Drug delivery systems based on trimethyl lock lactonization: Poly(ethylene glycol) prodrugs of amino-containing compounds

Article abstract of DOI:10.1021/jm990498j

A novel methodology for the synthesis of poly(ethylene glycol) (PEG) prodrugs of amino-containing compounds has been developed which is based on the trimethyl lock lactonization reaction. These PEG-modified double prodrugs are water soluble, and by select

Full text of DOI:10.1021/jm990498j

J . Med. Chem. 2000, 43, 475-487
475
Dr u g Deliver y System s Ba sed on Tr im eth yl Lock La cton iza tion : P oly(eth ylen e
glycol) P r od r u gs of Am in o-Con ta in in g Com p ou n d s
Richard B. Greenwald,* Yun H. Choe, Charles D. Conover, Kwok Shum, Dechun Wu, and Maksim Royzen
Enzon, Inc., 20 Kingsbridge Road, Piscataway, New J ersey 08854
Received September 30, 1999
A novel methodology for the synthesis of poly(ethylene glycol) (PEG) prodrugs of amino-
containing compounds has been developed which is based on the trimethyl lock lactonization
reaction. These PEG-modified double prodrugs are water soluble, and by selective modification
of the specifier or trigger, plasma half-lives can be adjusted at will to result in a wide range of
values. Facile syntheses of ester, carbonate, and carbamate functionalities were accomplished
and combined with greater or lesser degrees of steric hindrance in the spacer group, or on the
aromatic framework, to achieve predictable ranges of drug concentration in plasma. In vivo
screening of PEG prodrugs was done using a M109 syngeneic solid mouse tumor model. One
of the PEG-daunorubicin prodrugs, with a half-life of 2 h, was evaluated in an in vivo solid
tumor panel and found to be more efficacious against ovarian tumors (SKOV3) than equivalent
amounts of daunorubicin.
In tr od u ction
The lactonization of o-hydroxyphenylpropionic acid
derivatives was first investigated in depth by Cohen1
who defined the kinetic and structural parameters for
the ring-closure reaction (usually referred to as the
trimethyl lock (TML) effect). Subsequently, other groups
further refined the kinetic details of the lactonization.2
These findings ultimately formed the basis for an
important class of amino protecting groups and pro-
drugs based on the â,â-dimethylpropionic acid amide
side chain of the aromatic system. To effectively utilize
these reactive compounds as prodrugs, the ortho phe-
nolic group, similarly to 1,4- and 1,6-elimination-based
prodrugs,3 must necessarily first be modified to stabilize
the reactive system. This is accomplished by synthesiz-
ing acyl phenolic derivatives that can be predictably
hydrolyzed, or through a (bio)reductive mechanism
which generates a phenol, in essence, by forming a
double prodrug,4 or tripartate system,5 which consists
of a trigger6 (or specifier),5 a linker, and the drug (amino
or hydroxyl compound) to be released. Thus, regenera-
tion of the OH group becomes the rate-determining step
at which amine (drug) is generated (Figure 1). This
F igu r e 1. Mechanism of trimethyl lock (TML) prodrugs.
process has been referred to as an esterase-mediated
amide hydrolysis.7
the methodology has led to the release of a model
The application of the chemistry of this intriguing
hexapeptide.11 Most recently, Wang12 has shown that
cyclization reaction was initially investigated by Car-
cyclization (lactonization) of o-hydroxy-cis-cinnamic acid
pino,8 utilizing a reductive approach to the TML system
amides to coumarins, rapidly releasing amines, also
with simple amine models. Carpino first suggested in
lends itself to the list of feasible lactonization methods
this study that prodrugs of this sort would have
to consider for prodrug strategies. Prodrug strategies
potential applicability in treating solid tumors. Subse-
based on the TML approach as well as other intramo-
quent efforts have been reported which also utilize a
lecular cyclization reactions have recently been re-
reductive activation trigger.9 However, the TML double-
viewed.13
prodrug approach employing esterase-sensitive speci-
The use of the TML double-prodrug approach has also
fiers to generate amines has been extensively developed
been adapted to release alcohols when rates of hydroly-
by Borchardt and co-workers.7,10 By altering the enzyme-
sis, dissimilar to those of simple esters, are required.
sensitive specifier to an amino acid ester, extension of
Thus, an innovative use of the TML system which relied
on the alkaline phosphatase-mediated hydrolysis of an
* To whom correspondence should be addressed. Tel: 732-980-4924.
Fax: 732-980-5911. E-mail: RGREENWA@ENZON.com.
aromatic phosphate derivative was employed by Ueda
10.1021/jm990498j CCC: $19.00 © 2000 American Chemical Society
Published on Web 01/19/2000

476 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3
Greenwald et al.
Sch em e 1. Synthesis of TML Prodrugs with Ester
to produce a prodrug of paclitaxel (a secondary alcohol)
which demonstrated comparable activity to paclitaxel
in a M109 mouse model.14 Another unique prodrug
design incorporating ganciclovir (a primary alcohol) and
the TML system was shown to provide 5 times greater
oral bioavailability of the drug in mice.15
Triggersa
The enhanced permeation and retention (EPR) effect
has been established as a useful property with which
to target anticancer drugs.16 The underlying physiologi-
cal mechanism appears to be a combination of increased
tumor vascular permeability with insufficient lymphatic
drainage resulting in the greater accumulation of mac-
romolecular drugs. Thus, passive accumulation of high
molecular weight poly(ethylene glycol) (PEG) drug
conjugates in the interstitial tissue of tumors translates
into lower systemic toxicity while efficacy is enhanced,
thereby increasing the therapeutic index of the drug.
PEG-conjugated prodrugs (transport forms) based on
1,4- or 1,6-elimination reactions (BE system) have been
demonstrated to be an effective means for delivery of
the anticancer agent daunorubicin (DNR)17 and in
general provide a practical approach for solubilizing and
transporting amino-containing antitumor agents as well
as a variety of other drugs. In continuing our efforts to
define the limits of the PEG prodrug strategy, it was
apparent that the use of lactonization reactions could
be incorporated into the strategy and would provide a
practical alternative to elimination reactions. We have
now extended and refined existing PEG technology to
embrace the concept of the TML tripartate or double-
prodrug system (Figure 1). To utilize the TML system
for polymer-conjugated prodrugs it was necessary to
first establish various methodologies which allowed the
efficient synthesis of different acyl functionalities (trig-
gers) such as esters, carbonates, and carbamates on the
phenolic hydroxyl group. The acylating agents must by
necessity be bifunctional and offer a site for easy PEG-
ylation. Thus, introduction of large molecular weight
PEG into the TML system as part of the specifier or
trigger results in a neutral and highly water-soluble
tripartate polymeric prodrug capable of passive tumor
targeting. The PEG prodrug can be designed to attain
predictable rates of hydrolysis by changing the nature
of the trigger/linker bond, by adding steric hindrance
on the aromatic ring of the linker, and by the use of
different spacer groups (Figure 1). This approach offers
a versatile methodology for easily altering the final
design of the prodrug: it enables a “mix and match” of
spacers, triggers, and linkers that can be utilized in a
meaningful manner and ultimately provides optimal
pharmacokinetics for delivery of different types of drugs
in addition to antitumor agents.
a
(i) Methyl 3,3-dimethylacrylate, MeSO3H; (ii) LAH; (iii) TB-
DMS-Cl; (iv) EDC, DMAP, Boc-AA-OH; (v) HOAc; (vi) PCC; (vii)
NaClO2, H2O2; (viii) TFA; (ix) T-PEG, DIEA; (x) DNR‚HCl, EDC,
HOBT, NMM.
the N-4 amino to N-4 hydroxy and formation of inactive
ara-U.20 This was selected as a second model for
modification using amino conjugation to a TML system
at the N-4 position in order to demonstrate path A
chemistry (see below).
This paper illustrates the versatility of the TML
system using both DNR and ara-C. Synthesized PEG-
DNR derivatives were subsequently evaluated in vitro
and in vivo to examine their hydrolysis and biological
activity, respectively. The biological effects of PEG-
ara-C prodrug derivatives in the nude mouse bearing
human xenografts will be the subject of a separate
report.
Ch em istr y
Abbr evia tion s: A (R1 ) H, R2 ) CH3), B (R1 ) CH3,
R2 ) H), a (AA ) alanine), p (AA ) proline), â (AA )
â-alanine) [e.g. compound 5Aa : R1 ) H, R2 ) CH3, AA
) alanine], DCM (dichloromethane), DIEA (diisopropyl-
ethylamine), DIPC (1,3-diisopropylcarbodiimide), DNR
(daunorubicin), EDC (1-ethyl-3-(3-dimethylaminopro-
pyl)carbodiimide), LAH (lithium aluminum hydride),
HOBT (1-hydroxybenzotriazole), NMM (N-methylmor-
pholine), TBDMS-Cl (tert-butyldimethylsilyl chloride),
TEA (triethylamine).
Ester -Ba sed Tr igger s. The first entry into a hydro-
lytically labile PEG-TML system was accomplished
using the key intermediate 8, generated by the conve-
nient multistep synthesis developed by Borchardt and
co-workers11 (Scheme 1). Using 6Aa as a starting point
to build a cyclic prodrug, Borchardt accomplished the
elegant synthesis of a releasable peptide. For our
purposes, the alaninate ester intermediate 9Aa worked
admirably as the focal point for attachment of PEG as
shown in Scheme 1. When it was desired to alter the
Acylation of the primary amino group of DNR results
in loss of activity of the anthracycline drug.19 Therefore,
antitumor activity can be monitored as a function of
cleavage of the DNR amide prodrug conjugate in the
current PEG-TML development strategy, regardless of
which TML specifier cleaves.
Another drug that readily lends itself to the current
prodrug strategy is the antimetabolite 1-â-D-arabino-
furanosylcytosine (ara-C, cytarabine), a drug containing
a weakly basic aromatic amine which is used to treat
leukemia. Ara-C derivatization, as an amide, prevents
rapid inactivation via cytosine deaminase conversion of

Drug Delivery Systems via TML Lactonization
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3 477
Sch em e 2. Synthesis of TML Prodrugs with Carbonate
plasma half-life (t1/2) of the PEG prodrug, the phenolic
derivative 3 (Scheme 1) allowed the facile insertion of
other Boc-protected amino acids such as â-alanine, or
the more sterically hindered proline. In the course of
this work it was found that coupling of 3 with Boc amino
acids using EDC was more easily achieved than the
procedure employing Boc amino acid-activated esters.11
In tr od u ction of Ster ic Hin d r a n ce by Mod ifica -
tion of th e Rin g. As described in our earlier work on
drug delivery using the BE system,17 another effective
method which allowed adjustment of the rate of hy-
drolysis of the phenolic latentiating moiety (trigger) was
to introduce additional hindering alkyl groups into the
aromatic ring at the ortho position. This was ac-
complished for the TML system as depicted in Scheme
1. The known lactone derivative 2B was synthesized
starting from 2,5-dimethylphenol (1B) and further
converted to the aldehyde 7B in the usual manner
employing pyridinium chlorochromate (PCC). Aldehyde
7B was partially purified by simple silica gel filtration
and used directly in the next step. By employing NaClO2
as the oxidizing agent,21 a high yield of pure acid 8B
was easily obtained. Additional ortho position modifica-
tion, which substantially increased t1/2, was utilized with
different spacers and triggers (promoieties) leading to
a “mix and match” situation to achieve adjustments in
plasma hydrolysis. This is illustrated in Table 1 where
t1/2 and tumor efficacy of various PEG-TML prodrugs
can be seen to reflect the various steric (and electronic)
environments.
Triggersa
a
Ca r bon a te-Ba sed Tr igger s. Carbonate triggers for
tripartate TML systems are unknown. Coupling of
activated PEG (SC-PEG) directly to the phenolic OH of
4 to provide a carbonate bond was not attempted
because of potential concern for degradation of the
polymer during subsequent oxidation steps employing
PCC. The synthesis of this type of trigger ultimately
led to the use of a heterobifunctional spacer which was
found to provide the necessary means to attach PEG.
The design of the spacer chosen reflected an unambigu-
ous modification since it has been previously demon-
strated that anchimeric assistance to hydrolysis of
esters and carbonates can occur when NH groups are
placed in the δ and ꢀ positions to the carbonate carbo-
nyl.22 The desired PEG-TML compound was obtained
using the series of reactions outlined in Scheme 2.
Starting with the inexpensive and readily available
aminoethoxy alcohol (12), the activated succinimidyl
carbonate 14 was prepared in 95% overall yield in two
steps. Condensation of 14 with 4 gave the key TML silyl
intermediate 15 which was readily converted to the
amino derivative 19. Conversion of 19 to the key PEG
acid 20 was achieved using PEG thiazolidinethione
(T-PEG)23 via path B (discussed below). T-PEG is a
particularly useful reagent for acylation of amines and
phenols, in high yield, under mild conditions.17 In the
case of DNR, coupling to 20 was carried out with EDC
in the presence of 1-hydroxybenzotriazole (HOBT) and
the base N-methylmorpholine (NMM) to yield the
desired PEG transport form (prodrug), 21. HOBT cou-
plings11 were found to produce higher yields and cleaner
products than other commonly used reagents such as
DIPC or EDC.
(i) Boc2O, NaOH; (ii) DSC, pyridine; (iii) 4, DIEA; (iv) HOAc;
(v) PCC; (vi) NaClO2, H2O2; (vii) TFA; (viii) T-PEG, DIEA; (ix)
DNR‚HCl, EDC, HOBT, NMM.
functionality in the design of PEG-DNR prodrugs.17
This relatively stable trigger, like the carbonate, has
not yet been incorporated into TML tripartate prodrugs.
We therefore devised a synthetic strategy based on the
known monoprotected diamine, 2-amino-2′-(Boc-amino)-
ethylene glycol diethyl ether,24 which was condensed
with the chloroformate of 4, generated in situ using
triphosgene, to give the desired carbamate 22 in 78%
yield (Scheme 3). The usual sequence of reactions (22-
27, Scheme 3) gave the PEG linker 27, which was
coupled to DNR mediated by EDC/HOBT/NMM. The
final PEG-DNR conjugate 28 was purified by crystal-
lization from 2-propanol.
Deter m in a tion of P a th w a y. The use of amino acid
esters as bifunctional spacers enables facile attachment
of the TML system to PEG. As illustrated in Scheme 4
there are two viable routes, path A and path B, that
can be employed and lead to the final PEG prodrugs
(VII). Both routes involve trifluoroacetic acid (TFA)
cleavage of a Boc protecting group and generation of free
amine, but path A introduces the drug first followed by
PEG conjugation, while path B involves an initial
attachment to PEG and subsequently bonding the drug.
Path B is applicable for all candidates, but if the drug
chosen to be modified can survive TFA treatment, path
A is preferred because purification of the monomeric
adduct prior to introduction of PEG is a more facile
process and enables very high loading of parent drug
in the PEG conjugates to be realized. Drug can be
attached to the TML acid I by a number of methods:
the simplest of these is to directly couple with a
carbodiimide, usually EDC.
Ca r ba m a te-Ba sed Tr igger s. Our previous work
identified a carbamate specifier (trigger) as a desirable

