Selective phenolic acylation of 10-hydroxycamptothecin using poly (ethylene glycol) carboxylic acid

Article abstract of DOI:10.1016/S0960-894X(02)00926-5

Selective acylation of the phenolic hydroxyl group of 10-hydroxycamptothecin has been accomplished using phenyl dichlorophosphate. Additional modification of the 10-OH as an ether permits a 20-acyl derivative to be synthesized. This result along with data from a 6-hydroxyquinoline model strongly suggests that powerful intermolecular hydrogen bonding exists in the parent molecule.

Full text of DOI:10.1016/S0960-894X(02)00926-5

Bioorganic & Medicinal Chemistry Letters 13 (2003) 577–580
Selective Phenolic Acylation of 10-Hydroxycamptothecin Using
Poly (Ethylene Glycol) Carboxylic Acid
Richard B. Greenwald,* Yun H. Choe and Dechun Wu
Enzon, Inc., 20 Kingsbridge Road, Piscataway, NJ 08854, USA
Received 21 March 2002; accepted 22 August 2002
Abstract—Selective acylation of the phenolic hydroxyl group of 10-hydroxycamptothecin has been accomplished using phenyl
dichlorophosphate. Additional modification of the 10-OH as an ether permits a 20-acyl derivative to be synthesized. This result
along with data from a 6-hydroxyquinoline model strongly suggests that powerful intermolecular hydrogen bonding exists in the
parent molecule.
# 2002 Elsevier Science Ltd. All rights reserved.
10-Hydroxycamptothecin (10-HCPT)1 is a naturally-
occurring A-ring oxygenated derivative of camptothecin
and a minor component found in extracts of Camp-
totheca acuminata. 10-HCPT (1) has been reported to be
effective against certain human cancers, and has been
shown to be more active and less toxic than other
camptotheca alkaloids.2 Potent synthetic water soluble
10-HCPT analogues, CPT-11,3 a piperazine carbamate
derivative of 10-hydroxy-7-ethyl-camptothecin, and
topotecan,4 the 9-dimethylaminomethyl derivative of
10-HCPT, have recently been approved for clinical use.
Previously, we reported the preparation of a poly (eth-
ylene glycol) (PEG) ester prodrug of camptothecin
(CPT) by acylating the 20-OH moiety.5 The resulting
water soluble polymeric prodrugs (transport forms)
showed increased efficacy in a P388/0 mouse model, and
HPLC evidence5 strongly implied that these 20-acyl
derivatives are locked into the highly active lactone
form. More recent chemical, NMR and UV studies have
further validated this conclusion.6 The intriguing possi-
bility of preparing a 10-PEG ester transport form in
which the 20-OH group, necessary for activity, has not
been disabled, prompted us to explore the acylation of
this bifunctional alkaloid. Although 10-HCPT contains
two hydroxyls, one aromatic and one aliphatic, acyla-
tion of the phenolic moiety should be straightforward
since the aliphatic 20-hydroxyl is a hindered 3ꢀ alcohol
and accordingly, less reactive. Curiously, only a low
yield synthesis of CPT-11 and analogues resulted from
the direct acylation of 10-HCPT with N-piperazinyl
carbamoyl chloride.3 Perhaps that is why Takayama et
al.7 synthesized a series of potent anticancer long-chain
fatty acid esters of 10-HCPT by a lengthy construction
of the entire molecule and not by direct acylation. We
now wish to report a direct, efficient, and high yield
preparation of the PEG ester of 10-HCPT that we
believe will prove to be a general procedure for similar
bifunctional molecules. In addition we offer evidence
that both the 10- and 20-OH groups appear to influence
the reactivity of the other.
Chemistry
Acylation of the phenolic hydroxy moiety is generally
accomplished with carboxylic acids using standard cou-
pling agents such as DIPC or EDC.8 Acylation reac-
tions of 1 itself have been reported using only simple
acid chlorides or acid anhydrides in pyridine.9,10 Since 1
also has low solubility in most organic solvents, further
variation of reaction conditions is limited. Direct acyla-
tion of 1 with PEG was first attempted by employing the
acid chloride derived from mPEG-COOH (Mw 5000)
and resulted in incomplete conversion, while employing
the mPEG-NHS ester did not produce any product at
all. Using PEG acid per se with various coupling
reagents (DIPC, DCC, Mukaiyama reagent11), and dif-
ferent bases (pyridine, Et3N, DMAP, K2CO3) continued
to result in low yields of 10-position acylated products.
