaporated to leave an oil which was chromato-
graphed three times on silica gel (2 kg) eluting successively
with CH2Cl2, 1% MeOH/CH2Cl2, and 1.2% MeOH/CH2Cl2.
Fractions containing the desired product gave a yellow solid
which was recrystallized from acetonitrile to afford the desired
2-(trimethylsilyl)ethyl fumarate mixed ester of 1 (24.6 g, 33%
(c) A solution of succinic acid mono[2-(diethylamino)-2-
oxoethyl] ester (5.0 g, 21.6 mmol) and oxalyl chloride (2.0 mL,
23.5 mmol) in dry benzene (100 mL) was heated at reflux for
1 h. After cooling, the solution was concentrated to afford the
crude acid chloride as an oil. This was taken up in CH2Cl2
(20 mL) and added to a solution of 1 (3.40 g, 10.0 mmol) and
diisopropylethylamine (3.8 mL, 21.8 mmol) in CH2Cl2 (100
mL). The resulting mixture was stirred at room temperature
for 66 h and then diluted with CH2Cl2 and washed with
aqueous 1 N HCl and brine. After drying (MgSO4), the
solution was concentrated, and the residue was chromato-
graphed twice on silica gel eluting with CH2Cl2 and then 2.5%
MeOH/CHCl3 as eluants. Fractions containing the desired
product were combined and concentrated to a dark solid which
was triturated with 50% Et2O/CH2Cl2 and then recrystallized
from toluene to afford 42 as a yellow solid, 512 mg (5%). Mp:
165-167 °C. 1H NMR (250 MHz, CDCl3): δ 8.54 (br s, 1 H,
NH), 8.45 (d, J ) 7.0 Hz, 1 H, C7 H), 8.19 (dd, J ) 1.0, 3.9 Hz,
1 H, C3′ H), 7.71 (dd, J ) 1.0, 5.0 Hz, 1 H, C5′ H), 7.51 (d, J
) 9.3 Hz, 1 H, C4 H), 7.21 (dd, J ) 3.9, 5.0 Hz, 1 H, C4′ H),
5.32 (br s, 1 H, NH), 4.75 (s, 2 H, CH2O), 3.40 (q, J ) 7.1 Hz,
2 H, CH2N), 3.24 (q, J ) 7.1 Hz, 2 H, CH2N), 3.19-3.14 (m, 2
H, CH2CO), 2.97-2.92 (m, 2 H, CH2CO), 1.23 (t, J ) 7.1 Hz,
from 1) as a mixture of E and Z isomers. Anal. (C23H22
-
ClFN2O6SSi) C, H, N.
By careful chromatography on silica gel (using the above
procedure), it was possible to separate the geometric isomers
on a small scale for characterization by 1H NMR. 1H NMR (E
isomer) (250 MHz, acetone-d6): δ 8.45 (d, J ) 6.9 Hz, 1 H, C7
H), 8.32-8.30 (m, 2 H, NH, C3′ H), 8.05 (dd, J ) 1.1, 5.0 Hz,
1 H, C5′ H), 7.50 (d, J ) 9.5 Hz, 1 H, C4 H), 7.31 (dd, J ) 3.9,
4.9 Hz, 1 H, C4′ H), 7.28 (ABd, J ) 15.7 Hz, 1 H, fumarate
3 H, CH3), 1.14 (t, J ) 7.1 Hz, 3 H, CH3). Anal. (C24H23
ClFN3O7S) C, H, N.
-

Prodrugs of an Antirheumatic Oxindole
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 17
0.05 M
0.01 M
CH), 7.20 (ABd, J ) 15.7 Hz, 1 H, fumarate CH), 7.05 (br s, 1
H, NH), 4.42-4.36 (m, 2 H, CH2O), 1.06-1.09 (m , 2 H, CH2Si),
0.09 (s, 9 H, 3 × CH3). 1H NMR (Z isomer) (250 MHz, acetone-
d6): δ 8.43 (d, J ) 7.0 Hz, 1 H, C7 H), 8.19 (br s, 1 H, NH),
8.10 (d, J ) 4.9 Hz, 1 H, C5′ H), 8.01 (d, J ) 3.7 Hz, 1 H, C3′
H), 7.41 (d, J ) 9.7 Hz, 1 H, overlapped, C4 H), 7.40-7.37 (m,
1 H, C4′ H), 7.02 (s, 2 H, overlapped, 2 × fumarate CH), 7.01
(br s, 1 H, NH), 4.39-4.32 (m, 2 H, CH2O), 1.14-1.07 (m, 2 H,
CH2Si), 0.08 (s, 9 H, 3 × CH3).