478 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3
Greenwald et al.
Sch em e 3. Synthesis of TML Prodrugs with Carbamate
Sch em e 5. Synthesis of PEG-TML-Ara-C Prodrugsa
Triggersa
a
a
(i) (a) Triphosgene, pyridine, (b) 2-amino-2′-(Boc-amino)eth-
(i) Ara-C, EDC, HOBT, pyridine; (ii) TFA; (iii) T-PEG, DIEA.
ylene glycol diethyl ether, pyridine; (ii) HOAc; (iii) PCC; (iv)
NaClO2, H2O2; (v) TFA; (vi) T-PEG, DIEA; (vii) DNR‚HCl, EDC,
HOBT, NMM.
29 in the expected modest yield (60%; Scheme 5).
Purification of 29 was accomplished by silica gel column
chromatography followed by treatment with TFA which
removed the Boc protecting group and freed the R-amino
group of the alanine connector. Last, TFA salt 30 was
conjugated to PEG by employing T-PEG to give the
desired conjugate 31 in 84% weight yield after recrys-
tallization from 2-propanol. On the basis of the quan-
titation (UV method) of ara-C in the conjugate, the
conjugation yield of active drug was approximately 90%.
P a th B: Treatment of DNR with TFA appears to
result in cleavage of the O-glycosidic bond and loss of
the crucial amino sugar.25 Therefore, by necessity path
B (Scheme 4) must be the synthetic route utilized in
this particular case. This is illustrated in Scheme 1
where cleavage of 6 with TFA was done initially to
provide 9, followed by reaction with T-PEG to afford the
key intermediate 10. PEG acid 10 can be converted
directly to the final product 11 by reaction with DNR
in the presence of EDC/HOBT using DCM as the
solvent. However, this procedure is not always the
method of choice for other drugs.25 Several easily
prepared activated derivatives, VI (Scheme 4), X )
p-nitrophenyl (PNP) and N-hydroxysuccinimide (NHS)
esters, or the activated imide formed from the in situ
generation of acid chloride followed by reaction with
2-mercaptothiazoline in the presence of a tertiary
amine23 were often utilized effectively as alternate
reagents to accomplish drug conjugation when carbo-
diimide-mediated couplings of acids to amines were
inefficient.
Sch em e 4. Synthetic Routes to PEG-TML Prodrugs
Resu lts a n d Discu ssion
The synthetic methodology of the TML strategy
accomplished in this work resulted in a broad, utilitar-
ian PEG technology platform consisting of many vari-
able combinations of specifiers and linkers for designing
PEG amino prodrugs. A noteworthy feature of the TML
P a th A: Ara-C is a drug which exemplifies the facility
of using this route. Usually, ara-C acylation is a low-
yield reaction.20 Condensation of TML system 8 with
ara-C in the presence of EDC and pyridine gave adduct

Drug Delivery Systems via TML Lactonization
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3 479
Ta ble 1. In Vitro and in Vivo Results of PEG-TML Prodrugsa
a
b
All in vitro experiments were done at 37 °C in duplicate. Standard deviation of measurements ) (10%. 3 mg/kg/dose of active DNR
administered to balb/ c mice bearing subcutaneous Madison lung carcinoma on 1 & 4 (intraperitoneal) or 3 & 6 (intravenous) days after
inoculation. Percent treatment over control (% T/C) median tumor volumes were compared when control groups median tumor volume
reached ∼2000 mm3. c All the t1/2 values were calculated from the R phase except that for 28A which did not show a clear separation of
R and â phases.
prodrugs is that they are amides and as such can be
distinguished from the BE pathway which is based on
carbamates. Interestingly, in the BE series a carbamate
derivative of ara-C proved to be difficult to prepare while
TML amide derivatives were easily obtained; thus the
two methods appear to complement each other. In the
present study incorporating PEG and TML linkers, a
series of amino acid ester, carbonate, and carbamate
triggers were prepared and integrated into a DNR-
PEG-TML double prodrug. All these compounds dem-
onstrated prolonged stability (>24 h) in pH 7.4 phos-
phate buffer at 37 °C: thus, physically they have
potential clinical utility as injectable agents.
(11Aa ) to a proline (11Ap ) caused about a 10-fold
decrease in the rate of hydrolysis in rat plasma, while
removal of the R-substituent from spacer 11Aa resulted
in a 10-fold rate enhancement for compound 11Aâ.
Although a more rapid release rate is not desirable for
severely toxic anticancer drugs, it may be useful for the
rapid release of other classes of compounds. Thus the
steric requirements of the spacer group are of para-
mount importance in the design of TML cyclization
prodrugs. Interestingly, it has been reported that steric
bulkiness of the acyl group (trigger) has only a very
minor effect on the half-lives of esterase-mediated
release of amines from model cyclization prodrugs that
are coumarin based.12c Second, comparing 11Aa to 11Ba
shows that introduction of an o-methyl substituent into
the linker decreases the rate of plasma hydrolysis over
10-fold. Third, a carbonate trigger (promoiety), 21A,
hydrolyzes about 5-fold slower than an ester promoiety
Introduction or removal of substituents, either on the
spacer or on the aromatic linker, resulted in significant
changes of t1/2 in rat plasma. Some interesting observa-
tions regarding plasma hydrolysis can be gleaned by
examining Table 1. First, changing the alanine spacer

480 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3
Greenwald et al.
Ch a r t 1
plasma hydrolysis. However, when these compounds
were administered by the more clinically relevant
intravenous (iv) route, at the same dosage, they were
not toxic, suggesting rapid elimination of nascent DNR.
In fact, at the given dosage, none of the compounds
appeared toxic when administered iv, and compounds
11Aâ, 11Ba , and 11Bâ demonstrated the greatest
efficacy as compared to native DNR. Interestingly, the
in vitro plasma t1/2 for the active compounds ranged
from 0.2 to 24 h; thus understanding their efficacy in
relation to their rates of hydrolysis is problematic and
warrants an examination of their in vivo biodistribution.
However, a common thread that runs through these
derivatives, as well as for 11Aa which is also quite
effective, is the incorporation of an amino acid ester
spacer. The production of amino acid-conjugated species
has been implicated in enhanced cellular uptake of
DNR.28 Intriguingly, two recent papers have reported
on the cellular uptake of amino acid ester prodrugs by
a peptide transporter, although no peptide bond is
present in their structure.29,30 Further SAR relation-
ships are currently being explored in hopes of more
clearly defining the role of the spacer group in the drug
delivery of antineoplastic agents.
We previously reported17 that the type 1 carbamate
derivative of DNR had an in vitro plasma t1/2 of 4 h and
produced the best solid tumor (M109 and SKOV3)
results within the BE series. To compare the activities
of the BE and TML compounds, the TML derivative
11Aa appeared to approximate the BE type 1 carbamate
most closely (t1/2 of ca. 2 h) and was chosen to evaluate
chemotherapeutic activity against a panel of human
tumor xenografts, including SKOV3 (Table 2). SKOV3
tumors were the most sensitive to compounds in the
TML series; activity was exceptionally good and quite
similar to that for the BE derivative (%T/C 11 vs 5 for
BE and TML, respectively). PEG conjugation did not
appear to enhance the activity of DNR against tumor
lines, which were insensitive to DNR (MX-1 and PC-3).
Since significant antitumor activity was observed in the
SKOV3 tumor line (which was susceptible to DNR
treatment) we are currently planning to screen the
entire TML-DNR series against an ovarian tumor
panel.
(11Aâ). Furthermore, the hydrolysis of a carbamate
trigger (28A) with a TML linker was unexpectedly quite
slow (t1/2 > 24 h). In the case of a similar type carbamate
ester of a phenol in the BE system,17 a much shorter
t1/2 of 4 h was found which reflected the expected
mechanism of the reaction (base removal of a proton).26
To help clarify whether the innate steric hindrance (o-
tert-alkyl group) of the TML linker was responsible for
this difference, we prepared the corresponding simple
PEG linker 32 (Chart 1) based solely on phenol.
Compound 32 exhibited a plasma t1/2 ) 5.2 h, thus
demonstrating that ortho substitution may affect proton
abstraction in the crowded TML system. Finally, of the
seven compounds examined in Table 1, rat plasma
hydrolysis data shows only one derivative with a t1/2
between 2 and 17 h. While further combinations of
triggers and linkers should produce more intermediate
t1/2 values, it is evident that adjusting t1/2 of TML linkers
is not as facile as was the case for the BE system.
It has been our experience with PEG prodrugs that a
series usually shows a direct correlation between rapid
plasma t1/2 and low IC50 values. For example, PEG
prodrugs based on the BE system generally exhibited
low IC50 values when their plasma t1/2 vlaues were short
indicative that hydrolysis of the promoiety also takes
place in cell media.17 However, with the exception of
11Aa , compounds within the TML series do not appear
to readily break down in cell media and subsequently
had high IC50 values (Table 1). Fortunately, M109
results clearly established that these same compounds
are indeed biologically active. This finding again dem-
onstrates that in vitro results can be misleading in
predicting in vivo outcomes. This is the first time we
have observed this type of stability in cell media for a
PEG prodrug series; we are currently exploring this
enigma.
To exclude the possibility that any phenolic or lactone
species generated from the linker during the breakdown
of the PEG-drug conjugates possessed toxic properties
that might in any way affect the models, or was in part
responsible for any of the tumor regression observed,
we synthesized the simple isopropylamine conjugate of
10, compound 33Aa (Chart 1). This model compound
(33Aa ), which contained no anticancer agent, was tested
in normal mice at 3 times the concentration of PEG
linker used for the DNR-conjugated compound 11. No
adverse effects or toxicities were noted at this dose.
Since this TML latentiated simple amine also demon-
strated no in vitro cell inhibitory activity, it appears that
the anticancer agent and its delivery to the tumor target
are entirely responsible for any in vivo tumor growth
inhibitions observed.
The safety and efficacy of the PEG conjugates within
the M109 model varied according to the route of
administration and the rate of in vitro dissociation
(Table 1). When the compounds were dosed intraperi-
toneally (ip), the highest activity was observed for native
DNR followed by 21A and 11Aa . Similar efficacy for
DNR using ip dosages in this model has been reported
by others.27 However, both free DNR and 21A were also
toxic when given by this route, with DNR causing
lethality in 17% of the animals while 21A resulted in
66% lethality. In addition, 11Aâ produced fatalities in
83% of the animals. Not surprisingly the three com-
pounds that caused lethality displayed the most rapid
Con clu sion s
PEG conjugated to amino prodrugs that function via
a TML lactonization has been demonstrated to be a