Interestingly, no indication of 20-O-acylation (1H and
*Corresponding author. Tel.:+1-732-980-4924; fax:+1-732-980-5911;
e-mail: richard.greenwald@enzon.com
0960-894X/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00926-5

578
R. B. Greenwald et al. / Bioorg. Med. Chem. Lett. 13 (2003) 577–580
Scheme 1. (i) MeI, K2CO3, acetone; (ii) PEG-dicarboxylic acid (Mw 40,000, 3), 2-chloro-1-methylpyridinium iodide (4), DMAP, DCM; (iii) 3,
phenyldichlophosphate (5), pyridine, CHCl3.
Figure 1.
13C NMR analyses) could be detected although it has
been established that camptothecin readily forms a PEG
ester under these reaction conditions.5 Alkylation of 10-
HCPT with MeI and K2CO3 as base, on the other hand,
gave good yields (80%) of the known ether, 10-meth-
oxycamptothecin 2.1 Unlike 1, and in similar fashion to
native camptothecin, 2 underwent facile acylation at the
20-OH position employing the Mukaiyama reagent to
yield PEG ester prodrug 6 (Scheme 1). In order to
explain these results, we propose the existence of strong
intermolecular hydrogen bonding between the 10-OH
and the 20-OH moieties in the parent compound as
depicted in Figure 1 (see 1b). Methylation of 1 appears
to interrupt the hydrogen bonding between species, thus
allowing acylation to take place easily at the 20-OH
position. In 1978 Liu et al.12 reported that phenyl
dichlorophosphate in dimethoxyethane, with just suffi-
cient pyridine to accomplish neutralization of all acids,
could effect esterification of carboxylic acids with var-
ious alcohols as well as phenol at room temperature
under virtually neutral conditions. Later work13 indi-
cated that the reactive species in these acylations was
probably the mixed phenylphosphoric–carboxylic
anhydride or dicarboxylic acid anhydride. We also have
chosen 6-hydroxy quinoline (6-HQ, 8) as a model of the
phenolic-quinoline portion of 10-HCPT since the phe-
nolic 6-OH group cannot be influenced by an aliphatic
OH as is the case for the 10-HCPT (Scheme 2). The
results of several acylation experiments on 6-HQ using
acid chloride or carboxylic acid with GOC or the Liu
phosphorous reagents as condensing agent are sum-
marized in Table 1. Two important features for acyla-
tion of this model can be gleaned from these results.
First, phenyl dichlorophosphate and EDC give the best
yields. Secondly, because under the essentially neutral
conditions of the Liu reagent there is a sufficient con-
centration of the uncharged and/or non H-bonded spe-
cies, 8 and 8a in the equilibrium as shown in Scheme 2,
which is reactive towards acylation. Perhaps more infor-
mative is the result that EDC/base now is an effective
reagent for phenolic acylation, whereas in the presence
of the 20-OH in 10-HCPT this reaction does not occur.
Our earlier postulate that methylation of the 10-OH
moiety interrupts intermolecular hydrogen bonding
with the 20-OH position, which then allows acylation of
20-OH to take place is substantiated by the 6-HQ model
where there is no possibility of intermolecular hydrogen
bonding: acylation in the presence of EDC and pyridine
now takes place easily. However, employing phenyl
dichlorophosphate as coupling agent, 10-HCPT was
now successfully and selectively acylated with PEG
dicarboxylic acid5 (3, Mw 40,000) in the presence of
pyridine to produce the 10-PEG ester transport form 7
in 89% yield. The selective 10-acylation was also con-
firmed by NMR data. 1H NMR of the 20-acylated
camptothecin derivatives exhibited the C-19 protons as
two quartets14 (at 2.21 ppm for 6) compared to the
unacylated camptothecin as a single quartet. The PEG
acylated product 7 showed the C-19 protons as one
quartet at 1.83 ppm. Therefore, it appears that the 20-
OH is still present and that selective 10-OH aromatic
acylation was achieved. In vitro data for compound 7 is
shown in Table 2 with an in vivo evaluation using a
human colon tumor (LS174T) xenograft in nude mice in
comparison with PEG-ala-camptocthecin15 (Prothe-
can1) which currently is in clinical trial. The results show
that the PEGylated version of 10-HCPT was statistically
.
