Discovery and Optimization of Novel Hydrogen Peroxide Activated Aromatic Nitrogen Mustard Derivatives as Highly Potent Anticancer Agents

Article abstract of DOI:10.1021/acs.jmedchem.8b00559

We describe several new aromatic nitrogen mustards with various aromatic substituents and boronic esters that can be activated with H₂O₂ to efficiently cross-link DNA. In vitro studies demonstrated the anticancer potential of these compounds at lower concentrations than those of other clinically used chemotherapeutics, such as melphalan and chlorambucil. In particular, compound 10, bearing an amino acid ester chain, is selectively cytotoxic toward breast cancer and leukemia cells that have inherently high levels of reactive oxygen species. Importantly, 10 was 10-14-fold more efficacious than melphalan and chlorambucil for triple-negative breast-cancer (TNBC) cells. Similarly, 10 is more toxic toward primary chronic-lymphocytic-leukemia cells than either chlorambucil or the lead compound, 9. The introduction of an amino acid side chain improved the solubility and permeability of 10. Furthermore, 10 inhibited the growth of TNBC tumors in xenografted mice without obvious signs of general toxicity, making this compound an ideal drug candidate for clinical development.

Full text of DOI:10.1021/acs.jmedchem.8b00559

Subscriber access provided by Duquesne University - Gumberg Library
Article
Discovery and Optimization of Novel Hydrogen Peroxide Activated
Aromatic Nitrogen Mustard Derivatives as Highly Potent Anticancer Agents
WENBING CHEN, Heli Fan, Kumudha Balakrishnan, Yibin Wang, Huabing
Sun, Yukai Fan, Varsha Gandhi, Leggy A. Arnold, and Xiaohua Peng
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00559 • Publication Date (Web): 24 Sep 2018
Downloaded from http://pubs.acs.org on September 25, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
Discovery and Optimization of Novel Hydrogen Peroxide
Activated Aromatic Nitrogen Mustard Derivatives as Highly
Potent Anticancer Agents
8
9
Wenbing Chen a,+, Heli Fan a,+, Kumudha Balakrishnanb, Yibin Wang, Huabing Sun, Yukai Fan, Varsha
Gandhib, Leggy A. Arnolda, and Xiaohua Penga,*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aDepartment of Chemistry and Biochemistry and the Milwaukee Institute for Drug Discovery, University of
Wisconsin Milwaukee, 3210 N. Cramer St., Milwaukee. WI 53211, USA.
bDepartment of Experimental Therapeutics, MD Anderson Cancer Center, Houston, TX 77030, USA.
Keywords: ROSꢀactivation; DNA crossꢀlinking agents; reactive oxygen species; arylboronate esters;
targeted anticancer drugs
ABSTRACT
W
e describe several new aromatic nitrogen mustards with various aromatic substituents and boronic ester
that can be activated with H2O2 to efficiently crossꢀlink DNA. In vitro studies demonstrated the anticancer
potential of these compounds at lower concentrations than clinically used chemotherapeutics such as
mephalan and chlorambucil. Especially compound 10 bearing an amino acid ester chain is selectively
cytotoxic towards breast cancer and leukemia cells that inherently have high level of reactive oxygen species.
Importantly, 10 was 10ꢀ14 fold more efficacious than melphalan and chlorambucil for triple negative breast
cancer (TNBC) cells. Similarly, 10 is more toxic toward primary CLL cells than chlorambucil and the lead
compound 9. The introduction of an amino acid side chain improved the solubility and permeability of 10.
Furthermore, 10 inhibited the growth of TNBC tumors of xenografted mice without obvious signs of general
toxicity making this compound an ideal drug candidate for clinical development.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 35
INTRODUCTION
1
2
3
4
5
6
7
8
9
DNA crossꢀlinking occurs when various exogenous or endogenous compounds connect two different
positions of DNA.1,2 This reaction either takes place within the same DNA strand or at opposite strands,
blocking DNA replication and in turn causing cell cycle arrest and cell death. DNAꢀcrossꢀlinking agents are
effective against rapidly dividing cancer cells. However, many normal cells also divide very quickly, such as
bone marrow, the lining of the mouth and intestines, and the hair follicles. Most clinically useful alkylating
agents, such as cyclophosphamide, bendamustine, melphalan, and chlorambucil, have serious side effects.
Thus, new potent alkylating agents with improved selectivity and reduced adverse effects are needed. DNA
crossꢀlinking agents that are induced under specific physiological conditions have the potential to reduce
toxicity as these agents are not active by themselves but require specific conditions for activation. So far,
numerous methods have been developed for inducing DNA crossꢀlink formation under various conditions.
Among the existing methods, photoirradiation,3ꢀ11 reduction,12ꢀ17 oxidation,18ꢀ21 or fluoride induction22ꢀ24 are
well studied and proved to be highly efficient. For example, Greenberg and Zhou’s group employed phenyl
selenide precursors for producing DNA interstrand crossꢀlink (ICL) formation under mild oxidative
conditions.18ꢀ21 Rokita and coꢀworkers discovered several quinone methide precursors that induced DNA
crossꢀlinking in the presence of fluoride salts.22,23 The Freccero group introduced photogeneration of
bifunctional quinone methides by laser flash photolysis.3,5,6,10,25 Gates and coꢀworkers discovered
hypoxiaꢀselective DNAꢀalkylating agents built by grafting nitrogen mustards onto tirapazamine.16 A series
of dinitrobenzamide nitrogen mustard reagents with a nitrylꢀamidogen linker undergoes bioreduction to form
DNA crossꢀlinks.12ꢀ15 Recently, our group has shown that the aryl boronic acids and esters are ideal trigger
units for developing H2O2ꢀactivated DNA crossꢀlinking agents.26ꢀ32 Among these inducible DNA
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
crossꢀlinking agents, several classes of compounds showed anticancer activity and selectivity12,13,28,30
.
ACS Paragon Plus Environment