Drug Delivery Systems via TML Lactonization
Ta ble 2. Efficacy Comparison between DNR and PEG-DNR (11Aa ) Against Subcutaneous Human Tumorsa in Nude Mice
DNR PEG-DNR (11Aa )
% tumor growthc tumor inhibitiond growth delaye % tumor growth tumor inhibition growth delay
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3 481
tumor modelb
(∆ from initial)
(%T/C)
[%(T - C)/C]
(∆ from initial)
(%T/C)
[%(T - C)/C]
MX-1
mammary carcinoma
PC-3
prostate adenocarcinoma
SKOV3
1042
84
9
1671
124
0
1878
89
35
40
16
2176
79+*
84
5
44
3008+
>62f
ovarian adenocarcinoma
a
b
Mean baseline (initial) tumor volume was 75 mm3. 3 mg/kg/dose (DNR content) intravenously on days 1, 5, & 9. c % mean tumor
d
volume change from initial based on individual tumors at weeks 5, 4, and 6 in MX-1, PC-3, and SKOV3, respectively. The median
tumor volumes of treatment and control groups were measured and compared when the control group’s median tumor volume reached
∼1000 mm3. e % by which the treated median tumor volume was delayed in reaching 1000 mm3 as compared to the control. f Measurements
discontinued after day 60. + Significant versus control (P < 0.05). *Significant versus DNR (P < 0.05).
Co., NY). Boc-protected amino acids were purchased from
Advanced ChemTech (Louisville, KY). All PEG compounds
feasible methodology to deliver drugs-especially anti-
cancer agents-and is comparable to the BE system17
were dried under vacuum or by azeotropic distillation (toluene)
for introducing different trigger chemistries, spacer
groups, and steric hindrance in the linker portion in
1
prior to use. H NMR spectra were obtained with a J EOL FT
NMR System J NM GSX-270 instrument using deuteriochlo-
order to modify rates of drug release. It offers, in some
cases, the advantage of forming amides rather than
carbamates. By using the double-prodrug strategy and
changing promoieties (specifiers or triggers), alteration
of PEG prodrug pharmacokinetics can be accomplished
leading to greater drug efficacy. In a tripartate system
this can be achieved not only by changing the PEG
specifier or trigger but also by adding a spacer and/or
introducing steric hindrance either on the spacer or on
the linker, both of which will affect the rate of cleavage
of the specifier.
roform as solvent unless specified. 13C NMR spectra were
obtained at 67.80 MHz on the J NM GSX-270. Chemical shifts
(δ) are reported in parts per million (ppm) downfield from
tetramethylsilane and coupling constants (J values) are given
in hertz (Hz). TLC was performed on Whatman K6F silica gel
60 Å plates. Merck silica gel 60 (230-400 mesh) was used for
column chromatography. Elemental analyses were performed
by Quantitative Technologies, Inc. (Whitehouse, NJ ) and mass
analyses (ES and HRMS) were done at Yale Cancer Center
Mass Spectrometry Resource & W. M. Keck Roundation
Biotechnology Resource Laboratory (New Haven, CT) and The
Center for Advanced Food Technology at Rutgers, The State
University of New J ersey (New Brunswick, NJ ). All PEG-
conjugated DNR compounds were dissolved (∼15 mg/mL) in
sterile saline (0.9%) for injection prior to in vivo drug treat-
ments and were given as their DNR equivalents (absolute
amount of DNR given).
HP LC Meth od . Analytical HPLCs were performed using
a C8 reversed-phase column (Beckman, ultrasphere) under
isocratic conditions with an 80:20 mixture (v/v) of methanol-
water as mobile phase. Peak elutions were monitored at 254
nm using a UV detector. To detect the presence of any free
PEG and also to confirm the presence of PEGylated product,
an evaporative light scattering detector (ELSD), model PL-
EMD 950 (Polymer Laboratories), was employed. Based on
ELSD and UV analysis, all the final PEGylated products were
free of native drug and were g95% pure by HPLC.
It has been pointed out in previous work17 that most
amine drugs can be solubilized as acid salts, but their
rate of renal excretion is also high. When converted to
neutral small prodrug species, the ability to form salts
is lost, and solubility may again become problematic.
This is not so in the case of PEG-drug conjugates,
where PEG confers water solubility to insoluble small
organic compounds without the need for forming salts.
PEG-TML prodrug methodology can be accomplished in
a facile and reproducible manner, thus extending the
PEG prodrug strategy for amino-containing anticancer
compounds to drug release based on cyclization. These
results are also anticipated to be useful for drug delivery
of other difficulty soluble amine-containing pharmaceu-
tical agents. Additional representative details will be
reported soon. While we have not correlated rat plasma
An a lysis of DNR Con ten t in P EG Der iva tives.31 The
UV absorbance of native DNR in 86% EtOH was determined
at 475 nm for six different concentrations ranging from 0.02
to 0.08 µmol/mL. From the standard plot of absorbance vs
concentration, the absorption coefficient, ꢀ, for DNR was
calculated to be 21.6 (OD at 475 nm for 1 mg/mL with 1.0 cm
light path). PEGylated DNR derivatives were dissolved in
DMF at an approximate concentration of 0.015 µmol/mL (based
t
1/2 with in vivo efficacy for TML derivatives as favorably
as in our previous work, the PEG-TML drug delivery
approach ultimately resulted in the synthesis of a highly
potent conjugate derived from DNR that appeared even
more efficacious (on an equimolar basis) in an SKOV3
xenograft model than the parent drug or the BE system
type 1 carbamate PEG-DNR prodrug. Thus we con-
clude that TML lactonization indeed offers a very
practical approach for PEG modification and delivery
of drugs.
on
a MW of 40 kDa) and the UV absorbance of these
compounds at 475 nm was determined. Using this value and
employing the absorption coefficient, ꢀ, obtained from the
above, the concentration of DNR in the sample was deter-
mined. Dividing this value by the sample concentration
provided the percentage of DNR in the sample.
An a lysis of Ar a -C Con ten t in P EG Der iva tives. For the
determination of the ara-C content in PEG derivatives, N4-
acetylcytidine was used as the basis because of the absorbance
change due to the acylation of ara-C. The UV absorbance of
N4-acetylcytidine in H2O was determined at 257 nm for six
different concentrations ranging from 0.01 to 0.05 µmol/mL.
From the standard plot of absorbance vs concentration, the
absorption coefficient, ꢀ, of N4-acetylcytidine was calculated
to be 36.4 (OD at 257 nm for 1 mg/mL with 1.0 cm light path).
PEGylated ara-C derivatives were dissolved in H2O at an
approximate concentration of 0.015 µmol/mL (based on a MW
Exp er im en ta l Section
1. Ch em istr y. Gen er a l. For the reasons previously de-
scribed,17 all PEGs used in this study had a molecular weight
(MW) of 40 kDa. In addition, all reactions were run under an
atmosphere of dry nitrogen or argon. Commercial reagents
were used without further purification. DNR‚HCl was obtained
from Hande Tech USA, Inc. (Houston, TX) and ChemWerth
(Woodbridge, CT), ara-C from Sigma Chemical Co. (Madison,
WI), and PEG diol (40 kDa) from Serva (Crescent Chemical

482 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3
Greenwald et al.
of 40 kDa) and the UV absorbance of these compounds at 257
nm was determined. Using this value and employing the
absorption coefficient, ꢀ, obtained from the above, the concen-
tration of ara-C in the sample was determined. Dividing this
value by the sample concentration provided the percentage of
ara-C in the sample.
dissolved in DCM (140 mL) at 0 °C and a solution of TEA (62
mL) in DCM (60 mL) was added dropwise over 1 h. The
mixture was warmed to room temperature and stirred over-
night. The solvent was removed at 40 °C in vacuo and the
residue dissolved in DCM (500 mL). The solution was washed
with water (4 × 50 mL) and dried over anhydrous MgSO4. The
solvent was removed to give 32 g (91%) of product: 1H NMR
δ 0.08 (s, 6H, Si(CH3)2), 0.86 (s, 9H, SiC(CH3)3), 1.56 (s, 6H,
C(CH3)2), 2.14 (t, J ) 6.75, 2H, CH2CH2OSi), 2.15 (s, 3H,
ArCH3), 2.45 (s, 3H, ArCH3), 3.59 (t, J ) 6.75, 2H, CH2 CH2-
OSi), 5.80 (s, 1H, OH), 6.55 (d, J ) 6.75, 1H, ArH), 6.82 (d, J
) 6.75, 1H, ArH); 13C NMR δ 16.30, 18.19, 25.51, 25.62, 25.87,
25.87, 21.21, 39.67, 44.94, 61.66, 122.72, 125.18, 127.71,
132.04, 135.38, 153.79.
Deter m in a tion of Ra tes of Hyd r olysis of P EG P r o-
d r u gs. The rates of hydrolysis were obtained by employing a
C8 reversed-phase column (Zorbax SB-C8) using a gradient
mobile phase consisting of (a) 0.1 M triethylammonium acetate
buffer and (b) acetonitrile. A flow rate of 1 mL/min was used,
and chromatograms were monitored using a UV detector at
254 nm for DNR and 280 nm for ara-C. For hydrolysis in
buffer, PEG derivatives were dissolved in 0.1 M pH 7.4 PBS
at a concentration of 5 mg/mL, while for hydrolysis in plasma,
the derivatives were dissolved in distilled water at a concen-
tration of 20 mg/100 µL and 900 µL of rat plasma was added
to this solution. The mixture was vortexed for 2 min and
divided into 2-mL glass vials with 100 µL of aliquot/vial. The
solutions were incubated at 37 °C for various periods of time.
A mixture of methanol-acetonitrile (1:1, v/v, 400 µL) was
added to a vial at the proper interval and the mixture was
vortexed for 1 min, followed by filtration through 0.45-mm
filter membrane (optionally followed by a second filtration
through 0.2-mm filter membrane). An aliquot of 20 µL of the
filtrate was injected into the HPLC. On the basis of the peak
area, the amounts of native compound and PEG derivative
were estimated, and the half-life of each compound in different
media was calculated using linear regression analysis from
the disappearance of PEG derivative.
1-O-(ter t-Bu t yld im et h ylsilyl)-3-(2′-Boc-p r olin yl-4′,6′-
d im eth ylp h en yl)-3,3-d im eth ylp r op a n ol (5Ap ). DIPC (392
mg, 3.11 mmol) was added to a mixture of 4A10 (0.5 g, 1.55
mmol), DMAP (568 mg, 4.66 mmol), and Boc-Pro-OH (668 mg,
3.11 mmol) in anhydrous DCM (15 mL) at 0 °C. The mixture
was stirred at room temperature overnight, then filtered, and
concentrated. The residue was purified using silica gel column
chromatography (ethyl acetate-hexane ) 3:7, v/v) to give 700
mg (87%) of 5Ap : 1H NMR δ 0.04 (s, 3H, CH3Si), 0.14 (s, 3H,
CH3Si), 0.88 (s, 3H, CH3C), 0.92 (s, 3H, CH3C), 1.51 (s, 9H,
Si(CH3)3), 1.55 (s, 9H, C(CH3)3), 2.00 (m, 2H, CH2O), 2.07 (t,
2H, J ) 8.1, CH2CH2CH2), 2.25 (s, 3H, PhCH3), 2.30 (m, 2H,
CHCH2) 2.55 (s, 3H, PhCH3), 3.53 (t, 2H, J ) 8.1, CH2N), 4.57
(br s, 1H, COCHN), 6.65 (bs, 1H, ArH), 6.82 (s, 1H, ArH); 13
C
NMR δ -4.37, 18.23, 20.06, 25.14, 25.70, 25.98, 28.49, 32.00,
32.06, 39.51, 46.24, 46.62, 59.74, 60.93, 80.08, 122.73, 132.14,
134.56, 135.87, 138.30, 150.68, 171.84; EI MS m/z 542.37 (M+
+ Na, 70), 558.34 (M+ + K, 35); HRMS calcd for C29H49NO5-
SiNa (M+ + Na), 542.3278, found 542.3333.
4,4,5,8-Tet r a m et h yl-3,4-d ih yd r ocou m a r in (2B). This
compound was prepared by a modification of the existing
procedure:1a 2,5-Dimethylphenol (39.4 g, 0.33 mol) was added
to methanesulfonic acid (47 mL) followed by the addition of
methyl 3,3-dimethylacrylate (41 g, 0.36 mol). The mixture was
heated to 70 °C for 24 h. After cooling to room temperature,
the solution was poured into 1 L of water. The mixture was
extracted with ethyl acetate (1.5 L). The organic layer was
washed with 5% NaHCO3 and brine (1 L) and dried over
anhydrous MgSO4. The solvent was removed in vacuo, and
hexane added to the residue to precipitate a pale white solid
which was collected by filtration and washed with hexane to
give 32 g (48%) of product: 1H NMR δ 1.38 (s, 3H, C(CH3)2),
1.44 (s, 3H, C(CH3)2), 2.26 (s, 3H, ArCH3), 2.46 (s, 3H, ArCH3),
2.58 (s, 2H, CH2), 6.82 (d, J ) 6.75, 1H, ArH), 6.97 (d, J )
6.75, 1H, ArH); 13C NMR δ 16.09, 22.91, 27.47, 35.19, 45.46,
124.34, 127.86, 128.98, 129.35, 133.40, 137.62, 168.18.
1-O-(ter t-Bu tyld im eth ylsilyl)-3-(2′-Boc-â-a la n in yl-4′,6′-
d im eth ylp h en yl)-3,3-d im eth ylp r op a n ol (5Aâ). EDC‚HCl
(35 g, 0.182 mol) was added to the mixture of 4A (20 g, 0.062
mol), Boc-â-Ala-OH (23.5 g, 0.124 mol), and DMAP (53 g, 0.434
mol) in anhydrous DCM (150 mL) at 0 °C. The mixture was
stirred overnight and 600 mL of DCM was added. The solution
was washed with 1% NaHCO3 (3 × 150 mL) and 1 N HCl (3
× 150 mL) and dried over anhydrous MgSO4. The solvent was
removed in vacuo and the residue was purified by silica gel
column chromatography (10% EtOAc in hexane) to give 26 g
(85%) of product: 1H NMR δ 0.00 (s, 6H, 2 × CH3Si), 0.87 (s,
6H, 2 × CH3C), 1.48 (s, 18H, C(CH3)3 & Si(CH3)3), 2.05 (t, 2H,
J ) 5.4, CH2C), 2.26 (s, 3H, ArCH3), 2.54 (s, 3H, ArCH3), 2.79
(t, 2H, J ) 5.4, CH2C(dO)), 3.50 (t, 4H, J ) 8.1, NHCH2
&
CH2O,), 5.01 (br s, 1H, NH), 6.58 (s, 1H, ArH), 6.84 (s, 1H,
ArH); 13C NMR δ -4.42, 18.18, 20.11, 25.20, 25.88, 28.36,
31.78, 35.40, 36.00, 38.99, 45.94, 60.69, 79.35, 122.91, 132.38,
3-(2′-Hyd r oxy-3′,6′-d im et h ylp h en yl)-3,3-d im et h ylp r o-
p a n ol (3B). A solution of 2B (30 g, 0.15 mol) in anhydrous
THF (300 mL) was cooled in an ice bath and slowly added to
a suspension of LAH (10.8 g, 0.285 mol) in THF (150 mL)
under nitrogen. The mixture was stirred overnight 0 °C and
then allowed to warm to room temperature. The reaction was
monitored by TLC. 300 mL of HPLC grade THF was added
slowly with condenser attached, followed by 40 mL of saturated
aqueous ammonium chloride solution to quench excess LAH,
and the mixture was filtered. The solid was washed with THF
(1.5 L) and the solvent was removed in vacuo from the filtrate.
The residue was dissolved in DCM (700 mL) and washed with
water (2 × 150 mL). The organic layer was dried over
anhydrous MgSO4 and the solvent removed in vacuo. The
residue was purified by column chromatography (10-20%
EtOAC in toluene) to give 23 g (75%) of product: 1H NMR δ
1.55 (s, 6H, 2 × C(CH3)2), 2.14 (s, 3H, ArCH3), 2.20 (t, J )
6.75, 2H, CH2CH2OH), 2.46 (s, 3H, ArCH3), 3.52 (t, J ) 6.75,
2H, CH2CH2OH), 5.70 (s, 1H, OH), 6.56 (d, J ) 6.75, 1H, ArH),
6.83 (d, J ) 6.75, 1H, ArH); 13C NMR δ 16.22, 22.02, 31.95,
39.78, 44.81, 61.06, 122.23, 125.46, 127.86, 131.82, 135.63,
153.60; EI MS m/z 208 (M+, 50); HRMS calcd for C13H20O2 (M+)
208.1462, found 208.1463.
133.99, 135.92, 138.40, 149.46, 155.79, 171.73. Anal. (C27H47
NO5Si) C, H, N.
-
1-O-(ter t-Bu t yld im et h ylsilyl)-3-(2′-Boc-a la n in yl-3′,6′-
d im eth ylp h en yl)-3,3-d im eth ylp r op a n ol (5Ba ). Prepared
by reaction of 4B and Boc-Ala-OH in 85% yield as described
for 5Aâ: 1H NMR δ 0.03 (s, 6H, 2 × Si(CH3)2), 0.91 (s, 9H,
SiC(CH3)3), 1.49 (s, 9H, OC(CH3)3), 1.54 (s, 6H, 2 × C(CH3)2),
1.61 (t, 2H, J ) 6.75, CH2CH2OSi), 2.14 (m, 4H, CH(CH3)3 &
ArCH3), 2.45 (s, 3H, ArCH3), 3.50 (t, 2H, J ) 6.75, CH2CH2-
OSi), 4.64 (m, 1H, CH(CH3)3), 5.28 (m, 1H, NH), 6.95 (m, 2H,
2 × ArH); 13C NMR δ -5.35, 17.25, 18.16, 18.68, 25.22, 25.88,
28.28, 31.53, 31.88, 39.52, 45.85, 46.10, 49.54, 49.93, 60.70,
79.79, 128.51, 128.88, 131.35, 136.49, 137.52, 148.36, 154.97,
171.70.
1-O-(ter t-Bu tyld im eth ylsilyl)-3-(2′-Boc-â-a la n in yl-3′,6′-
d im et h ylp h en yl)-3,3-d im et h ylp r op a n ol (5Bâ). Prepared
by reaction of 4B and Boc-â-Ala-OH in 90% yield as described
for 5Aâ: 1H NMR δ -0.05 (s, 6H, Si(CH3)2), 0.82 (s, 9H, SiC-
(CH3)3), 1.42 (s, 9H, OC(CH3)3), 1.46 (s, 6H, 2 × C(CH3)2), 2.00
(s, 3H, ArCH3), 2.05 (m, 2H, CH2CH2OSi), 2.50 (s, 3H, ArCH3),
2.78 (m, 2H, NHCH2), 3.50 (m, 4H, CH2CH2OSi & CH2C(d
O)O), 5.13 (br s, 1H, NH), 6.90 (dd, 2H, J ) 13.36 & 7.76, 2 ×
ArH); 13C NMR δ -5.45, 17.11, 18.13, 25.14, 25.84, 28.30,
31.15, 31.82, 34.81, 35.91, 39.35, 46.00, 60.62, 79.22, 128.35,
1-O-(t er t -Bu t yld im e t h ylsilyl)-3-(2′-h yd r oxy-3′,6′-d i-
m eth ylp h en yl)-3,3-d im eth ylp r op a n ol (4B). Compound
3B (23 g, 0.11 mol) and TBDMS-Cl (18.6 g, 0.12 mol) were