Scheme 2. (i) mPEG carboxylic acid (3a, Mw 5000), EDC HCl,
DMAP, DCM; (ii) 3, 5, pyridine, CHCl3.

R. B. Greenwald et al. / Bioorg. Med. Chem. Lett. 13 (2003) 577–580
Table 1. Acylation reactions of 6-hydroxyquinoline
579
Entry
Acid or acid chloride
Condition
% Acylationa
1
2
3
4
5
BnOCH2C(¼O)Cl
mPEG (5000) CO2H
PEG (40,000) CO2H
PEG (40,000) CO2H
mPEG (5000) CO2H
Pyridine, CH2Cl2
PhOP(¼O)Cl2, pyridine, CHCl3
PhOP(¼O)Cl2, pyridine, CHCl3
Me2NP(¼O)Cl2, pyridine, CHCl3
EDC, pyridine
0
>95
>95
20%
>95
aEstimated by NMR and HPLC.
Table 2. In vitro and in vivo data of PEG 10-HCPT derivative (7)
t1=2 (h, PBS)a
t1=2 (h, rat plasma)a
IC50 (nM, P388/0)b
%TGI (LS174) solid tumorc
Camptothecin
10-HCPT (1)
Prothecan1
PEG 10-HCPT (7)
—
—
100
5
—
—
15
0.12
7
4.9
12
16
55
36.5
71.5
57.6
aAll in vitro experiments were done at 37 ꢀC in duplicate with standard deviation of ꢂ10%.
bFor method, see ref 5.
cPercent tumor growth inhibition (%TGI) was calculated from the quotient of the median tumor volume of the treatment group divided by the
median tumor volume of the control group {(1-T/C)ꢁ100} when the latter reached 3000 mm3. All compounds were administered at 12 mg/kg/inj
(CPT content) intravenously qd7ꢁ3 in established (300 mm3) colorectal LS174 subcutaneous tumors.
better than control and demonstrated enhanced activity
over the unmodified drug as well in this model.
Preparation of 10-methoxycamptothecin (2). Amixture
of 10-hydroxycamptothecin (1, 50 mg, 0.14 mmol),
potassium carbonate (100 mg, 0.72 mmol), and iodo-
methane (50 mL, 0.80 mmol) in acetone (8 mL) was
refluxed overnight. The mixture was filtered and the fil-
trate concentrated. The residue was recrystallized from
MeOH–Et2O to give 2 (53 mg, 80%) as a yellow solid:
1H NMR d 1.04 (t, 3H, J=13.5, H-18), 1.94 (q, 2H,
J=13.5, H-19), 3.99 (s, 3H, 10-OCH3), 5.25 (s, 2H, H-
5), 5.31 (d, 1H, J=16.2, H-17), 5.64 (d, 1H, J=16.2, H-
17), 7.21 (d, 1H, J=2.7, H-11), 7.48 (m, 1H, H-9), 7.66
(s, 1H, H-14), 8.04 (d, 1H, J=8.1, H-12), 8.35 (s, 1H, H-
7); 13C NMR d 7.10, 30.89, 49.79, 55.15, 65.37, 72.51,
97.79, 105.14, 117.90, 123.56, 128.82, 129.45, 129.71,
129.87, 144.24, 145.84, 149.20, 150.99, 157.49, 158.70,
173.33.
Conclusion
It is probable that 10-HCPT exists as a strongly H-
bonded intermolecular species such as 1b depicted in
Figure 1, and does not easily undergo acylation at either
position. On the other hand, compound 2 (10-methoxy-
CPT), unlike10-HCPT (1), undergoes facile acylation at
the 20-OH position clearly demonstrating the effect of
blocking the phenolic OH and, in turn, intermolecular
hydrogen bonding. Selective high yield acylation of the
10-OH position of 1 can be achieved with carboxylic
acids under neutral conditions using phenyl dichlor-
ophosphate (5) as condensing agent. It is not clear at
this time whether 5 is superior for the acylation of 1 due
to the neutrality of the conditions or due to the proper-
ties of the intermediates, but it is clear that 5 has the
capability of acylation even in the presence of the H-
bonding form (1b) compared to other acylating agents.
