Protease-mediated fragmentation of p-amidobenzyl ethers: A new strategy for the activation of anticancer prodrugs

Article abstract of DOI:10.1021/jo016187+

A new anticancer prodrug activation strategy based on the 1,6-elimination reaction of p-aminobenzyl ethers is described. Model studies were undertaken with the N-protected peptide benzyloxycarbonylvaline-citrulline (Z-val-cit), which was attached to the amino groups of p-aminobenzyl ether derivatives of 1-naphthol and N-acetylnorephedrine. The amide bond that formed was designed for hydrolysis by cathepsin B, a protease associated with rapidly growing and metastatic carcinomas. Upon treatment with the enzyme, the Z-val-cit-p-amidobenzyl ether of 1-naphthol (2) underwent peptide bond hydrolysis with the rapid release of 1-naphthol. The aliphatic Z-val-cit-p-amidobenzyl ether of N-acetylnorephedrine (5) also underwent amide bond hydrolysis, but without the ensuing elimination of N-acetylnorephedrine. On the basis of these results, the phenolic anticancer drugs etoposide (6) and combretastatin A-4 (7) were attached to the Z-val-cit-p-amidobenzyl alcohol through ether linkages, forming the peptide-drug derivatives 8 and 9, respectively. Both compounds were stable in aqueous buffers and serum and underwent ether fragmentation upon treatment with cathepsin B, resulting in the release of the parent drugs in chemically unmodified forms. The released drugs were 13-50 times more potent than were the prodrug precursors on a panel of cancer cell lines. In contrast, the corresponding carbonate derivative of combretastatin A-4 (13) was unstable in aqueous environments and was as cytotoxic as combretastatin A-4. This result extends the use of the self-immolative p-aminobenzyl group for the fragmentation of aromatic ethers and provides a new strategy for anticancer prodrug development.

Full text of DOI:10.1021/jo016187+

1866
J . Org. Chem. 2002, 67, 1866-1872
P r otea se-Med ia ted F r a gm en ta tion of p-Am id oben zyl Eth er s: A
New Str a tegy for th e Activa tion of An tica n cer P r od r u gs
Brian E. Toki, Charles G. Cerveny, Alan F. Wahl, and Peter D. Senter*
Seattle Genetics, 21823 30th Drive SE, Bothell, Washington 98021
psenter@seagen.com
Received October 9, 2001
A new anticancer prodrug activation strategy based on the 1,6-elimination reaction of p-aminobenzyl
ethers is described. Model studies were undertaken with the N-protected peptide benzyloxycarbonyl-
valine-citrulline (Z-val-cit), which was attached to the amino groups of p-aminobenzyl ether
derivatives of 1-naphthol and N-acetylnorephedrine. The amide bond that formed was designed
for hydrolysis by cathepsin B, a protease associated with rapidly growing and metastatic carcinomas.
Upon treatment with the enzyme, the Z-val-cit-p-amidobenzyl ether of 1-naphthol (2) underwent
peptide bond hydrolysis with the rapid release of 1-naphthol. The aliphatic Z-val-cit-p-amidobenzyl
ether of N-acetylnorephedrine (5) also underwent amide bond hydrolysis, but without the ensuing
elimination of N-acetylnorephedrine. On the basis of these results, the phenolic anticancer drugs
etoposide (6) and combretastatin A-4 (7) were attached to the Z-val-cit-p-amidobenzyl alcohol through
ether linkages, forming the peptide-drug derivatives 8 and 9, respectively. Both compounds were
stable in aqueous buffers and serum and underwent ether fragmentation upon treatment with
cathepsin B, resulting in the release of the parent drugs in chemically unmodified forms. The
released drugs were 13-50 times more potent than were the prodrug precursors on a panel of
cancer cell lines. In contrast, the corresponding carbonate derivative of combretastatin A-4 (13)
was unstable in aqueous environments and was as cytotoxic as combretastatin A-4. This result
extends the use of the self-immolative p-aminobenzyl group for the fragmentation of aromatic ethers
and provides a new strategy for anticancer prodrug development.
In tr od u ction
ment of new cancer chemotherapeutics, which has led to
several promising orally active protease inhibitors having
both preclinical and clinical antitumor activities.4 An
additional line of research involves the conscription of
proteases for anticancer prodrug activation. Toward this
end, peptide-containing anticancer prodrugs have been
developed that are activated by proteases within solid
tumors.6-19 Several of these agents have led to significant
in vitro and in vivo antitumor activities.
There are two general approaches for attaching drugs
to peptides for intratumoral proteolytic activation. The
drugs can be appended directly to the peptide, leading
to prodrugs that can either release the parent drug or a
drug that contains vestiges of the bound peptide.16-18 In
the latter case, the released drug may have impaired
cytotoxic activity. An additional consideration for direct
drug attachment to peptides is the negative influence the
drug can have on the kinetics of peptide hydrolysis.
Metastatic carcinomas often express proteolytic en-
zymes including the cysteine protease cathepsin B,1-3
matrix metalloproteinases such as collagenases and
stromelysins,4 and serine proteases represented by plas-
minogen activator and plasmin.5 These enzymes are
thought to be critically involved in the events that lead
to metastasis because they are capable of degrading the
basement membrane and extracellular matrix around
tumor tissue, allowing the tumor cells to migrate and
invade into the surrounding stroma and endothelium.
Cathepsin B in particular has been shown to be clinically
relevant in cancer progression on the basis of studies
showing that cytosolic enzyme levels (856 ng cathepsin
B/mg of protein, determined from 167 breast cancer
patients)were 11 times higher than those in benign
breast tissue specimens.3 Patients with high intratumoral
cathepsin B levels suffer a significantly worse prognosis
than do patients with low levels.3 Additional activities
associated with these proteases include participation in
protease cascades, activation of enzymes and growth
factors, and tumor angiogenic stimulation.1-3
(5) Andreasen, P. A.; Egelund, R.; Petersen, H. H. Cell. Mol. Life
Sci. 2000, 57, 25-40.
(6) Dubowchik, G. M.; Walker, M. A. Pharmacol. Ther. 1999, 83,
67-123.
(7) Carl, P. L.; Chakravarty, P. K.; Katzenellenbogen, J . A.; Weber,
M. J . Proc. Natl. Acad. Sci. U.S.A. 1980, 77, 2224-2228.
(8) Chakravarty, P. K.; Carl, P. L.; Weber, M. J .; Katzenellenbogen,
J . A. J . Med. Chem. 1983, 26, 633-638.