Drug Delivery Systems via TML Lactonization
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3 483
135.87, 131.15, 136.21, 137.23, 148.52, 155.72, 171.06. Anal.
(C27H47NO5Si) C, H, N.
pH was adjusted to 2.0 with 1 N HCl, followed by extraction
with ethyl acetate (500 mL). The organic layer was washed
with brine (2 × 200 mL) and dried over anhydrous MgSO4.
The solvent was removed in vacuo and the residue was purified
with silica gel column chromatography (5% MeOH in DCM)
to give the product 8Ap (4.5 g, 83%): 1H NMR δ 1.41, 1.46,
1.50, 1.52, 1.61, 1.97, 2.21, 2.30, 2.33, 2.54, 2.85, 3.56, 4.56,
6.57, 6.79; 13C NMR δ 20.09, 24.13, 25.17, 28.14, 28.56, 29.87,
31.40, 31.57, 39.12, 46.79, 47.40, 59.76, 80.68, 122.56, 132.43,
134.04, 136.27, 138.29, 150.21, 171.87.
3-(2′-Boc-â-a la n in yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth -
ylp r op ion ic Acid (8Aâ). Prepared from 7Aâ in 78% yield as
described for 8Ap : 1H NMR δ 1.44 (s, 9H, t-Bu), 1.57 (s, 6H,
2 × C(CH3)), 2.23 (s, 3H, ArCH3), 2.54 (s, 3H, ArCH3), 2.76 (d,
2H, J ) 5.94, CH2CH2NH), 2.81 (s, 2H, CH2COOH), 3.48 (d,
2H, J ) 5.94, CH2CH2NH), 5.14 (br s, 1H, NH), 6.58 (s, 1H,
ArH), 6.82 (d, 1H, J ) 1.97, ArH); 13C NMR δ 14.03, 20.07,
20.86, 25.10, 28.24, 28.52, 31.22, 35.30, 35.90, 38.50, 38.50,
47.30, 60.27, 79.80, 122.81, 132.38, 133.27, 136.05, 138.00,
149.09, 155.83, 171.07, 171.47, 176.28. Anal. (C21H31NO6‚1/
2EtOAc) C, H, N.
3-(2′-Boc-p r olin yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth yl-
p r op a n ol (6Ap ). A mixture of 5Ap (2.82 g, 5.43 mmol), THF
(10 mL), H2O (10 mL), and HOAc (30 mL) was stirred at room
temperature for 1 h. The solvent was removed to give the
product as a colorless oil (2.2 g, 100%). The product was used
without further purification: 1H NMR δ 1.38, 1.40, 1.42, 1.93,
1.97, 2.01, 2.13, 2.38, 2.43, 3.47, 4.48, 6.50, 6.70; 13C NMR δ
19.95, 20.29, 23.19, 25.06, 25.20, 25.59, 28.44, 31.78, 32.04,
32.27, 33.21, 39.33, 42.96, 46.01, 46.53, 46.62, 59.62, 60.22,
64.95, 66.59, 80.44, 103.87, 122.59, 132.04, 132.27, 134.23,
135.79, 135.89, 138.14, 150.37, 150.55, 154.52, 171.86, 175.10.
3-(2′-Boc-â-a la n in yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth -
ylp r op a n ol (6Aâ). Prepared from 5Aâ in 95% yield as
described for 6Ap : 1H NMR δ 1.45 (s, 9H, t-Bu), 1.48 (s, 6H,
2 × C(CH3)), 2.04 (t, 2H, J ) 7.25, CH2CH2OH), 2.23 (s, 3H,
ArCH3), 2.53 (s, 3H, ArCH3), 2.78 (t, 2H, J ) 5.94, CH2CH2-
NH), 3.51 (m, 4H, CH2OH & CH2CH2NH), 5.13 (bs, 1H, NH),
6.55 (d, 1H, J ) 1.97, ArH), 6.83 (d, 1H, J ) 1.97, ArH).
3-(2′-Boc-a la n in yl-3′,6′-d im eth ylp h en yl)-3,3-d im eth yl-
p r op a n ol (6Ba ). Prepared from 5Ba in 90% yield as described
for 6Ap : 13C NMR δ 17.77, 17.92, 25.48, 28.27, 31.82, 32.00,
32.08, 32.26, 39.59, 45.67, 49.54, 50.08, 60.10, 80.39, 128.54,
128.70, 128.95, 131.43, 136.21, 137.41,137.85, 148.59, 155.04,
155.56, 171.61.
3-(2′-Boc-â-a la n in yl-3′,6′-d im eth ylp h en yl)-3,3-d im eth -
ylp r op a n ol (6Bâ). Prepared from 5Bâ in 90% yield as
described for 6Ap : 13C NMR δ 17.07, 25.20, 25.57, 28.27, 31.99,
34.84, 35.87, 39.37, 45.68, 53.36, 60.13, 79.35, 128.48, 129.01,
131.26, 136.30, 137.08, 148.49, 155.75, 171.52.
3-(2′-Boc-p r olin yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth yl-
p r op a n a l (7Ap ). A solution of 6Ap (2.5 g, 6.2 mmol) in
anhydrous DCM (125 mL) was slowly added to a suspension
of pyridinium chlorochromate (2.88 g, 13.4 mmol) in anhydrous
DCM (125 mL) at room temperature and allowed to stir
overnight at room temperature. The reaction mixture was
concentrated and the residue dissolved in 30 mL of DCM
followed by filtration through a short (∼2 in.) silica gel pad.
The silica gel was flushed several times with ethyl ether. The
filtrate was concentrated in vacuo to give the product as a
viscous oil (2.3 g, 88%). The aldehyde was used in the next
step without further purification.
3-(2′-Boc-â-a la n in yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth -
ylp r op a n a l (7Aâ). Prepared from 6Aâ in 90% yield as
described for 7Ap : 1H NMR δ 1.45 (s, 9H, t-Bu), 1.55 (s, 6H,
2 × C(CH3)), 2.24 (s, 3H, ArCH3), 2.54 (s, 3H, ArCH3), 2.78 (d,
2H, J ) 5.94, CH2CH2NH), 2.81 (d, 2H, J ) 2.30, CH2CHO),
3.50 (d, 2H, J ) 5.94, CH2CH2NH), 5.09 (br s, 1H, NH), 6.61
(s, 1H, ArH), 6.86 (d, 1H, J ) 1.97, ArH), 9.53 (t, 1H, J ) 2.62,
C(dO)H).
3-(2′-Boc-a la n in yl-3′,6′-d im eth ylp h en yl)-3,3-d im eth yl-
p r op a n a l (7Ba ). Prepared from 6Ba in 67% yield as described
for 7Ap : 1H NMR δ 1.45 (s, 9H, t-Bu), 1.61 (m, 9H, 2 × CH3
& CHCH3), 2.04 (s, 3H, ArCH3), 2.54 (s, 3H, ArCH3), 2.88 (ABq,
J ) 101.27 & 16.20, 2H, CH2CHO), 4.59 (m, 1H, CHCH3), 5.13
(m, 1H, NH), 6.97 (q, J ) 5.4, 2H, 2 × ArH), 9.53 (s, 1H, C(d
O)H); 13C NMR δ 17.21, 18.26, 25.36, 28.22, 31.40, 31.87, 38.48,
49.56, 49.98, 56.52, 56.78, 80.02, 129.19, 131.72, 135.89,
136.37, 148.02, 155.10, 171.86, 202.72.
3-(2′-Boc-â-a la n in yl-3′,6′-d im eth ylp h en yl)-3,3-d im eth -
ylp r op a n a l (7Bâ). Prepared from 6Bâ in 99% yield as
described for 7Ap : 13C NMR δ 15.07, 16.97, 285.13, 25.49,
28.18, 31.45, 34.72, 35.78, 38.24, 56.51, 65.61, 79.23, 129.01,
129.24, 131.45, 135.57, 135.79, 148.06, 155.60, 170.86, 202.40.
3-(2′-Boc-p r olin yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth yl-
p r op ion ic Acid (8Ap ). A solution of 80% sodium chlorite (4
g, 0.032 mol) in water (37 mL) was slowly added to a solution
of 7Ap (5.45 g, 0.013 mol) and sodium dihydrogen phosphate,
NaH2PO4 (0.98 g, 0.008 mol), in CH3CN (27 mL) and water
(11 mL) that had been cooled to 0 °C in an ice-salt water bath.
The mixture was stirred for 1 h at 0 °C and then allowed to
reach room temperature. Sodium sulfite (1.73 g, 0.013 mol)
was added to the reaction to decompose HOCl and H2O2. The
3-(2′-Boc-a la n in yl-3′,6′-d im eth ylp h en yl)-3,3-d im eth yl-
p r op ion ic Acid (8Ba ). Prepared from 7Ba in 95% yield as
described for 8Ap . NMR resonances showed multiplicity
presumably because of the presence of two ortho substitu-
ents: 1H NMR δ 1.43 (s, 9H, t-Bu), 1.59 (m, 9H, 2 × CH3
&
CHCH3), 2.01 (s, 3H, ArCH3), 2.53 (s, 3H, ArCH3), 2.84 (ABq,
J ) 101.27 & 16.20, 2H, CH2COOH), 4.59 (m, 1H, CHCH3),
5.20 (m, 1H, NH), 6.92 (q, J ) 5.4, 2H, 2 × ArH); 13C NMR δ
17.21, 18.14, 25.28, 28.27, 28.78, 31.05, 31.56, 39.06, 47.16,
47.74, 50.00, 80.02, 128.84, 131.53, 136.20, 137.17, 148.16,
155.44, 171.79, 175.81; EI MS m/z 416.23 (M+ + Na, 100),
432.21 (M+ + K, 85); HRMS calcd for C21H31NO6Na (M+ + Na)
416.2049, found 416.2047.
3-(2′-Boc-â -a la n in yl-3′,6′-d im eth ylp h en yl)-3,3-d im eth -
ylp r op ion ic Acid (8Bâ). Prepared from 7Bâ in 78% yield as
described for 8Ap : 13C NMR δ 16.95, 24.97, 28.47, 31.13, 34.66,
35.78, 38.77, 47.39, 79.23, 128.54, 128.81, 131.17, 135.76,
136.71, 148.13, 155.80, 170.95, 176.41; EI MS m/z 416.23 (M+
+ Na, 95), 432.19 (M+ + K, 90); HRMS calcd for C21H31NO6-
Na (M+ + Na) 416.2049, found 416.2046.
3-(2′-Ala n in yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth ylp r op i-
on ic Acid (9Aa ). Trifluoroacetic acid (10 mL) was added to a
solution of 8Aa 11 (1.0 g, 2.54 mmol) in DCM (10 mL) and the
mixture was stirred at room temperature for 2 h. Solvent was
removed completely in vacuo followed by addition of ethyl ether
to precipitate the product (as the TFA salt, 0.6873 g, 67%):
1H NMR δ 1.41 (3H, s, C(CH3)2), 1.42 (3H, s, C(CH3)2), 1.54
(3H, d, J ) 8.1, CHCH3), 2.21 (3H, s, ArCH3), 2.45 (3H, s,
ArCH3), 2.65 (2H, s, CH2COOH), 4.36 (1H, d, J ) 8.1, CHCH3),
6.57 (1H, s, ArH), 6.81 (1H, s, ArH); 13C NMR δ 15.27, 19.57,
24.66, 30.92, 38.28, 47.32, 48.45, 122.23, 132.30, 133.97,
135.61, 138.05, 149.10, 169.50, 172.50.
3-(2′-P r olin yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth ylp r op i-
on ic Acid (9Ap ). Prepared from 8Ap in 77% yield as described
for 9Aa : 1H NMR δ 1.53, 1.55, 2.03, 2.19, 2.48, 2.59, 2.76, 3.32,
4.61, 6.54, 6.85, 9.80; 13C NMR δ 13.86, 23.73, 24.86, 28.27,
31.70, 31.92, 38.99, 46.43, 47.76, 60.33, 122.25, 133.08, 133.74,
136.65, 138.89, 149.30, 168.35.
3-(2′-â-Ala n in yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth ylp r o-
p ion ic Acid (9Aâ). Prepared from 8Aâ in 69% yield as
described for 9Aa : 1H NMR (CDCl3 + CD3OD) δ 1.55 (s, 6H,
2 × C(CH3)), 2.21 (s, 3H, ArCH3), 2.53 (s, 3H, ArCH3), 2.78 (s,
2H, CH2CH2NH), 2.85 (s, 2H, CH2COOH), 3.12 (br s, 2H,
CH2CH2NH), 6.55 (s, 1H, ArH), 6.87 (s, 1H, ArH), 10.19 (s,
COOH); 13C NMR (CDCl3 + CD3OD) δ 19.35, 24.44, 30.93,
31.61, 34.88, 38.21, 47.17, 122.26, 132.17, 133.20, 135.83,
137.99, 148.55, 169.91, 174.15.
3-(2′-Ala n in yl-3′,6′-d im eth ylp h en yl)-3,3-d im eth ylp r op i-
on ic Acid (9Ba ). Prepared from 8Ba in 50% yield as described
for 9Aa : 1H NMR (DMSO-d6) δ 1.41 & 1.43 (s, 3H, CH3), 1.56
(s, 3H, CH3), 1.65 & 1.67 (d, J ) 6.59, 3H, CHCH3), 1.99 (s,
3H, ArCH3), 2.51 (s, 3H, ArCH3), 2.70 (m, 2H, CH2COOH), 4.52
(m, 1H, CHCH3), 6.98 (m, 2H, 2 × ArH), 9.65 (br s, 2H, NH2);

