Studies on synthesis and activation mechanism of mitomycin dimers connected by 1,2-dithiolane and diol linkers

Article abstract of DOI:10.1016/j.bmc.2012.08.015

We report the synthetic and mechanistic studies on a new cyclic disulfide mitomycin dimer, 7-N,7′-N′-(1″,2″-dithiolanyl-3″, 5″-dimethylenyl)bismitomycin C (8), and a diol mitomycin dimer, 7-N,7′-N′-(2″,4″-dihydroxy-1″,5″- pentanediyl)bismitomycin C (9). Mitomycin 8 is a dimer connected by a 1,2-dithiolane (a five-membered cyclic disulfide) linker, and was specifically designed to undergo nucleophilic activation and double DNA alkylations leading to efficient production of DNA interstrand cross-link (DNA ISC) adducts. Disulfide cleavage in 8 would generate two thiol groups that could serve as probes to activate two mitomycin rings. At first, the target mitomycin 8 was synthesized using mitomycin A (1) and the key intermediate, cyclic disulfide (10), which was prepared through a seven-step synthetic sequence. Diol mitomycin 9 was also synthesized from 1 and diamine salt 13. Next, kinetic studies using solvolysis reaction revealed that the activation rates of 8 were much higher than those of 9 and mitomycin C (2) under nucleophilic conditions provided by Et₃P presumably due to the presence of a cyclic disulfide unit in 8. These findings led us to propose a nucleophilic activation pathway for 8. Then, DNA ISC experiments further revealed that the levels of DNA ISC caused by 8 in the presence of Et₃P were much higher (97%) than those by 9 (5%) and 2 (4%). More importantly, mitomycin 8 underwent much faster activation and produced slightly higher levels of DNA ISC than the previously reported mitomycins 5-7. Overall, we concluded that 8 was highly efficient for both nucleophilic activation and corresponding DNA ISC formation, and that this differentiation came from the crucial function of the cyclic disulfide unit in 8.

Full text of DOI:10.1016/j.bmc.2012.08.015

Bioorganic & Medicinal Chemistry 20 (2012) 5720–5729
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Studies on synthesis and activation mechanism of mitomycin dimers
connected by 1,2-dithiolane and diol linkers
⇑
Hyoung Rae Kim, Jae Jin Kim, Jung Jae Park, Sang Hyup Lee
College of Pharmacy, Duksung Women’s University, 419 Ssangmun-dong, Dobong-gu, Seoul 132-714, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
We report the synthetic and mechanistic studies on a new cyclic disulfide mitomycin dimer, 7-N,70-N0-(100,200-
dithiolanyl-300,500-dimethylenyl)bismitomycin C (8), and a diol mitomycin dimer, 7-N,70-N0-(200,400-dihydroxy-
100,500-pentanediyl)bismitomycin C (9). Mitomycin 8 is a dimer connected by a 1,2-dithiolane (a five-membered
cyclic disulfide) linker, and was specifically designed to undergo nucleophilic activation and double DNA alky-
lations leading to efficient production of DNA interstrand cross-link (DNA ISC) adducts. Disulfide cleavage in 8
would generate two thiol groups that could serve as probes to activate two mitomycin rings. At first, the target
mitomycin 8 was synthesized using mitomycin A (1) and the key intermediate, cyclic disulfide (10), which was
prepared through a seven-step synthetic sequence. Diol mitomycin 9 was also synthesized from 1 and diamine
salt 13. Next, kinetic studies using solvolysis reaction revealed that the activation rates of 8 were much higher
than those of 9 and mitomycin C (2) under nucleophilic conditions provided by Et3P presumably due to the
presence of a cyclic disulfide unit in 8. These findings led us to propose a nucleophilic activation pathway
for 8. Then, DNA ISC experiments further revealed that the levels of DNA ISC caused by 8 in the presence of
Et3P were much higher (97%) than those by 9 (5%) and 2 (4%). More importantly, mitomycin 8 underwent much
faster activation and produced slightly higher levels of DNA ISC than the previously reported mitomycins 5–7.
Overall, we concluded that 8 was highly efficient for both nucleophilic activation and corresponding DNA ISC
formation, and that this differentiation came from the crucial function of the cyclic disulfide unit in 8.
Ó 2012 Elsevier Ltd. All rights reserved.
Article history:
Received 13 July 2012
Revised 8 August 2012
Accepted 9 August 2012
Available online 16 August 2012
Keywords:
Disulfide mitomycin
Mitomycin dimer
1,2-Ditholane
Nucleophilic activation
DNA interstrand cross-link
1. Introduction
disulfide mitomycins 3 (KW-2149)6 and 4 (BMS-181174)7 were
developed. Significantly, the disulfide mitomycins 3 and 4 con-
Mitomycins are DNA alkylating agents that constitute a class of
potent antitumor agents. Although mitomycin A (MMA, 1)1 was
the first compound in this series, mitomycin C (MMC, 2) is consid-
ered a prototype compound and has been used as a potent agent of
clinical significance.1,2 Mitomycins must be converted to good
electrophiles through proper activation processes in order to exhi-
bit biological activities. Studies on activation mechanism revealed
that the reduction of the quinone ring initiates activation leading
to the generation of electrophilic C(1) and C(10) sites that react
with DNA,3 which was reductive activation. Generally, the C(1) site
is 10–100 times more reactive than the C(10) site.4 Among plausi-
ble DNA alkylation of mitomycins, the DNA interstrand cross-link
(DNA ISC) adducts by bis-alkylation are the most lethal, and in par-
ticular, the DNA ISC adducts of 2 display ꢀ60 times greater lethal-
ity than the corresponding monoadducts of 2.5 This clearly implies
the significance of DNA ISC adducts in this series of compounds.
However, long-term administration of 2 has been found to in-
duce severe drug resistance and side effects.2 Therefore, subse-
quent efforts have focused on finding mitomycin derivatives that
could be activated by a different activation mechanism. As a result,
tained, in particular, a C(7) disulfide ethylamino unit instead of
the C(7) amino unit in 2, and were believed to work by a different
activation mechanism than 2 based on enhanced pharmacological
activities in 2-resistant tumor cell lines and in non-hypoxic cells.8–
10
Interestingly, at the core of the activation mechanism lies the
disulfide group.6,7,10 Disulfide cleavage in 3 and 4 provides corre-
sponding thiols that could trigger the activation of the mitomycin
ring by intramolecular cyclization to the quinone ring,11,12 and in-
duce the corresponding DNA adduction.11–13 These processes are
considered a nontraditional nucleophilic activation mechanism
that differs from those of 2.
O
10
OC(O)NH2
OC(O)NH2
O
RS
H
N
7
6
R
9
OMe
8
OMe
NH
9a
1
N
NH
N
O
O
O
1
2
S
S
CO2H
NH2
R = OCH3
R = NH2
3 R =
4 R =
N
H
NO2
⇑
Corresponding author. Tel.: +82 2 901 8393; fax: +82 2 901 8386.
E-mail address: sanghyup@duksung.ac.kr (S.H. Lee).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.08.015

H. R. Kim et al. / Bioorg. Med. Chem. 20 (2012) 5720–5729
5721
linker based on the same carbon skeleton as 8, and could therefore
serve as a good reference to identify the net effect of the disulfide
unit in 8. The objectives of this work were to study the design and
synthesis, activation mechanism, and the DNA ISC formation for 8
compared to 9, to compare 8 with 5–7 and to elucidate the linker
effect. Considering that 8 and 9 retained all the structure of mito-
mycin C (2), we believed that 8 and 9 would undergo correspond-
ing activations under reductive3 and acidic conditions19 and that
these mitomycin dimers would display basic levels of cytotoxicity.
Furthermore, since we were mainly interested in the activation
mechanism of these peculiar mitomycins, we did not include the
cytotoxicity data of 8 and 9 in the scope of our studies.
As mentioned above, compound 8 was designed to combine two
effects: a disulfide effect and a dimerization effect. The disulfide ef-
fect represents nontraditional nucleophilic activation by the disul-
fide unit and the dimerization effect represents double C(1)
activations at two C(1) sites of higher reactivity provided by a di-
meric structure, which could eventually induce enhanced DNA
ISC formation. The disulfide cleavage or reduction in 8 may provide
a crucial dithiol species (8r: reduced form of 8) that contains two
thiols in the linker part. The two thiols in 8r were both expected
to work as good activation probes to activate two mitomycin rings
in a dimer by intramolecular cyclization to quinone ring and sub-
sequent activation processes,17,20 leading to the corresponding
DNA adduction.21 Unfortunately, the thiol that is required for acti-
vation was very unstable12 due to the presence of adjacent quinone
ring. Thus, the disulfide unit might be an appropriate, stable pre-
cursor for the required thiol, which was a key aspect in our design.
Despite comprehensive investigations on mitomycin activation,
much is still unclear about the reasonable structure of activated
products when reacting with DNA.