Preparation of 10-methoxy-20-PEG-camptothecin ester
(6). PEG carboxylic acid (3, Mw 40,000, 0.5 g, 0.0124
mmol) was dissolved in anhydrous methylene chloride
(20 mL) at room temperature followed by the addition
of 2 (14 mg, 0.04 mmol). The reaction mixture was
cooled to 0 ꢀC in an ice bath. 2-Chloro-1-methylpyr-
idinium iodide (4, 17 mg, 0.067 mmol) and DMAP (17
mg, 0.14 mmol) were added to the mixture at 0 ꢀC. The
reaction mixture was allowed to warm to room tempera-
ture and left for 24 h. The solution was washed with 0.1 N
HCl (10 mLꢁ2), dried over anhydrous sodium sulfate, and
concentrated under reduced pressure to give a light yellow
syrup. The crude product was crystallized from 2-propa-
nol to give 6 as a white solid (0.39 g, 80%). The amount
Experimental
General
All reactions were run under an atmosphere of dry
nitrogen or argon. Commercial reagents were used
without further purification. 10-Hydroxycamptothecin
was obtained from Hande Tech USA, Inc. (Houston,
1
of 1 in 6 measure by UV assay14 was 1.8% (wt/wt): H
1
TX, USA). H spectra were obtained with a JEOL FT
NMR d 0.98 (t, 3H, J=5.4, H-18), 2.21 (m, 2H, H-19),
3.37–3.93 (m, PEG) 3.99 (s, 3H, 10-OCH3), 4.36 (s, 2H,
PEG–CH2CO2–), 5.28 (s, 2H, H-5), 5.41 (d, 1H,
J=16.2, H-17), 5.68 (d, 1H, J=16.2, H-17), 7.14 (s, 1H,
H-11), 7.41 (s, 1H, H-9), 7.51 (s, 1H, H-14), 8.09 (d, 1H,
J=8.1, H-12), 8.33 (s, 1H, H-7); 13C NMR d 7.15,
31.28, 49.58, 55.33, 66.70–70.77 (PEG), 94.66, 105.15,
NMR System JNM GSX-270 instrument using deuterio-
chloroform as solvent. 13C NMR spectra were obtained
at 67.80 MHz on the JNM GSX-270. Chemical shifts (d)
are reported in parts per million (ppm) downfield from
tetramethylsilane and coupling constants (J values) are
given in hertz (Hz).

580
R. B. Greenwald et al. / Bioorg. Med. Chem. Lett. 13 (2003) 577–580
119.08, 123.33, 128.64, 129.25, 130.50, 144.56, 144.97,
146.33, 169.61, 156.87, 158.56, 166.80, 169.18.
67.92–72.85 (PEG), 118.21, 120.38, 123.32, 128.42,
130.69, 135.65, 146.13, 148.09, 150.15, 168.35.
Preparation of 10-PEG-camptothecin ester (7). Amix-
ture of 3 (Mw 40,000, 1.0 g, 0.025 mmol) and 1 (35 mg,
0.096 mmol) in toluene (30 mL) was dried by azeo-
tropic distillation of 15 mL of toluene. The mixture
was cooled to room temperature and the solvent
removed by distillation in vacuo. Anhydrous chloro-
form (30 mL) was added to the mixture followed by
addition of pyridine (1.0 mL, 12.4 mmol) and phenyl
dichlorophosphate (5, 0.15 mL, 1.0 mmol). The reac-
tion mixture was stirred at room temperature for 18 h.
The solution was washed with ice-cold 1 N HCl (25
mLꢁ2), dried over anhydrous sodium sulfate, and
concentrated under reduced pressure to give the pro-
duct as a light yellow solid. After recrystallization from
2-propanol 7 was obtained as a white solid (0.8918 g,
89% yield). The amount of 1 in 7 measured by UV
assay14 was 1.7% (wt/wt): 1H NMR d 0.98 (t, 3H,
J=11.2 Hz, H-18), 1.83 (q, 2H, J=8.6, H-19), 2.94 (s,
H2O in PEG), 3.38–3.90 (bs, PEG), 4.53 (s, PEG–CH2–
CO2–), 5.31 (s, 2H, H-5), 5.19–5.69 (ABq, 2H,
J=108.21 & 27.06, H-17), 7.2 (d, 1H, J=11.21, H-11),
7.64 (s, 1H, H-9), 7.82 (s, 1H, H-14), 8.18 (d, 1H,
J=11.88, H-12), 8.44 (s, 1H, H-7); 13C NMR d 7.33,
31.10, 49.53, 65.65, 68.13, 70.03–70.97 (PEG), 77.92,
97.45, 118.24, 118.42, 125.05, 127.94, 128.85, 130.45,
130.78, 145.57, 146.35, 148.55, 149.64, 152.07, 156.97,
168.30, 172.98.
