Synthesis and stereochemical preference of peptide 4-aminocyclophosphamide conjugates as potential prodrugs of phosphoramide mustard for activation by prostate-specific antigen (PSA)

Article abstract of DOI:10.1016/j.bmcl.2009.03.009

In an effort to develop proteolytically activated prodrugs of phosphoramide mustard by prostate-specific antigen (PSA), a series of tetrapeptide (Cbz-Ser-Ser-Phe-Tyr)-conjugated 4-aminocyclophosphamide (4-NH₂-CPA) isomers were synthesized and e

Full text of DOI:10.1016/j.bmcl.2009.03.009

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2587–2590
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and stereochemical preference of peptide
4-aminocyclophosphamide conjugates as potential prodrugs
of phosphoramide mustard for activation by prostate-specific antigen (PSA)
Yongying Jiang a, , Robert S. DiPaola b, Longqin Hu a,b,
*
a Department of Pharmaceutical Chemistry, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA
b The Cancer Institute of New Jersey, New Brunswick, NJ 08901, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
In an effort to develop proteolytically activated prodrugs of phosphoramide mustard by prostate-specific
antigen (PSA), a series of tetrapeptide (Cbz-Ser-Ser-Phe-Tyr)-conjugated 4-aminocyclophosphamide (4-
NH2-CPA) isomers were synthesized and evaluated as substrates of PSA. The cleavage of the conjugates
by PSA were found to be stereoselective as only the two isomers with 4R-configuration were efficiently
cleaved by PSA. The cis-(2R,4R)-isomer was the best substrate of PSA with a half-life of 12 min. LC/MS
analysis of the incubation solution of this isomer with PSA suggests that 4-NH2-CPA is released upon pro-
teolysis and quickly degrades to cytotoxic phosphoramide mustard. These results clarified the stereo-
chemical requirements of PSA on the peptide conjugates of 4-NH2-CPA and demonstrated the potential
of these conjugates as potential PSA-activated prodrugs targeting prostate cancer.
Received 16 January 2009
Revised 3 March 2009
Accepted 4 March 2009
Available online 9 March 2009
Keywords:
Cyclophosphamide
Prostate-specific antigen
Proteolytic
Ó 2009 Elsevier Ltd. All rights reserved.
Prodrug
Prostate cancer is the most commonly diagnosed cancer and the
second-leading cause of cancer-related death in both American and
European men.1,2 Currently, androgen deprivation therapy remains
the mainstay of treatment for patients with metastatic prostate
cancer. However, it eventually fails due to the development of hor-
mone-refractory cancer.3 Chemotherapy has primarily been used
for the palliation of symptoms in patients with metastatic prostate
cancer.4 Recently, it has also been reported that chemotherapy im-
proved overall survival in this group of patients.5 Given the po-
tency of many anticancer drugs against cancer cells, there has
been a persisting interest to target these drugs selectively on can-
cer cells through a prodrug strategy to increase their therapeutic
indices.6 Several biological features characteristic to prostate can-
cer cells have been extensively explored for this purpose.7–9 In par-
ticular, prostate-specific antigen (PSA) has been considered an
attractive target because of the following characteristics:9 (1) it is
a serine protease selectively expressed in prostate tissue and carci-
noma in prostate cancer patients; (2) PSA is expressed at a high le-
vel (up to mg/g) in prostate carcinoma;10 (3) the enzymatic activity
of PSA is confined to the prostate and prostate-derived cancer cells;
those leaked into systemic circulation are inactivated by plasma
protease inhibitors. A number of prodrugs using PSA as the activat-
ing enzyme have been reported and some have entered clinical
trials.9
Cyclophosphamide is an alkylating antitumor agent with activ-
ity against a broad spectrum of human cancers including slow-
growing tumors.11 However, the clinical application of cyclophos-
phamide is limited due to its dose-related toxic side effects. This
has resulted in considerable efforts in designing prodrugs that spe-
cifically release cytotoxic phosphoramide mustard at tumor
sites.12–14 We have recently proposed and demonstrated that 4-
aminocyclophosphamide (4-NH2-CPA) decomposes into phospho-
ramide mustard under physiological conditions and this feature
enables its application as a prodrug moiety.15,16 In this Letter, we
report the synthesis of two pairs of diastereomers of a tetrapeptide
(Cbz-Ser-Ser-Phe-Tyr-) conjugated 4-aminocyclophosphamide (1)
and the stereoselectivity of PSA-mediated cleavage of these conju-
gates. This is our initial report of our research toward developing
phosphoramide mustard prodrugs utilizing PSA as the activating
enzyme.