In addition, considering the significance of DNA ISC, we were
interested in dimerization of mitomycin units to enhance the for-
mation of DNA ISC. Mitomycin dimers would provide two C(1)
sites of higher reactivity and, as a result, would be expected to pro-
duce higher levels of DNA ISC by bis-alkylation at the two distal
C(1) sites. We had previously started a program to investigate
disulfide mitomycin dimers that contained a cyclic disulfide as a
linker. From the initial results of this program, we have already re-
ported the studies on the mitomycin dimers, 5,14 6,15,16 and 7.17,18
The structural differences among these compounds were the ring
size of the cyclic disulfide linkers. Comparing the three compounds
5–7 each other, we found that all compounds displayed high levels
of DNA ISC formation (P83%), and in detail, the levels of DNA ISC
were similar or slightly decreased (e.g., DNA ISC by Et3P:
94% ? 86% ? 83%) as the ring size of the cyclic disulfide linker de-
creased (5 ? 6 ? 7, respectively). However, the activation rates for
these compounds in the presence of nucleophiles (e.g., Et3P or
L-
dithiothreitol ( -DTT)) showed different patterns. For example,
L
the activation rates were decreased by almost half (e.g., kobs by
Et3P 10 equiv: 8.7 dꢁ1 ? 4.6 dꢁ1 ? 2.7 dꢁ1) as the ring size of the
cyclic disulfide linker decreased (5 ? 6 ? 7, respectively). These
differences between the activation rates and DNA ISC formation
according to the ring size of the linkers seemed to be quite mean-
ingful and led us to focus on the influence of the ring size of the lin-
ker on the activation and DNA ISC formation.
S
S
H2NC(O)O
MeO
OC(O)NH2
OMe
O
O
( )n
NH
HN
8
1'
1
HN
N
NH
N
O
O
5
n= 3
6 n= 2
7
n= 1
OH OH
H2N(O)CO
MeO
OC(O)NH2
OMe
H2N(O)CO
MeO
OC(O)NH2
OMe
O
8'
O
O
8'
O
8
S
S
H
N
H
N
H
N
H
N
8
1'
1
1
1'
HN
N
N
NH
HN
N
N
NH
O
8
O
O
O
9
disulfide cleavage
quinone ring activation
quinone ring activation
SH SH
H2N(O)CO
OC(O)NH2
OMe
O
8'
O
8
H
N
H
N
MeO
1'
1
HN
N
N
NH
O
O
8r
2. Results and discussion
2.1. Design and structural features of 8 and 9
When comparing compound 8 with 5–7, the only difference is
the ring size of the linker. Disulfide cleavage of compounds 5–8
provides the corresponding dithiol species 5r–8r, respectively,
which induce activation of mitomycin rings. Therefore, we as-
sumed that the formation and lifetime of thiol(s) in dithiol species,
and the flexibility and distance between the two C(1) sites would
be crucial for the activation and subsequent DNA adduction. In or-
der to verify this assumption, we investigated the effect of the ring
size of the linker using 8 and compared the results with those for
5–7. The distance between the two C(1) sites in 8 (or 8r) may be
slightly shorter than those in 5–7. The maximum distances in 8
were 23ꢀ24 Å depending on their conformations (Sybyl 6.0,
The results above prompted us to examine the influence of the
ring size of the cyclic disulfide linker on mitomycin activation. We
report herein the studies on new mitomycin dimers 8 and 9 and
the comparison of these compounds with the precedent examples
(e.g., 5–7). Compound 8 contains a 1,2-dithiolane (a five-mem-
bered cyclic disulfide) and was designed to undergo efficient nucle-
ophilic activation and corresponding facile DNA adduction.
Compound 9 contains a diol linker instead of a cyclic disulfide

5722
H. R. Kim et al. / Bioorg. Med. Chem. 20 (2012) 5720–5729
HyperChem 7.1) and differed from those in 5–7. In addition, the
lifetimes of corresponding thiol(s) in dithiol species would be dif-
ferent, which may directly affect the activation of mitomycins.
Based on the previous studies22,23 regarding thiol-disulfide inter-
change and their stabilities, it was believed that the cyclization
tendency of dithiol species would be 7r>>6rP8r>>5r.
product that showed a higher Rf value (0.53) than that (0.50) of
the desired product 16a. In our previous studies,15,17 we conducted
similar reactions of 15b and 15c using KSAc and obtained the
diacetylthio derivatives 16b and 16c in reasonably good yields,
60% and 74%, respectively. Considering these results, the reaction
of only 15a with KSAc might undergo some side reaction or
SH
SH
SH
H2N(O)CO
MeO
O
disulfide
cleavage
O
O
OC(O)NH2
H
N
H
N
5
6
7
OMe
SH
cyclization
1'
HN
N
N
N
1
N
NH
O
O
5r
SH
H2N(O)CO
MeO
O
O
OC(O)NH2
OMe
disulfide
cleavage
H
N
H
N
1
cyclization
1'
HN
N
NH
O
O
6r
H2N(O)CO
MeO
disulfide
cleavage
O
O
OC(O)NH2
OMe
H
N
H
N
SH
1'
HN
cyclization
1
N
NH
O
O
7r
SH SH
H2N(O)CO
MeO
O
OC(O)NH2
OMe
disulfide
cleavage
H
N
H
N
8
1
cyclization
1'
HN
N
N
NH
O
O
8r
2.2. Chemistry
2.2.1. Synthesis of cyclic disulfide 10
overreaction. The side product contained only one acetylthio group
and no mesylate group relative to two Boc groups based on the
NMR analysis. We suggest that the Boc-amino group probably par-
ticipates in removing the mesylate or acetylthio group. We also
found that the side product might be in equilibrium with 16a since
we observed that treatment of the side product with KSAc partly
produced 16a. Unfortunately, intensive efforts to clarify the reac-
tion and identify the structure of the side product were inconclu-
sive. Further attempts to improve the yield of the desired
product 16a were conducted by modulating the reaction condi-
tions such as amount of KSAc, temperature, and reaction time, fi-
nally affording 16a in 32% yield.
In order to synthesize the target mitomycin 8, we first synthe-
sized the required key intermediate, cyclic disulfide 10, as shown
in Scheme 1. Compound 10 and all of the related precursors shown
in Scheme 1 contain a symmetrical structure with two stereocen-
ters and could therefore exist as three stereoisomers (meso-(R,S)
and enantiomeric pair, threo-(R,R) and threo-(S,S)). Thus, we ana-
lyzed the diastereomeric ratio (dr) of compounds by 13C NMR.
At first, we prepared diepoxide 11 in good yield (90%, dr = 1.1:1)
according to previous procedures.24 Then, treatment of 11 with
phthalimide (PhtNH) afforded diphthalimide derivative 12 in rela-
tively low yield (30–40%) with a varying ratio of diastereomers
(1:1–3:1). We applied stepwise temperature control in order to
avoid loss of the starting diepoxide through facile evaporation,
and careful work-up including precipitation using diethyl ether
(Et2O) to provide 12 in good yield (76%, dr = 1.4:1). We further at-
tempted to improve the diastereomeric ratio by recrystallization in
DMF-Et2O and consequently obtained 12 in relatively low yield
(21%) but in high diastereomeric ratio (dr = 4:1). Subsequent
deprotection of phthalimide groups in 12 using hydrazine hydrate,
followed by treatment of hydrochloric acid, provided amine salt 13
(dr = 5:1) in 80% yield. Treatment of amines in 13 with di-t-butyl
dicarbonate (Boc2O) afforded 14 (dr: an apparent single set of sig-
nals in 13C NMR) in 70% yield. We then reacted 14 with methane-
sulfonyl chloride (MsCl) to obtain the dimesyl derivative 15a in
93% yield. Replacement of the mesylate units with acetylthio-
groups to give diacetylthio derivative 16a through the use of potas-
sium thioacetate (KSAc) in DMF was problematic. Surprisingly, in
this reaction we observed formation of a major unknown side
SAc SAc
BocHN
NHBoc
OMs OMs
16a
KSAc
(32%)
BocHN
NHBoc
+
15a
side product
(major)
R
R
R
BocHN
NHBoc
BocHN
NHBoc
R
R = OMs
R = SAc
15c
16c
R = OMs
R = SAc
15b
16b
KSAc
(74%)
KSAc
(60%)

H. R. Kim et al. / Bioorg. Med. Chem. 20 (2012) 5720–5729
5723
OH OH
O
O
OC(O)NH2
OMe
O
O
a
PhtN
NPht
TFA
.
MeO
TFA
.
a
S
S
H2N
NH2
8
+
11
12
N
NH
10
b
1
(MMA)
OH OH
OH OH
c
HCl
.
HCl
.
BocHN
BocHN
NHBoc
H2N
NH2
O
OC(O)NH2
OMe
MeO
HCl
.
OH OH HCl
.
b
14
13
9
H2N
NH2
+
N
NH
d
13
O
1
SAc SAc
(MMA)
OMs OMs
e
g
BocHN
NHBoc
NHBoc
Scheme 2. Synthesis of mitomycins 8 and 9. Reagents and conditions: (a) Et3N,
MeOH, room temperature, 2 d, 66%; (b) Et3N, MeOH, room temperature, 2 d, 67%.
15a
16a
f
2.2.2. Synthesis of mitomycins 8 and 9
S
S
R
R
The target mitomycins 8 and 9 were synthesized using mitomy-
cin A (MMA, 1, KyowaHakko Kirin Co.) and the corresponding
intermediates 10 and 13, respectively, as shown in Scheme 2.