Several investigators have explored the possibility of
exploiting tumor-associated proteases for the develop-
(9) Chakravarty, P. K.; Carl, P. L.; Weber, M. J .; Katzenellenbogen,
J . A. J . Med. Chem. 1983, 26, 638-644.
* Corresponding author.
(1) Demchik, L. L.; Sloane, B. F. In Proteases: New Perspectives;
Turk, A., Ed.; Birkhauser Verlag: Basel, 1999; pp 109-124.
(2) Mai, J .; Waisman, D. M.; Sloane, B. F. Biochim. Biophys. Acta
2000, 1477, 215-230.
(3) Thomssen, C.; Schmitt, M.; Goretzki, L.; Oppelt, P.; Pache, L.;
Dettmar, P.; J anicke, F.; Graeff, H. Clin. Cancer Res. 1995, 7, 741-746.
(4) Davidson, A. H.; Drummond, A. H.; Galloway, W. A.; Whittaker,
M. Chem. Ind. (London) 1997, 258-261.
(10) Dubowchik, G. M.; Firestone, R. A. Bioorg. Med. Chem. Lett.
1998, 8, 3341-3346.
(11) Dubowchik, G. M.; Mosure, K.; Knipe, J . O.; Firestone, R. A.
Bioorg. Med. Chem. Lett. 1998, 8, 3347-3352.
(12) de Groot, F. M. H.; de Bart, A. C. W.; Verheijen, J . H.; Scheeren,
H. W. J . Med. Chem. 1999, 42, 5277-5283.
(13) de Groot, F. M. H.; van Berkon, L. W. A.; de Bart, A. C. W.;
Scheeren, H. W. J . Med. Chem. 2000, 43, 3093-3102.
10.1021/jo016187+ CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/12/2002

Fragmentation of p-Amidobenzyl Ethers
J . Org. Chem., Vol. 67, No. 6, 2002 1867
Sch em e 1
To circumvent these potential shortcomings, an alter-
native approach for drug attachment incorporates the use
of self-immolative spacers that spatially separate the
drug from the site of enzymatic cleavage. The subsequent
collapse of the incorporated linkers allows for the elimi-
nation of the fully active, chemically unmodified drug
from the conjugate upon amide bond hydrolysis. One of
the most commonly used spacers is the bifunctional
p-aminobenzyl alcohol group, which is linked to the
peptide through the amine moiety, forming an amide
bond. Amine-containing drugs are attached through
carbamate functionalities to the benzylic hydroxyl group
of the linker. The resulting prodrugs are activated upon
protease-mediated cleavage, leading to a 1,6-elimination
reaction20 that releases the unmodified drug, carbon
dioxide, and remnants of the linker group (Scheme 1).
This methodology, based on the work of Sartorelli,
Katzenellenbogen, and co-workers,21,22 has been applied
to the plasmin-catalyzed release of phenylenediamine
mustard8 and anthracyclines9,12-14 from their correspond-
ing peptide-p-amidobenzyl carbamate derivatives and
also to the release of doxorubicin and mitomycin C from
peptide-p-amidobenzyl carbamate peptide derivatives by
lysosomal enzymes and cathepsin B.10,11,15 The same
linkage system has also been applied to the activation of
anthracyclines in cells that were transfected with car-
boxypeptidase G2.23
had already been reported for other paclitaxel carbon-
ates.24,25 Many anticancer agents containing reactive
hydroxyl groups would not be expected to exhibit such
high carbonate stability. For that reason, we explored
new methodologies for attaching alcohol-containing an-
ticancer drugs to self-immolative spacers, which would
lead to high serum stability and conditional drug release
upon peptide bond hydrolysis. Here we describe a novel
drug release strategy involving the fragmentation of
peptide-p-amidobenzyl ether derivatives (Scheme 1). The
synthesis, stabilities, fragmentation, and in vitro cyto-
toxic activities of some representative compounds are
described.
Resu lts
The peptide derivative, Z-valine-citrulline-p-aminoben-
zyl alcohol (1, Z-val-cit-PAB-OH) has previously been
used for the preparation of Z-val-cit-PAB-doxorubicin
carbamate, a compound that released active doxorubicin
upon treatment with cathepsin B.10,11 The drug was
attached to the peptide through a carbamate linkage, as
shown in Scheme 1. To explore the potential of using this
drug-elimination pathway for cleaving less labile bonds,
ether derivatives of 1 were prepared using either the
Mitsunobu reaction to form the naphthol ether 2 or the
two-step imidate-substitution reaction to form the N-
acetylnorephedrine derivative 5 (Scheme 2). These com-
pounds were designed to model anticancer drugs that
contain chemically related moieties. HPLC analysis
indicated that the naphthol ether 2 was a substrate for
bovine spleen cathepsin B and that the products formed
were naphthol and Z-val-cit-COOH. The reaction pro-
ceeded rapidly (350 nmol/min/mg cathepsin B), and by
HPLC, there was no evidence of the build up of the
deacylated p-aminobenzyl ether. In the absence of added
enzyme, there was no breakdown of the starting material
after 1 week at 37 °C at pH 5.1 or 7.2 or in pooled human
serum (Table 1). These results provide the first indication
that p-aminobenzyl ethers are capable of undergoing 1,6-
elimination reactions. Similar studies undertaken with
the N-acetylnorephedrine ether 5 demonstrated that the
compound was hydrolyzed by cathepsin B, leading to the
release of Z-val-cit-COOH as expected. However, no
N-acetylnorephedrine was detected, suggesting that the
p-aminobenzyl ether formed after peptide bond cleavage
did not undergo further fragmentation. Thus, the nature
The chemistry used for drug attachment has generally
been restricted to amine-containing drugs, with the
exception of paclitaxel, which was linked through carbon-
ates formed from hydroxyl groups at the 2′- or 7-posi-
tion.11,13,14 Unlike many carbonates that are hydrolytically
unstable, such paclitaxel 2′- and 7-carbonates were quite
stable in aqueous environments, consistent with what
(14) de Groot, F. M. H.; Loos, W. J .; Koekkoek, R.; van Berkon, L.
W. A.; Busscher, G. F.; Seelen, A. F.; Albrecht, C.; de Bruijn, P.;
Scheeren, H. W. J . Org. Chem. 2001, 66, 8815-8830.
(15) Greenwald, R. B.; Pendri, A.; Conover, C. D.; Zhao, H.; Choe,
Y. H.; Martinez, A.; Shum, K.; Guan, S. J . Med. Chem. 1999, 42, 3657-
3667.
(16) Putnam, D. A.; Shiah, J . G.; Kopecek, J . Biochem. Pharmacol.