Page 3 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. H2O2ꢀactivated alkylating agents
Compared with their normal counterparts, most cancer cells have increased levels of reactive oxygen
species (ROS) such as superoxide, H2O2, and hydroxyl radicals,33ꢀ35 which offers a possible way to develop
selective anticancer agents. Pinacol boronate esters and boronic acids are excellent trigger groups responsive
to hydrogen peroxide.36 Our group discovered the first example of ROSꢀactivated DNA alkylating agents
with selective toxicity toward particular cancer cells.28,30 Although ROSꢀactivated anticancer reagents have
been sought for some time, very few agents effective in vivo have been developed so far.37ꢀ39 Aromatic
nitrogen mustards are powerful DNA crossꢀlinking agents. In particular, structure 1 (1aꢀf) with various
leaving groups in the nitrogen mustard exhibit potent cellular activity, and selectively kill primary leukemic
lymphocytes isolated from chronic lymphocytic leukemia patients (40−80% apoptosis) over normal
lymphocytes from healthy donors (20% apoptosis). In this work, we modified 1a and 1f to develop
analogues with increased anticancer activity by introducing different substituents on the aromatic ring.
Several analogues of 1f have been identified to be more toxic towards cancer cells than clinically used
chemotherapy drugs. One of which (10) showed much higher activity than the lead compound (9). More
importantly, in vivo efficacy was evaluated in a triple negative breast cancer xenograft mouse model with the
results that 10 exhibited excellent antitumor properties.
RESULTS AND DISCUSSION
Chemistry. We modified aromatic nitrogen mustard 3 to identify more efficacious prodrugs. Therefore, ring
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 35
substitution patterns and pinacol boronate ester groups were investigated. Eight compounds (4ꢀ11) were
designed and synthesized, which contained either a boronic acid group or a pinacol boronate ester as a
H2O2ꢀresponsive trigger in combination with a nitrogen mustard as a DNA crossꢀlinker (Scheme 2). Both
electron donating (4 and 5) and withdrawing groups (6) were introduced for investigating the electronic
effects in regard to their biological activity. Several amino acid side chains (7, 8, 10, and 11) and pinacol
boronate groups were employed to improve the biophysical and biochemical properties of these reagents.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Chemical structures of compounds 3ꢀ11.
Compounds 4 and 5 were synthesized starting from orthoꢀsubstituted pꢀbromoaniline (12a, b) (Scheme 3).
Alkylation of the amino group was performed with 2ꢀchloroethanol in the presence of sodium iodide and
calcium carbonate yielding 13a and 13b. Dichloromustards 14a and 14b were prepared from 13a and 13b
via mesylation followed by chlorination. Classic palladiumꢀcatalyzed borylation of 14a or 14b did not
produce the desired products (Scheme 3, path b). Thus, an alternative borylation strategy was employed
involving the synthesis of the boronic acid analogues and subsequent esterification (Scheme 3, path a).
Reaction of 14a or 14b with butyllithium and triisopropyl borate followed by hydrolysis yielded boronic
acid derivatives, which were converted into the desired products 4 and 5 by esterification in the presence of
pinacol and magnesium sulfate.
ACS Paragon Plus Environment

Page 5 of 35
Journal of Medicinal Chemistry
Scheme 3. Synthesis of compounds 4 and 5
1
2
3
4
Compound
6
was synthesized according to Scheme 4. Nucleophilic substitution of
5
6
7
8
4ꢀbromoꢀ1ꢀfluoroꢀ2ꢀnitrobenzene 15 with diethanolamine (16) generated diol 17 with 90% yield. The
hydroxyl groups of 17 were converted to a mesylate and replaced by chlorine using LiCl to yield
nitroꢀsubstituted aromatic nitrogen mustard 19 in 80% yield for two steps. Finally, the desired product 6 was
obtained by classic palladiumꢀcatalyzed borylation of 19.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Synthesis of compound 6.
Compounds 7 and 10 with a cysteine methyl ester chain were synthesized from benzoic acid derivative 20.
Reduction of 20 by lithium aluminum hydride yielded the corresponding hydroxylmethyl analogue 21.
Similar to the synthesis of 14, compound 21 was converted to the nitrogen mustard analogue 24 by
Nꢀalkylation (→22), mesylation (→23), and chlorination (→24). In comparison with 12a and 12b,
compound 21 was obtained in a lower yield (40%) and required a longer reaction time (3 days) for
Nꢀalkylation possibly due to the presence of an orthoꢀhydroxymethyl group that dramatically decreasing the
nucleophilicity of the amino group. The cysteine methyl ester was selectively introduced at the benzylic
position of 24 using one equivalent of Nꢀtertꢀbutoxycarbonyl cysteine methyl ester leading to 25. Borylation
of 25 generated 7 that in turn was deprotected with TFA providing the boronic acid ester 26 and the boronic
acid 10.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 35
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 5. Synthesis of compounds 7 and 10.
Similarly, a tryptophan side chain was selectively introduced at the benzylic position of 24 with
Nꢀtertꢀbutoxycarbonyl tryptophan ester 27 and sodium hydride as a base generating 28. Subsequent
borylation of 28 yielded the boronic acid ester 8, which was converted to 11 by using TFA for deprotection
and hydrolysis.
Scheme 6. Synthesis of compounds 8 and 11.
ACS Paragon Plus Environment

Page 7 of 35
Journal of Medicinal Chemistry
Stability of the lead compounds
.
The stability is an essential factor of quality, safety, and efficacy of a drug
1
2
3
4
5
6
7
8
9
product. Initially, the chemical stability of these ROSꢀactivated DNA alkylating agents were determined in
phosphate buffered saline (PBS) at physiological pH (7.4) and endosomal pH (6.0) as well as in the presence
of GSH. The lead compounds 3, 9, and 10 were chosen for this study. To ensure the solubility of all
compounds, a mixture of DMSOꢀd6/PBS buffer was used. PBS buffer with pH 6.0 or pH 7.4 was prepared
in D2O, then diluted with DMSOꢀd6 (the ratio of DMSOꢀd6/PBS is 9:1) to provide the final buffer solution.
A solution of 10 mM 3, 9, or 10 in 100 mM PBS buffer was incubated at room temperature for 3 to 10 days
and monitored by NMR analysis (Figure S1AꢀC). Both 9 and 10 have good stability under the test
conditions. After two days, >99% of 9 and 92% of 10 remained unchanged under both conditions (pH 6.0
and pH 7.4, respectively). There was still 99% of 9 and 85% of 10 remained undecomposed after
incubation at rt for 4 days. Although compound 3 is less stable than 9 and 10, it was mainly hydrolyzed to 9
in PBS buffer. Hydrolysis of 3 was complete within 48 hours.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Glutathione (GSH) is an important antioxidant in plants and animals. It is also a strong nucleophile that may
compete with DNA to react with nitrogen mustards, therefore affecting the stability of these molecules. Thus,
we further investigate the stability of 9 and 10 in the presence of GSH. Initially, a mixture of 9 or 10 and
GSH with a ratio of 1:2 were incubated in DMSOꢀd6/PBS buffer (9:1, pH 6.0). 90% DMSO was used
because of poor solubility of 9 and 10 in phosphate buffer. For better comparison, two control samples with
compound or GSH only were prepared. The results showed that the stability of 9 and 10 in the presence of
GSH was similar to that in the absence of GSH (Figure S1D,E), indicating that GSH did not affect the
stability of these compounds. Considering that DMSO may deactivate GSH, a lower percentage of DMSO
(DMSOꢀd6/PBS buffer 1:1) was used for the stability study of 9 and 10. When DMSO percentage decreased
to lower than 50%, 9 and 10 could not dissolve well, leading to a suspension. The results showed that the
addition of GSH didn’t affect the stability of 9 and 10 in 1:1 DMSOꢀd6/PBS buffer. However, we did
observe decreased stability of 9 and 10 in a solvent mixture with higher percentage of PBS buffer (50%)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 35
(Figure S1F,G). In addition, compound 10 is less stable than 9. More than 96% of 9 remained unchanged
after 48 h incubation, while about 90% of 10 remained undecomposed after 24 h incubation and 70%
undecomposed after 48 h incubation. Based on these results, we conclude that these ROSꢀactivated DNA
alkylating agents are relatively stable under cellular conditions as well as in the presence of GSH.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In order to further investigate the reactivity of these compounds at molecular level, the reaction mixture of 3,
9, and 10 after incubation in DMSOꢀd6/PBS buffer at rt for 10 days was analyzed by ITꢀTOF mass analysis.
Compounds 9 and 31 were detected from compound 3, which was caused by hydrolysis (Scheme 7 and
Scheme S3). Similarly, the hydrolysis product 32 was observed for 9. In addition to 32, deboronation
product 33 was also formed. GSH was converted to its oxide form 34 after 10 days incubation. For 10, apart
from the hydrolysis product 35, the adducts (36 and 37) caused by intramolecular attack were observed.
Similar adducts were observed with or without the addition of GSH. No adducts were formed between GSH
and 9 or 10. These results provide the evidence for that GSH does not affect the stability of these
H2O2ꢀactivated nitrogen mustard analogues.
Scheme 7. The proposed structure for the adducts generated from 3, 9, or 10 after incubation in PBS buffer
for 10 days.
DNA crossꢀlinking assay. The primary mechanism for the cytotoxicity of nitrogen mustard reagents entails
ACS Paragon Plus Environment