Previously, we have shown that four diastereomers exist for 4-
NH2-CPA and its conjugates (Fig. 1).15–17 Results from the
a-chy-
motrypsin digestion of N-phenylalanyl-4-aminocyclophosphamide
isomers indicate that one of the cis isomers, the (2R,4R)-isomer, is
the best substrate for
a
-chymotrypsin.15 Although PSA is a mem-
ber of the serine protease family and shares similar substrate spec-
ificity to chymotrypsin,18 the stereochemical preference for PSA
still needs to be confirmed. Among a number of amino acid se-
quences screened for PSA on its affinity and specificity, the se-
* Corresponding author. Tel.: +1 732 445 5291; fax: +1 732 445 6312.
E-mail address: LongHu@rutgers.edu (L. Hu).
Present address: Hoffmann-La Roche Inc. Non-Clinical Safety, Nutley, NJ 07110,
USA.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.03.009

2588
Y. Jiang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2587–2590
1
5
O
O
6
H
N
H
N
H
H
O
3
N(CH2CH2Cl)2
N(CH2CH2Cl)2
2
Cbz-Ser-Ser-Phe-Tyr
Cbz-Ser-Ser-Phe-Tyr
N
N
4
P
P
N(CH2CH2Cl)2
N(CH2CH2Cl)2
O
cis-(2R, 4R)-1
trans-(2S, 4R)-1
O
P
P
O
O
O
N
N
Cbz-Ser-Ser-Phe-Tyr NH
cis-(2S, 4S)-1
Cbz-Ser-Ser-Phe-Tyr NH
trans-(2R, 4S)-1
H
H
Figure 1. Four stereoisomers of Cbz-Ser-Ser-Phe-Tyr-NH-CPA (1).
quence SS(Y/F)Y;S(G/S) was identified to be cleaved the fastest by
PSA, with a catalytic efficiency as high as 3100 Mꢀ1sꢀ1 19
Therefore,
we used this sequence and designed the tetrapeptide Cbz-Ser-Ser-
Phe-Tyr-conjugate (1) of 4-NH2-CPA in our initial efforts to screen
for the stereochemical requirements of PSA-mediated cleavage of
peptide conjugates of 4-NH2-CPA (Fig. 1).
overnight (ꢁ18 h). As a tertiary amide, DMF is stable under the
Hofmann reaction conditions and has been used as a solvent for
the BTI-mediated Hofmann rearrangement reaction.20 The TBDPS
.
group in 6 was removed by TBAF and the resulting
c-aminoalcohol
7 was used for the subsequent cyclization reaction.
Cyclization of (S)-7 with bis(2-chloroethyl)phosphoramidic
dichloride was initially carried out using TEA as a base in anhy-
drous EtOAc or THF at room temperature for 48 h. This condition
has been generally used for the synthesis of cyclophosphamide
Synthesis of the four stereoisomers of 1 is outlined in Scheme 1.
The two diastereomers of 1 with 4R-configuration were synthe-
sized from
L-homoserine (Hse) and the other two with 4S-configu-
ration were from
D-Hse. Homoserine (2) was reacted with the pre-
analogs from c
-aminoalcohols.17,21 Under this condition the fas-
activated HOSu ester of Cbz-Tyr(Bn)-OH to give the dipeptide 3
which was converted to its amide 4 through HOBt/EDC activation
followed by aminolysis. The side chain hydroxyl group of 4 was
protected with a TBDPS group using TBDPS-Cl/imidazole in DMF.