AcOH MeOH MeCN i-PrOH H2O H3PO4
compd (%, v/v) (%, v/v) (%, v/v) (%, v/v) (%, v/v) (%, v/v)
other
pH 2.5
2, 24-34
4, 12
5, 6, 35
7, 10
16
25
40
30
35
30
50
35
45
35
25
25
25
30
55
35
pH 3.0
pH 3.0
pH 3.0
pH 3.0
0.1% (v/v)
Et3N,
pH 3.0
pH 3.0
5 mmol/L
15
18
65
(b) The 2-(trimethylsilyl)ethyl fumarate mixed ester of 1
(70.8 g, 0.131 mol) was added in three batches to HF/pyridine
complex (900 mL) in a poly(ethylene) vessel at 0 °C. The
mixture was occassionally manually agitated over 70 min and
was then poured into 3 L of ice/water. The precipitated solid
was collected by filtration, washed with water (1 L), and dried
over P2O5 in a vacuum dessicator. The solid was mixed with
ethyl acetate (6 L), and the mixture was warmed, with stirring,
to 60 °C. After cooling to room temperature, the slurry was
filtered to collect 45 as a fine yellow solid (43.8 g, 76%). Mp:
197-198 °C dec. 1H NMR (250 MHz, acetone-d6): δ 8.48 (d,
J ) 7.0 Hz, 1 H, C7 H), 8.34-8.33 (m, 2 H, NH, C3′ H), 8.07
(d, J ) 4.5 Hz, 1 H, C5′ H), 7.55 (d, J ) 9.6 Hz, 1 H, C4 H),
7.33 (t, J ) 4.5 Hz, 1 H, C4′ H), 7.31 (ABd, J ) 15.7 Hz, 1 H,
fumarate CH), 7.22 (ABd, J ) 15.7 Hz, 1 H, fumarate CH),
7.05 (br s, 1 H, NH). Anal. (C18H10ClFN2O6S) C, H, N.
It was possible to recrystallize 45 from anhydrous acetic acid
giving analytically pure material with mp 229-232 °C. The
E stereochemistry for the major isomer was assigned based
on the close similarity of its 1H NMR spectrum to that of the
E isomer of the 2-(trimethylsilyl)ethyl ester precursor in
acetone-d6. The E stereochemistry for the latter was assigned
22, 44, 45
37-39
35
35
30
35
35
35
30
C6H13SO3H,
pH 3.0
pH 3.0
43
15
40
10
The aqueous solubility of compounds was determined as
follows. The test compound (1 mg) was placed in a screw cap
vial to which unbuffered distilled water (6 mL) was added.
The tightly capped vial was sonicated for 5 min,and the
solution was filtered through a Millex HV13 0.45 µm filter,
discarding the first 0.5 mL. The solution was immediately
measured for pH and assayed by HPLC. Solubility was
determined by comparison of the peak area of the compound
with that obtained from a reference solution of the compound
dissolved in the mobile phase.
Solution state stabilities were routinely measured at pH 2.0
and 7.4. Limiting solubility usually precluded testing under
100% aqueous conditions. For solution state stability at pH
2.0, a test compound (0.5 mg) was typically dissolved in 1:1:2
(v/v/v) 0.02 M HCl/MeOH/MeCN (10 mL); for testing at pH
7.4, 1:1:2 (v/v/v) 0.01 M NaHPO4/MeOH/MeCN (adjusted to
pH 7.4 with 0.2 M NaOH) was typically used. After 10-fold
dilution in the same system, samples (kept protected from light
at room temperature) were assayed by HPLC immediately and
thereafter at regular time intervals. The first-order rate
constant and t10% for decomposition were determined by linear
regression of a plot of time versus ln(F) where F is the fraction
of compound remaining. Dissolution of very unstable com-
pounds was carried out in MeCN prior to dilution into the
mixed aqueous/organic systems above. Compounds 18 and 45
were compared in 100% aqueous phosphate buffer systems
except at low pH where the solubility of 45 was limiting.