Page 9 of 35
Journal of Medicinal Chemistry
DNA ICL formation.1,2 ICLs are deleterious to cells, because they inhibit DNA replication and transcription.
Additionally, it has been reported that cancer cells have higher levels of hydrogen peroxide than normal
cells.33ꢀ35,40ꢀ44 Therefore, the mode of action of the synthesized reagents can be evaluated by their ability to
induce DNA crossꢀlinks or DNA alkylations in the presence or absence of H2O2. In this work, the DNA
crossꢀlinking and selectivity of 3ꢀ11 were investigated using a 49ꢀmer DNA duplex 30 that contained one
consensus sequence for nitrogen mustards. ICL formation and yields were analyzed via denaturing
polyacrylamide gel electrophoresis (PAGE) with phosphor image analysis (Image Quant 5.2).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Without hydrogen peroxide
70
60
50
40
30
20
10
0
With hydrogen peroxide
3
4
5
6
9
7
10
8
11
Compounds
Figure 1. Comparison of the H2O2ꢀinduced activity and selectivity of 3ꢀ11. The DNA crossꢀlinking yields of
these compounds were determined by treatment of a 32Pꢀlabelled 49 mer DNA duplex with 3ꢀ11 (1.0 mM) in
the presence or absence of H2O2 (1.5 mM) at room temperature for 16 h, then subjected to 20% denaturing
PAGE analysis.
As expected, most compounds (3ꢀ6 and 9ꢀ11, 1.0 mM) in the absence of H2O2 (1.5 mM) induced very few
DNA ICL formation (1.5 ꢀ15%), while addition of H2O2 greatly enhanced their crossꢀlinking yields (up to
65%) (Figure 1 and Figure S2). Compounds with weak donating group such as (4) increased the selectivity
(defined as the ratio of DNA ICL yield with H2O2 to that without H2O2). In contract, compounds with strong
donating group (5) or strong withdrawing group (6) decreased this ratio (Figure 1). A strong donating group
offsets the deactivating effect of the boronic acid or ester group leading to increased crossꢀlinking yield (8%
for 5) in the absence of H2O2. A withdrawing group suppresses the reactivity of the liberated compound
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 35
causing low enhancement on the H2O2ꢀinduced crossꢀlinking efficiency (9% for 6). These results suggested
that electronic effects can tune the crossꢀlinking ability and selectivity of the compounds. We also
investigated compounds with amino acid side chains (7, 8, 10 and 11). Surprisingly, very low DNA ICL
yields (less than 5%) were observed for 7 and 8 either with or without H2O2. The low DNA crossꢀlinking
capability might have been caused by steric hindrance of the BOC group that prevented 7 and 8 from
effectively interacting with major or minor groves of DNA. Further investigation showed that removal of the
BOC group (10 or 11) lead to compounds with greatly enhanced DNA ICL formation. Thus, in addition to
electronic properties, steric hindrance greatly affect ICL formation of these compounds. The results further
indicated that a strong electron donating group or an amino acid side chain (5, 10, and 11) slightly increased
the DNA ICL yield in the absence of H2O2.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Next, we compared the ICL yields and the selectivity of these ROSꢀactivated DNA alkylating agents with
the same type of chemotherapy drugs Chlorambucil and Melphalan. Duplex DNA 30 was incubated with
increasing concentrations (200 ꢀM, 500 ꢀM, 1 mM) of Chlorambucil, Melphalan, 9, or 9 with H2O2
respectively at 25 oC for 16 h. As shown in figure S3, at all concentrations, 9 induced the fewest ICL yield in
the absence of H2O2 while the highest ICL yield was observed for 9 with the addition of H2O2. In particular,
at the concentration of 1 mM, the ICL yield of 9 (1.7%) was less than one third of those induced by
Chlorambucil or Melphalan (5.7% or 4.9% respectively). In contrast, 9 with H2O2 induced much higher
yield of DNA ICLs than Chlorambucil or Melphalan at the same dose (8 folds more).
Evaluation of cytotoxicity in 60 human cancer cell lines. Initially, some compounds were tested against
60 human cancer cell lines, which was performed at the National Cancer Institute (NCI Developmental
Therapeutics Program). The cell lines were derived from different cancers including leukemia,
nonꢀsmallꢀcell lung cancer, colon cancer, CNS, melanoma, ovarian, renal, prostate, and breast cancer. The
procedure can be found at: https://dtp.cancer.gov/default.htm. At a single dose of 10 µM, compounds 3ꢀ5, 9,
and 10 induced significant growth inhibition in the 60 cancer cell lines with less than 50% growth
ACS Paragon Plus Environment