BTI-mediated Hofmann rearrangement of the amide 5 was carried
out in acetonitrile/water (1:1) with 20% DMF as a cosolvent, afford-
ing the gem-diamine derivative 6 in nearly quantitative yields.
DMF was added to increase the solubility of 5 in the reaction sol-
vents in order to expedite the reaction to reach completion within
4 h. Otherwise, the starting material would remain in the reaction
mixture even after significantly extending the reaction time to
ter-eluting diastereomer (2R,4R)-8 (the cis isomer) was isolated
in 16% yield and the slower-eluting diastereoisomer (2S,4R) -8
(the trans isomer) in 5% yield after flash column chromatography.
However, the degradation product from the starting gem-diamine,
Cbz-Tyr(Bn)-NH2, was isolated as the major side product in 25%
yield. Assignment of configurations to these isomers has been
extensively discussed in our previous publications.15,17 The spec-
troscopic data used to assign the configuration of the four diaste-
reomers of 8 are summarized in Table 1.
To improve the cyclization yield, we investigated several condi-
tions which used NaH or n-BuLi in place of TEA. The stronger base
OH
OH
OH
OTBDPS
NH2
O
O
O
OH
OH
H
N
H
N
H
N
Cbz
Cbz
NH2
Cbz
i
ii
iii
N
H
*
N
H
*
N
H
*
H2N
*
O
O
O
O
OBn
OBn
OBn
(S)-2 H-
L
-Hse-OH
(S)-3 90%
(R)-3 100%
(S)-4 84%
(S)-5 94%
2
(R)- H-
D
-Hse-OH
4
5
(R)- 80%
(R)- 90%
O
OTBDPS
v
O
OH
O
O
O
P
H
H
N
H
iv
vi
Cbz
N
Cbz
Cbz N
*
*
N
H
NH2
N
NH2
N
N
N(CH2CH2Cl)2
*
H
H
H
OBn
OBn
OBn
cis-(2R, 4R)-8
7
7
(S)-7 54%
21% from (S)-
15% from (S)-
17% from (R)-
12% from (R)-
(S)-6 100%
7
trans-(2S, 4R)-8
(R)- 65%
6
(R)-
81%
8
8
cis-(2S, 4S)-
7
7
trans-(2R, 4S)-
OH
O
O
O
P
O
O
O
P
O
H
N
H
O
H2N
N
vii
viii
*
N
N
H
N(CH2CH2Cl)2
N
H
*
N
N
H
N(CH2CH2Cl)2
Cbz
N
H
H
H
O
OH
OH
OH
9
1
cis-(2R, 4R)- 100%
cis-(2R, 4R)- 52%
9
95%
100%
96%
1
47%
55%
33%
trans-(2S, 4R)-
trans-(2S, 4R)-
cis-(2S, 4S)-1
trans-(2R, 4S)-1
cis-(2S, 4S)-9
trans-(2R, 4S)-9
Scheme 1. Synthesis of diastereomers of Cbz-Ser-Ser-Tyr-Phe-NH-CPA (1) from D- or L-homoserine. Reagents and conditions: (i) Cbz-Tyr(Bn)-OSu, 1 M KHCO3, THF, rt; (ii)
HOBt, EDC, THF, rt, then satd NH3 (aq); (iii) TBDPS-Cl, imidazole, DMF, rt; (iv) BTI, CH3CN/H2O (1/1), rt; (v) TBAF, THF, rt; (vi) n-BuLi, HMAP, THF, ꢀ70 °C for 2 h, then
Cl2PON(CH2CH2Cl)2, TEA, rt, 48 h; (vii) H2, Pd/C (10%), MeOH, rt; (viii) Cbz-Ser-Ser-Phe-OSu, DIEA, DME (dimethoxyethane).