484 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3
Greenwald et al.
13C NMR (DMSO-d6) δ 15.05, 16.76, 24.57, 30.90, 38.70, 48.03,
48.58, 114.93, 119.35, 128.17, 128.44, 131.31, 135.94, 137.44,
147.31, 157.98, 158.43, 168.64, 172.37.
3-(2′-â-Ala n in yl-3′,6′-d im eth ylp h en yl)-3,3-d im eth ylp r o-
p ion ic Acid (9Bâ). Prepared from 8Bâ in 65% yield as
described for 9Aa : 13C NMR δ 14.81, 16.59, 24.81, 30.83, 31.83,
35.39, 39.23, 47.07, 65.91, 128.67, 129.12, 131.52, 136.60,
136.76, 147.94, 170.74, 175.81.
3-(2′-P EG-a la n in yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth yl-
p r op ion ic Acid (10Aa ). Diisopropylethylamine (36.6 µL, 0.20
mmol) was added to a solution of T-PEG23 (1 g, 0.025 mmol)
and 9Aa (29 mg, 0.099 mmol) in anhydrous DCM (15 mL) and
the mixture was stirred at room temperature overnight. The
solvent was removed in vacuo and the residue recrystallized
from 2-propanol to give 0.8 g (80%) of product: 13C NMR δ
16.90, 19.48, 24.45, 30.81, 38.34, 46.83, 47.87, 68.52-72.82
(PEG), 76.19, 77.83, 122.00, 131.77, 133.76, 135.40, 137.51,
149.46, 169.28, 170.88, 171.55.
3-(2′-P EG-p r olin yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth yl-
p r op ion ic Acid (10Ap ). Prepared from 9Ap in 85% yield as
described for 10Aa : 13C NMR δ 19.61, 24.39, 27.34, 30.72,
38.11, 45.86, 59.13, 68.40-71.45 (PEG), 122.18, 131.78, 133.40,
135.04,137.18, 149.36, 168.65, 170.67, 171.18.
32.09, 32.11, 34.13, 38.74, 44.16, 44.39, 46.82, 47.71, 47.97,
55.62, 66.06-69.50 (PEG), 75.44, 99.90, 110.12, 110.17, 117.92,
118.53, 119.63, 127.30, 127.59, 130.48, 133.40, 134.21, 134.82,
135.52, 135.93, 136.86, 147.85, 154.59, 155.11, 155.31, 155.62,
160.03, 169.03, 169.18, 170.30, 170.33, 170.94, 185.42, 185.59,
210.77.
Com p ou n d 11Bâ. Prepared from 10Bâ in 90% yield as
described for 11Aa . The amount of DNR present in this
compound as measured by UV assay was 2.5 wt %: 13C NMR
δ 16.13, 16.31, 23.97, 34.82, 28.35, 30.95, 31.61, 32.40, 33.50,
33.79, 34.28, 39.40, 44.35, 48.49, 55.88, 66.34, 67.86-75.73
(PEG), 78.28, 100.12, 110.28, 110.42, 117.93, 118.85, 119.89,
128.23, 128.43, 130.75, 133.50, 133.66, 134.53, 135.05, 135.58,
147.97, 154.92, 155.60, 160.21, 169.32, 169.64, 185.68, 186.00,
211.03.
2-(2-Boc-a m in oeth oxy)eth a n ol (13). This compound was
prepared by a modification of the existing procedure.32
A
solution of di-tert-butyl carbonate (10.27 g, 47.2 mmol) in
chloroform (40 mL) was added to a solution of 2-(2-amino-
ethoxy)ethanol (12; 5.0 g, 47.62 mmol) in chloroform (40 mL)
and the mixture was stirred at room temperature for 1.5 h.
The solution was washed with water (30 mL) and the organic
layer dried over anhydrous MgSO4 and concentrated in vacuo
to give the product (9.6 g, 99%): 1H NMR δ 1.45 (s, 9H, t-Bu),
3.32 (m, 2H, NHCH2), 3.43 (bs, 1H, OH), 3.56 (m, 4H, CH2-
OCH2), 3.73 (t, 2H, J ) 5.4, CH2OH), 5.43 (bs, 1H, NH); 13C
NMR δ 28.21, 40.21, 61.31, 70.08, 72.14, 77.18, 79.06, 156.06.
3-(2′-P EG-â-a la n in yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth -
ylp r op ion ic Acid (10Aâ). Prepared from 9Aâ in 90% yield
as described for 10Aa : 13C NMR δ 19.44, 24.47, 30.70, 33.56,
34.13, 37.90, 46.62, 69.77-70.25 (PEG), 122.17, 131.47, 133.06,
134.92, 137.25, 148.54, 169.29, 170.08, 171.76.
2-(2-Boc-a m in oeth oxy)eth a n ol NHS Ca r bon a te (14). A
mixture of 13 (1.0 g, 4.88 mmol), N,N-disuccinimidyl carbonate
(1.5 g, 5.86 mmol), and anhydrous pyridine (474 mg, 6.0 mmol)
in anhydrous chloroform (25 mL) was stirred at 25-30 °C
overnight. The reaction mixture was washed with 0.5 N HCl
then dried over anhydrous MgSO4. Evaporation of the solvent
gave the product (1.5 g, 96%): 1H NMR δ 1.45 (s, 9H, t-Bu),
2.84 (bs, 4H, NHS), 3.32 (m, 2H, NHCH2), 3.56 (t, 2H, J )
4.9, CH2O), 3.74 (t, 2H, J ) 4.3, OCH2), 4.47 (t, 2H, J ) 4.3,
CH2OC(dO)ONHS), 5.01 (br s, 1H, NH); 13C NMR δ 25.00,
27.92, 39.81, 67.57, 69.59, 69.89, 77.20, 78.56, 151.18, 155.58,
168.60.
3-(2′-P EG-a la n in yl-3′,6′-d im eth ylp h en yl)-3,3-d im eth yl-
p r op ion ic Acid (10Ba ). Prepared from 9Ba in 87% yield as
described for 10Aa : 13C NMR δ 16.25, 16.75, 24.27, 30.12,
30.51, 37.95, 46.84, 47.08, 69.51-70.30 (PEG), 127.61, 130.33,
134.94, 136.54, 147.22, 168.78, 169.64, 171.23.
3-(2′-P EG-â-a la n in yl-3′,6′-d im eth ylp h en yl)-3,3-d im eth -
ylp r op ion ic Acid (10Bâ). Prepared from 9Bâ in 96% yield
as described for 10Aa : 13C NMR δ 16.42, 24.48, 30.70, 33.47,
38.25, 46.65, 68.17-71.90 (PEG), 127.80, 128.15, 130.49,
135.17, 136.51, 147.61, 169.29, 169.68, 171.64.
P EG-TML-DNR. Com p ou n d 11Aa . EDC‚HCl (94.5 mg,
0.492 mmol) was added to a mixture of 10Aa (2.5 g, 0.0615
mmol), DNR‚HCl (208 mg, 0.3688 mmol), NMM (99.4 mg,
0.984 mmol), and HOBT hydrate (49.8 mg, 0.369 mmol) in 50
mL of anhydrous methylene chloride at 0 °C. The reaction
mixture was stirred at room temperature overnight and
filtered, and the filtrate was concentrated in vacuo. The
residue was recrystallized from 2-propanol (200 mL) to give
2.3 g (92%) of product. The amount of DNR present in this
compound as measured by UV assay was 2.4 wt %: 13C NMR
δ 16.14, 19.31, 23.55, 24.63, 28.80, 31.15, 32.78, 34.77, 39.23,
44.97, 48.12, 56.18, 66.80-75.95 (PEG), 100.14, 110.93, 118.68,
119.13, 121.81, 131.88, 133.86, 134.02, 134.96, 135.03, 137.96,
149.56, 155.20, 160.73, 169.73, 170.22, 171.34, 186.05, 210.41.
Com p ou n d 11Ap . Prepared from 10Ap in 84% yield as
described for 11Aa . The amount of DNR present in this
compound as measured by UV assay was 2.5 wt %: 13C NMR
δ 16.03, 19.33, 20.65, 21.57, 23.76, 24.41, 24.86, 26.96, 26.94,
27.83, 28.71, 29.89, 31.04, 31.30, 32.60, 34.38, 39.02, 44.16,
44.63, 45.85, 46.82, 55.73, 58.91, 58.97, 59.72, 66.77, 67.76-
71.43 (PEG), 75.59, 77.90, 99.74, 110.11, 110.28, 117.92,
118.64, 119.46, 119.69, 120.84, 131.25, 133.46, 134.30, 134.93,
137.46, 148.39, 154.72, 155.42, 160.09, 167.99, 169.38, 170.55,
171.13, 185.53, 185.73, 205.25, 210.54.
1-O-(ter t-Bu t yld im et h ylsilyl)-3-(2′-2′′-(2′′-Boc-a m in o-
eth oxy)eth oxyca r bon yloxy-4′,6′-d im eth ylp h en yl)-3,3-d i-
m eth ylp r op a n ol (15A). A mixture of 14 (1.0 g, 3.13 mmol),
4A (503 mg, 1.56 mmol), and DIEA (600 µL, 3.51 mmol) in
chloroform (20 mL) was refluxed overnight. The reaction
mixture was washed with 0.5 N HCl (2 × 20 mL) and dried
over anhydrous MgSO4 and the solvent was removed in vacuo.
The residue was purified by column chromatography (10-30%
ethyl acetate in hexane) to give the product (470 mg, 28%):
1H NMR δ -0.07 (s, 6H, Si(CH3)2), 0.81 (s, 9H, SiC(CH3)3),
1.42 (s, 9H, t-Bu), 1.46 (s, 6H, C(CH3)2), 2.01 (t, 2H, J ) 7.60,
CH2OSi), 2.21 (s, 3H, ArCH3), 2.49 (s, 3H, ArCH3), 3.30 (q,
2H, J ) 5.12, NHCH2), 3.46 (t, 2H, J ) 7.43, CH2CH2OSi),
3.52 (t, 2H, J ) 5.28, CH2CH2O), 3.70 (t, 2H, J ) 4.62, OCH2),
4.35 (t, 2H, J ) 4.78, CH2OC(dO)O), 4.93 (br s, 1H, NH), 6.63
(s, 1H, ArH), 6.79 (s, 1H, ArH); 13C NMR δ -5.43, 18.10, 20.05,
25.03, 25.83, 28.32, 31.64, 39.11, 40.30, 45.83, 60.65, 67.16,
68.57, 70.25, 79.16, 122.44, 132.43, 134.05, 135.98, 138.27,
150.08, 154.05, 155.84. Anal. (C29H51NO7Si) C, H, N.
3-(2′-2′′-(2′′-Boc-a m in oeth oxy)eth oxyca r bon yloxy-4′,6′-
d im et h ylp h en yl)-3,3-d im et h ylp r op a n ol (16A). Prepared
from 15A in 82% yield as described for 6Ap : 1H NMR δ 1.42
(s, 9H, t-Bu), 1.49 (s, 6H, C(CH3)2), 2.05 (t, 2H, J ) 7.43, CH2-
OH), 2.22 (s, 3H, ArCH3), 2.50 (s, 3H, ArCH3), 3.31 (q, 2H, J
) 5.28, NHCH2), 3.52 (m, 4H, CH2CH2OH & CH2O), 3.72 (m,
2H, OCH2), 4.39 (m, 2H, CH2OC(dO)O), 5.12 (br s, 1H, NH),
6.67 (s, 1H, ArH), 6.82 (d, 1H, J ) 1.32, ArH); 13C NMR δ
20.18, 25.26, 28.43, 31.87, 39.31, 40.44, 46.11, 60.48, 67.28,
68.73, 70.41, 79.23, 122.65, 132.75, 133.90, 136.38, 138.42,
150.11, 154.34, 155.99.
Com p ou n d 11Aâ. Prepared from 10Aâ in 90% yield as
described for 11Aa . The amount of DNR present in this
compound as measured by UV assay was 2.5 wt %: 13C NMR
δ 16.10, 19.38, 23.92, 24.73, 28.46, 31.09, 32.46, 33.65, 34.32,
38.97, 44.36, 48.53, 55.90, 66.70-69.77 (PEG), 75.76, 100.15,
110.40, 117.98, 118.87, 122.26, 131.74, 133.62, 134.57, 135.05,
135.60, 137.62, 148.92, 154.94, 155.60, 160.26, 169.39,169.76,
170.72, 185.76, 210.97.
3-(2′-2′′-(2′′-Boc-a m in oeth oxy)eth oxyca r bon yloxy-4′,6′-
d im et h ylp h en yl)-3,3-d im et h ylp r op a n a l (17A). Prepared
from 16A as described for 7Ap in 97% yield: 1H NMR δ 1.40
(s, 9H, t-Bu), 1.54 (s, 6H, C(CH3)2), 2.21 (s, 3H, ArCH3), 2.50
Com p ou n d 11Ba . Prepared from 10Ba in 90% yield as
described for 11Aa . The amount of DNR present in this
compound as measured by UV assay was 2.5 wt %: 13C NMR
δ 15.79, 15.90, 16.13, 23.57, 24.41, 24.79, 28.18, 30.99, 31.21,