Page 11 of 35
Journal of Medicinal Chemistry
percentages for most cell lines (Figure 2 and supporting information S5). In contrast, no obvious
cytotoxicity was observed with compounds containing either a strong withdrawing group (6) or an amino
acid side chain with Boc protecting group (7 and 8). This is consistent with our chemistry observation that
6ꢀ8 are poor H2O2ꢀinducible DNA crossꢀlinking agents with less than 9% ICL products. To our surprise, 11
with strong crossꢀlinking capability did not show any activity towards these cancer cell lines. Among the
active compounds, boronic acids 9 and 10 showed much higher growthꢀinhibitory effect than the boronic
acid esters 3ꢀ5. It is worth noting that although 10 didn’t show the best DNA crossꢀlinking efficiency, it
exhibited the most significant cytotoxicity towards most of the cell lines tested. For examples, the negative
growth percentages (ꢀ11% to ꢀ23%) were observed with 10ꢀtreated cancer cell lines (Figure 2), such as SR,
MOLTꢀ4, HLꢀ60, SFꢀ539, CAKIꢀ1, and MDAꢀMBꢀ468, indicating that 10 not only inhibited cancer cell
growth but led to cancer cell death. However, such phenomena was not observed with other compounds at
the same dose. These results suggested that DNA crossꢀlinking ability might not be the only parameter that
affect the cytotoxicity and efficacy of these compounds. Introduction of an amino acid side chain and/or
boronic acid group may improve the drugꢀlike properties of these compounds.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Growth % of the cell lines after incubating with 3-11 (10 ꢀM)
100
90
80
70
60
50
40
30
20
10
0
3
4
5
6
7
8
9
10
11
-10
-20
-30
SR
MOLT-4
SF-539
CAKI-1
MDA-MB-468
Figure 2. Comparison of cellular activity of 3ꢀ11 at the concentration of 10 ꢀM. 60 human tumor cell lines
representing leukemia, melanoma, and cancers of the lung, colon, brain, ovary, breast, prostate, and kidney
were incubated with 10 µM of 3ꢀ11 for 48 h at 37°C, 5 % CO2, 95 % air, and 100 % relative humidity. The
Oneꢀdose data is reported as a mean graph of the percent growth of treated cells. The number reported is
growth relative to the noꢀdrug control, and relative to the time zero number of cells.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 35
The GI50 of 3, 4, 9, and 10 was further evaluated against the 60 human cancer cell lines at five concentration
levels (supporting information S5). Consistent with the data obtained from one dose screening, 10 exhibited
the lowest GI50 in most cancer cell lines, such as CCRFꢀCEM, Kꢀ562, MOLTꢀ4, RPMIꢀ8226, SR, HOPꢀ92,
HTꢀ29, SNBꢀ75, NALMEꢀ3M, IGROV1, and MDAꢀMBꢀ468. Most of the cancer cell lines that are sensitive
to these compounds belong to leukemia or breast cancer cells that proliferate under conditions of oxidative
stress and have high intracellular concentrations of ROS (e.g. superoxide, peroxides, and hydroxyl
radicals).18 The boronic acid analogue 9 is slightly more toxic than the pinacol ester 3. Compound 10 with
an amino acid chain is 1.7 to 18 folds more toxic than the parent compound 9, 2.3 to 9.4 folds more toxic
than 3, and 1.7 to 23.8 folds more toxic than 4. Although 10 is not the most powerful DNA crossꢀlinking
agent, the presence of the amino acid ester chain improves the biophysical properties such as solubility and
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
permeability therefore making it the best drugꢀlike molecule (Table 2). The boronic acids (9 and 10) with
higher activity have better solubility in PBS (328 µM for 9 and 231 µM for 10) than 3ꢀ5 (24 µM to 167 µM).
Among these agents tested, although compound 10 with an amino ester chain didn’t show the highest ICL
yield in the presence of hydrogen peroxide, the in vitro results demonstrated that this molecule showed the
highest inhibition ability towards most cancer cell lines. We are aware of that cellular activity is also related
to drug delivery and is more complex than DNA crossꢀlinking capability. Nevertheless, the investigation of
the H2O2ꢀinduced DNA crossꢀlinking ability and selectivity gave us a good sign that these agents are
potential anticancer reagents, which encourage us to continue the further study.
Table 1. GI50 of compounds 3, 4, 9, and 10.a
GI50 (µM)
Tumor type
Cell Line
3
4
9
10
CCRFꢀCEM
Kꢀ562
MOLTꢀ4
RPMIꢀ8226
SR
3.34
17.2
3.48
10.9
0.63
9.24
2.90
5.48
2.60
12.0
0.49
13.3
3.27
15.8
2.90
8.59
0.48
10.5
0.54
3.25
0.6
2.62
0.28
0.56
leukemia
nonꢀsmallꢀcell lung
HOPꢀ92
ACS Paragon Plus Environment