Consequently, for the pH 2.0 stability of 45, various amounts
of MeOH or MeCN were used with the aqueous buffer. From
the observed rate constants in these systems and the respec-
tive dielectric constant of each mixture, the first-order rate
constants and t10% values for decomposition of 45 were
extrapolated to pure water at pH 2.0.
To determine solid state stability, a test sample (0.5-0.8
mg) was placed in a clear screw cap vial and stored in a
humidity chamber over a saturated solution of sodium acetate
(75% relative humidity) at 50 °C for 4 weeks. The sample was
dissolved in 10 mL of the mobile phase for HPLC analysis
(ultrasonication was often required). After 10-fold dilution in
mobile phase and filtration through a Millipore HV13 filter,
the sample was analyzed by HPLC. The stability of the test
compound was quantitated based on the percent of its peak
area relative to the total detected area assuming the compound
and its degradation products to have the same absorptivity.
Peaks occurring in the solvent front were excluded from the
total peak area count. The percentage of test compound
remaining was compared to that obtained from a reference
sample stored at 0 °C for 4 weeks.
In Vitr o Sta bility of 45 in Biologica l F lu id s. Solution
A was prepared by dissolving AcOH (100 µL) and H3PO4 (50
µL) in MeCN (800 mL); solution B was prepared by dissolving
45 (2.63 mg) in solution A (10 mL); solution C was prepared
by dissolving the internal standard 5-chloro-N-ethyl-2,3-dihy-
dro-3-(hydroxy-2-furanylmethylene)-2-oxo-1H-indole-1-carbox-
amide (2.0 mg) in MeCN (500 mL) containing H3PO4 (2.5 mL).
Fresh plasma, gastric fluids, and intestinal homogenates (2
mL) were preincubated at 37 °C for 5-10 min and then spiked
with solution B (100 µL). The samples were incubated at
37 °C over a period of 30 min; aliquots (50 µL) were removed
1
based on correlation of the H NMR spectrum in CDCl3 with
that of the (E)-octyl carbonate isomer 34.
In Vivo Eva lu a tion . The relative bioavailabilities of 1
from test compounds were determined in all animals (Spra-
gue-Dawley rats, beagle dogs, and cynomolgus monkeys) at
a dose equivalent to 3 mg of 1/kg administered by oral gavage
in 0.1% methyl cellulose (10 mL/kg for rats, 1 mL/kg for dogs,
and 1 mL/kg for monkeys). Each experiment was carried out
in three animals fasted overnight prior to dosing. Plasma
samples were obtained from rats at 1, 3, and 6 h postdose via
sinus orbital bleeding; dogs were bled from the jugular vein
at 0, 0.5, 1, 2, 4, 6, 8, 12, and 24 h postdose, and plasma
samples from monkeys were obtained at 0, 0.5, 1, 2, 4, 6, and
8 h postdose. For determination of plasma concentrations of
1, plasma samples were precipitated with acetonitrile fortified
with tenidap, the internal standard. The suspensions were
centrifuged, and the supernatants were transferred to vials
containing tris buffer. Analysis was carried out by reverse-
phase HPLC with UV detection at 360 nm. The analytical
column was a Waters Novapak C18 column (150 × 3.9 mm i.d.)
fitted with a 2 cm LC-Si guard column by Supelco Pelliguard.
The mobile phase consisted of 1:1 MeOH/tris buffer adjusted
to pH 6.3 with H3PO4; a flow rate of 1 mL/min was used.
Under these conditions, the retention times of 1 and tenidap
were 4.7 and 3.3 min, respectively. Concentrations of 1 were
measured from standard curves based on the peak height ratio
of 1 to tenidap. The relative bioavailabilities were calculated
by comparison of the averaged plasma AUCs of 1 with
averaged plasma AUCs achieved of 1 following oral dosing of
the free drug (3 mg/kg) to controls. A separate control group
of three animals was used for determining rat bioavailabilities;
for dogs and monkeys, each animal served as its own control.
Experiments in hepatic portal vein-cannulated dogs (beagles)
were carried out using 45 and 1 dosed orally as above in 0.1%
methyl cellulose. Dogs were bled from the portal and jugular
veins at 0, 0.25, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, and 4 h postdose.