1996, 52, 957-962.
(17) Harada, M.; Sakakibara, H.; Yano, T.; Suzuki, T.; Okuno, S. J .
Controlled Release 2000, 69, 399-412.
(18) Denmeade, S. R.; Nagy, A.; Gao, J .; Lilja, H.; Schally, A. V.;
Isaacs, J . T. Cancer Res. 1998, 58, 2537-2540.
(19) Loadman, P. M.; Bibby, M. C.; Double, J . A.; Al-Shakhaa, W.
M.; Duncan, R. Clin. Cancer Res. 1999, 5, 3682-3688.
(20) Wakselman, M. Nouv. J . Chim. 1983, 7, 439-447.
(21) Teicher, B. A.; Sartorelli, A. C. J . Med. Chem. 1980, 23, 955-
960.
(22) Carl, P. L.; Chakravarty, P. K.; Katzenellenbogen, J . A. J . Med.
Chem. 1981, 24, 479-480.
(23) Niculescu-Duvaz, I.; Niculescu-Duvaz, D.; Fiedlos, F.; Spooner,
R.; Martin, J .; Marais, R.; Springer, C. J . J . Med. Chem. 1999, 42,
2485-2489.
(24) Ueda, Y.; Matiskella, J . J .; Mikkilineni, A. B.; Farina, V.; Knipe,
J . O.; Rose, W. C.; Casazza, A. M.; Vyas, D. M. Bioorg. Med. Chem.
Lett. 1995, 5, 247-252.
(25) Senter, P. D.; Marquardt, H.; Thomas, B. A.; Hammock, B. D.;
Frank, I. S.; Svensson, H. P. Cancer Res. 1996, 56, 1471-1474.

1868 J . Org. Chem., Vol. 67, No. 6, 2002
Toki et al.
Sch em e 2
erization of tubulin.27 Treatment of 8 and 9 with cathe-
psin B led to the release of etoposide (6) and combret-
astatin A-4 (7), respectively (Table 1). As with the
naphthol ether 2, there was no detectable p-aminobenzyl
ether, consistent with the fact that peptide hydrolysis is
rate-determining in the fragmentation of the aromatic
ether bond. Both peptide derivatives were stable at pH
5.1 and 7.2 and in human serum.
Ta ble 1. Com p ou n d Sta bility in Aqu eou s Solu tion s a n d
in th e P r esen ce of Ca th ep sin B
stabilitya
specific activity
human
serum
cathepsin Bb
compound
pH 5.1
pH 7.2
(nmol/min/mg)
2
5
8
0% loss 7 d 0% loss 7 d 0% loss 7 d
0% loss 7 d 0% loss 7 d 0% loss 7 d
0% loss 7 d 0% loss 7 d t1/2 48 hd
0% loss 7 d 0% loss 7 d 0% loss 7 d
350
145c
160
61
9
For comparison, the carbonate derivatives 12 and 13
were prepared from the activated p-nitrophenyl carbon-
ate derivative of N-acetylnorephedrine (4) and combre-
tastatin A-4 (7), respectively (Scheme 5). Both carbonates
12 and 13 proved to be unstable in aqueous environ-
ments, in contrast to the corresponding ethers 5 and 9,
respectively (Table 1). As expected, enzymatic hydrolysis
of 12 and 13 led to the formation of 4 and 7. It is
noteworthy that there were no significant kinetic differ-
ences in cathepsin B-mediated hydrolyses between the
peptide-carbonate and peptide-aromatic ether deriva-
tives. Thus, peptide derivatives of p-aminobenzyl aro-
matic ethers are stable in neutral or slightly acidic
buffers and undergo facile ether fragmentation upon
treatment with an enzyme that cleaves the amide bond.
In vitro cytotoxicity studies were performed on cancer
cell lines to determine if the peptide derivatives acted
as prodrugs. The cell lines L2987 (human lung adeno-
carcinoma), WM266/4 (human melanoma), and IGR-39
(human melanoma) were exposed to the agents for 24 h
and then washed, and viability was determined 2 days
later by measuring the incorporation of 3H-thymidine
compared to that in the untreated controls. There were
significant differences in the cytotoxic activity of etopo-
side (6) and that of the corresponding peptide ether
derivative (8) on all three cell lines (Figure 1A-C). At
the concentration required for 50% cell kill, etoposide (6)
was 20-50 times more active than 8, a result consistent
with the loss in cytotoxic activity that has been reported
with another phenol derivative of etoposide.27 Similarly,
the combretastatin ether (9) was less potent than com-
bretastatin A-4 (7) by a factor of 13 on L2987 human lung
adenocarcinoma cells (Figure 1D). Significantly, the
12
13
t1/2 104 h
t1/2 62 h
t1/2 79 h
t1/2 55 h
t1/2 9 d
t1/2 45 h
150
32
a
Measured as the loss of starting material and the appearance
of the released alcohol at 37 °C in phosphate-buffered saline at
pH 7.2, acetate buffer at pH 5.1, or pooled human serum.
b
Measured as the loss of starting material at 37 °C in pH 5.1
acetate buffer. c Measured as the loss of starting material, which
correlated with the appearance of Z-val-cit-COOH. HPLC analysis
indicated that N-acetylnorephedrine (4) was not released upon
d
amide bond hydrolysis. Etoposide (6) and the etoposide moiety
of 8 were unstable in serum. There was no apparent breakdown
of the peptide-linker in 8. No Z-val-cit-COOH, 1, or 6 were
detected.
Sch em e 3
of the leaving group attached to the p-aminobenzyl group
significantly affects the 1,6-elimination reaction.
On the basis of these results, the anticancer drugs
etoposide (6) and combretastatin A-4 (7) (Scheme 3) were
linked to 1 using the coupling conditions shown in
Scheme 4. Etoposide is a clinically approved topoisom-
erase inhibitor that has demonstrated utility in chemo-
therapeutic combinations for the treatment of leukemia,
lymphoma, germ cell tumors, small cell lung tumors, and
several other carcinomas.26 Combretastatin A-4 is a
promising antiangiogenic agent that inhibits the polym-
(26) Hande, K. R. Eur. J . Cancer 1998, 34, 1514-1521.
(27) Horsman, M. R.; Murata, R.; Breidahl, T.; Nielson, F. U.;
Maxwell, R. J .; Stodkiled-Horgensen, H.; Overgaard, J . J . Adv. Exp.
Med. Biol. 2000, 476, 311-323.