Page 13 of 35
Journal of Medicinal Chemistry
colon cancer
CNS
melanoma
HTꢀ29
SNBꢀ75
MALMEꢀ3M
IGROV1
SKꢀOVꢀ3
MDAꢀMBꢀ468
MCF
13.8
7.98
17.5
15.6
8.03
1.60
4.12
14.7
10.1
13.6
16.1
4.62
7.88
3.86
11.9
3.21
19.7
18.1
3.81
0.51
1.89
4.03
0.85
2.39
3.32
1.46
1.49
0.72
1
2
3
4
5
6
7
8
Ovarian
Breast Cancer
a60 human tumor cell lines were incubated with 5ꢀdose (100 µM, 10 µM, 1 µM, 100
nM, and 10 nM) of compounds 3, 4, 9 and 10 for 48 h at 37°C, 5 % CO2, 95 % air, and
100 % relative humidity. The number reported for 5ꢀdose assay is growth relative to the
noꢀdrug control, and relative to the time zero number of cells which is the same as
oneꢀdose assay.
9
10
11
12
13
14
15
16
Table 2. Biophysical properties of 3ꢀ5, 9 and 10.
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
4
5
6
9
10
Solubility (µM)a
Permeabilityb
167
ꢀ6.31
24
ꢀ
50
380
ꢀ6.63
328
ꢀ6.67
231
ꢀ6.34
ꢀ5.38
aSolubilities were determined in phosphate buffered saline at pH 7.4. Permeabilities were measured using
the parallel artificial membrane permeation assay (PAMPA) at neutral pH (pH 7.4). The following
permeability standards (log Pe) were used: ranitidine (−8.02 ± 0.074 cm/s) low permeability, carbamazepine
(−6.81 ± 0.0011 cm/s) medium permeability, and verapamil (−5.93 ± 0.015 cm/s) high permeability. The
solubility and permeability assay conditions reflect conditions required for activity in cellꢀbased assays.
b
Cytotoxicity comparison of H2O2ꢀactivated DNA alkylating agents with clinically useful drugs. Given
that these compounds showed significant cytotoxicity in several cancer cell lines, we compared their efficacy
with two of the clinically used DNA alkylating agents, chlorambucil and melphalan, which belong to the
aromatic nitrogen mustard analogues acting on cancer cells by alkylating DNA. Chlorambucil is a
chemotherapy drug for chronic lymphocytic leukemia (CLL)45 and melphalan is used for treatment of
advanced breast cancer.46 Thus, we investigated their effects towards breast cancer MDAꢀMBꢀ468 cells and
primary CLL cells. It was reported that breast cancer cells and CLL cells have high intracellular
concentrations of ROS.43 The doseꢀdependent response analysis with MDAꢀMBꢀ468 cells showed that
among all H2O2ꢀactivated compounds, 10 was the most cytotoxic compound with an IC50 of 3.43 µM, which
was 10ꢀ14 folds more toxic than chlorambucil (IC50 48.7 µM) and melphalan (IC50 34.44 µM) (Figure 3).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 35
1
2
3
4
5
6
7
8
MDA-MB-468
MDA-MB-468
9
3
4
5
6
10
120
100
80
60
40
20
0
120
100
80
60
40
20
0
Chlorambucil
Melphalan
IC50 (ꢀM)
IC50 (ꢀM)
9: 16.7±1.4
10: 3.4±0.3
Chlo: 34.4±6.5
Melph: 48.7±6.6
3: 24.6±4.1
4: 48.3±7.2
5: 34.4±6.5
6: 92.8±25.7
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
-1
0
1
2
3
-1
0
1
2
3
log [Conc.] (ꢀM)
log [Conc.] (ꢀM)
Figure 3. Comparison of inhibition activity of 3ꢀ6 and 9ꢀ10 with that of chlorambucil and melphalan. The
Fluorescence polarization (FP) assay was conducted in 384ꢀwell white polystyrene microplates (Corning,
#3570). Compounds transfer into 20 ꢁl assay solution was accomplished using a stainless steel pin tool
(V&P Scientific) delivering 200 nl of compound at different concentrations. Incubation time is 48 h.
Inhibition of binding was detected by fluorescence polarization using a M1000, Tecan reader at
excitation/emission wavelength of 630/685 nm. IC50 value (ꢁM) was calculated using the following
nonꢀlinear regression equation: Y = Bottom + (TopꢀBottom)/(1+10^((LogIC50ꢀX)*HillSlope)).
Similarly, doseꢀdependent experiments on evaluation of apoptosis using primary CLL cells with three
representative compounds (5, 10, and chlorambucil) side by side in the same patient sample (n=6)
demonstrated that 10 is more toxic to CLL lymphocytes than chlorambucil and H2O2ꢀactivated reagent 5
(Figure 4). The IC50 of 10 in CLL lymphocytes ranged from 3 nM to 778 nM, while for chlorambucil and
compound 5 the IC50 were not achieved under the dosages tested. Our previous study showed that the IC50 of
the lead compound 9 was 5000ꢀ6000 nM, demonstrating less activity than 10 (Note: five samples obtained
from five patients were tested. Due to heterogeneous response between samples, the IC50 range was wide)28.
Collectively, these data demonstrate that among these compounds, compound 10 is highly potent to induce
apoptosis in CLL lymphocytes compared to chemotherapy drugs, indicating this could be the best drugꢀlike
molecule.
(A)
(B)
(C)
ACS Paragon Plus Environment

Page 15 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
Figure 4. Doseꢀdependent apoptosis of CLL cells with compounds 5, 10, and chlorambucil. CLL
lymphocytes were obtained from CLL patients and incubated with series of concentrations of compounds 5,
10, and chlorambucil for 24 h, and the apoptosis induction was measured by annexin/PI binding assay: (A)
compound 5 (n = 5); (B) compound 10 (n = 5); (C) Chlorambucil (n = 2). Each colored line represents a
patient and the data point on the line denotes the concentration of the drug. The survival curve is drawn
using GraphPad Prism 6.0.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To evaluate the selectivity of these compounds to cancer cells, normal lymphocytes were tested for toxicity
with two representative compounds (5 and 10, n=3). The statistical evaluation demonstrated that even at
highest concentrations of 10 µM, no obvious toxicity observed with either compound (Figure 5).
Figure 5. Evaluate apoptosis in normal lymphocytes: Normal PBMCs were obtained from normal donors
(n=3) and the toxicity of compounds 5 and 10 (10 uM) were evaluated by annexin / PI binding method.
Mechanism of action. H2AX is a protein that is responsible for DNA damage repair.47 Formation of γH2AX
is a measure of DNA damage response. One DNA damage, such as DNA strand break or DNA crossꢀlink,
leads to 200 phosphorylation of H2AX forming γH2AX. The reagentꢀtreated CLL primary cells that were
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 35
tested for apoptosis were also used to assess H2AX phosphorylation by flow cytometry. As expected, there
was an increase in the γH2AX formation in prodrugꢀtreated CLL cells (Figure 6; n=3). Among 3 compounds
tested on the CLL leukemic lymphocytes from 3 subjects, 10 led to the most severe DNA damage as well as
apoptosis because the maximum H2AX phosphorylation was observed for 10ꢀtreated CLL cells. Compound
2a, a positive control, also led to an increase in H2AX phosphorylation in 2aꢀtreated CLL cells, while 29 (a
negative control) did not have such effect on CLL cells. Collectively these data are consistent with our
hypothesis.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OH
O
O
B
N
Cl
Cl
OH
2a
29
Negative control
Positive control
A.
B.
n = 3
15
10
5
0
n
M
M
u
o
u
C
1
0
,
1
0
,
1
5
d
p
m
p
m
o
o
C
C
Figure 6. Induction of DNA damage response measured by H2AX phosphorylation. A. CLL lymphocytes
(n=3) were incubated with compounds 5 (10 µM) and compound 10 (1 µM); B. CLL lymphocytes (n=5) were
incubated with 2a and 29 at increasing concentrations (5 – 20 µM) for 24 hrs and H2AX phosphorylation was
measured by flow cytometry
.
In vivo efficacy of the ROSꢀactivated anticancer reagents. Suspension cells like lymphocytes often show
a higher sensitivity to chemotherapeutics due to their higher exposure in solution when compared to
adherent cancer cells. Therefore, we focus on MDAꢀMBꢀ468 cell line and use them for in vivo breast cancer
ACS Paragon Plus Environment