Additional plasma samples were obtained from the jugular
vein at 6 and 8 h postdose.
Aqu eou s Solu bilty, Solu tion Sta te Sta bility, a n d Solid
Sta te Sta bility Deter m in a tion s. Compounds selected for
measurements of solution state stability, solid state stability,
and/or aqueous solubilty were assayed by reverse-phase HPLC
on a Waters Novapak C18 column (150 × 3.9 mm i.d.) using a
UV detector at 214 nm and a flow rate of 1 mL/min. The
mobile phase systems used were as follows:

18 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Robinson et al.
R. Pharmacological profile of droxicam. Gen. Pharmacol. 1988,
19, 49-54. (c) Esteve, A.; Martinez, L.; Roser, R.; Sagarra, R.
Pharmacokinetics of droxicam in rat and dog. Methods Find.
Exp. Clin. Pharmacol. 1986, 8, 423-429.
at 1, 2, 5, 10, 20, and 30 min after spiking and added to
microvials containing solution C (100 µL). Zero-hour time
point measurements were carried out by addition of solution
C (2 mL) to the biological fluid (1 mL, preincubated at 37 °C)
followed by spiking with solution B (50 µL). After centrifuga-
tion, aliquots of supernatant (75 µL) were removed, added to
injection vials, and diluted with solution A (75 µL). Analysis
of the samples was carried out by reverse-phase HPLC with
UV detection at 360 nm. The analytical column was a
Phenomenex Ultracarb 5 ODS (30) column (150 × 4.6 mm i.d.)
fitted with a Phenomenex Ultracarb 5 ODS (30) guard column.
The mobile phase consisted of 4:1 MeCN/H2O containing 0.1%
(v/v) AcOH and 0.05% (v/v) H3PO4; a flow rate of 1 mL/min.
was used. Under these conditions, the retention times of 1,
45, and internal standard were 5.4, 2.3, and 11.4 min,
respectively. Concentrations of 1 and 45 were measured from
standard curves based on the peak height ratios of 1 and 45
to internal standard.
(9) (a) Che´rie´-Lignie`re, G.; Montagnani, G.; Alberici, M.; Acerbi; D.
Plasma and synovial fluid concentrations of piroxicam during
prolonged treatment with piroxicam pivalic ester. Arzneim.-
Forsch./ Drug Res. 1987, 37, 560-563. (b) Cinnoxicam, piroxicam
cinnamate. Drugs Future 1991, 16, 164.
(10) Prodrug reviews: (a) Bundgaard, H. Design of prodrugs: bio-
reversible derivatives for various functional groups and chemical
entities. In Design of Prodrugs; Bundgaard, H., Ed., Elsevier:
Amsterdam, 1985; Chapter 1, pp 1-92. (b) Bundgaard, H. Novel
chemical approaches in prodrug design. Drugs Future 1991, 16,
443-458. For a recent example of an enol prodrug, see: Burkhart,
J . P.; Koehl, J . R.; Mehdi, S.; Durham, S. L.; J anusz, M. J .;
Huber, E. W.; Angelastro, M. R.; Sunder, S.; Metz, W. A.; Shum,
P. W.; Chen, T.-M.; Bey, P.; Cregge, R. J .; Peet, N. P. Inhibition
of human neutrophil elastase. 3. An orally active enol acetate
prodrug. J . Med. Chem. 1995, 38, 223-233.
(11) The hemiterephthalate was also prepared but proved to be
extremely insoluble in most solvents (H2O, DMSO, etc.) and
could not be purified to a level acceptable for screening.
(12) As a corollary, the hemisuccinate ester cannot be isolated after
exposure of 1 to succinic anhydride.
Ack n ow led gm en t . We thank Dr. J on Bordner
(Pfizer Inc., Central Research, Groton, CT) for the X-ray
crystallographic analyses of compounds 12 and 18.
(13) Bundgaard, H.; Falch, E.; J ensen, E. A novel solution-stable,
water-soluble prodrug type for drugs containing a hydroxyl or
an NH-acidic group. J . Med. Chem. 1989, 32, 2503-2507.
(14) Robinson, R. P.; Donahue, K. M. Synthesis of N-alkoxycarbonyl
and N-carboxamide derivatives of anti-inflammatory oxindoles.