Treatment of 1 with intermediate 10 in MeOH provided the target
mitomycin 8 in 66% yield, and it was found that 8 was a mixture of
diastereomers (dr = 1.2:1) by analysis of 13C NMR signals. In addi-
tion, treatment of 1 with amine salt 13 afforded diol mitomycin 9
in 67% yield, and 9 was also found to be a mixture of diastereomers
(dr = 1.1:1) by analysis of 13C NMR signals. Mitomycin 9 was used
as a reference one for the studies on 8.
SH SH
BocHN
NHBoc
18 R = NHBoc
h
17
.
10 R = NH2 TFA
Scheme 1. Synthesis of cyclic disulfide 10. Reagents and conditions: (a) PhtNH,
DMF, 100 °C /1 h ? 115 °C /1 h ? 135 °C /1 h, 76%; then, recrystallization in DMF,
21%; (b) NH2NH2ꢂH2O, EtOH, reflux, 3.5 h; then HCl, reflux, 1 h, 80%; (c) Boc2O,
DMF–H2O (1:1), room temperature, 1 d, 70%; (d) MsCl, Et3N, CH2Cl2, 0 °C, 5 h, 93%;
(e) KSAc, DMF, 60 °C, 8 h, 32%; (f) K2CO3, MeOH-H2O (5:1), room temperature, 1 h;
(g) I2, Et3N, CHCl3, room temperature, 3 h, 69% (for two steps); (h) TFA, room
temperature, 1 h, 100%.
2.3. Activation studies on mitomycins 8 and 9
2.3.1. Methanolysis of mitomycin 8 to give mitosene product 19
We employed a methanolysis reaction of mitomycin as a chem-
ical model system to mimic the activation and subsequent reaction
with nucleophiles in biological systems. Methanolysis of 8 was
conducted for ꢀ2 d under acidic conditions (MeOH–CHCl3 (1:1)
solution at nonaqueous effective pH (‘pH’) 2.5–3.0) to afford the
activated mitosene product, C(1) methoxymitosene 19. As the
reaction proceeded, starting mitomycin 8 disappeared and mono-
activated intermediate was generated; the intermediate was even-
tually converted to diactivated mitosene product 19. However, the
diactivated mitosene product was highly polar and unstable, even
at room temperature, and displayed long tailing on TLC. Accord-
ingly, it was very difficult to isolate and purify the product. How-
ever, careful monitoring on TLC and HPLC enabled us to identify
the progress of the reaction. We purified the product by prepara-
tive thin layer chromatography (PTLC) with poor results (45%
yield). Further efforts to improve the results were inconclusive.
We believe that, as mentioned above, the mitosene product is
highly polar due to the presence of two newly generated primary
amine groups and is composed of a mixture of many diastereomers
due to the uncontrolled stereochemistry at two C(1) sites, without
considering the stereochemistry of cyclic disulfide linker. Despite
these complications, we succeeded in partially characterizing 19
using HPLC, UV–vis, 1H NMR, and mass spectroscopy. The HPLC
chromatogram showed three peaks (tR = 30.1, 31.3, 32.5 min,
ꢀ1:2:1) that might indicate the formation of corresponding diaste-
reomers of two unidentified stereocenters at C(1) and C(10). More
conclusively, the UV–vis spectra confirmed the activation of mito-
mycin leading to the generation of a mitosene product based on
absorption pattern.19,25 We observed an absorption maximum at
ꢀ313 nm for the mitosene product 19, not at ꢀ376 nm for starting
mitomycin 8. Furthermore, we observed the expected resonance
(d 3.52) for the C(1) methoxy units and the downfield resonance
Hydrolysis of 16a using potassium carbonate (K2CO3) was con-
ducted to give dithiol derivative 17 that was directly used for the
next reaction. Considering the instability of the dithiol species, we
performed the two reactions, hydrolysis and oxidation, in situ
without isolation of dithiol 17. At first, intramolecular cyclization
of dithiol units was achieved by oxidation using I2/Et3N in varying
concentrations to afford cyclic disulfide 18. In order to avoid the
formation of unwanted side products through intermolecular
disulfide formation,14 we then applied reaction conditions of high
dilution (ꢀ5 mM concentration) and obtained product 18 in rela-
tively good yield. However, we believed that the intramolecular
cyclization of 1,3-dithiol might be more favorable than that of
1,6-dithiol. We further investigated the reaction conditions of
higher concentrations and eventually established the reaction
condition of 50 mM concentration to give the desired product
18 in good yield (69%, for two steps). The product 18 was identi-
fied by NMR (1H and 13C), Mass, and IR, and more significantly, 18
was verified by reasonable Rf value (0.53, EtOAc/hexanes = 1:2) on
TLC by comparison with the values of 3,6-bis(tert-butyloxycarb-
onylaminomethyl)-1,2-dithiane (0.55),17 3,7-bis(tert-butyloxy-
carbonylaminomethyl)-1,2-dithiepane (0.56)15 and 3,8-bis(tert-
butyloxycarbonylaminomethyl)-1,2-dithiocane (0.56).14 Then,
deprotection of Boc groups in 18 using trifluoroacetic acid (TFA)
was executed to afford amineꢂTFA salt 10 in near quantitative
yield. Interestingly, compounds 14, 15a, 16a, 17, 18 and 10
showed an apparent single set of signals in 13C NMR and there-
fore, we believed that, among three possible stereoisomers, these
compounds including 10 existed as either a meso-isomer or an
enantiomeric pair, not as diastereomeric mixtures. Consequently,
the key intermediate, cyclic disulfide 10 was synthesized
from 11 through a seven-step synthetic sequence in 2.4% overall
yield.

5724
H. R. Kim et al. / Bioorg. Med. Chem. 20 (2012) 5720–5729
and specific coupling pattern (d 5.73 and 5.78, 1/2ABq) for the
C(10) methylene protons in the 1H NMR spectra for 19. These sig-
nals were known to be characteristic for the formation of C(1)
methoxymitosenes.26,27 Taken together, the mitosene product 19
was believed to be appropriately identified and was then employed
as an authentic sample for subsequent kinetic experiments.
10 and 20 equiv of
tinct decrease (less than 10% of original amount) of starting mito-
mycins 8 and 9 in 3 d, which indicated that neither -DTT nor GSH
significantly induced the nucleophilic activation of mitomycin di-
mers 8 and 9.
L-DTT, and 20 equiv of GSH, we observed no dis-
L
Then, we investigated the effect of Et3P as a phosphine nucleo-
phile on activation of
8 compared with 9. We found that
5–50 equiv of Et3P significantly induced a decrease of disulfide
mitomycin 8, which implies that 8 underwent rapid nucleophilic
activation by Et3P. The half-lives (t1/2) of the activation reactions
ranged from 0.075 d (5 equiv) to 0.011 d (50 equiv), and the activa-
tion rates were roughly proportional to the amount of Et3P. How-
ever, we found that Et3P (until 50 equiv) did not induce an
appreciable decrease of diol mitomycin 9 in 3 d, which indicated
that 9 did not undergo appreciable activation by Et3P. We believed
that the rapid nucleophilic activation for only 8 but not 9 by Et3P
must be induced by the presence of the disulfide unit in 8. So,
Et3P was expected to attack the disulfide unit in 8 leading to the
generation of thiol (or thiolate) that could trigger the activation
by attacking the quinone ring of mitomycin. When we monitored
the activation reactions by Et3P using UV–vis spectroscopy, we ob-
served a gradual decrease of starting mitomycin 8 at 373 nm and a
subsequent increase of mitosene products at 313 nm, which is in
accord with the known absorption patterns.19,25 The HPLC chro-
matograms for these activation reactions displayed multiple peaks
between ꢀ30 and ꢀ33 min. All these peaks showed absorption
maxima at 313 nm which again confirmed the formation of mito-
sene products. However, identification of all these peaks by coin-
jection with the authentic sample 19 was inconclusive. In reality,
the peak pattern was complex probably due to the many diastereo-
mers of mitosene products. In addition, we had to consider the low
concentrations of reaction solutions (ꢀ0.03 mM) and the possibil-
ity of precipitation of polar mitosene products during the reaction
period. Further efforts did not lead to successful results.
According to the information obtained above we propose a plau-
sible mechanism for nucleophilic activation of 8 by Et3P as shown in
Scheme 3. Mitomycin 8 undergoes disulfide cleavage by Et3P28 and
subsequent decomposition of thiophosphonium group by H2O (or
MeOH)28 to provide a dithiol species 8r that has two thiols. As men-
tioned above, the generation of thiol is significant since the gener-
ated thiol group would serve as a key probe to initiate mitomycin
activation. Both of the thiols in 8r are expected to attack the C(8)
carbon of the quinone ring by intramolecular cyclization to give
hemi-thioketal 20. Once 20 is formed the stabilization of the non-
bonding electron pair at N(4) through N(4)–C(4a)–C(8a)–C(8)–O
conjugated system may be disturbed due to loss of the C(8) car-
bonyl group. This destabilization would facilitate elimination of
methoxide at C(9a) to give an iminium ion at N(4)–C(9a) that
may undergo subsequent shift of the double bond to C(9)–C(9a)
leading to the generation of mitosene structure 21. The structure
21 may be slightly unstable and easily undergo aziridine ring open-
ing. Consequently, the C(1) site becomes highly electrophilic and
reacts with nucleophile (e.g., MeOH or DNA) to afford 22,21 which
eventually represents C(1)–C(10) double activations and adduc-
tions. Then, the facile decomposition of the hemi-thioketal struc-
ture in 22 would give dithiol and finally disulfide 23 by proper
oxidation processes. Thus, through the action of Et3P, mitomycin
8 underwent double activations and adductions at the two distal
C(1) and C(10) sites, which might constitute a kind of nucleophilic
activation mechanism that differs from traditional mechanisms.