Page 17 of 35
Journal of Medicinal Chemistry
xenograft. Toxicity study with mice showed that no obvious toxicity was observed for 10 within three
months period when the mice were injected with 1.5 mg /kg of 10 per day and 5 days per week. Thus, a
dosage of 1.5 mg/kg was used for determining the in vivo efficacy of 10 in a MDAꢀMBꢀ468ꢀderived
xenograft mouse model. Tumorꢀbearing mice were treated i.p. with vehicle or 10 (1.5 mg/kg) (Figure 7). All
mice were weighed regularly and the tumor size measured using a caliper every week. After 7 weeks of
treatment, tumor sizes in 10ꢀtreated mice ranged from 58% to 170% of the initial size (Figure 7A). In
contrast, the control mice showed significant tumor growth, reaching 278%ꢀ657% of the original tumor size
in that period. Normal increase of body weight was observed for all mice (Figure 7B). Together, these
results indicated that 10 efficiently suppressed tumor growth in mice without obvious side effects.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. Compound 10 inhibits tumor growth in athymic nude mice. A dosage of 1.5 mg/kg was used for
determining the in vivo efficacy of 10 in a MDAꢀMBꢀ468ꢀderived xenograft mouse model. Tumorꢀbearing
mice were treated i.p. with vehicle (DMSO only) or 10 (1.5 mg/kg) (11 mice per group). (A) In vivo tumor
volumes after injection of DMSO only (control) or 1.5 mg/kg of 10; (B) comparison of mice weight in
DMSOꢀtreated mice and 10ꢀtreated mice.
CONCLUSION
In summary, our investigation of a series of novel aromatic nitrogen mustard arylboronate esters revealed
that the aromatic substituents significantly affected their DNA crossꢀlinking ability and cellular activity. A
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 35
weak electron donating group introduced in the benzene ring increases the selectivity (defined as the ratio of
ICL yield with H2O2 to that without H2O2) while a strong donating group or a withdrawing group decreases
the selectivity. Bulky groups block the interaction between these compounds and DNA therefore inhibiting
DNA crossꢀlinking and leading to low cellular activity. Introduction of a boronic acid and an amino acid
side chain greatly improved the solubility and cell membrane permeability, leading to higher cellular activity.
Compound 10 containing a boronic acid and a cysteine methyl ester side chain showed the highest cellular
activity towards most cell lines tested. In particular, breast cancer cell lines and leukemia cell lines are
highly sensitive to 10 that is 10ꢀfold more potent than Mephalan, 14ꢀfold more potent than Chlorambucil
and 5ꢀfold more potent than the lead compound 9. Compound 10 exhibited an IC50 of 3 nM to 778 nM
towards primary CLL cells obtained from peripheral blood of patients, while much higher IC50 was observed
for the lead compound 9 (IC50 = 5ꢀ6 µM) and the IC50 was not achieved for chlorambucil under tested
dosages. More importantly, compound 10 efficiently inhibited tumor growth in a MDAꢀMBꢀ468ꢀderived
xenograft mouse model while no obvious side effects were observed, such as no weight loss and other
unusual behavior. Normal increase of body weight was observed for both compoundsꢀtreated mice and
control mice. Collectively, these data demonstrated the selectivity and therapeutic utility of these agents,
which inspires further and effective application as potential cancer chemotherapies.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
Chemistry. Unless otherwise specified, chemicals were purchased from Aldrich or Fisher Scientific and
were used as received without further purification. T4 polynucleotide kinase was purchased from New
England Biolabs. Oligonucleotides were synthesized via standard automated DNA synthesis techniques
using an Applied Biosystems model 394 instrument in a 1.0 ꢁmole scale using commercial 1000 Å
CPGꢀsuccinylꢀnucleoside supports. Deprotection of the nucleobases and phosphate moieties and cleavage of
the linker were carried out under mild deprotection conditions using a mixture of 40% aqueous MeNH2 and
28% aqueous NH3 (1:1) at room temperature for 2 h. [γꢀ32P]ATP was purchased from PerkinꢀElmer Life
ACS Paragon Plus Environment

Page 19 of 35
Journal of Medicinal Chemistry
Sciences. Quantification of radiolabeled oligonucleotides was carried out using a Molecular Dynamics
1
2
3
4
1
phosphorimager equipped with ImageQuant, version 5.2, software. H NMR and 13C NMR spectra were
5
6
7
8
taken on either a Bruker DRX 300 or DRX 500 MHz spectrophotometer. Silicon reagents were used in
CDCl3 as internal standard. Highꢀresolution mass spectrometry ITꢀTOF was used for measurement of
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
molecular mass. The purity was determined by RPꢀHPLC on a 4.6
×
250 mm RPꢀC18 column with 254 nm
detection, which confirmed that all compounds had
structures are shown in support information S7.
≥ 95% purity. Systematic numbering of chemical
N,NꢀBis(2ꢀchloroethyl)ꢀ3ꢀmethylꢀ4ꢀ(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)aniline (4). Step 1. A
mixture of 4ꢀbromoꢀ2ꢀmethylaniline (0.1 mol), 2ꢀchloroethanol (40 mL), CaCO3 (20.0 g), and NaI (2.8 g) in
250 mL water was heated to reflux overnight for 24 h, then extracted with dichloromethane and washed with
water. After evaporation of the solvent, the residue was purified by column chromatography (Hexane/Ethyl
Acetate = 1:2) to afford 2,2'ꢀ((4ꢀBromoꢀ3ꢀmethylphenyl)azanediyl)diethanol (13a) as a white solid (14.8 g,
55%). mp 46.0ꢀ47.5 ℃; 1H NMR (DMSOꢀd6, 300 MHz): δ 2.25 (s, 3H, H3ꢀC(15')), 3.38 (t, J = 6.0 Hz, 4H,
H2ꢀC(10'), H2ꢀC(13')), 3.53 (t, J = 6.0 Hz, 4H, H2ꢀC(9'), H2ꢀC(12')), 3.73 (t, J = 5.4 Hz, 2H, HꢀO(11'),
HꢀO(14')), 6.45 (dd, J = 3.0 Hz, J = 9.0 Hz, 1H, HꢀC(2')), 6.65 (d, J = 3.0 Hz, 1H, HꢀC(4')), 7.24 (d, J = 9.0
13
Hz, 1H, HꢀC(1')); C NMR (CDCl3, 75 MHz): δ 23.3, 53.6, 58.5, 109.2, 111.5, 114.3, 132.5, 137.5, 147.9;
HRMSꢀES (m/z) [M+H]+ calcd for C11H16BrNO2: 274.0425, found: 274.0437.
Step 2. A solution of MsCl (10 mL, 57.5 mmol) in DCM (10 mL) was added dropwise to a mixture of 13a
(10.8 g, 39.7 mmol) and Et3N (7.5 mL, 57.5 mmol) in dry CH2Cl2 (50mL) at 0 °C. After 30 min, the mixture
was extracted with CH2Cl2 twice and the combined organic phase was washed with brine water, dried over
Na2SO4 and then concentrated under vacuum. The residue was used in the next step without any further
purification. The residue was dissolved in DMF (40 mL) and LiCl (4 g, 115 mmol) was added. After stirred
at 70 °C for 3 h, the mixture was extracted with CH2Cl2 and washed with brine water, dried over Na2SO4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 35
and concentrated under vacuum. The residue was purified by column chromatography (Hexane/Ethyl
1
2
3
4
Acetate = 10:1) to afford 4ꢀBromoꢀN,Nꢀbis(2ꢀchloroethyl)ꢀ3ꢀmethylaniline (14a) as a white foam (10.5 g,
5
6
1
85%). H NMR (CDCl3, 300 MHz): δ 2.28 (s, 3H, H3ꢀC(15')), 3.71ꢀ3.72 (m, 8H, H2ꢀC(10'), H2ꢀC(13'),
7
8
9
H2ꢀC(9'), H2ꢀC(12')), 6.53 (dd, J = 3.0 Hz, J = 9.0 Hz, 1H, HꢀC(2')), 6.75 (d, J = 3.0 Hz, 1H, HꢀC(4')), 7.31
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
(d, J = 9.0 Hz, 1H, HꢀC(1')); C NMR (CDCl3, 75 MHz): δ 23.3, 41.5, 52.5, 111.3, 112.1, 114.9, 132.9,
138.2, 146.3; HRMSꢀES (m/z) [M+H]+ calcd for C11H14BrNCl2: 309.3748, found: 309.9759, chlorine
isotopic peak [M+H+2]+ found 311.9717.
o
n
Step 3. A solution of 14a (1.6 g, 5.2 mmol) in dry THF (60 mL) was cooled to ꢀ78 C under Ar. BuLi (8.0
mL, 2.6 M in Hexane) was added slowly at the same temperature within 10 min. After 30 min, B(OiPr)3 (2.9
g, 15 mmol) was added. The mixture was allowed to warm to room temperature and stirred overnight, then
quenched by NH4Cl solution at 0 oC. The mixture was extracted with CH2Cl2, washed with water, dried over
Na2SO4 and concentrated under vacuum. The residue was purified by column chromatography
(Hexane/Ethyl Acetate = 1:1) to afford white solid (4ꢀ(bis(2ꢀchloroethyl)amino)ꢀ2ꢀmethylphenyl)boronic
1
acid (1.0 g, 70%). H NMR (DMSOꢀd6+D2O, 300 MHz): δ 2.36 (s, 3H, H3ꢀC(19')), 3.70ꢀ3.71 (m, 8H,
H2ꢀC(9'), H2ꢀC(10'), H2ꢀC(12'), H2ꢀC(13')), 6.47ꢀ6.49 (m, 2H, HꢀC(2'), HꢀC(4')), 7.39 (d, J = 9.0 Hz, 1H,
H2ꢀC(1')); 13C NMR (DMSOꢀd6+D2O, 75 MHz): δ 23.3, 41.7, 52.3, 108.6, 113.4, 136.4, 144.7, 147.6;
HRMSꢀEI (m/z) [M+H]+ calcd for C11H16BNO2Cl2: 276.0713, found: 276.0726. To a mixture of
(4ꢀ(bis(2ꢀchloroethyl)amino)ꢀ2ꢀmethylphenyl)boronic acid (1.0 g, 3.63 mmol) and magnesium sulfate (1 g)
in dry Et2O (50mL) was added pinacol (857 mg, 7.26 mmol) at room temperature. After stirring overnight,
the mixture was concentrated under vacuum and purified by column chromatography (Hexane/Ethyl Acetate
1
= 10:1) to afford a white solid 4 (1.1 g, 85%). mp 89.4ꢀ90.2
℃
; H NMR (CDCl3, 300 MHz): δ 1.34 (s, 12H,
H3ꢀC(19'), H3ꢀC(20') , H3ꢀC(21'), H3ꢀC(23')), 2.54 (s, 3H, H3ꢀC(22')), 3.66 (t, J = 6.3 Hz, 4H, H2ꢀC(10'),
H2ꢀC(13')), 3.77 (t, J = 6.3 Hz, 4H, H2ꢀC(9'), H2ꢀC(12')), 6.49ꢀ6.52 (m, 2H, HꢀC(2'), HꢀC(4')), 7.71 (d, J =
8.1 Hz, 1H, HꢀC(1')); 13C NMR (CDCl3, 75 MHz): δ 22.9, 24.9, 40.4, 53.2, 82.9, 108.2, 112.9, 138.2, 147.3,
ACS Paragon Plus Environment