Fragmentation of p-Amidobenzyl Ethers
J . Org. Chem., Vol. 67, No. 6, 2002 1869
Sch em e 4
Sch em e 5
combretastatin A-4 carbonate derivative 13 was as cy-
totoxic as combretastatin A-4 (7), reflecting the inherent
instability of carbonate compared to that of the ether
linkages (Table 1). These results, taken together with the
enzyme hydrolysis studies, indicate that the peptide ether
drug derivatives are prodrugs that can be activated by
cathepsin B.
formed as a result of disulfide bond reduction.31,32 The
purpose of the work described here was to extend this
linker methodology to include hydroxyl-based drug at-
tachment, with the goal of producing highly stable
anticancer prodrugs that can be activated by tumor-
associated proteases. To our knowledge, the finding that
combretastatin A-4 and etoposide p-aminobenzyl ethers
undergo facile drug release constitutes the first demon-
stration that a 1,6-elimination reaction can effect the
fragmentation of an ether bond. Furthermore, the ap-
plication of this reaction pathway for the release of an
anticancer drug from a stable peptide prodrug is new.
The advantage of using ethers rather than carbonates
for drug attachment to peptides is evident from the
stability results obtained with the combretastatin A-4
derivatives, 9 and 13. While enzyme-catalyzed hydrolysis
of the two compounds proceeded at similar rates, the
carbonate 13 proved to be hydrolytically unstable and
consequently was significantly more cytotoxic than 9. It
Discu ssion
Self-immolative linkers based on the p-aminobenzyl
group have shown considerable utility for the attachment
of amine-containing anticancer drugs to carriers that are
cleaved by enzymes within tumors.6-15,23 Variations on
this methodology include the activation of p-nitrobenzyl
carbamates upon biological reduction29,30 and the 1,6-
elimination reaction of p-mercaptobenzyl carbamates
(28) Senter, P. D.; Saulnier, M. G.; Schreiber, G. J .; Hirschberg, D.
L.; Brown, J . P.; Hellstro¨m, I.; Hellstro¨m, K. E. Proc. Natl. Acad. Sci.
U.S.A. 1988, 85, 4842-4846.
(29) Mauger, A. B.; Burke, P. J .; Somani, H. H.; Friedlos, F.; Knox,
R. J . J . Med. Chem. 1994, 37, 3452-3458.
(31) Senter, P. D.; Pearce, W. E.; Greenfield, R. S. J . Org. Chem.
1990, 55, 2975-2978.
(32) Zalipsky, S.; Qazen, M.; Walker, J . A.; Mullah, N.; Quinn, Y.
P.; Huang, S. K. Bioconjugate Chem. 1999, 10, 703-707.
(30) Hay, M. P.; Wilson, W. R.; Denny, W. A. Bioorg. Med. Chem.
Lett. 1995, 5, 2829-2834.

1870 J . Org. Chem., Vol. 67, No. 6, 2002
Toki et al.
F igu r e 1. Cytotoxic effects on L2987 human lung adenocarcinoma (A and D), WM266/4 (B), and IGR-39 (C) human melanoma
cell lines. The cells were exposed to various concentrations of the drugs for 24 h, washed, and incubated for another 48 h, and the
3
cytotoxic activities were quantified through the incorporation of H-thymidine relative to that in untreated control cells.
was expected that unstable carbonates33 such as 13 would
display nonspecific toxicities in vivo and would have
limited application for selective activation by endogenous
tumor proteases. On the other hand, the peptide ether
derivatives 2, 5, 8, and 9 showed no evidence of drug
fragmentation after at least 1 week of incubation in
human serum or in aqueous buffers at 37 °C. Treatment
of 2, 8, and 9 with cathepsin B, an enzyme that is
associated with growing and metastatic tumors,1-3 led
to the release of 1-naphthol or active anticancer drugs,
with a concomitant increase in cytotoxic activity, which
validates the application of protease-sensitive p-amido-
benzyl ethers for the development of anticancer prodrugs.
The failure of N-acetylnorephedrine (5) to undergo 1,6-
elimination may be attributed to the high pKa of the
aliphatic alcohol-containing leaving group.
Further studies are necessary to explore the scope and
the biological applications of the enzymatically induced
reaction that leads to ether fragmentation. We demon-
strated the feasibility of the reaction using cathepsin B,
a lysosomal enzyme that is highly up-regulated in the
cytosol and on the membranes of several metastatic
cancers.1-3 This enzyme is known to cleave the serum-
stable val-cit sequence used in these studies but cleaves
other peptide sequences such as val-lys and phe-lys much
more rapidly.10 Interestingly, the latter sequence is a
substrate not only for cathepsin B but also for plasmin,8-14
another lysosomal enzyme associated with various ma-
lignancies. We are investigating several such peptides,
with the goal of finding particular sequences that lead
to efficient intratumoral drug release and minimal
release in nontarget tissues. In addition, investigation
of other phenolic cytotoxic agents, such as the campto-
thecin family member SN-38,34 phenol mustard,35 minor
groove binders related in structure to CC-1065,36 strep-
tonigrin,37 and elliptininium acetate,38 will expand the
methodology to include drugs with widely different
mechanisms of activities and structures. Some of these
agents are among the most potent small molecular
weight drugs known, and their corresponding prodrugs
may be therapeutically effective in tumor types that are
insensitive for kinetic reasons to prodrugs of etoposide
and combretastatin A-4. In summary, we have demon-
strated that p-aminobenzyl ethers undergo a facile 1,6-
elimination reaction and that conditionally stable anti-
cancer prodrugs can be developed on the basis of this
reaction.
Exp er im en ta l Section
Gen er a l Meth od s. Commercially available reagents and
solvents were obtained as follows: HPLC-grade solvents,
Fisher; anhydrous solvents, Aldrich; diisopropyl azodicarboxy-
late (DIAD, 95%), Lancaster; 4-aminobenzyl alcohol, Alfa
(34) Senter, P. D.; Beam, K. S.; Mixan, B.; Wahl, A. F. Bioconjugate
Chem. 2001, 12, 1074-1080.
(35) Wallace, P. M.; Senter, P. D. Bioconjugate Chem. 1991, 2, 349-
352.
(36) Denny, W. A. Curr. Med. Chem. 2001, 8, 533-544.
(37) Bolzan, A. D.; Bianchi, M. S. Mutat. Res. 2001, 488, 25-37.
(38) Malonne, H.; Atassi, G. Anti-Cancer Drugs 1997, 8, 811-822.
(33) Shaikh, A. A. G.; Sivaram, A. Chem. Rev. 1996, 96, 951-976.