J . Heterocycl. Chem. 1994, 1541-1544.
Su p p or tin g In for m a tion Ava ila ble: X-ray crystal struc-
ture experimental and data for compounds 12 and 18 (26
pages). Ordering information is given on any current mast-
head page.
(15) [(Aminoacyl)oxy]methyl esters have been examined as water-
soluble prodrugs of cephalosporins; see: Wheeler, W. J .; Preston,
D. A.; Wright, W. E.; Huffman, G. W.; Osborne, H. E.; Howard,
D. P. Orally active esters of cephalosporin antibiotics. 3.
Synthesis and biological properties of aminoacyloxymethyl esters
of 7-[D-(-)-mandelamido]-3-[[(1-methyl-1H-tetrazol-5-yl)thio]-
methyl]-3-cephem-4-carboxylic acid. J . Med. Chem. 1979, 22,
657-661.
Refer en ces
(1) (a) Robinson, C. Tenidap sodium. Drugs Future 1990, 15, 898-
901. (b) Tenidap sodium, Enable, Enablex. Drugs Future 1993,
18, 875-878.
(2) (a) Kraska, A. R.; Wilhelm, F. E.; Kirby, D. S.; Loose, L. D.; Ting,
N.; Shanahan, W. R.; Weiner, E. S. Tenidap vs. piroxicam vs.
piroxicam plus hydroxychloroquine in rheumatoid arthritis: a
24-week interim analysis. Arthritis Rheum. 1993, 36 (Suppl. 9),
S57. (b) Leeming, M. R. G. A double-blind randomized compari-
son of tenidap vs. auranofin plus diclofenac in early rheumatoid
arthritis (RA). Arthritis Rheum. 1993, 36 (Suppl. 9), S111.
(3) (a) Carty, T. J .; Showell, H. J .; Loose, L. D.; Kadin, S. B.
Inhibition of both 5-lipoxygenase and cyclooxygenase pathways
of arachidonic acid (AA) metabolism by CP-66,248, a novel anti-
inflammatory compound. Arthritis Rheum. 1988, 31 (Suppl. 4),
S89. (b) Moilamen, E.; Alanko, J .; Asmawi, M. Z.; Vapaatalo, H.
CP-66,248, a new anti-inflammatory agent is a potent inhibitor
of leukotriene B4 and prostanoid synthesis in human polymor-
phonuclear leucocytes in vitro. Eicosanoids 1988, 1, 35-39.
(4) (a) LaLiberte, R.; Perregaux, D.; Svensson, L.; Pazoles, C. J .;
Gabel, C. A. Tenidap modulates cytoplasmic pH and inhibits
anion transport in vitro. I. Mechanism and evidence of functional
significance. J . Immunol. 1994, 153, 2180-2193. (b) LaLiberte,
R.; Perregaux, D.; Svensson, L.; Pazoles, C. J .; Gabel, C. A.
Tenidap modulates cytoplasmic pH and inhibits anion transport
in vitro. II. Inhibition of IL-1â production from ATP-treated
monocytes and macrophages. J . Immunol., in press. (c) McNiff,
P. A.; LaLiberte, R.; Svensson, L.; Pazoles, C. J .; Gabel, C. A.
Inhibition of cytokine activation processes in vitro by tenidap, a
novel anti-inflammatory agent. Cytokine, in press. (d) Otterness,
I. G.; Bliven, M. L.; Downs, J . T.; Natoli, E. J .; Hanson, D. C.
Inhibition of interleukin 1 synthesis by tenidap: a new drug for
arthritis. Cytokine 1991, 3, 277. (e) Sipe, J . D.; Bartle, L. M.;
Loose, L. D. Modification of proinflammatory cytokine production
by the antirheumatic agents tenidap and naproxen: a possible
correlate with clinical acute phase response. J . Immunol. 1992,
148, 480.
(16) This stability profile resembles that of cefamandole (aminoacyl)-
oxy ester prodrugs (ref 15).
(17) (a) Bundgaard, H.; Nielsen, N. M. Esters of N,N-disubstituted
2-hydroxyacetamides as a novel highly biolabile prodrug type
for carboxylic acid agents. J . Med. Chem. 1987, 30, 451-454.