8
activation
MeOH / H+
H2N(O)CO
OC(O)NH2
O
8'
O
8
S
S
H
N
H
N
MeO
1'
OMe
1
N
N
NH2
H2N
O
O
19
2.3.2. Kinetic studies on nucleophilic activation of 8 and 9
The rates of methanolysis were believed to directly represent
the efficiency of activation and adduction of mitomycins. There-
fore, kinetic studies on nucleophilic activations were conducted
by measuring the rate of methanolysis of mitomycins in the pres-
ence of appropriate nucleophiles. We wished to see if the target
mitomycin 8 undergoes efficient nucleophilic activation compared
with reference diol mitomycin 9, and particularly, to verify the crit-
ical function of the disulfide group in 8. We used Et3P, L-DTT and
glutathione (GSH) as proper nucleophiles to induce disulfide cleav-
age and subsequent activation. The activations were induced in
buffered methanolic solutions (0.1 M TrisꢂHCl, ‘pH’ 7.4) at room
temperature and monitored by UV–vis spectroscopy (200–
500 nm) for more than two half-lives. The absorption was
monitored at ꢀ373 nm for starting mitomycins and ꢀ313 nm for
mitosene products (activated products). At the end of the reactions,
the solutions were analyzed by HPLC and TLC by comparison with
authentic samples of 8, 9, and mitosene product 19. Generally, the
reactions well followed pseudo first-order kinetics, and the kobs
(dꢁ1) and t1/2 (d) were calculated. The reactions were run in
duplicate, and the results averaged.
As shown in Table 1, we measured the methanolysis rates of 8
and 9 in the absence of nucleophile and observed no distinct de-
crease (less than 10% of original amount) of starting mitomycins
in 5 d, which means there was no appreciable activation without
nucleophile. Next, we measured the effect of thiol nucleophiles
such as L-DTT and GSH on the activation of 8 and 9. When we used
Table 1
Activation rates for 8 and 9 at ‘pH’ 7.4a
Reagents
8
9
Nu
equiv
kobs(dꢁ1
)
t1/2(d)
kobs(dꢁ1
)
t1/2(d)
b
c
c
c
c
b
c
c
c
c
b
c
c
c
b
c
c
c
No Nu
10
20
20
2
L
-DTT
GSH
Et3P
—
—
5
9.2
16
23
63
0.075
0.043
0.030
0.011
—
—
c
c
10
20
50
c
c
c
c
a
Reactions were run in buffered methanolic solution (0.1 M TrisꢂHCl, ‘pH’ 7.4) at
25 °C. The reactions were run in duplicate and the values averaged. The results were
obtained using a Shimatzu UV-1800 spectrophotometer and the reactions moni-
tored at ꢀ373 nm unless otherwise indicated. The concentration of the mitomycin
was 0.030 mM.
2.4. Studies on the formation of DNA ISC for 8 and 9
We next examined the ability of mitomycin dimer 8 to produce
DNA ISC under nucleophilic activation conditions at pH 7.4
compared to reference 9 in order to verify and highlight the
b
No appreciable change in 5 d (less than 10% of the original amount).
No appreciable change in 3 d (less than 10% of the original amount).
c

H. R. Kim et al. / Bioorg. Med. Chem. 20 (2012) 5720–5729
5725
Et3P
H2N(O)CO
MeO
OC(O)NH2
OMe
O
O
H2N(O)CO
MeO
OC(O)NH2
OMe
O
O
SH SH
O
S
S
Et3P / H+
H2O
H
N
H
N
H
N
H
N
8
8'
8
1'
1
1'
1
HN
N
N
NH
NH
HN
N
N
-Et3P=O
O
8
O
O
-H+
8r
H2N(O)CO
MeO
H2N(O)CO
HO
S
OC(O)NH2
OMe
OC(O)NH2
S OH
HO
S
S OH
-MeOH
NH
HN
NH
HN
8
1'
1
1'
1
HN
HN
N
N
NH
N
NH
N
O
O
O
O
20
21
NuH
H2N(O)CO
HO
S
OC(O)NH2
H2N(O)CO
OC(O)NH2
S
OH
O
O
O
O
S
S
H
N
H
N
NH
HN
Nu
Nu
Nu
Nu
1'
1
1
1'
N
N
N
N
O2/-H2
H2N
NH2
H2N
O
NH2
O
22
23
C(1') Activation
C(1) Activation
where NuH: MeOH, DNA
Scheme 3. Proposed mechanism of nucleophilic activation of 8 by Et3P.
L-DTT
GSH
2
9
8
2
9
8
23kb -
9.4kb -
6.6kb -
4.4kb -
Figure 1. Denaturing 1.2% Alkaline Agarose Gel for 2, 9 and 8 using Et3P (5 equiv).
DNA cross-linking experiments using 0.1 mM concentration of mitomycins except
for 2 (0.2 mM) and EcoRI-linearized pBR322 plasmid DNA with Et3P (5 equiv). All
reactions were incubated at room temperature (2 h). Lane 1:
molecular weight marker. Lane 2: control (only linearized pBR322). Lane 3: 2 + Et3P.
Lane 4: 9 + Et3P. Lane 5: 8 + Et3P. Lane 6: only Et3P.
k Hind III DNA
1
2
3
4
5
6
7
Figure 2. Denaturing 1.2% Alkaline Agarose Gel for 2, 9 and 8 using
(5 equiv). DNA cross-linking experiments using 0.1 mM concentration of mitomyc-
ins except for 2 (0.2 mM) and EcoRI-linearized pBR322 plasmid DNA with -DTT
(5 equiv) and GSH (5 equiv). All reactions were incubated at room temperature
(2 h). Lane 1: k Hind III DNA molecular weight marker. Lane 2: 2 + -DTT. Lane 3:
9 + -DTT. Lane 4: 8 + -DTT. Lane 5: 2 + GSH. Lane 6: 9 + GSH. Lane 7: 8 + GSH.
L-DTT and GSH
L
function of the disulfide in 8. For this purpose we employed the
methods by Cech,29 and Tepe and Williams30 with minor modifica-
tions. Based on these methods pBR322 plasmid DNA was linearized
using EcoRI enzyme. The linearized pBR322 DNA was treated with
mitomycins under nucleophilic activation conditions and applied
to denaturing alkaline agarose gel electrophoresis to analyze the
formation of DNA ISC along with k DNA digested with HindIII as
a molecular weight marker. Based on the information obtained
from our previous studies with 717 we selected appropriate condi-
tions such that the reactions were run at 0.1 mM concentrations of
mitomycin dimers with 5 equiv of nucleophiles for 2 h at room
temperature to efficiently differentiate the ability to form DNA
ISC. To keep the concentrations of mitomycin units per experiment
the same, we used 0.2 mM concentration of 2 since it is a mono-
meric structure not a dimeric structure.
L
L
L
Remarkably, mitomycin 8 produced much higher levels of DNA ISC
(97%) under these conditions compared to references 9 (5%) and 2
(4%). As expected, mitomycin 8 underwent fast activation by the
function of Et3P on the disulfide unit. Accordingly, we concluded
that target mitomycin 8 was highly efficient for nucleophilic acti-
vation and DNA ISC formation compared to references 9 and 2. In
addition, we compared the results for DNA ISC formation with
the rate data of kinetic studies (Table 1). Among the mitomycins
tested, only mitomycin 8 underwent facile activation by Et3P
(5 equiv: tR = 0.27 d) while references 9 and 2 did not, which en-
abled us to confirm a good correlation.
First, we checked the formation of DNA ISC for 2, 9, and 8 in the
presence of Et3P (5 equiv), and the results were shown in Figure 1.
We then checked the formation of DNA ISC for 2, 9, and 8 by
-DTT and GSH as thiol nucleophiles, and the results were shown
L

5726
H. R. Kim et al. / Bioorg. Med. Chem. 20 (2012) 5720–5729
Table 2
activation than all three mitomycins 5–7, which seemed to be
unexpected, but interesting. Furthermore, the levels of DNA ISC
for 8 were also higher than those for 5–7. Despite the trend for
5–7, mitomycin 8 that has the shortest distance between the two
C(1) sites provided the highest level of DNA ISC (97%). This fact
might imply that shorter distance between the two reaction sites,
in a limited range, could provide better condition for the formation
of DNA ISC, which could constitute an important piece of informa-
tion on DNA ISC. However, we believed that much is still unknown
and many other factors might be involved in controlling the activa-
tion and DNA adduction, and so, the working mechanism needs to
be better elucidated by further studies. Despite these complexities,
mitomycin 8, taken together, was found to be more efficient than
5–7 for both nucleophilic activation and DNA ISC formation.