Fragmentation of p-Amidobenzyl Ethers
J . Org. Chem., Vol. 67, No. 6, 2002 1871
Aesar; Z-val-OSu, Advanced ChemTech; L-citrulline, Novabio-
chem; (1S,2R)-(+)-norephedrine and other commercially avail-
able reagents, Aldrich. Z-val-cit-PAB-OH (1)10,11 and combre-
tastatin A-4 (7)39 were synthesized as previously described.
1H NMR spectra were recorded on a Varian Gemini spectro-
photometer at 300 MHz. Flash column chromatography was
performed using 230-400 mesh ASTM silica gel from EM
Science. Analtech silica gel GHLF plates were used for thin-
layer chromatography. HPLC was performed using a Waters
Alliance system with a photodiode array detector. Combustion
analyses were determined by Quantitative Technologies, Inc.,
Whitehouse, NJ .
Gen er a l P r oced u r e for th e Mitsu n obu Rea ction . Pep-
tide 1 (1.0 equiv), triphenylphosphine (1.1 equiv) and the
appropriate phenol (1.0-1.1 equiv) were dissolved in DMF/
toluene (1:1) and evaporated to dryness under high vacuum.
The residue was taken up in dry DMF while under N2 and
cooled to 0 °C. DIAD (1.1 equiv) was added dropwise over 1
min while stirring. The yellow-brown solution was monitored
by TLC (9:1 CH2Cl2-MeOH). An additional 1.1 equiv of PPh3
and DIAD was added after 4 h. The solution was stirred for a
total of 16-24 h, followed by solvent removal in vacuo. The
resulting product was purified by chromatography on silica
gel (eluent gradient: 100% CH2Cl2 to 9:1 CH2Cl2-MeOH). The
desired fractions were pooled and concentrated to a white or
an off-white solid. Further purification could be obtained by
triturating with ether.
Z-Va l-cit-P AB-1-O-n a p h th ol (2): Rf 0.26 (9:1 CH2Cl2-
MeOH); mp 175 dec; UV λmax 215, 242, 305 nm; LRMS (ESI+)
m/z 640.3 (M + H)+, 662.2 (M + Na)+, 678.2 (M + K)+; 1H NMR
(DMSO-d6) δ 8.19-8.24 (2 H, m, aromatic), 7.80-7.87 (2 H,
m, aromatic), 7.06-7.50 (13 H, m, aromatic), 6.76 (1 H, d, J )
7.8 Hz, aromatic), 5.07 (2 H, s, Z-CH2), 4.44-4.50 (1 H, m,
val-CH), 4.28 (2 H, s, PAB-CH2), 3.95 (1 H, d, J ) 6.9 Hz, cit-
CH), 3.03-3.20 (2 H, m, cit-NCH2), 1.95-2.10 (1 H, m, val-
CH), 1.28-1.78 (4 H, m, cit-CH2’s), 0.96 (3 H, d, J ) 6.9 Hz,
val-CH3), 0.93 (3 H, d, J ) 6.9 Hz, val-CH3). Anal. (C36H41N5O6‚
H2O) C, H, N.
Z-Va l-cit-P AB-O-tr ich lor oa ceta m id a te (3). Peptide 1
(100 mg, 0.19 mmol) was dissolved in anhydrous DMF, and
cesium carbonate (13 mg, 4 µmol, 0.2 equiv) was added. While
under N2, trichloroacetonitrile (0.2 mL, 1.9 mmol, 10 equiv)
was added, and the contents were stirred while being moni-
tored by TLC (9:1 CH2Cl2-MeOH) . The reaction was complete
after 16 h. The mixture was filtered, and the filtrate was
concentrated and subjected to chromatography on SiO2 (eluent
gradient 100% CH2Cl2 to 9:1 CH2Cl2-MeOH containing 1%
triethylamine). The desired fractions were pooled and evapo-
rated to an off-white powder (99 mg, 77%): Rf 0.44 (9:1
CH2Cl2-MeOH); UV λmax 215, 250 nm; LRMS (ESI+) m/z 679.3
(M + Na)+, 681.2 (M + 2 + Na)+, 683.2 (M + 4 + Na)+, 685.2
(M + 6 + Na)+, 695.2 (M + K)+, 697.2 (M + 2 + K)+, 699.2 (M
+ 4 + K)+; 1H NMR (DMSO-d6) δ 10.08 (1 H, s, PAB-NH),
9.37 (1 H, s, CdNH), 8.10 (1 H, d, J ) 7.8 Hz, amide NH),
7.60 (2 H, d, J ) 8.4 Hz, PAB-CH ×2), 7.21-7.38 (8 H, m,
aromatic), 5.96 (1 H, t, J ) 5.1 Hz, cit-NH), 5.40 (2 H, s, cit-
NH2), 5.21 (2 H, s, PAB-CH2), 5.02 (2 H, s, Z-CH2), 4.40 (1 H,
dd, J ) 13.2, 7.8 Hz, val-CH), 3.90 (1 H, t, J ) 8.4 Hz, cit-
CH), 2.85-3.15 (2 H, m, cit-CH2), 1.90-2.05 (1 H, m, val-CH),
1.28-1.74 (4 H, m, cit-CH2), 0.86 (3 H, d, J ) 6.6 Hz, val-
CH3), 0.82 (3 H, d, J ) 6.9 Hz, val-CH3). Anal. (C28H35Cl3N6O6‚
0.2H2O, 0.4Et3N) C, H, N, Cl.
was purified by chromatography on SiO2 (1:1 CH2Cl2-EtOAc),
and the combined fractions were concentrated to a clear oil
that solidified. Recrystallization from EtOAc-hexanes gave a
white cottonlike solid as the desired product (4.95 g, 79%): mp
123 °C; Rf 0.14 (1:1 CH2Cl2-EtOAc); UV λmax 215, 256 nm; 1H
NMR (CDCl3) δ 7.28-7.39 (5 H, m, aromatic), 5.59 (1 H, br d,
J ) 8.4 Hz, NH), 4.87 (1 H, d, J ) 3.6 Hz, H-1), 4.34 (1 H, dp,
J ) 3.0, 6.9 Hz, H-2), 3.48 (1 H, br s, OH), 2.01 (3 H, s, Ac),
1.02 (3 H, d, J ) 6.9 Hz, H-3). Anal. (C11H15NO2) C, H, N.