Drug Delivery Systems via TML Lactonization
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3 485
(s, 3H, ArCH3), 2.82 (d, 2H, J ) 2.64, CH2CHO), 3.29 (q, 2H,
J ) 5.12, NHCH2), 3.52 (t, 2H, J ) 5.12, CH2O), 3.69 (m, 2H,
OCH2), 4.36 (m, 2H, CH2OC(dO)O), 4.96 (bs, 1H, NH), 6.69
(s, 1H, ArH), 6.82 (d, 1H, J ) 1.32, ArH), 9.50 (t, 1H, J ) 2.48,
CH2CHO); 13C NMR δ 20.11, 25.22, 28.30, 31.48, 40.27, 56.59,
67.37, 68.49, 70.22, 79.18, 133.67, 132.56, 132.73, 136.88,
137.72, 149.68, 153.81, 155.84, 202.57.
3.35-3.56 (m, 13H, 5 × CH2 & CH2CH2OSi & OH), 5.21 (bs,
1H, NH), 5.91 (bs, 1H, NH), 6.57 (s, 1H, ArH), 6.71 (s, 1H,
ArH); 13C NMR δ 20.04, 25.19, 28.27, 29.08, 31.80, 39.05, 40.21,
40.93, 45.92, 53.32, 60.15, 69.91, 70.11, 79.15, 123.22, 131.92,
134.28, 135.78, 137.98, 149.79, 155.54, 155.98.
3-[2′-(2′′-Am in o-2-Boc-a m in oet h ylen e glycol d iet h yl
eth er ca r ba m a te)-4′,6′-d im eth ylp h en yl]-3,3-d im eth ylp r o-
p a n a l (24A). Prepared from 23A in 68% yield as described
for 7Ap : 1H NMR δ 1.41 (s, 9H, OC(CH3)3), 1.54 (s, 6H, 2 ×
C(CH3)2), 2.20 (s, 3H, ArCH3), 2.50 (s, 3H, ArCH3), 2.77 (br s,
2H, CH2CHO), 3.29-3.62 (m, 12H, 6 × CH2), 5.07 (br s, 1H,
NH), 5.66 (br s, 1H, NH), 6.64 (s, 1H, ArH), 6.79 (s, 1H, ArH),
9.53 (br s, 1H, CHO); 13C NMR δ 21.12, 25.20, 28.30, 31.44,
38.05, 40.24, 41.05, 53.35, 56.71, 69.90, 70.12, 70.28, 79.18,
123.40, 132.20, 132.98, 136.54, 137.48, 149.44, 154.76, 155.90,
203.36.
3-[2′-(2′′-Am in o-2-Boc-a m in oet h ylen e glycol d iet h yl
eth er ca r ba m a te)-4′,6′-d im eth ylp h en yl]-3,3-d im eth ylp r o-
p ion ic Acid (25A). Prepared from 24A in 80% yield as
described for 8Ap : 1H NMR δ 1.43 (s, 9H, OC(CH3)3), 1.58 (s,
6H, 2 × C(CH3)2), 2.22 (s, 3H, ArCH3), 2.52 (s, 3H, ArCH3),
2.79 (br s, 2H, CH2COOH), 3.30-3.63 (m, 12H, 6 × CH2), 5.28
(br s, 1H, NH), 6.02 (br s, 1H, NH), 6.68 (s, 1H, ArH), 6.78 (s,
1H, ArH); 13C NMR δ 20.10, 25.08, 28.26, 31.15, 38.60, 40.09,
40.93, 47.92, 69.83, 70.11, 79.29, 122.93, 131.85, 133.85,
135.89, 137.52, 149.47, 155.05, 156.07, 175.09; CI MS m/z 497
(M+ + H, 20); HRMS calcd for C25H41N2O8 (M+ + H) 497.2863,
found 497.2865.
3-[2′-(2′′-Am in o-2-a m in oeth ylen e glycol d ieth yl eth er
ca r ba m a te)-4′,6′-d im eth ylp h en yl]-3,3-d im eth ylp r op ion -
ic Acid (26A). Prepared from 25A in 99% yield as described
for 9Aa : 1H NMR δ 1.52 (s, 6H, 2 × C(CH3)2), 2.18 (s, 3H,
ArCH3), 2.50 (s, 3H, ArCH3), 2.79 (br s, 2H, CH2COOH), 2.91
(br s, 2H, CH2NH2), 3.36 (br s, 1H, NH), 3.56 (br s, 10H, 5 ×
CH2), 6.32 (br s, 1H, NH), 6.59 (s, 1H, ArH), 6.77 (s, 1H, ArH);
13C NMR δ 19.90, 24.99, 31.27, 31.67, 38.66, 39.55, 40.79,
47.98, 66.23, 69.89, 122.91, 131.94, 133.99, 135.89, 138.03,
149.41, 155.76, 160.90, 175.60.
3-(2′-2′′-(2′′-Boc-a m in oeth oxy)eth oxyca r bon yloxy-4′,6′-
d im eth ylp h en yl)-3,3-d im eth ylp r op ion ic Acid (18A). Pre-
pared from 17A as described for 8Ap in 97% yield: 1H NMR
δ 1.42 (s, 9H, t-Bu), 1.61 (s, 6H, C(CH3)2), 2.24 (s, 3H, ArCH3),
2.55 (s, 3H, ArCH3), 2.86 (s, 2H, CH2COOH), 3.31 (bs, 2H,
NHCH2), 3.56 (m, 2H, CH2O), 3.74 (m, 2H, OCH2), 4.40 (m,
2H, CH2OC(dO)O), 5.06 (bs, 1H, NH), 6.72 (s, 1H, ArH), 6.83
(s, 1H, ArH); EI MS m/z 476.22 (M+ + Na, 95), 492.21 (M+
+
K, 90); HRMS calcd for C25H40N2O8Na (M+ + Na) 476.2260,
found 476.2278.
3-(2′-2′′-(2′′-Am in oet h oxy)et h oxyca r b on yloxy-4′,6′-d i-
m eth ylp h en yl)-3,3-d im eth ylp r op ion ic Acid (19A). Pre-
pared from 18A in 99% yield as described for 9Aa : 1H NMR
δ 1.59 (s, 6H, C(CH3)2), 2.24 (s, 3H, ArCH3), 2.55 (s, 3H,
ArCH3), 2.86 (s, 2H, CH2COOH), 3.09 (s, 2H, NHCH2), 3.68
(s, 2H, CH2O), 3.78 (s, 2H, OCH2), 4.41 (s, 2H, CH2OC(dO)O),
5.85 (br s, 1H, NH2), 6.67 (s, 1H, ArH), 6.85 (s, 1H, ArH); 13C
NMR δ 20.05, 24.93, 31.51, 38.77, 39.61, 47.43, 66.18, 66.87,
69.10, 122.44, 132.57, 133.47, 136.22, 138.21, 149.71, 154.22,
169.75, 175.20.
3-(2′-2′′-(2′′-P EG-a m in oeth oxy)eth oxyca r bon yloxy-4′,6′-
d im eth ylp h en yl)-3,3-d im eth ylp r op ion ic Acid (20A). Pre-
pared from 19A in 92% yield as described for 10Aa : 13C NMR
δ 19.35, 24.36, 30.54, 37.66, 37.98, 46.40, 65.67, 67.68-70.61
(PEG), 121.55, 131.53, 133.01, 134.99, 137.17, 149.06, 152.91,
169.23, 171.84.
P EG-ca r bon a te-TML-DNR (21A). Prepared from 20A in
90% yield as described for 11Aa . The amount of DNR present
in this compound as measured by UV assay was 2.4 wt %: 13C
NMR δ 16.04, 19.28, 23.89, 24.59, 28.38, 30.93, 32.32, 34.23,
37.63, 38.91, 44.47, 55.82, 66.32, 66.84, 67.71-71.74 (PEG),
75.66, 78.33, 100.06, 110.20, 110.36, 117.90, 118.77, 119.81,
121.68, 131.85, 132.82, 133.46, 134.44, 135.00, 135.64, 137.49,
149.20, 153.46, 154.84, 155.52, 160.15, 169.38, 185.62, 185.91,
210.92.
3-[2′-(2′′-P EG-a m in o-2-a m in oeth ylen e glycol d ieth yl
eth er ca r ba m a te)-4′,6′-d im eth ylp h en yl]-3,3-d im eth ylp r o-
p ion ic Acid (27A). Prepared from 26A in 97% yield as
described for 10Aa : 13C NMR δ 19.21, 24.27, 31.35, 37.53,
37.75, 39.98, 46.95, 69.51-70.27 (PEG), 122.04, 130.63, 133.35,
134.39, 136.53, 148.81, 154.08, 169.13, 172.13.
1-O-(ter t-Bu t yld im e t h ylsilyl)-3-[2′-(2′′-a m in o-2-Boc-
a m in oeth ylen e glycol d ieth yl eth er ca r ba m a te)-4′,6′-
d im eth ylp h en yl]-3,3-d im eth ylp r op a n ol (22A). Triphos-
gene (308.2 mg, 1.04 mmol) and pyridine (307.1 mg, 3.88
mmol) were added to a solution of 4A (500 mg, 1.55 mmol) in
chloroform (30 mL) and the mixture stirred at 35-40 °C for 3
h followed by cooling to room temperature. 2-Amino-2-[(tert-
butoxycarbonyl)amino]ethylene glycol diethyl ether24 (967.7
mg, 3.88 mmol) and pyridine (307.1 mg, 3.88 mmol) were
added to the reaction solution and the mixture stirred at 35-
40 °C overnight. The reaction solution was washed with 0.5
N HCl (3 × 10 mL) and water (10 mL) and dried over
anhydrous MgSO4, followed by removal of the solvent in vacuo.
The residue was purified by silica gel column chromatography
(30-40% EtOAc in hexane) to give product (720 mg, 78%): 1H
NMR δ -0.01 (s, 6H, 2 × Si(CH3)2), 0.86 (s, 9H, SiC(CH3)3),
1.45 (s, 9H, OC(CH3)3), 1.49 (s, 6H, 2 × C(CH3)2), 2.05 (t, J )
7.43, 2H, CH2CH2OSi), 2.23 (s, 3H, ArCH3), 2.52 (s, 3H,
ArCH3), 3.34 (m, 2H, CH2NHC(dO)O), 3.45-3.65 (m, 12H, 5
× CH2 & CH2CH2OSi), 5.10 (br s, 1H, NH), 5.62 (bs, 1H, NH),
6.65 (s, 1H, ArH), 6.78 (s, 1H, ArH); 13C NMR δ -5.40, 18.11,
20.08, 25.13, 25.84, 28.30, 31.68, 39.00, 40.24, 41.00, 46.04,-
60.80, 70.14, 79.15, 123.20, 131.78, 134.36, 135.61, 137.89,
149.85, 155.03, 155.86; CI MS m/z 597 (M+ + H, 80); HRMS
calcd for C31H57N2O7Si (M+ + H) 597.3935, found 597.3933.
P EG-ca r ba m a te-TML-DNR (28A). Prepared from 27A
in 88% yield as described for 11Aa . The amount of DNR
present in this compound as measured by UV assay was 2.4
wt %: 13C NMR δ 16.02, 19.21, 23.80, 24.67, 28.06, 31.28,
31.45, 32.25, 34.20, 37.59, 39.40, 40.22, 44.42, 48.61, 55.76,
66.23, 67.50, 67.76, 69.62-70.46 (PEG), 75.59, 100.00, 110.11,
110.28, 117.92, 118.70, 119.75, 122.30, 131.07, 132.77, 133.55,
134.37, 134.95, 135.52, 137.34, 149.61, 154.77, 155.17, 155.47,
160.10, 169.19, 169.76, 185.56, 185.82, 210.86.
8Aa -Ar a -C (29Aa ). A mixture of 8Aa (700 mg, 1.78 mmol),
ara-C (1.73 g, 7.12 mmol), HOBT (0.96 g, 7.12 mmol), and
EDC‚HCl (2.73 g, 14.25 mmol) in anhydrous pyridine (50 mL)
was stirred at room temperature for 2 h followed by stirring
at 40 °C overnight. The solvent was removed and DCM (50
mL) was used to dissolve the residue followed by water wash
(3 × 30 mL) and 0.1 N HCl wash (2 × 30 mL). The organic
layer was dried over anhydrous MgSO4 and the solvent was
removed in vacuo to give the crude product which purified by
silica gel column chromatography (5 to 10%, v/v, MeOH in
DCM) to give 638.8 mg (52%) of product as a white solid: 1H
NMR δ 1.42, 1.55, 2.17, 2.26, 2.46, 2.79, 3.84, 3.91, 4.14, 4.33,
4.53, 5.49, 6.07, 6.17, 6.52, 6.76, 7.31, 7.67, 8.16, 8.62; 13C NMR
δ 17.77, 20.11, 25.36, 28.32, 31.51, 31.96, 39.57, 50.18, 50.45,
61.88, 74.50, 80.15, 85.90, 88.58, 96.25, 122.51, 132.82, 133.34,
136.73, 138.22, 146.57, 149.90, 155.65, 155.96, 162.08, 171.89,
174.06.
3-[2′-(2′′-Am in o-2-Boc-a m in oet h ylen e glycol d iet h yl
eth er ca r ba m a te)-4′,6′-d im eth ylp h en yl]-3,3-d im eth ylp r o-
p a n ol (23A). Prepared in 99% yield from 22A as described
for 6Ap : 1H NMR δ 1.37 (s, 9H, OC(CH3)3), 1.41 (s, 6H, 2 ×
C(CH3)2), 2.05 (t, J ) 6.77, 2H, CH2CH2OH), 2.15 (s, 3H,
ArCH3), 2.43 (s, 3H, ArCH3), 3.25 (m, 2H, CH2NHC(dO)O),
9Aa -Ar a -C (30Aa ). Prepared from 32Aa in 82% yield as
described for 9Aa : 1H NMR (DMSO-d6) δ 1.52 (s, 3H, (CH3)2-
CH) 1.55 (s, 3H, (CH3)2CH), 1.62 (d, 1 H, J ) 8.1, (CH3)2CH),
2.22 (s, 3H, ArCH3), 2.57 (s, 3H, ArCH3), 2.97 (s, 2H, CH2C-