Comparison of 8 with 5–7 in activation rates and DNA ISC formations using Et3P
Compounds
5c
8.7
94
6d
4.6
86
7e
2.7
83
8
Activation ratesa kobs (dꢁ1
)
16
97
DNA ISCb (%)
a
b
c
Activation rates with 10 equiv of Et3P.
DNA ISC formations with 5 equiv of Et3P.
Ref. [14].
Ref. [16].
Ref. [17].
d
e
in Figure 2. By L-DTT(5 equiv), 8 produced medium levels (45%) of
DNA ISC while 9 and 2 produced trace amount of DNA ISC (5% and
3%, respectively). By GSH (5 equiv), 8 produced low levels (15%) of
DNA ISC while 9 and 2 produced trace amount of DNA ISC (less
than 5%). These results implied that 8 underwent moderate or
3. Conclusions
low activation by L-DTT and GSH, and that thiol nucleophiles were
not as efficient as the phosphine nucleophile for activation of mito-
mycin 8. We also compared these results with kinetic data of 8 in
We report the studies on the synthesis and modes of action for
mitomycins 8 and 9, and the comparison of 8 with previously re-
ported 5–7. Mitomycin 8 is a dimer tethered through a 1,2-dithio-
lane (a five-membered cyclic disulfide) linker, which may provide
the facile nontraditional nucleophilic activation and the two C(1)
activations by the dimeric structure, leading to efficient formation
of DNA ISC.
the presence of thiols (Table 1). Neither
apparent activation of 8 in 3 d, so the formation of medium levels
of DNA ISC for 8 by -DTT was not clearly understood. We believe
L-DTT nor GSH induced
L
that this discrepancy might originate from the different reaction
conditions between the two sets of experiments and that the
DNA experiments were more sensitive than the kinetic
experiments.
First, we established an efficient seven-step synthetic sequence
to prepare the key intermediates, a cyclic disulfide (10) and a diol
intermediate (13). Then the new target mitomycin 8 and the diol
mitomycin 9 were synthesized using the obtained intermediates.
Next, we performed mechanistic studies by measuring the activa-
tion rates of 8 by appropriate nucleophiles compared with the ref-
erence 9, and found that the target mitomycin 8 underwent facile
activation by Et3P (5–50 equiv) while 9 did not. This indicated that
8 underwent initial activation by the instantly generated thiol
group upon disulfide cleavage and advanced to subsequent activa-
tion. As a result, we propose a nucleophilic activation pathway for
8. Then, we evaluated the ability of mitomycin 8 to produce DNA
ISC by proper nucleophiles, and found that 8 displayed high effi-
ciency in forming DNA ISC (98%) compared to 9 (5%) and 2 (4%)
in the presence of Et3P. We therefore concluded that the target
mitomycin 8 is more efficient for both nucleophilic activation
and the corresponding DNA ISC formation than references 9 and
2. This finding highlighted the key function of the disulfide group
on mitomycin activation. More significantly, mitomycin 8 under-
went much faster activation and provided slightly higher levels
of DNA ISC than previously reported mitomycins 5–7.
2.5. Comparison of mitomycin 8 with mitomycins 5–7
As discussed above, one of our main interests lies in the effects
of the ring size of the linkers in a dimer on the activation rates and
the formation of DNA ISC. Therefore, we compared mitomycin 8
with mitomycins 5–7 connected through a different cyclic disulfide
linker with respect to the rate and the DNA ISC data as shown in
Table 2. We used the rate constants (kobs) instead of the half-lives
for activation rates in order to more easily discern the differences.
In general, all mitomycins displayed high activation rates and DNA
ISC formation in the presence of Et3P through nucleophilic activa-
tion pathways. When we compared the two sets of data we found
that the DNA ISC data reasonably correlated with the rate data.
However, when we compared the two sets of data according to
the corresponding compounds we found unexpected results for
mitomycin 8. For mitomycins 5–7, as the ring size of the cyclic
disulfide linker decreased from an eight-membered ring (mitomy-
cin 5) to a seven-membered ring (mitomycin 6) and then to a six-
membered ring (mitomycin 7), the activation rates decreased (kobs
:
8.66 ? 4.62 ? 2.67 dꢁ1, respectively) and the levels of DNA ISC
slightly decreased (DNA ISC: 94 ? 86 ? 83%, respectively),14,16,17
which seemed to be consistent with the ring size of the linkers in
dimers. Nevertheless, when we compared the activation rate data
4. Experimental
4.1. General
of 8 with 5–7, the activation rate of 8 was suddenly increased (kobs
:
Melting points were measured in open capillary tubes using Bu-
chi B-545 melting point apparatus and are uncorrected. FT–IR
spectra were obtained by a Perkin–Elmer Spectrum GX spectrom-
eter. 1H (300 MHz) and 13C (75 MHz) NMR spectra were recorded
on a Bruker DRX 300 spectrometer. Mass spectra were obtained
by CI, EI or FAB ionization method. UV–vis spectra were obtained
by Shimatzu UV-1800 Spectrophotometer. The pH of aqueous solu-
tions were measured using an IQ Scientific Instruments IQ-240 me-
ter. The nonaqueous effective pH (‘pH’) of the buffered methanolic
solutions was similarly measured. Thin layer chromatography was
run on silica gel plates (20 ꢃ 20 cm; Aldrich No. Z12272-6). HPLC
analyses were conducted using the following Waters Associate
Units: 515 A pump, 515 B pump, dual k absorbance 2487 detector,
717 plus autosampler, and Hypersil ODS column (4.6 ꢃ 300 mm).
The product analyses were performed using linear gradient
16 dꢁ1), which was almost six- and two-fold higher than those of 7
and 5, respectively. The sudden increase in activation rates for 8
was surprising and we could not clearly understand it. We previ-
ously suggested that the lifetime of generated thiol (e.g., 5r–7r)
upon disulfide cleavage would be significant since the thiol group
itself was known to initiate the activation.14 The lifetime of thiol
would be inversely proportional to the rate of disulfide formation
(cyclization), thus, as mentioned above, the lifetimes of thiols
would be 7r<<6r68r<<5r.22,23 These results reasonably explained
the trend for the activations of mitomycins 5–7, but not for activa-
tion of mitomycin 8. The lifetime of thiol in 8r would be in between
those of 6r and 5r. If this factor mainly governs the activations,
mitomycin 8 should have shown faster activation than 6, and
slower activation than 5. However, 8 did undergo much faster

H. R. Kim et al. / Bioorg. Med. Chem. 20 (2012) 5720–5729
5727
condition: 90% A (aqueous 0.025 M triethylammonium acetate, pH
6.5), 10% B (acetonitrile) isocratic for 5 min, then from 90% A, 10% B
to 45% A, 55% B in 30 min. The flow rate was 1 mL/min, and the elu-
ent was monitored from 200 to 400 nm. The HPLC solvents were
filtered (aqueous solution with Millipore HVLP, 0.45 mm; acetoni-
trile with Millipore HV, 0.45 mm) and degassed before utilization.
1.37 (s, 18H, OC(CH3)3), 1.50–1.60 (m, 2H, CH2CH), 3.00–3.13 (m,
2H, CHH’N), 3.15–3.28 (m, 2H, CHH’N), 3.65 (br s, 2H, CHOH),
3.91 (br s, 2H, CHOH), 4.99 (br s, 2H, NHCO). 13C NMR (75 MHz;
CDCl3) d: 28.6 (OC(CH3)3), 37.9 (CH2CH), 47.1 (CH2N), 69.5 (CHOH),
80.1 (OC(CH3)3), 157.4 (NHCO). MS m/z 335 [M+H]+. HRMS (+FAB)
calcd for C15H31N2O6 [M+H]+: 335.2182; found 335.2182.
4.2. 1,5-Bis(phthalimido)-2,4-pentanediol (12)
4.5. 1,5-Bis(tert-butyloxycarbonylamino)-2,4-pentanediol
dimethanesulfonate (15a)
To a DMF solution (1.2 mL) of phthalimide (2.1 g, 14 mmol) was
added 1,4-pentadiene diepoxide (11, 0.70 g, 7.0 mmol) with vigor-
ous stirring. The reaction mixture was heated (100 °C /1 h ?
115 °C/1 h ? 135 °C /1 h) and then cooled. The precipitate was fil-
tered and washed with Et2O (50 mL) to give the crude title com-
pound (2.1 g, 76%, dr = 1.4:1, 13C NMR analysis) as a white solid.
Recrystallization of crude product in DMF–Et2O provided the title
compound of higher diastereomeric ratio (0.55 g, 21%, dr = 4:1,
13C NMR analysis). Mp 229–231 °C. Rf 0.25 (2:1 EtOAc/hexanes).
IR (KBr) 3437, 2940, 1724, 1391, 1187, 1052, 889, 802, 720, 640,
To a cooled (0 °C) solution of 1,5-bis(tert-butyloxycarbonylami-
no)-2,4-pentanediol (14, 262 mg, 0.78 mmol) in CH2Cl2 (4 mL) was
added Et3N (0.3 mL, 2.4 mmol) and methanesulfonylchloride
(MsCl) (0.1 mL, 1.9 mmol). After stirring at the same temperature
(5 h) the solvent was removed in vacuo. H2O (50 mL) was added
to the residue and then the mixture was extracted with EtOAc
(2 ꢃ 50 mL). The combined organic layers were washed with satu-
rated aqueous NaHCO3 (50 mL) and H2O (50 mL). The organic layer
was dried (MgSO4) and concentrated in vacuo. Crystallization of
crude product in EtOAc/hexanes mixture afforded the title com-
pound (355 mg, 93%, dr: an apparent single set of signals in 13C
NMR) as a white solid. Mp 88–90 °C. Rf 0.6 (2:1 EtOAc/hexanes).