Z-Va l-cit-P AB-O-(N-Ac)-Nor (5). The trichloroacetamidate
3 (1 equiv) and alcohol 4 (1 equiv) were suspended in
anhydrous CH2Cl2 and cooled to 0 °C. Dropwise addition of
trifluoromethanesulfonic acid (0.5 equiv) gave an immediate
gummy precipitate. TLC analysis (9:1 CH2Cl2-MeOH) showed
a product (Rf 0.28) and some decomposition of 3 to Z-val-cit-
PAB-OH 1 (Rf 0.14). The contents were evaporated to a yellow
solid and purified by chromatography on SiO2 (eluent gradient
100% CH2Cl2 to 9:1 CH2Cl2-MeOH). The desired ether (5) was
isolated as an off-white solid after triturating with diethyl
ether: Rf 0.28 (9:1 CH2Cl2-MeOH); UV λmax 215, 256 nm;
LRMS (ESI+) m/z 688.4 (M + H)+, 711.4 (M + Na)+; 1H NMR
(DMSO-d6) δ 10.02 (1 H, s, PAB-NH), 8.10 (1 H, d, J ) 7.2
Hz, amide NH), 7.81 (1 H, d, J ) 9.0 Hz, amide NH), 7.56 (2
H, d, J ) 8.7 Hz, PAB-CH ×2), 7.21-7.39 (12 H, m, aromatic),
5.98 (1 H, t, J ) 5.1 Hz, cit-NH), 5.41 (2 H, s, cit-NH2), 5.03 (2
H, s, Z-CH2), 4.06-4.45 (4 H, m, val-CH, Nor-CH, PAB-CH2),
3.84-3.94 (2 H, m, cit-CH, Nor-CH), 2.85-3.15 (2 H, m, cit-
CH2), 1.87-2.04 (1 H, m, val-CH), 1.67 (3 H, s, Nor-Ac), 1.28-
1.75 (4 H, m, cit-CH2’s), 0.98 (3 H, d, J ) 6.6 Hz, Nor-CH3),
0.86 (3 H, d, J ) 6.6 Hz, val-CH3), 0.82 (3 H, d, J ) 6.9 Hz,
val-CH3). Anal. (C37H48N6O7‚H2O) C, H, N.
Z-Va l-cit-P AB-O-etop osid e (6). Following the Mitsunobu
procedure described above, the pure fractions from chroma-
tography on SiO2 gave the ether as a white solid (64%): Rf
0.29 (9:1 CH2Cl2-MeOH); UV λmax 215, 250, 290 nm; LRMS
(ESI+) m/z 1084.6 (M + H)+, 1106.6 (M + Na)+, 1122.6 (M +
K)+; 1H NMR (DMSO-d6) δ 10.01 (1 H, s, PAB-NH), 8.08 (1 H,
d, J ) 7.2 Hz, amide NH), 7.57 (2 H, d, J ) 8.1 Hz, PAB-CH
×2), 7.29-7.40 (7 H, m, aromatic), 7.00 (2 H, s, etop aromatic),
6.53 (1 H, s, etop aromatic), 6.23 (2 H, s, etop aromatic), 6.01
(1 H, d, J ) 3.3 Hz, etop-CH2), 5.96 (1 H, t, J ) 5.1 Hz, cit-
NH), 5.40 (2 H, s, cit-NH2), 5.24 (1 H, s, etop-OH), 5.22 (1 H,
s, etop-OH), 5.02 (2 H, s, Z-CH2), 4.92 (1 H, d, J ) 3.0 Hz,
etop-CH), 4.74 (2 H, s, PAB-CH2), 4.70 (1 H, dd, J ) 9.9, 4.8
Hz, etop-CH), 4.56 (1 H, d, J ) 7.8 Hz, etop-CH), 4.54 (1 H, d,
J ) 5.1 Hz, etop-CH), 4.36-4.44 (1 H, m, val-CH), 4.25 (2 H,
dd, J ) 9.0 Hz, etop-CH ×2), 4.06 (1 H, dd, J ) 11.1, 4.8 Hz,
etop-CH), 3.90 (1 H, t, J ) 6.9 Hz, cit-CH), 3.62 (6 H, s, etop-
OCH3 ×2), 3.49 (1 H, t, J ) 9.6 Hz, etop-CH), 2.81-3.30 (9 H,
m, etop-CH ×7, cit-NCH2), 1.88-2.05 (1 H, m, val-CH), 1.30-
1.74 (4 H, m, cit-CH2’s), 1.22 (3 H, d, J ) 4.8 Hz, etop-CH3),
0.86 (3 H, d, J ) 6.6 Hz, val-CH3), 0.82 (3 H, d, J ) 6.9 Hz,
val-CH3). Anal. (C55H65N5O18‚2H2O) C, H, N.
Z-Va l-cit-P AB-3′-O-com br eta sta tin A-4 (9). Using the
Mitsunobu reaction conditions described above, the compound
was isolated as an amorphous solid after trituration. Rf 0.42
(9:1 CH2Cl2-MeOH); mp 169-172 dec; UV λmax 215, 248, 300
nm; LRMS (ESI+) m/z 812.4 (M + H)+, 834.4 (M + Na)+, 850.4
(M + K)+; 1H NMR (DMSO-d6) δ 10.06 (1 H, s, PAB-NH), 8.11
(1 H, d, J ) 7.2 Hz, amide NH), 7.57 (2 H, d, J ) 8.4 Hz, PAB-
CH ×2), 7.24-7.45 (6 H, m, aromatic), 7.21 (2 H, d, J ) 8.4
Hz, PAB-CH ×2), 6.82-6.98 (2 H, m, CSA4-H-5′, -6′), 6.56 (2
H, s, CSA4-H-2), 6.48 (2 H, d, J ) 12.3 Hz, CSA4-cis-CH), 6.44
(2 H, d, J ) 12.3 Hz, CSA4-cis-CH), 5.97 (1 H, t, J ) 5.1 Hz,
cit-NH), 5.41 (2 H, s, cit-NH2), 5.03 (2 H, s, Z-CH2), 4.76 (2 H,
s, PAB-CH2), 4.36-4.45 (1 H, m, val-CH), 3.92 (1 H, t, J ) 7.2
Hz, cit-CH), 3.82 (3 H, s, CSA4-3′-OCH3), 3.61 (9 H, s, CSA4-
3,4,5-OCH3), 2.88-3.07 (2 H, m, cit-NCH2), 1.90-2.03 (1 H,
m, val-CH), 1.28-1.78 (4 H, m, cit-CH2’s), 0.86 (3 H, d, J )
6.6 Hz, val-CH3), 0.82 (3 H, d, J ) 6.9 Hz, val-CH3). Anal.
(C44H53N5O10‚H2O) C, H, N.