Y. Jiang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2587–2590
2589
Table 1
PSA, aliquots were withdrawn at various time intervals, quenched
with acetonitrile, and analyzed by HPLC. The half-life of each
compound under the reaction conditions was calculated based
on the disappearance of the substrate. As shown in Figure 2, a sig-
nificant difference of substrate activity exists between the stereo-
mers of 1. The diastereomers with 4R-configuration were
efficiently hydrolyzed by PSA, with a half-life of 12 min for cis-
(2R,4R)-1 and 18 min for trans-(2S,4R)-1. On the other hand, the
two diastereomers with the 4S-configuration were resistant to
PSA hydrolysis. Under the same incubation conditions, no hydro-
lysis of cis-(2S,4S)-1 was observed and only ꢁ10% of trans-(2R,4S)
-1 was hydrolyzed after 60 min-incubation with PSA. These re-
sults demonstrate that the configuration of 4-NH2-CPA is critical
for the efficient hydrolysis of its peptide conjugate by PSA with
strong preference for the (2R,4R)-configuration. Given that PSA
is chymotrypsin-like serine protease, these results are consistent
Analytical data of diastereomers of Cbz-Tyr-NH-CPA (8) and Cbz-Ser-Ser-Phe-Tyr-NH-
CPA (1)
Compound
MS
NMR d (ppm)
1H (C-5, 2H)c
1H (C-4, 1H)c
31Pd
Cis-(2R,4R)-8
Trans-(2S,4R)-8
Cis-(2S,4S)-8
Trans-(2R,4S)-8
Cis-(2R,4R)-1
Trans-(2S,4R)-1
Cis-(2S,4S)-1
663.1884a
663.1878a
663.1926ª
663.1918ª
894.2739b
894.2736b
894.2785b
894.2761b
5.38–5.08
5.41–5.30
5.30–5.10
5.45–5.25
5.34–5.20
5.45–5.30
5.40–5.25
5.40–5.30
2.30–1.80
2.10–1.70
2.02–1.50
1.90–1.44
2.21–1.80
2.18–1.90
2.20–1.76
2.00–1.70
9.8
13.4
9.2
11.0
8.8
12.1
10.7
13.3
Trans-(2R,4S)-1
a
HRMS (FAB), m/z calcd for C31H38Cl2N4O6P [MH]+ 663.1906.
HRMS (FAB), m/z calcd for C39H51Cl2N7O11P [MH]+ 894.2761.
b
c
The 1H chemical shift was recorded at 200 MHz Varian Gemini spectrometer
using residual undeuterated solvents as the internal reference.
d
The 31P chemical shift was recorded at 121 MHz using 5% H3PO4 in D2O in a
with our previous observation that
a-chymotrypsin has the same
coaxial insert as an external standard.
stereochemical preference on the phenylalanine conjugate of
4-NH2-CPA.15 This is also consistent with the observation that
substrate recognition by PSA is mediated by an extended binding
pocket and subsites beyond subsite 1 (S1) play a critical role in
defining its substrate specificity.19 When the incubation mixture
of cis-(2R,4R) -1 with PSA was analyzed by LC/MS, the only
product with UV absorption identified was the tetrapeptide
Cbz-Ser-Ser-Phe-Tyr-OH, indicating that the PSA cleavage oc-
curred only after tyrosine residue and 4-NH2-CPA was released.
However, we were not able to detect 4-NH2-CPA using LC/MS.
Instead, 4-hydroxylcyclophosphamide (or aldophosphoramide)
was unambiguously identified. The degradation mechanism of
4-NH2-CPA has been investigated and discussed in our previous
publications.15,16
was thought to facilitate the phosphorylation reaction of the hy-
droxyl group in 7 and thus the subsequent cyclization. One set of
conditions using n-BuLi was found to increase the combined cycli-
zation yield from 21% under the TEA conditions to 36%. Under the
new conditions, (S)-7 was treated with one equivalent of n-BuLi in
a solvent mixture of HMPA and THF at ꢀ70 °C for 30 min before the
addition of bis(2-chloroethyl)phosphoramidic dichloride. After 2 h
at ꢀ70 °C, the reaction was allowed to warm up to room tempera-
ture and TEA (1.1 equiv) was added to complete the cyclization.