IR (KBr) 3748, 2978, 1713, 1520, 1455, 1353, 1251, 1172, 909,
533 cmꢁ1 1H NMR (300 MHz; DMSO-d6) d: 1.33–1.45 (m, 2H,
.
CH2CH), 3.44–3.56 (m, 4H, CH2N), 3.91–4.06 (m, 2H, CHOH), 4.86
(d, J = 5.7 Hz, 2H, CHOH), 7.79–7.88 (m, 8H, Pht). 13C NMR (75
MHz; DMSO-d6) d: 35.8 (CH2CH), 44.5 (CH2N), 64.1 (CHOH),
122.9 (Pht, CHCHC), 131.7 (Pht, C), 134.2 (Pht, CHC), 168.0 (NCO),
for the minor diastereomer d: 122.8 (Pht, CHCHC), 131.8 (Pht, C),
134.1 (Pht, CHC), 168.1 (NCO). MS m/z 395 [M+H]+.
787, 526 cmꢁ1 1H NMR (300 MHz; CDCl3) d: 1.38 (s, 18H,
.
OC(CH3)3), 1.87–1.97 (m, 2H, CH2CH), 3.08 (s, 6H, OMs), 3.21–
3.40 (m, 2H, CHH’N), 3.41–3.55 (m, 2H, CHH’N), 4.75–4.90 (m,
2H, CHOMs), 4.97 (br s, 2H, NHCO). 13C NMR (75 MHz; CDCl3) d:
28.5 (OC(CH3)3), 34.6 (CH2CH), 38.8 (OMs), 44.6 (CH2N),
77.7(CHOMs), 80.3 (OC(CH3)3), 156.3 (NHCO). MS m/z 491
[M+H]+; HRMS (+FAB) calcd for C17H35N2O10S2 [M+H]+: 491.1733;
found 491.1730.
4.3. 1,5-Diamino-2,4-pentanediol dihydrochloride (13)
A mixture of 1,5-bis(phthalimido)-2,4-pentanediol (12, 1.4 g,
3.6 mmol), EtOH (32 mL) and NH2NH2ꢂH2O (0.44 mL, 8.9 mmol)
was heated under reflux (3.5 h). After cooling to room temperature,
the solvent was removed in vacuo. H2O (33 mL) and concentrated
hydrochloric acid (17 mL) were added to the residue and the mix-
ture was heated under reflux (1 h). After cooling to 0 °C, the precip-
itate was filtered off and the filtrate was concentrated in vacuo.
The residue was dissolved in H2O (50 mL) and the insoluble matter
was removed by filtration. The clear filtrate was concentrated in
vacuo to give the title compound (595 mg, 80%, dr = 5:1, 13C NMR
analysis) as a white solid. Mp 175–180 °C. Rf 0.10 (4:6 MeOH/
CHCl3). IR (KBr) 3233, 1709, 1504, 1465, 1396, 1280, 1117, 1044,
4.6. 1,5-Bis(tert-butyloxycarbonylamino)-2,4-
bis(acetylthio)pentane (16a)
To a stirred solution of 1,5-bis(tert-butyloxycarbonylamino)-
2,4-pentanediol dimethanesulfonate (15a, 100 mg, 0.20 mmol) in
DMF (2 mL) was added KSAc (57 mg, 0.50 mmol). After warming
to 60 °C, stirring was continued (8 h) and then the solvent was re-
moved in vacuo. The residue was treated with H2O (50 mL) and the
resulting mixture was extracted with EtOAc (2 ꢃ 50 mL). The com-
bined organic layers were dried (MgSO4) and concentrated in va-
cuo. Purification by column chromatography (1:4 ? 1:3 EtOAc/
hexanes) afforded the title compound (29 mg, 32%) as a brown so-
lid. Mp 75–76 °C. Rf 0.50 (1:2 EtOAc/hexanes). IR (KBr) 3748, 2927,
891, 835, 799, 724, 628, 532 cmꢁ1 1H NMR (300 MHz; CD3OD) d:
.
1.52–1.82 (m, 2H, CH2CH), 2.70–2.98 (m, 2H, CHH’N), 3.00–3.20
(m, 2H, CHH’N), 3.96–4.20 (m, 2H, CHOH). 13C NMR (75 MHz;
CD3OD) d: 40.6 (CH2CH), 46.4 (CH2N), 65.5 (CHOH), for the minor
diastereomer d: 40.3 (CH2CH), 45.8 (CH2N), 66.0 (CHOH). MS m/z
135 [M-2HCl+H]+. HRMS (+FAB) calcd for C5H15N2O2 [M-2HCl+H]+:
135.1135; found 135.1134.
1698, 1511, 1455, 1366, 1251, 1168, 955, 872, 759, 632 cmꢁ1 1H
.
NMR (300 MHz; CDCl3) d: 1.43 (s, 18H, OC(CH3)3), 1.77–1.90 (m,
2H, CH2CH), 2.32 (s, 6H, SAc), 3.16–3.34 (m, 2H, CHH’N), 3.35–
3.47 (m, 2H, CHH’N), 3.65–3.82 (m, 2H, CHSAc), 4.85 (br s, 2H,
NHCO). 13C NMR (75 MHz; CDCl3) d: 28.3 (OC(CH3)3), 30.7 (COCH3),
33.4 (CH2CH), 42.9 (CH2N), 44.7 (CHSAc), 79.6 (OC(CH3)3), 155.9
(NHCO), 195.2 (COCH3). MS m/z 451 [M+H]+. HRMS (+FAB) calcd
for C19H35N2O6S2 [M+H]+: 451.1936; found 451.1937.
4.4. 1,5-Bis(tert-butyloxycarbonylamino)-2,4-pentanediol (14)
To a stirred solution of 1,5-diamino-2,4-pentanediol dihydro-
chloride (13, 468 mg, 2.3 mmol) and Et3N (1.9 mL, 14 mmol) in
H2O–DMF (1:1, 40 mL) was added a solution of di-tert-butyl dicar-
bonate (Boc2O) (1.2 g, 5.7 mmol) in DMF (4 mL). After stirring at
room temperature (1 d), the solvent was removed in vacuo. The
remaining residue was treated with H2O (80 mL) and then the mix-
ture was extracted with EtOAc (2 ꢃ 80 mL). The combined organic
layers were successively washed with aqueous 0.1 N HCl (80 mL),
saturated aqueous NaHCO3 (80 mL) and H2O (80 mL). The organic
layer was dried (MgSO4) and concentrated in vacuo. Purification
by column chromatography (1:1 ? 2:1 EtOAc/hexanes) afforded
the title compound (535 mg, 70%, dr: an apparent single set of sig-
nals in 13C NMR) as a white solid. Mp 93–94 °C. Rf 0.35 (2:1 EtOAc/
hexanes). IR (KBr) 3748, 2977, 1693, 1524, 1455, 1366, 1251, 1170,
4.7. 3,5-Bis(tert-butyloxycarbonylaminomethyl)-1,2-dithiolane
(18)
To a stirred solution of 1,5-bis(tert-butyloxycarbonylamino)-
2,4-bis(acetylthio)-pentane (16a, 20 mg, 0.040 mmol) in MeOH–
H2O (5:1, 0.4 mL, c = 100 mM) was added K2CO3 (36 mg,
0.25 mmol). After stirring at room temperature (1 h), Et3N (13 lL,
0.10 mmol) was added. A saturated CHCl3 solution of iodine
(0.4 mL) was then added dropwise at room temperature until
slight excess of iodine was evidenced by its color. The resulting
solution (c = 50 mM) was stirred at room temperature (3 h) and
then treated with saturated aqueous Na2S2O3 (50 mL). The mixture
1042, 892, 781, 737, 672, 616 cmꢁ1 1H NMR (300 MHz; CDCl3) d:
.