(1S,2R)-N-Acetyl-n or ep h ed r in e (4). (1S,2R)-(+)-norephe-
drine (5.0 g, 32.4 mmol) was partially suspended in water (65
mL, 0.5 M). Acetic anhydride (6.2 mL, 64.8 mmol, 2.0 equiv)
was added, and the resulting yellow solution was stirred for 1
h. EtOAc was added, the layers were separated, and the
aqueous layer was further washed with EtOAc (2×). The
combined organic extracts were washed with brine and dried
(MgSO4). Filtration, followed by removal of solvent, led to a
yellow oil that slowly formed yellow crystals. The crude product
(1S,2R)-N-Acet yl-O-(4-n it r op h en yloxyca r b on yl)n or -
ep h ed r in e (10). Compound 4 (1.0 g, 5.17 mmol, 1.0 equiv)
and p-nitrophenylchloroformate (1.61 g, 7.76 mmol, 1.5 equiv)
were dissolved in anhydrous THF (12 mL, 0.5 M) while under
(39) Pettit, G. R.; Singh, S. B.; Boyd, M. R.; Hamel, E.; Pettit, R.
K.; Schmidt, J . M.; Hogan, F. J . Med. Chem. 1995, 38, 1666-1672.

1872 J . Org. Chem., Vol. 67, No. 6, 2002
Toki et al.
N2. Dry pyridine (0.63 mL, 7.76 mmol, 1.0 equiv) was added
via syringe over a 3 min period. The resulting turbid mixture
contained no starting material after 15 min, according to TLC
(1:1 CH2Cl2-EtOAc). Solids were filtered off and washed with
THF. The filtrate was concentrated to a yellow oil that was
purified by chromatography on SiO2 (1:1 hexanes-EtOAc). The
desired product 10 was an off-white solid (1.43 g, 78%) that
was stored in the dark at <0 °C: Rf 0.16 (1:1 hexanes-EtOAc);
UV λmax 215, 270 nm; 1H NMR (CDCl3) δ 8.24 (2 H, d, J ) 9.3
Hz, Pnp-CH ×2), 7.38 (2 H, d, J ) 9.0 Hz, Pnp-CH ×2), 7.32-
7.44 (5 H, m, aromatic), 5.78 (1 H, d, J ) 3.3 Hz, H-1), 5.42 (1
H, br d, J ) 8.4 Hz, NH), 4.61 (1 H, dp, J ) 3.3, 7.2 Hz, H-2),
2.00 (3 H, s, Ac), 1.11 (3 H, d, J ) 7.2 Hz, H-3).
H, t, J ) 7.2 Hz, cit-CH), 3.73 (3 H, s, CSA4-3′-OCH3), 3.61 (3
H, s, CSA4-4-OCH3), 3.58 (6 H, s, CSA4-3,5-OCH3), 2.88-3.07
(2 H, m, cit-NCH2), 1.88-2.04 (1 H, m, val-CH), 1.28-1.178
(4 H, m, cit-CH2’s), 0.86 (3 H, d, J ) 6.9 Hz, val-CH3), 0.82 (3
H, d, J ) 6.6 Hz, val-CH3). Anal. (C45H53N5O12‚H2O) C, H, N.
Gen er a l P r oced u r e for Ca th ep sin B Assa ys. Bovine
spleen cathepsin B was obtained from Sigma-Aldrich. SDS-
PAGE analysis revealed bands corresponding to molecular
masses of 24 and 29 kDa, as expected for the fully processed
protein.40 Enzyme activity was assayed using the cathepsin B
substrate Z-lysine nitrophenyl ester, according to a published
procedure.41 The average specific activity was 26 µmol/min/
mg of enzyme, which fell within the indicated range of 18-35
µmol/min/mg. The enzyme, dissolved in phosphate-buffered
saline (pH 7.2, 1 mg/mL final concentration), was activated
with dithiothreitol and EDTA, as previously described.41 A 1.0
mM stock solution of the peptide substrate in DMSO was
added to acetate buffer (25 mM) containing 1 mM EDTA (pH
5.1) to a final concentration of 0.08-0.14 mM, and to this
solution was added the activated enzyme (12-15 µg/mL). In
the case of the naphthol ether 2, a 5.0 mM solution in MeOH
was diluted to a final concentration of 0.22 mM. Periodically,
aliquots were taken, quenched with an equal volume of MeCN,
and centrifuged, and 100 µL injections were analyzed by HPLC
(4.6 mm × 15 cm C18 column) with detection between 210 and
400 nm. The mobile phase consisted of (A) 5 mM sodium
phosphate (pH 7) and (B) either MeOH (for compounds 2, 9,
and 13) or MeCN (for compounds 5, 8, and 12). The gradient
elution was 90% to 10% A over 10 min, followed by 5 min at
10% A, and the flow rate was 1.0 mL/min. The disappearance
of substrate and the appearances of released alcohol and Z-val-
cit were recorded. Cathepsin B hydrolysis rates were calculated
according to the disappearance of substrate (Table 1).
Gen er a l P r oced u r e for All Sta bility Stu d ies. Solutions
of the substrates (0.08-0.14 mM in DMSO, and 0.22 mM in
MeOH for 2) were diluted 10-20-fold in PBS, acetate buffer
(25 mM, pH 5.1), or pooled human serum, and incubation was
carried out at 37 °C. For the serum studies, equal volumes of
MeCN were added, and the samples were centrifuged prior to
HPLC analysis. The other samples were injected directly into
the HPLC.
In Vitr o Cytotoxicity Assa ys. L2987 human lung adeno-
carcinoma cells were obtained as previously described.42
WM266/4 and IGR-39 human melanoma cells were obtained
from ATCC (Manassas, VA) and DSMZ (Braunschweig, Ger-
many), respectively. L2987 and WM266/4 cells were grown in
Roswell Park Memorial Institute (RPMI) medium containing
10% fetal bovine serum, 10 units/mL penicillin G, and 10 µg/
mL streptomycin sulfate. Dulbecco’s modified Eagle’s medium
was used in place of RPMI for the IGR-39 cells. The cells (2500
cells in 0.1 mL of medium) were plated into 96-well plates,
and after 24 h at 37 °C, various concentrations of the drugs in
the medium (50 µL) were added in triplicate. Incubation was
continued for an additional 24 h, the cultures were washed,
and fresh medium (0.15 mL) was added. After 48 h at 37 °C,
3H-thymidine (25 µL, 0.5 µCi/well) was added, and the cultures
were frozen, thawed, and harvested onto glass fiber filters 4
h later. Incorporation of the label into nascent DNA was
measured using a â-counter.