The n-BuLi conditions produced (2R,4R) -8 in 21% yield and
(2S,4R)-8 in 15% yield, which were 5–10% higher than that under
the earlier TEA conditions. The side product, Cbz-Tyr(Bn)-NH2, also
increased slightly from 25% under the TEA conditions to 31%. The
unreacted starting material was not recovered under both sets of
conditions. The cis-( 2R,4R)-8 and trans-(2S,4R)-8 isomers were
readily separated by flash column chromatography on silica gel.
However, we found that purification of the cis-(2R,4R)-8 was often
contaminated by Cbz-Tyr(Bn)-NH2, the major side product of the
cyclization reaction. This is because of their similar polarity on
TLC. We finally resorted to BTI-mediated Hoffmann rearrangement
to purify cis-(2R,4R)-8; this was accomplished by treating the mix-
ture first collected with BTI to convert the amide side product Cbz-
Tyr(Bn)-NH2 to an unstable gem-diamine carbamate that would
further decompose22 before a second column purification. The
Cbz and benzyl groups in 8 were removed by hydrogenolysis over
10% Pd–C, affording tyrosine-conjugated 4-aminocyclophospha-
mide 9 in nearly quantitative yields. The tripeptide Cbz-Ser-Ser-
Phe-OH was synthesized through the standard solution peptide
synthesis procedure using EDC/HOBt. For conjugation of the tri-
peptide with 9, the HOSu ester of the tripeptide was synthesized
using DCC/HOSu in DME and purified by recrystallization. The acti-
vated ester was then reacted with individual isomers of 9 in DMF
and the reaction was monitored by LC/MS. The final product (1)
was purified by flash column chromatography on silica gel and
its identity was confirmed by its spectroscopic data. Other meth-
ods of conjugating the tripeptide with 8 using EDC/HOBt or
HBTU/DIEA in DMF failed to give the desired peptide conjugate 1
as the major product. This unexpected result could be due to the
involvement of the peptide side chain hydroxyl and phenol groups
in the conjugation reaction. These side reactions were avoided by
using the less active HOSu ester.
In summary, the four stereoisomers of Cbz-Ser-Ser-Phe-Tyr-
conjugated 4-NH2-CPA were synthesized stereospecifically from
the
D- and L-homoserine and were evaluated as substrates for
PSA. Our results demonstrate that the configuration of 4-NH2-
CPA is critical for the proteolysis of the peptide conjugates by
PSA. Only the stereoisomers with 4R-configuration were efficiently
hydrolyzed by PSA and cis-(2R,4R)-1 was most efficiently cleaved
by PSA with a half-life of 12 min followed by trans-(2S,4R)-1. LC–
MS analysis of the proteolytic products of cis-(2R,4R)-1 indicates
that 4-NH2-CPA was released upon proteolysis and disintegrates
into cytotoxic phosphoramide mustard. These results suggest that
peptide-conjugated 4-aminocyclophosphamides could potentially
be used as prodrugs for proteolytic activation to improve the ther-
apeutic effectiveness of cyclophosphamide in the treatment of
cancer.
1.0
0.8
0.6
0.4
0.2
0.0
0
10
20
30
40
50
60
The four stereoisomers of 1 were evaluated as substrates of
PSA by incubating each individual compound with purified PSA
(Calbiochem, La Jolla, CA) in Tris/HCl buffer (50 mM, pH 8.0,
0.1% TWEEN-20Ò) at 37 °C. The enzyme/substrate molar ratio
was 1/100. After the reaction was initiated by the addition of
Time (min)
Figure 2. PSA digestion of compounds cis-(2R,4R)-1 (d, t1/2 = 12 min), trans-(2S,4R)-
1 (s, t1/2 = 18 min), cis-(2R,4S)-1 (N, no cleavage within 60 min), and trans-(2S,4S)-1
(4, <10% cleavage within 60 min). Shown is the disappearance of the substrates.