5728
H. R. Kim et al. / Bioorg. Med. Chem. 20 (2012) 5720–5729
was extracted with EtOAc (2 ꢃ 40 mL), and the combined organic
layers were washed with saturated aqueous NaHCO3 (40 mL) and
H2O (40 mL). The organic layer was dried (MgSO4) and concen-
trated in vacuo. Purification by column chromatography (1:3
EtOAc/hexanes) afforded the title compound (10 mg, 69%) as a yel-
low solid. Mp 128–129oC. Rf 0.53 (1:2 EtOAc/hexanes). IR (KBr)
3748, 2927, 1698, 1511, 1455, 1366, 1251, 1168, 955, 872, 759,
(3:7 MeOH/CHCl3) afforded the title compound (2.2 mg, 67%,
dr = 1.1:1, 13C NMR analysis) as a dark blue solid. Rf 0.38 (3:7
MeOH/CHCl3). HPLC tR 21.1 min. UV–vis (MeOH) kmax
: 222,
370 nm. 1H NMR (300 MHz; pyridine-d5) d: 1.48–1.82 (m, 2H,
C(3’’)H2), 2.18 (br s, 6H, C(6)CH3), 2.74 (d, J = 3.6 Hz, 2H, C(2)H),
3.13 (d, J = 4.5 Hz, 2H, C(1)H), 3.19 (br s, 6H, C(9a)OCH3), 3.52–
3.64 (m, 2H, C(3)HH’), 3.69–3.82 (m, 2H, C(1’’)HH’), 3.83–3.96 (m,
6H, C(1’’)HH’, C(2’’)H, C(9)H), 4.54 (dd, J = 12.9, 3.6 Hz, 2H,
C(3)HH’), 5.00–5.10 (m, 2H, C(10)HH’), 5.26–5.45 (m, 2H,
C(10)HH’), 7.62–7.90 (m, 2H, C(7)NH), the signals for the N(1a) H,
C(10)OC(O)NH2 and C(2’’)OH protons were not detected and are
believed to overlap with the observed peaks. 13C NMR (75 MHz;
pyridine-d5) d: 11.8 (C(6)CH3), 31.2 (C(3’’)), 33.7 (C(2)), 38.3
(C(1)), 45.8 (C(9)), 51.2 (C(9a)OCH3), 52.2 (C(3)), 53.3 (C(1’’)),
64.0 (C(10)), 69.1 (C(2’’)), 104.8 (C(6)), 108.5 (C(9a)), 112.0
(C(8a)), 149.3 (C(7)), 157.2 (C(5a)), 159.8 (C(10a)), 178.3 (C(8)),
181.0 (C(5)), for the minor diastereomer d: 38.2 (C(1)). MS m/z
791 [M+Na]+.
632 cmꢁ1 1H NMR (300 MHz; CDCl3) d: 1.38 (s, 18H, OC(CH3)3),
.
1.98–2.09 (m, 2H, CH2CH), 3.03–3.18 (m, 2H, CHH’N), 3.29–3.42
(m, 2H, CHH’N), 3.69–3.83 (m, 2H, CHS), 4.90 (br s, 2H, NHCO).
13C NMR (75 MHz; CDCl3) d: 28.3 (OC(CH3)3), 39.9 (CH2CH), 43.2
(CH2N), 56.1 (CHS), 79.7 (OC(CH3)3), 155.8 (NHCO). MS m/z 364
[M]+. HRMS (+FAB) calcd for C15H28N2O4S2 [M]+: 364.1490; found
364.1491.
4.8. 3,5-Bis(aminomethyl)-1,2-dithiolaneꢂ2TFA (10)
3,5-Bis(tert-butyloxycarbonylaminomethyl)-1,2-dithiolane (18,
4.1 mg, 0.010 mmol) was dissolved in trifluoroacetic acid (TFA)
(0.50 mL) and stirring was continued at room temperature (1 h).
The reaction mixture was concentrated in vacuo to give the title
compound (4.4 mg, ꢀ100%) as a brown solid. Mp 154–156 °C. Rf
0.13 (4:6 MeOH/CHCl3). IR (KBr) 3433, 1675, 1519, 1433, 1390,
4.11. C(1) Methoxymitosenes (19)
Mitomycin 8 (2.0 mg, 2.5 lmol) was dissolved in MeOH–CHCl3
(1:1, 2 mL) and then the ‘pH’ was adjusted to 2.5–3.0 with a meth-
anolic 0.02 M HCl solution. The reaction solution was stirred at
room temperature (2 d) along with continual adjustment of pH
and then the solvent was removed under reduced pressure. Purifi-
cation of the reaction mixture by PTLC (3:7 MeOH/CHCl3) afforded
the title compound (0.9 mg, 45%) as a red solid. Rf 0.24 (3: 7 MeOH/
1342, 1201, 837, 799, 722, 597, 519 cmꢁ1 1H NMR (300 MHz;
.
CD3OD) d: 2.07–2.23 (m, 2H, CH2CH), 2.97–3.13 (m, 4H, CH2 N),
3.85–3.98 (m, 2H, CHS). 13C NMR (75 MHz; CD3OD) d: 41.6
(CH2CH), 42.5 (CH2 N), 54.4 (CHS). MS m/z 165 [M-2TFA+H]+.
HRMS (+FAB) calcd for C5H13N2S2 [M-2TFA+H]+: 165.0520, found
165.0520
CHCl3). HPLC tR 30.1, 31.3, 32.7 min (ꢀ1:2:1). UV–vis (MeOH) kmax
:
253, 313 nm. 1H NMR (300 MHz; pyridine-d5) d: 2.13 (br s, 6H,
C(6)CH3), 2.16–2.26 (m, 2H, C(400)H2), 3.52 (br s, 6H, C(1)OCH3),
3.72–4.04 (m, 8H, C(300)H, C(7)NHCH2, C(2)H), 4.09–4.45 (m, 4H,
C(3)H2), 4.52–4.74 (m, 2H, C(1)H), 5.73 (1/2ABq, J = 10.3 Hz, 2H,
C(10)HH’), 5.78 (1/2ABq, J = 10.3 Hz, 2H, C(10)HH’), 6.78–6.95 (m,
2H, C(7)NH), the signals for the C(2)NH2 and C(10)OC(O)NH2 pro-
tons were not detected and are believed to overlap with the ob-
served peaks. MS m/z 821 [M+Na]+. HRMS (+FAB) calcd for
4.9. 7-N,70-N0-(100,200-Dithiolanyl-300,500-
dimethylenyl)bismitomycin C (8)
To a solution of 3,5-bis(aminomethyl)-1,2-dithiolaneꢂ2TFA (10,
2.5 mg, 6.4 lmol) and Et3N (4.9 lL, 35 lmol) in MeOH (0.4 mL)
was added MMA (1, 4.1 mg, 12 lmol). The reaction solution was
stirred at room temperature (2 d) and then the solvent was re-
moved in vacuo. Purification of the reaction mixture by PTLC (2:8
MeOH/CHCl3) afforded the title compound (3.3 mg, 66%,
dr = 1.2:1, 13C NMR analysis) as a dark blue solid. Rf 0.65 (3:7
C
35H42N8NaO10S2 [M+Na]+: 821.2363, found 821.2326.
4.12. General procedure for the mitomycin activation studies
MeOH/CHCl3). HPLC tR 27.2 min. UV–vis (MeOH) kmax
: 223,
373 nm. 1H NMR (300 MHz; pyridine-d5) d: 2.11 (s, 6H, C(6)CH3),
2.21–2.29 (m, 2H, C(4’’)H2), 2.75 (d, J = 3.6 Hz, 2H, C(2)H), 3.15
(d, J = 4.2 Hz, 2H, C(1)H), 3.25 (s, 6H, C(9a)OCH3), 3.61 (d,
J = 12.6 Hz, 2H, C(3)HH’), 3.76–3.95 (m, 4H, C(7)NHCH2), 3.97–
4.12 (m, 4H, C(9)H, C(3’’)H), 4.54 (d, J = 12.6 Hz, 2H, C(3)HH’),
5.09–5.16 (m, 2H, C(10)HH’), 5.36–5.48 (m, 2H, C(10)HH’), 7.25–
7.40 (m, 2H, C(7)NH), the signals for the N(1a)H and
C(10)OC(O)NH2 protons were not detected and are believed to
overlap with the observed peaks. 13C NMR (75 MHz; pyridine-d5)
d: 10.0 (C(6)CH3), 32.6 (C(2)), 36.7 (C(1)), 40.0 (C(4’’)), 44.3 (C(9)),
47.5 (C(7)NHCH2), 49.6 (C(9a)OCH3), 50.9 (C(3)), 57.2 (C(3’’)),
62.4 (C(10)), 104.7 (C(6)), 106.9 (C(9a)), 110.9 (C(8a)), 146.9
(C(7)), 155.7 (C(5a)), 158.1 (C(10a)), 176.9 (C(8)), 179.3(C(5)), for
the minor diastereomer d: 10.1 (C(6)CH3), 44.4 (C(9)), 57.1
(C(3’’)), 147.0 (C(7)). MS m/z 799 [M+H]+. HRMS (+FAB) calcd for
To a buffered methanolic solution (0.1 M TrisꢂHC l ’pH’ 7.4) (fi-
nal volume 1.5 mL) maintained at 25 °C containing the mitomycins
(45
lL
of 1.0 mM methanolic solution, final concentration
0.03 mM) was added a methanolic solution (18–113
lL) of the
nucleophile of choice (stock solution: 5–20 mM, final nucleophile
concentration 0.06–1.5 mM). The reaction was monitored by UV–
vis spectroscopy (200–500 nm), and generally followed for more
than two half-lives. The ‘pH’ of the solution was measured at the
conclusion of the reaction and found to be within 0.1 pH units
of the original solution. The reaction solutions were analyzed by
HPLC and unreacted staring materials and products (e.g., 8, 9, 19)
were determined by coinjection of authentic samples in the HPLC
and cospotting of authentic samples in the TLC. The kmax of mito-
mycin (ꢀ373 nm) was plotted versus time and found to decrease
in a first-order decay (exponential decay) process. Nonlinear
regression analysis by SigmaPlot Program (SigmaPlot, 2001) was
used to fit the observed exponential decay and yielded pseudo-
first-order rate constants (kobs) and half-lives (t1/2). The reactions
were done in duplicate and the results averaged.