Page 21 of 35
Journal of Medicinal Chemistry
148.0; HRMSꢀES (m/z) [M+H]+ calcd for C17H26NO2Cl2B: 358.1500, found: 358.1510, chlorine isotopic
1
2
3
4
peak [M+H+2]+ found 360.1467.
5
6
7
8
9
N,NꢀBis(2ꢀchloroethyl)ꢀ3ꢀmethoxyꢀ4ꢀ(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)aniline (5). Step 1.
2,2'ꢀ((4ꢀBromoꢀ3ꢀmethoxyphenyl)azanediyl)diethanol (13b) was prepared according to representative
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
example 13a from 4ꢀbromoꢀ2ꢀmethoxyaniline 12b (20 g). White solid (14.5 g, 50%); mp 107.0ꢀ108.6
℃;
1H NMR (CDCl3, 300 MHz): δ 3.62 (t, J = 4.8 Hz, 4H, H2ꢀC(10'), H2ꢀC(13')), 3.88 (t, J = 4.8 Hz, 4H,
H2ꢀC(9'), H2ꢀC(12')), 3.90 (s, 3H, H3ꢀC(15')), 6.76 (d, J = 8.4 Hz, 2H, HꢀC(2'), HꢀC(4')), 7.38 (d, J = 8.4 Hz,
1H, HꢀC(1')); 13C NMR (CDCl3, 75 MHz): δ 55.3, 60.5, 108.8, 114.2, 131.9, 146.8; HRMSꢀES (m/z)
[M+H]+ calcd for C11H16BrNO3: 290.0380, found: 290.0386.
Step 2. 4ꢀBromoꢀN,Nꢀbis(2ꢀchloroethyl)ꢀ3ꢀmethoxyaniline (14b) was synthesized from 13b (11.5 g) in a
similar procedure to the synthesis of 14a. White solid (11.4 g , 88%); mp 92ꢀ93
℃
; 1H NMR (CDCl3, 300
MHz): δ 3.63ꢀ3.68 (m, 4H, H2ꢀC(10'), H2ꢀC(13')), 3.72ꢀ3.77 (m, 4H, H2ꢀC(9'), H2ꢀC(12')), 3.90 (s, 3H,
H3ꢀC(15')), 6.22 (dd, J = 2.1 Hz, J = 9.0 Hz, 1H, HꢀC(2')), 6.29 (d, J = 2.1 Hz, 1H, HꢀC(4')), 7.37 (d, J = 9.0
13
Hz, 1H, HꢀC(1')); C NMR (CDCl3, 75 MHz): δ 40.3, 53.8, 56.2, 97.4, 99.7, 105.9, 133.7, 146.8, 156.9;
HRMSꢀES (m/z) [M+H]+ calcd for C11H14OBrNCl2: 325.9700, found: 325.9709, chlorine isotopic peak
[M+H+2]+ found 327.9685.
Step 3. Compound 5 was prepared from 14b (1.7 g) in a similar procedure to the synthesis of 4. White solid
(0.78 g, 40%); mp 107.0ꢀ108.2
℃; 1H NMR (CDCl3, 300 MHz): δ 1.35 (s, 12H, H3ꢀC(19'), H3ꢀC(20') ,
H3ꢀC(21'), H3ꢀC(23')), 3.66 (t, J = 6.3 Hz, 4H, H2ꢀC(10'), H2ꢀC(13')), 3.78 (t, J = 6.3 Hz, 4H, H2ꢀC(9'),
H2ꢀC(12')), 3.85 (s, 3H, H3ꢀC(22')), 6.16 (d, J = 2.1 Hz, 1H, HꢀC(4')), 6.27 (dd, J = 2.1 Hz, J = 8.4 Hz, 1H,
HꢀC(2')), 7.62 (d, J = 8.4 Hz, 1H, HꢀC(1')); 13C NMR (CDCl3, 75 MHz): δ 24.8, 40.4, 53.5, 56.0, 82.9, 95.1,
104.0, 138.8, 150.1, 166.5; HRMSꢀES (m/z) [M+H]+ calcd for C17H26BNO3Cl2: 374.1465, found: 374.1459,
chlorine isotopic peak [M+H+2]+ found 376.1443.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 35
1
2
3
4
5
6
7
8
9
N,NꢀBis(2ꢀchloroethyl)ꢀ2ꢀnitroꢀ4ꢀ(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)aniline (6). Step 1.
4ꢀBromoꢀN,Nꢀbis(2ꢀchloroethyl)ꢀ2ꢀnitroaniline (19) was prepared from compound 17 (12.1 g) in a similar
procedure to the synthesis of 14a. Yellow liquid (11.2 g, 83% for two steps). 1H NMR (CDCl3, 300 MHz): δ
3.49ꢀ3.59 (m, 8H, H2ꢀC(10'), H2ꢀC(11'), H2ꢀC(12'), H2ꢀC(13')), 7.27 (d, J = 8.7 Hz, 1H, HꢀC(2')), 7.62 (dd, J
= 2.4 Hz, J = 8.7 Hz, 1H, HꢀC(1')), 7.88 (d, J = 2.4 Hz, 1H, HꢀC(5')); 13C NMR (CDCl3, 75 MHz): δ 41.4,
55.3, 115.7, 127.5, 128.3, 136.3, 141.9, 145.6; HRMSꢀES (m/z) [M+H]+ calcd for C10H11BN2O2Cl2:
340.9430, found: 340.9454, chlorine isotopic peak [M+H+2]+ found 342.9407.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Step 2. A mixture of 19 (1.36 g, 4.0 mmol), bis(pinacolato)diboron (1.0 g, 4.0 mmol), KOAc (0.8 g, 8.0 mol),
and PdCl2(dppf) (0.33 g, 0.4 mmol) in dioxane (50 ml) was flushed with argon for 10 minutes and heated to
reflux overnight under argon. After cooling to room temperature, the mixture was extracted with ethyl
acetate and washed with brine. The organic layers was collected, dried over Na2SO4 and evaporated under
vacuum, the residue was purified by column chromatography (Hexane/Ethyl Acetate = 10:1) to afford 6 as a
yellow solid (1.3 g, 84%). mp 56.6ꢀ58.6
℃; 1H NMR (CDCl3, 300 MHz): δ 1.36 (s, 12H, H3ꢀC(20'),
H3ꢀC(21'), H3ꢀC(22'), H3ꢀC(23')), 3.55ꢀ3.58 (m, 8H, H2ꢀC(9'), H2ꢀC(10'), H2ꢀC(12'), H2ꢀC(13')), 7.29 (d, J =
9.1 Hz, 1H, HꢀC(2')), 7.89 (d, J = 9.1 Hz, 1H, HꢀC(1')), 8.15 (s, 1H, HꢀC(5')); 13C NMR (CDCl3, 75 MHz): δ
24.8, 41.2, 54.8,84.4, 123.4, 132.3, 139.2, 144.0, 144.7; HRMSꢀES (m/z) [M+H]+ calcd for C16H23N2O4BCl2:
389.1215, found: 389.1204, chlorine isotopic peak [M+H+2]+ found 3491.1187.
Methyl
3ꢀ((2ꢀ(bis(2ꢀchloroethyl)amino)ꢀ5ꢀ(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)benzyl)thio)ꢀ2ꢀ((tertꢀbu
toxycarbonyl)amino)propanoate (7). Step 1. To a suspension of LiAlH4 (1.75 g, 46.3 mmol) in dry Et2O
(50 mL) cooled at 0 °C, a solution of 20 (15 g, 23.15 mmol) in dry Et2O (50 mL) was added dropwise in 15
minutes. The resulting mixture was stirred at room temperature for 2 h and then it was quenched by slowly
ACS Paragon Plus Environment