3′-O-(4-Nit r op h en yloxyca r bon yl)com br et a st a t in A-4
(11). Using the same procedure as described above, combre-
tastatin A-4 (120 mg, 0.38 mmol) was converted to the
4-nitrophenyl carbonate in quantitative yield (183 mg) and
isolated as a yellow oil: Rf 0.47 (3:2 hexanes-EtOAc); 1H NMR
(CHCl3) δ 8.30 (2 H, d, J ) 9.3 Hz, Pnp-CH ×2), 7.45 (2 H, d,
J ) 9.3 Hz, Pnp-CH ×2), 7.18 (2 H, d, J ) 1.5 Hz, H-2′), 6.88-
6.94 (2 H, m, H-5′,-6′), 6.51 (1 H, d, J ) 12.0 Hz, cis-CH), 6.49
(2 H, s, H-2), 6.48 (1 H, d, J ) 12.0 Hz, cis-CH), 3.89 (3 H, s,
3′-OCH3), 3.84 (3 H, s, 4-OCH3), 3.70 (3 H, s, 3,5-OCH3).
Z-Val-cit-P AB-OCO-(1S,2R)-(N-acetyl)n or eph edr in e (12).
The activated carbonate 10 (90 mg, 0.25 mmol) and Z-val-cit-
PAB-OH 1 (130 mg, 0.25 mmol) were suspended in dry
CH2Cl2 (8 mL), followed by the addition of DMAP (34 mg, 0.28
mmol, 1.1 equiv). The reaction was stopped after 26 h by the
addition of EtOAc and 10% citric acid. The layers were
separated, and the organic phase was further washed with
water and brine. A precipitate formed that was filtered and
added to the separated EtOAc layer and concentrated. The
resulting yellow solid was subjected to chromatography on SiO2
(gradient eluent 95:5 to 9:1 CH2Cl2-MeOH). The desired
product eluted first and was concentrated to a white flaky solid
(35 mg, 19%), while Z-val-cit-PAB-OH (1) was recovered as
the second eluate: Rf 0.17 (9:1 CH2Cl2-MeOH); UV λmax 215,
256 nm; LRMS (ESI+) m/z 792.5 (M + H)+, 814.5 (M + Na)+,
830.4 (M + K)+; 1H NMR (DMSO-d6) δ 10.08 (1 H, s, PAB-
NH), 8.10 (1 H, d, J ) 7.2 Hz, amide NH), 8.00 (1 H, d, J )
7.8 Hz, amide NH), 7.59 (2 H, d, J ) 8.7 Hz, PAB-CH ×2),
7.26-7.39 (10 H, m, aromatic), 7.24 (2 H, d, J ) 8.7 Hz, PAB-
CH ×2), 5.96 (1 H, t, J ) 5.1 Hz, cit-NH), 5.61 (1 H, d, J ) 4.2
Hz, Nor-CH), 5.40 (2 H, s, cit-NH2), 5.07 (2 H, s, Z-CH2), 5.03
(2 H, s, PAB-CH2), 4.40 (1 H, dd, J ) 13.2, 7.8 Hz, val-CH),
4.03-4.14 (1 H, m, Nor-CH), 3.92 (1 H, t, J ) 7.8 Hz, cit-CH),
2.85-3.06 (2 H, m, cit-CH2), 1.90-2.02 (1 H, m, val-CH), 1.74
(3 H, s, Nor-Ac), 1.28-1.75 (4 H, m, cit-CH2’s), 0.96 (3 H, d,
J ) 6.9 Hz, Nor-CH3), 0.87 (3 H, d, J ) 6.9 Hz, val-CH3),
0.83 (3 H, d, J ) 7.2 Hz, val-CH3). Anal. (C38H48N6O9‚1/2H2O)
C, H, N.
Z-Va l-cit-P AB-OCO-com br eta sta tin A-4 (13). Activated
combretastatin A-4 11 (120 mg, 0.25 mmol) and Z-val-cit-PAB-
OH 1 (130 mg, 0.25 mmol) were suspended in dry CH2Cl2/
pyridine (3 mL each), followed by the addition of DMAP (34
mg, 0.28 mmol, 1.1 equiv). The reaction was sonicated for 2 h,
followed by stirring for 20 h. Evaporation of the reaction
mixture, followed by purification by chromatography on SiO2
(gradient eluent 100% CH2Cl2 to 9:1 CH2Cl2-MeOH) and
concentration of the appropriate fractions, resulted in a yellow
oil that was precipitated from CH2Cl2 (1 mL) through the
addition of ether, which led to a yellow solid (83 mg, 38%): Rf
0.47 (9:1 CH2Cl2-MeOH); mp 155-158 dec; UV λmax 215, 245,
285 nm; LRMS (ESI+) m/z 856.5 (M + H)+, 878.5 (M + Na)+,
894.5 (M + K)+; 1H NMR (DMSO-d6) δ 10.10 (1 H, s, PAB-
NH), 8.11 (1 H, d, J ) 7.8 Hz, amide NH), 7.62 (2 H, d, J )
8.1 Hz, PAB-CH ×2), 7.25-7.40 (8 H, m, aromatic), 7.15 (1 H,
dd, J ) 8.7, 1.8 Hz, CSA4-H-6′), 7.07 (1 H, d, J ) 8.1 Hz, CSA4-
H-5′), 7.06 (1 H, d, J ) 2.4 Hz, CSA4-H-2′), 6.51 (2 H, s, CSA4-
H-2), 6.48 (2 H, s, CSA4-CHdCH), 5.97 (1 H, t, J ) 5.1 Hz,
cit-NH), 5.42 (2 H, s, cit-NH2), 5.14 (2 H, s, Z-CH2), 5.02 (2 H,
s, PAB-CH2), 4.40 (1 H, dd, J ) 12.9, 8.1 Hz, val-CH), 3.92 (1
Ack n ow led gm en t. This work was supported in part
by Grant 1R43 CA 88583-01A1 from the National
Cancer Institute.
J O016187+
(40) Otto, H.-H.; Schirmeister, T. Chem. Rev. 1997, 97, 133-171.
(41) Bajkowski, A. S.; Frankfater, A. Anal. Biochem. 1975, 68, 119-
127.
(42) Svensson, H. P.; Vrudhula, V. M.; Emsweiler, J . E.; MacMaster,
J . F.; Cosand, W. L.; Senter, P. D.; Wallace, P. M. Cancer Res. 1995,
55, 2357-2365.