C
35H43N8O10S2 [M+H]+: 799.2543, found 799.2544.
4.10. 7-N,70-N0-(200,400-Dihydroxy-100,500-
pentanediyl)bismitomycin C (9)
To a solution of 1,5-diamino-2,4-pentanediol dihydrochloride
4.13. General procedure for alkaline agarose gel
electrophoresis29,30
(13, 0.90 mg, 4.3
lmol) and Et3N (3.6
lL, 25
lmol) in MeOH
(0.3 mL) was added MMA (1, 3.0 mg, 8.6
l
mol). The reaction solu-
tion was stirred at room temperature (2 d) and then the solvent
was removed in vacuo. Purification of the reaction mixture by PTLC
The agarose gels were prepared by adding 1.20 g of agarose to
100 mL of an aqueous 100 mM NaCl and 2 mM EDTA solution

H. R. Kim et al. / Bioorg. Med. Chem. 20 (2012) 5720–5729
5729
2. Mitomycin, C. In Current Status and New Developments; Carter, S. K., Crooke, S.
T., Eds.; Academic Press: New York, 1979.
3. (a) Iyer, V. N.; Szybalski, W. Science 1964, 145, 45; (b) Szybalski, W.; Iyer, V. N.
In Antibiotics: Mechanism of Action; Gottlieb, D., Shaw, P. D., Eds.; Springer: New
York, 1967; Vol. 1, pp 211–245; (c) Moore, H. W.; Czerniak, R. Med. Res. Rev.
1981, 1, 249.
4. Hornemann, U.; Keller, P. J.; Kozlowski, J. F. J. Am. Chem. Soc. 1979, 101, 7121.
5. (a) Keyes, S. R.; Loomis, R.; DiGiovanna, M. P.; Pritsos, C. A.; Rockwee, S.;
Sartorelli, A. C. Cancer Commun. 1991, 3, 351; (b) Ramos, L. A.; Lipman, R.;
Tomasz, M.; Basu, A. K. Proc. Am. Assoc. Cancer Res. 1997, 38, 182; (c) Tomasz,
M.; Palom, Y. Pharmacol. Ther. 1997, 76, 73.
6. Kono, M.; Saitoh, Y.; Kasai, M.; Sato, A.; Shirahata, K.; Morimoto, M.; Ashizawa,
T. Chem. Pharm. Bull. 1989, 37, 1128.
7. Vyas, D. M.; Chiang, Y.; Benigni, D.; Rose, W. C.; Brander, W. T. In Recent
Advances in Chemotherapy. Anticancer Section; Tshigami, J., Ed.; University of
Tokyo Press: Tokyo, 1985; p 485.
8. Tsuruo, T.; Sudo, Y.; Asami, N.; Inaba, M.; Morimoto, M. Cancer Chemother.
Pharmacol. 1990, 27, 89.
9. Morimoto, M.; Ashizawa, T.; Ohno, H.; Azuma, M.; Kobayashi, E.; Okabe, M.;
Gomi, K.; Kono, M.; Saitoh, Y.; Arai, H.; Sato, A.; Kasai, M.; Tsuruo, T. Cancer Res.
1991, 51, 110.
(pH 8.0). The suspension was heated in a microwave oven until all
of the agarose was dissolved (1 min). The hot solution was poured
and then allowed to cool and solidify at room temperature (1 h).
The gel was soaked in an aqueous alkaline running buffer solution
(50 mL) containing 40 mM NaOH and 1 mM EDTA (1 h) and then
the comb was removed. The buffer solution was refreshed prior
to electrophoresis.
To an aqueous solution of ꢀ80
1 M TrisꢂHCl (pH 7.4) was added a solution of linearized pBR322
(5 L, 5 g) in 10 mM Tris solution containing 1 mM EDTA (pH
8.0). After deaeration with N2 gas (15 min), the mitomycin (2–
L of 5 mM DMSO solution, final concentration 0.1–0.2 mM)
and the nucleophile (2–10 L of 5–25 mM DMSO solution, final
concentration 0.5 mM) were added and the resulting solution (final
volume 100 L) was incubated at room temperature (2 h). The
solution was washed with 1:1 PhOH/CHCl3 (100 L) and CHCl3
L), and precipitated (12.1 L of 3 M NaOAc and 250
lL of H2O (sterile) and 2.5 lL of
l
l
4
l
l
l
l
10. Kobayashi, E.; Okabe, M.; Kono, M.; Arai, H.; Kasai, M.; Gomi, K.; Lee, J.-H.;
Inaba, M.; Tsuruo, T. Cancer Chemother. Pharmacol. 1993, 32, 20.
11. He, Q.-Y.; Maruenda, H.; Tomasz, M. J. Am. Chem. Soc. 1994, 116, 9349.
12. Wang, S.; Kohn, H. J. Med. Chem. 1999, 42, 788.
(2 ꢃ 100
l
l
lL
of EtOH, ꢁ70 °C (10 min)). The mixture was centrifuged at 0 °C
(15 min), and the EtOH was decanted off and evaporated in vacuo.
13. (a) Masters, J. R. W.; Know, R. J.; Hartley, J. A.; Kelland, L. R.; Hendricks, H. R.;
Conners, T. Biochem. Pharmacol. 1997, 53, 279; (b) McAdam, S. R.; Knox, R. J.;
Hartley, J. A.; Masters, J. R. W. Biochem. Pharmacol. 1998, 55, 1777.
14. Park, H. J.; Kim, J. J.; Kim, H. R.; Lee, E. K.; Kim, E. S.; Jeong, C. S.; Moon, A.; Lee, S.
H. Bioorg. Med. Chem. 2011, 19, 4004.
15. Kim, J. J.; Kim, H. R.; Arai, H.; Lee, S. H. Arch. Pharm. Res. 2012, 35, 1413.
16. Kim, J. J.; Kim, H. R.; Lee, S. H., Arch. Pharm. Res. 2012, 35, in press.
17. Lee, S. H.; Kohn, H. Org. Biomol. Chem. 2005, 3, 471.
18. Lee, S. H. Synthesis and Mechanistic Studies of Novel Mitomycins, Ph.D. Thesis,
University of North Carolina, North Carolina, August 2003.
19. (a) Stevens, C. L.; Taylor, K. G.; Munk, M. E.; Marshall, W. S.; Noll, K.; Shah, G. D.;
Shah, L. G.; Uzu, K. J. Med. Chem. 1965, 8, 1; (b) Tomasz, M.; Lipman, R. J. Am.
Chem. Soc. 1979, 101, 6063; (c) McClelland, R. A.; Lam, K. J. Am. Chem. Soc. 1985,
107, 5182.
The remaining DNA was dissolved in 25
solution containing 1 mM EDTA (pH 8.0).
Agarose loading dye (5 L) was added to the sample (5 lL) and
the samples were loaded onto the wells. The gel was run at 75 mA/
25 V (30 min) and then at 145 mA/38 V (3–4 h). The gel was then
neutralized for 45 min in an aqueous 100 mM Tris pH 7.0 buffer
solution containing 150 mM NaCl, which was refreshed every
15 min. The gel was stained with an aqueous 100 mM Tris pH 7.5
lL of aqueous 10 mM Tris
l
buffer solution (100 mL) containing ethidium bromide (20 lL of
an aqueous ethidium bromide stock solution (10 mg/10 mL)) and
150 mM NaCl for 20 min. The background staining was then re-
moved by soaking the gel in aqueous 50 mM NH4OAc and 10 mM
b-mercaptoethanol solution (3 h). The gel was analyzed with a
Bio-Rad SmartspecTM 3000 and/or UV Trans Illuminator with a dig-
ital camera, and quantitative analyses of each band were also
performed.
20. Lee, S. H.; Kohn, H. J. Am. Chem. Soc. 2004, 126, 4281.
21. Wang, S.; Kohn, H. J. Org. Chem. 1997, 62, 5404.
22. Houk, J.; Whitesides, G. M. J. Am. Chem. Soc. 1987, 109, 6825.
23. Burns, J. A.; Whitesides, G. M. J. Am. Chem. Soc. 1990, 112, 6296.
24. Baylon, C.; Heck, M.-P.; Mioskowski, C. J. Org. Chem. 1999, 64, 3354.
25. Tomasz, M.; Lipman, R. Biochemistry 1981, 20, 5056.
26. Hong, Y. P.; Kohn, H. J. Am. Chem. Soc. 1991, 113, 4634.
27. Han, I.; Kohn, H. J. Org. Chem. 1991, 56, 4648.
28. Lee, S. H.; Kohn, H. Heterocycles 2003, 60, 47.
29. Cech, T. R. Biochemistry 1981, 20, 1431.
Acknowledgments
30. Tepe, J. J.; Williams, R. M. J. Am. Chem. Soc. 1999, 121, 2951.
This work was supported by a National Research Foundation of
Korea Grant funded by the Korean Government (2009–0066477).
References and notes
1. Hata, T.; Hoshi, T.; Kanamori, K.; Matsumae, A.; Sano, Y.; Shima, T.; Sugawara,
R. J. Antibiot. 1956, 9, 141.