(b) Bundgaard, H.; Nielsen, N. M. Glycolamide esters as a novel
biolabile prodrug type for nonsteroidal anti-inflammatory car-
boxylic acid drugs. Int. J . Pharm. 1988, 43, 101-110. (c)
Bundgaard, H.; Nielsen, N. M. Glycolamide esters as biolabile
prodrugs of carboxylic acid agents: synthesis, stability, biocon-
version, and physicochemical properties. J . Pharm. Sci. 1988,
77, 285-298.
(18) Several glycolamide derivatives in the ether series (e.g., R ) CH-
(Me)OCO2CH2CONEt2, Table 1) were also prepared but were
likewise poorly active as prodrugs of 1 in the rat.
(19) Fujimoto, K.; Ishihara, S.; Yanagisawa, H.; Ide, J .; Nakayama,
E.; Nakao, H.; Sugawara, S.-I.; Iwata, M. Studies on orally active
cephalosporin esters. J . Antibiot. 1987, 40, 370-384.
(20) Yoshimura, Y.; Hamagushi, N.; Yashiki, T. Preparation of
1-acyloxyethyl esters of 7-[2-(2-aminothiazol-4-yl)acetamido]-3-
[[[1-(dimethylaminoethyl)-1H-tetrazol-5-yl]thio]methyl]ceph-3-
em-4-carboxylic acid (cefotiam) and their oral absorption in mice.
J . Antibiot. 1986, 39, 1329-1342.
(21) The chloromethyl esters used for the preparation of compounds
5, 6, 17, 19, 20, and 22 are known in the literature. (a)
Chloroethyl pivalate: Rasmussen, M.; Leonard, N. J . The
synthesis of 3-(2′-deoxy-D-ribofuranosyl)adenine. Application of
a new protecting group, pivaloyloxymethyl (Pom). J . Am. Chem.
Soc. 1967, 89, 5439-5444. (b) Chloromethyl benzoate: Bodor,
N.; Kaminski, J . J .; Worley, S. D.; Gerson, S. H. Quantitative
evaluation of the reactivity of alkylating agents. Z. Naturforsch.,
B. Anorg. Chem. Org. Chem. 1980, 35, 758-763. (c) N-t-BOC-
glycine chloromethyl ester, N-t-BOC-L-valine chloromethyl ester,
and N-t-BOC-L-leucine chloromethyl ester: see ref 15.
(22) Lutz, R. E. The synthesis and configurations of unsaturated 1,4-
diketones and ketonic acids, and the stereochemical mechanism
of the addition of bromine. Studies on unsaturated 1,4-diketones.
VI. J . Am. Chem. Soc. 1930, 52, 3423.
(5) Carty, T. J . Personal communication.
(6) (a) Carty, T. J .; Marfat, A.; Moore, P. F.; Falkner, F. C.; Twomey,
T. M.; Weissman, A. Ampiroxicam, an anti-inflammatory agent
which is a prodrug of piroxicam. Agents Actions 1993, 39, 157-
165. (b) Hopkins, S. J .; Rabasseda, X. Ampiroxicam: A prodrug
of piroxicam devoid of gastrointestinal toxicity. Drugs Today
1994, 30, 557-563. (c) Falkner, F. C.; Twomey, T. M.; Borger,
A. P.; Garg, D.; Weidler, D.; Gerber, N.; Browder, I. W. Disposi-
tion of ampiroxicam, a prodrug of piroxicam, in man. Xenobiotica
1990, 20, 645-652. (d) Ampiroxicam. Drugs Future 1992, 17,
451-454.
(7) Marfat, A. Ether prodrugs of anti-inflammatory oxicams. US
Patent 4,551,452.
(8) (a) Farre, A. J .; Colombo, M.; Fort, A.; Gutie´rrez, B.; Rodriguez,
L.; Roser, R. Pharmacological properties of droxicam, a new
nonsteroidal anti-inflammatory agent. Methods Find. Exp. Clin.
Pharmacol. 1986, 8, 407-422. (b) Esteve, A.; Farre, A. J .; Roser,
(23) Sharma, S. K.; Miller, M. J .; Payne, S. M. Spermexatin and
spermexatol: new spermidine-based siderophore analogues. J .
Med. Chem. 1989, 32, 357-367.
J M950575K