Acknowledgements
The authors would like to thank Dr. Chyi Lee and and
analytical group for their assistance in the analysis of
the camptothecin derivatives and Dr. Charles D. Con-
over and pharm/tox group of Enzon for in vitro and in
vivo testing.
References and Notes
1. (a) Wani, M. C.; Wall, M. E. J. Org. Chem. 1969, 34, 1364.
(b) Wani, M. C.; Ronman, P. E.; Lindley, J. T.; Wall, M. E. J.
Med. Chem. 1969, 1980, 23 554.
2. Zhang, R.; Li, Y.; Cai, Q.; Liu, T.; Sun, H.; Chambless, B.
Cancer Chemother. Pharmacol. 1998, 41, 257.
3. Sawada, S.; Okajima, S.; Aiyama, R.; Nokata, K.; Furuta,
T.; Yokokura, T.; Sugino, E.; Yamaguchi, K.; Miyasaka, T.
Chem. Pharm. Bull. 1991, 39, 1446.
4. Kingsbury, W. D.; Boehm, J. C.; Jakas, D. R.; Holden,
K. G.; Hecht, S. M.; Gallagher, G.; Caranfa, M. J.; McCabe,
F. L.; Faucette, L. F.; Johnson, R. K.; Hertzberg, R. P. J.
Med. Chem. 1991, 34, 98.
5. Greenwald, R. B.; Pendri, A.; Conover, C.; Gilbert, C.;
Yang, R.; Xia, J. J. Med, Chem. 1996, 39, 1938.
6. Zhao, H.; Lee, C.; Sai, P.; Choe, Y. H.; Pendri, A.; Guan,
S.; Greenwald, R. B. J. Org. Chem. 2000, 65, 4601.
7. Takayama, H.; Watanabe, A.; Hosokawa, M.; Chiba, K.;
Satoh, T.; Aimi, N. Bioorg. Med. Chem. Lett. 1998, 8, 415.
8. Kopple, K. D.; Nitecki, D. E. J. Am. Chem. Soc. 1962, 84,
4457.
9. Miyasaka, T.; Sawada, S.; Nokata, K.; Sugino, E.; Mutai,
M. US Patent 4,604,463, 1986.
10. Berges, D. A.; Taggart, J. J. US Patent 5,663,177, 1997.
11. Mukaiyama, T.; Usai, M.; Shimada, E.; Saigo, K. Chem.
Lett. 1975, 1045.
Preparation of 6-mPEG-hydroxyquinoline ester (9). A
mixture of 6-Hydroxyquinoline (8, 129 mg, 0.89 mmol),
mPEG acid (3a, Mw 5,000, 1.5 g, 0.3 mmol), and
.
EDC HCl (228 mg, 1.19 mmol) in anhydrous pyridine
(20 mL) was stirred at room temperature overnight. The
mixture was concentrated in vacuo followed by crystal-
lization of the residue from 2-propanol to give 9 (1.2 g,
80%): 13C NMR d 57.44, 66.80–71.81 (PEG), 117.19,
120.53, 123.22, 127.13, 129.63, 134.54, 144.84, 146.72,
149.05, 167.59.
12. Liu, H. J.; Chan, W. H.; Lee, S. P. Tetrahedron Lett. 1978,
4461.
13. Arrieta, A.; Garcia, T.; Lago, J. M.; Palomo, C. Syn.
Comm. 1983, 13, 471.
14. Greenwald, R. B.; Pendri, A.; Conover, C. D.; Lee, C.;
Choe, Y. H.; Gilbert, C.; Martinez, A.; Xia, J.; Wu, D.; Hsue,
M. Bioorg. Med. Chem. 1998, 6, 551.
15. Conover, C. D.; Greenwald, R. B.; Pendri, A.; Shum,
K. L. Anti-Cancer Drug Des. 1999, 14, 499.
Preparation of 6-PEG-hydroxyquinoline ester (10).
Compound 8 was subjected to the same conditions for
the conversion of 1 to 7 to give 10 in 89%: 13C NMR d