486 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3
Greenwald et al.
(dO)), 3.41-4.27 (m, 5 H, ara-C′s H-2′-H-5′), 6.09 (d, 1H, J )
5.4, ara-C′s H-1′), 6.67 (s, 1H, ArH), 6.90 (s, 1H, ArH), 7.12
(d, J ) 5.4, ara-C’s H-6), 8.05 (d, J ) 8.1, ara-C′s H-5), 8.67
(br s, 1H, TFA); 13C NMR (DMSO-d6) δ 15.45, 19.67, 24.97,
31.05, 31.23, 38.56, 40.41, 48.53, 49.02, 61.02, 64.94, 74.64,
76.14, 85.74, 86.95, 94.32, 122.32, 132.41, 134.08, 135.67,
138.09, 146.71, 149.20, 154.50, 158.21, 158.72, 162.02, 169.68,
171.87.
P EG-TML-Ar a -C (31Aa ). A solution of T-PEG (778 mg,
0.019 mmol), 30 (40 mg, 0.077 mmol), and DIEA (20 mg, 0.15
mmol) in anhydrous DCM (10 mL) was stirred at room
temperature overnight. The solvent was removed in vacuo and
the residue crystallized from 2-propanol to give 650 mg (84%
yield) of the product 31Aa as a white solid. The amount of
ara-C present in this compound as measured by UV assay was
1.0 wt %: 13C NMR δ 14.87, 18.58, 24.54, 31.30, 31.36, 38.66,
47.68, 49.09, 61.12, 67.54-71.05 (PEG), 74.76, 85.19, 86.54,
94.69, 121.25, 131.85, 133.08, 134.73, 137.35, 144.94, 148.58,
154.39, 160.88, 170.96, 171.26, 172.38.
Com p ou n d 32. Phenol was converted to carbamate 32 in
87% yield using the same procedures as described for 28A:
13C NMR δ 37.75, 40.22, 60.68, 67.40-69.40 (PEG), 78.28,
120.78, 124.24, 128.37, 148.80, 150.54, 169.15.
P EG-TML-Isop r op yla m in e (33Aa ). Prepared from 10Aa
and isopropylamine in 88% yield as described for 11Aa : 13C
NMR δ 16.00, 19.50, 21.44, 21.54, 25.00, 31.32, 31.60, 39.33,
39.83, 47.94, 48.10, 67.48-71.55 (PEG), 77.92, 120.80, 131.85,
132.0, 133.6, 135.7, 138.2, 149.5, 169.34, 169.91, 171.56.
2. Biologica l R esu lt s. Cell Lin es a n d Cyt ot oxicit y
Assa ys. Studies using P388/0 cell lines for IC50 (drug concen-
tration inhibiting growth of cells by 50%) were maintained and
conducted as previously reported.22 Briefly, for IC50 determi-
nation, cells were seeded into the microwell plates at a density
of 2 × 103 cells/50 µL/well. Plates were incubated at 37 °C in
a humidified incubator with 5% CO2 for 3 days. Cell growth
was measured by the addition of 10 µL/well of Alamar Blue
(Alamar Biosciences, Inc., Sacramento, CA) and the plates
were incubated a further 4 h at 37 °C. The IC50 values for each
compound were determined from absorbance vs dilution factor
plots.
All cell cultures for animal implantation were maintained
at 37 °C in a humidified atmosphere of 5% CO2/95% O2 and
subcultured once a week. All cell lines were periodically tested
for Mycoplasma and were Mycoplasma free. M109 (Madison
109 murine lung carcinoma, National Cancer Institute (NCI),
Bethesda, MD) was adapted to cell culture and grown in
EMEM (Eagle’s modified essential medium) with 10% fetal
bovine serum (FBS). SKOV3 (human, ovarian adenocarcinoma,
ATCC/HTB77) was raised in McCoy’s 5a medium supple-
mented with 15% FBS. PC-3 (human, prostate adenocarci-
noma, ATCC/CRL1435) was maintained in Ham’s F12K
medium with 7% FBS. MX-1 (human, mammary carcinoma)
was obtained from the NCI and maintained in serial subcu-
taneous passages in nude mice.
33Aa , ICR mice (Harlan Sprague-Dawley, Madison, WI) were
intravenously treated (25 mg/mouse) with this simple PEG
linker conjugate. Mice (4/group) were observed daily and
weighed twice weekly for 2 weeks. All animals received
humane care in compliance with the Principles of Laboratory
Animal Care formulated by the National Society of Medical
Research and the Guide for the Care and Use of Laboratory
Animals published by the NIH. These experimental protocols
were approved by the Institutional Animal Care and Use
Committee of UMDNJ -Robert Wood J ohnson Medical School.
Sta tistics. The differences between treatment groups were
assessed by one-way ANOVA. Multiple comparisons, when
significant differences existed, were determined by least
significant differences techniques. Statistical analysis was
conducted using the StatView software program (Abacus
Concepts, Inc., Berkeley, CA).
Ack n ow led gm en t. We express our appreciation for
the analytical efforts of Shuiyun Guang, Michelle Boro,
Tao Wu, Dr. Prakash Sai, and Dr. Chyi Lee; the in vivo
evaluations of Zuliang Yao, Virna Guillen, and J enny
Hsu of the pharm/tox group; Dr. J eff McGuire for his
support and encouragement; and Carol Aitken for her
administrative assistance.
Refer en ces
(1) (a) Milstein, S.; Cohen, L. A. Stereopopulation Control. I. Rate
Enhancement In The Lactonizations Of o-Hydroxyhydrocin-
namic Acids. J . Am. Chem. Soc. 1972, 94, 9158-9165. (b)
Borchardt, R. T.; Cohen, L. A. Stereopopulation Control. II. Rate
Enhancement Of Intramolecular Nucleophilic Displacement. J .
Am. Chem. Soc. 1972, 94, 9175-9182.
(2) (a) Danforth, C.; Nicholson, A. W.; J ames, J . C.; Loudon, G. M.
Steric Acceleration Of Lactonization Reactions: An Analysis Of
“Stereopopulation Control”. J . Am. Chem. Soc. 1976, 98, 4275-
4280. (b) Winans, R. E.; Wilcox, C. F. J r. Comparison Of
Stereopopulation Control With Conventional Steric Effects In
Lactonization Of Hydrocoumarinic Acids. J . Am. Chem. Soc.
1976, 98, 4281-4258.
(3) (a) Wakselman, M. 1,4- And 1,6-Eliminations From Hydroxy And
Amino-Substituted Benzyl Systems: Chemical And Biochemical
Applications. Nouveau J . Chem. 1983, 7, 439-447. (b) Waksel-
man, M.; Guibe-J ampel, E. J . An Alkali-Labile Benzyloxycar-
bonyl Amino-Protecting Group. J . Chem. Soc. Chem. Commun.
1973, 593-594. (c) Le Corre, G.; Guibe-J ampel, E.; Wakselman,
M. Groupements Protecteurs Benzyloxycarbonyles Substitues
Alcalinolabiles. Tetrahedron 1978, 34, 3105-3112. (d) Guibe-
J ampel, E.; Wakselman, M. Selective Cleavage Of p-Nitrobenzyl
Esters With Sodium Dithionite. Synth. Commun. 1982, 12, 219-
223.
(4) Bundgaard, H. The Double Prodrug Concept And Its Applica-
tions. Adv. Drug Delivery Rev. 1989, 3, 39-65.
(5) Carl, P. L.; Chakravarty, P. K.; Katzenellenbogen J . A. A Novel
Connector Linkage Applicable In Prodrug Design. J . Med. Chem.
1981, 24, 479-480.
(6) Denny, W. A.; Wilson, W. R. The Design Of Selectively-Activated
Anti-Cancer Prodrugs For Use In Antibody-Directed And Gene-
Directed Enzyme-Prodrug Therapies. J . Pharm. Pharmacol.
1998, 50, 387-394.
In Vivo An titu m or Activity a n d P EG Lin k er Toxicity.
Primary screening for antitumor activity against M109 tumors
was conducted in 20-25 g, female balb/ c mice (Hilltop Labs,
Scottdale, PA). Mice were implanted (day 0) subcutaneously
at the left flank with approximately 1 × 106 cells in 100 mL.
Intraperitoneal treatments were administered on days 1 and
4, while intravenous dosages were given on days 3 and 6.
Treatment groups (6/group) consisted of untreated controls,
DNR, and seven PEG-DNR derivatives. For solid tumor
xenograft studies, female nu/ nu mice (Harlan Sprague-
Dawley, Madison, WI), 18-24 g, were inoculated subcutane-
ously at the left flank with either a fragment of donor tumor
tissue (MX-1) or a tumor cell suspension (1 × 106 cells) in 0.1
mL of PBS (SKOV3 and PC-3). Treatments began when the
mean tumor volume reached approximately 75 mm3 (day 1).
The xenograft panel compared the efficacy of free DNR and
PEG-TML-DNR (11Aa ) administered intravenously at 3 mg/
kg/dose (6/group) on days 1, 5, and 9 with control (untreated).
Antitumor activity for all studies was calculated and deter-
mined as previously described.17 To determine the toxicity of
(7) Amsberry, K. L.; Gerstenberger, A. E.; Borchardt, R. T. Amine
Prodrugs Which Utilize Hydroxy Amide Lactonization II. A
Potential Esterase-Sensitive Amide Prodrug. Pharm. Res. 1991,
8, 455-461.
(8) Carpino, L. A.; Tiolo, S. A.; Berglund, R. A Reductive Lacton-
ization Of Strategically Methylated Quinone Propionic Acid
Esters And Amides. J . Org. Chem. 1989, 54, 3303-3310.
(9) Amsberry, K. L.; Borchardt, R. T. Amine Prodrugs Which Utilize
Hydroxy Amide Lactonization. I. A Potential Redox-Sensitive
Amide Prodrug. Pharm. Res. 1991, 8, 323-330.
(10) Amsberry, K. L.; Borchardt, R. T. The Lactonization of 2′-
Hydroxyhydrocinnamic Acid Amides: A Potential Prodrug For
Amines. J . Org. Chem. 1990, 55, 5867-5877.
(11) Wang, B.; Gangwar, S.; Pauletti, G. M.; Siahaan, T. J .; Bor-
chardt, R. T. Synthesis Of A Novel Esterase-Sensitive Cyclic
Prodrug System For Peptides That Utilizes A “Trimethyl Lock”-
Facilitated Lactonization Reaction. J . Org. Chem. 1997, 62,
1363-1367.
(12) (a) Wang, B.; Zhang, H.; Wang, W. Chemical Feasibility Studies
Of A Potential Coumarin-Based Prodrug System. Bioorg. Med.
Chem. Lett. 1996, 6, 945-950. (b) Wang, B.; Wang, W.; Zhang,
H.; Shan, D.; Smith, T. D. Coumarin-Based Prodrugs. Part 2.
Synthesis and Bioreversibility Studies of an Esterase Cyclic