2590
Y. Jiang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2587–2590
5. de Wit, R. BJU Int. 2008, 101, 11.
Acknowledgements
6. Kratz, F.; Muller, I. A.; Ryppa, C.; Warnecke, A. ChemMedChem 2008, 3, 20.
7. Jayaprakash, S.; Wang, X.; Heston, W. D.; Kozikowski, A. P. ChemMedChem
2006, 1, 299.
8. Graeser, R.; Chung, D. E.; Esser, N.; Moor, S.; Schachtele, C.; Unger, C.; Kratz, F.
Int. J. Cancer 2008, 122, 1145.
9. Atkinson, J. M.; Siller, C. S.; Gill, J. H. Br. J. Pharmacol. 2008, 153, 1344.
10. Lilja, H.; Abrahamsson, P. A.; Lundwall, A. J. Biol. Chem. 1989, 264, 1894.
11. Colvin, O. M. Curr. Pharm. Des. 1999, 5, 555.
12. Mulcahy, R. T.; Gipp, J. J.; Schmidt, J. P.; Joswig, C.; Borch, R. F. J. Med. Chem.
1994, 37, 1610.
We gratefully acknowledge the financial support of grant SNJ-
CCR 700-009 from the State of New Jersey Commission on Cancer
Research and grant RSG-03-004-01-CDD from the American Cancer
Society. This work was also partially supported by a pilot grant
from the Gallo Prostate Cancer Center of the Cancer Institute of
New Jersey that was supported by grant DAMD17-01-1-0755 from
the Department of Defense.
13. Hernick, M.; Flader, C.; Borch, R. F. J. Med. Chem. 2002, 45, 3540.
14. Jain, M.; Kwon, C. H. J. Med. Chem. 2003, 46, 5428.
15. Jiang, Y.; Hu, L. Bioorg. Med. Chem. Lett. 2007, 17, 517.
16. Jiang, Y.; Hu, L. Bioorg. Med. Chem. Lett. 2008, 18, 4059.
17. Jiang, Y.; Zhang, Z.; DePaola, R.; Hu, L. Tetrahedron 2007, 63, 10637.
18. Sokoll, L. J.; Chan, D. W. Urol. Clin. North. Am. 1997, 24, 253.
19. Coombs, G. S.; Bergstrom, R. C.; Pellequer, J. L.; Baker, S. I.; Navre, M.; Smith, M.
M.; Tainer, J. A.; Madison, E. L.; Corey, D. R. Chem. Biol. 1998, 5, 475.
20. Yu, C.; Jiang, Y.; Liu, B.; Hu, L. Tetrahedron Lett. 2001, 42, 1449.
21. Jiang, Y.; Han, J.; Yu, C.; Vass, S. O.; Searle, P. F.; Browne, P.; Knox, R. J.; Hu, L. J.
Med. Chem. 2006, 49, 4333.
References and notes
1. Jemal, A.; Siegel, R.; Ward, E.; Hao, Y.; Xu, J.; Murray, T.; Thun, M. J. CA Cancer J.
Clin. 2008, 58, 71.
2. Ferlay, J.; Autier, P.; Boniol, M.; Heanue, M.; Colombet, M.; Boyle, P. Ann. Oncol.
2007, 18, 581.
3. Petrylak, D. P.; Tangen, C. M.; Hussain, M. H.; Lara, P. N., Jr.; Jones, J. A.; Taplin,
M. E.; Burch, P. A.; Berry, D.; Moinpour, C.; Kohli, M.; Benson, M. C.; Small, E. J.;
Raghavan, D.; Crawford, E. D. N. Engl. J. Med. 2004, 351, 1513.
4. Gilligan, T.; Kantoff, P. W. Urol. 2002, 60, 94.
22. Fletcher, M. D.; Campbell, M. M. Chem. Rev. 1998, 98, 763.