Drug Delivery Systems via TML Lactonization
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 3 487
Prodrug of DADLE, an Opioid Peptide. Bioorg. Med. Chem. Lett.
1996, 6, 2823-2826. (c) Wang, B.; Zhang, H.; Zheng, A.; Wang,
W. Coumarin-based Prodrugs. Part 3: Structural Effects on the
Release Kinetics of Esterase-sensitive Prodrugs of Amines.
Bioorg. Med. Chem. 1998, 6, 417-426.
(23) Greenwald, R. B.; Pendri, A.; Martinez, M.; Gilbert, C.; Bradley,
P. PEG Thiazolidine-2-Thione, A Novel Reagent For Facile
Protein Modification Of Bovine Hemoglobin. Bioconjugate Chem.
1996, 7, 638-641.
(24) Zuckermann, R. N.; Martin, E. J .; Spellmeyer, D. C.; Staruber,
G. B.; Shoemaker, K. R.; Kerr, J . M.; Figliozzi, G. M.; Goff, D.
A.; Siani, M. A.; Simon, R. J .; Banville, S. C.; Brown, E. G.;
Wang, L.; Richter, L. S.; Moos, W. H. Discovery of Nanomolar
Ligands For 7-Transmembrane G-Protein-Coupled Receptors
From A Diverse N-(Substituted) glycine Peptoid Library. J . Med.
Chem. 1994, 37, 2678-2685.
(13) (a) Shan, D.; Nicholaou, M. G.; Borchardt, R. T.; Wang, B.
Prodrug Strategies Based On Intramolecular Cyclization Reac-
tions. J . Pharm. Sci. 1997, 86, 765-767. (b) Testa, B.; Mayer,
J . M. Design of Intramolecularly Activated Prodrugs. Drug
Metab. Rev. 1998, 30, 787-807. (c) Wang, W.; J iang, J .; Ballard,
C. E.; Wang, B. Prodrug Approaches to the Improved Delivery
of Peptide Prodrugs. Curr. Pharm. Des. 1999, 5, 265-287.
(14) Ueda, Y.; Matiskella, J . D.; Mikkilineni, A. B.; Farina, V.; Knipe,
J . O.; Rose, W. C.; Casazza, A. M.; Vyas, D. M. Novel, Water-
Soluble Phosphate Derivatives Of 2′-Ethoxycarbonylpaclitaxel
As Potential Prodrugs Of Paclitaxel: Synthesis And Antitumor
Evaluation. Bioorg. Med. Chem. Lett. 1995, 5, 247-252.
(15) Dillon, M. P.; Cai, H.; Maag, H. Applications Of The “Trimethyl
Lock” To Ganciclovir. A Pro-Prodrug With Increased Oral
Bioavailability. Bioorg. Med. Chem. Lett. 1996, 6, 1653-1656.
(16) Maeda, H. SMANCS and polymer-conjugated macromolecular
drugs: advantages in cancer chemotherapy. Adv. Drug Delivery
Rev. 1991, 181-202.
(17) Greenwald, R. B.; Pendri, A.; Conover, C. D.; Zhao, H.; Choe, Y.
H.; Martinez, A.; Shum, K.; Guan, S. Drug Delivery Systems
Employing 1,4- or 1,6-Elimination: Poly(ethylene glycol) Pro-
drugs of Amine-Containing Compounds. J . Med. Chem. 1999,
42, 3657-3667.
(18) A full description of this and other additional novel work on PEG
amino prodrugs will be discussed in forthcoming reports.
(19) Chakravarty, P. K.; Carl, P. L.; Weber, M. J .; Katzenellenbogen,
J . A. Plasmin-Activated Prodrugs for Cancer Chemotherapy 2.
Synthesis and Biological Activity of Peptidyl Derivatives of
Doxorubicin. J . Med. Chem. 1983, 26, 638-644 and references
therein.
(25) Unpublished findings.
(26) (a) Christenson, I. Alkaline Hydrolysis of Some Carbamic Acid
Esters. Acta Chem. Scand. 1964, 18, 904-922. (b) Bender, M.
L.; Homer, R. B. The Mechanism of the Alkaline Hydrolysis of
p-Nitrophenyl N-Methylcarbamate. J . Org. Chem. 1965, 30,
3975-3978.
(27) Rose, W. C. Evaluation of Madison 109 Lung Carcinoma as a
Model for Screening Antitumor Drugs. Cancer Treat. Rep. 1981,
65, 299-312.
(28) Baurain, R.; Masquelier, M.; Deprez-De Campeneere, D.; Trouet,
A. Amino acid and dipeptide derivatives of daunorubicin. 2.
Cellular pharmacology and antitumor activity on l1210 leukemic
cells in vitro and in vivo. J . Med. Chem. 1980, 23, 1171-1174.
(29) Balimane, P. V.; Tamai, I.; Guo, A.; Nakanishi, T.; Kitada, H.,
Leibach, F. H.; Tsuji, A.; Sinko, P. J . Direct evidence for peptide
transporter (PepT1)-mediated uptake of a nonpeptide prodrug,
valacyclovir. Biochem. Biophys. Res. Commun. 1998, 250, 246-
251.
(30) Han, H.; Oh, D.; Amidon, G. L. Cellular uptake mechanism of
amino acid ester prodrugs in Caco-2/hPEPT1 cells overexpress-
ing a human peptide transporter. Pharm. Res. 1998, 15, 1382-
1386.
(31) This method was adapted from an earlier published procedure:
Greenwald, R. B.; Gilbert, C. W.; Pendri, A.; Conover, C. D.; Xia,
J .; Martinez, A. Drug Delivery Systems: Water Soluble Taxol
2′-Poly(ethylene glycol) Ester Prodrugs-Design and In Vivo
Effectiveness. J . Med. Chem. 1996, 39, 424-431.
(32) Lebeau, L.; Oudet, P.; Mioskowski, C. Synthesis of New Phos-
pholipids Linked to Steroid-Hormone Derivatives Designed for
Two-Dimensional Crystallization of Proteins. Helv. Chim. Acta
1991, 74, 1697-1706.
(20) (a) Ichikawa, H.; Onishi, H.; Takahata, T.; Machida, Y.; Nagai,
T. Evaluation of the conjugate between N4-(4-carboxybutyryl)-
1-â-D-arabinofuranosylcytosine and chitosan as a macromolecu-
lar prodrug of 1-â-D-arabinofuranosylcytosine. Drug Des. Dis-
covery 1993, 10, 343-353. (b) Wipf, P.; Li, W.; Adeyeye, C. M.;
Rusnak, J . M.; Lazo, J . S. Synthesis of Chemoreversible Pro-
drugs of ara-C with Variable Time-Release Profiles. Biological
Evaluation of Their Apoptotic Activity. Bioorg. Med. Chem. 1996,
4, 1585-1596.
(21) Decanale, E.; Montanari, F. Selective Oxidations of Aldehydes
to Carboxylic Acids with Sodium Chlorite-Hydrogen Peroxide.
J . Org. Chem. 1986, 51, 567-569.
(22) Greenwald, R. B.; Pendri, A.; Conover, C. D.; Lee, C.; Choe, Y.
H.; Gilbert, C.; Martinez, M.; Xia, J .; Wu, D.; Hsue, M. Camp-
tothecin-20-PEG Ester Transport Forms: the Effect of Spacer
Groups on Antitumor Activity. Bioorg. Med. Chem. 1998, 6, 551-
562.
J M990498J