Page 23 of 35
Journal of Medicinal Chemistry
adding EtOAc (50 mL). The reaction mixture was poured into H2O (300 mL) and extracted with EtOAc (3 x
250 mL). The combined organic layers were washed with brine (200 mL), dried over Na2SO4 and
evaporated under reduced pressure. The crude product was purified by column chromatography
(Hexane/Ethyl Acetate = 1:2) to obtain (2ꢀaminoꢀ5ꢀbromophenyl)methanol (21) as a white solid (4.28 g,
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
92%); mp 112ꢀ113
℃
; 1H NMR (DMSOꢀd6, 300 MHz) δ 4.37 (d, J = 5.7 Hz, 2H, H2ꢀC(8')), 5.23ꢀ5.26 (m,
3H, H2ꢀC(9'), HꢀO(10')), 6.38 (dd, J = 2.1 Hz, J = 8.4 Hz, 1H, HꢀC(1')), 6.80 (d, J = 2.1 Hz, 1H, HꢀC(5')),
13
7.11 (d, J = 8.4 Hz, 1H, HꢀC(2')); C NMR (CDCl3, 75 MHz): δ 63.1, 105.8, 114.1, 114.3, 132.3, 141.4,
148.8, 148.9; HRMSꢀES (m/z) [M+H]+ calcd for C7H8NOBr: 201.9849, found: 201.9862.
Step 2. 2,2'ꢀ((4ꢀBromoꢀ2ꢀ(hydroxymethyl)phenyl)azanediyl)diethanol (22) was prepared according to
1
representative example 13a from compound 21 (20.2 g). White solid (11.6 g, 40%); mp 106ꢀ107
℃
; H
NMR (CDCl3, 300 MHz): δ 3.14 (d, J = 5.1 Hz, 4H, H2ꢀC(11'), H2ꢀC(13')), 3.38 (s, 2H, HꢀO(12') , HꢀO(15')),
3.64 (d, J = 5.1 Hz, 4H, H2ꢀC(10'), H2ꢀC(14')), 4.74 (s, 2H, H2ꢀC(8')), 7.20 (d, J = 8.7 Hz, 1H, HꢀC(2')),
13
7.44ꢀ7.48 (m, 2H, HꢀC(1'), HꢀC(5')); C NMR (CDCl3, 75 MHz): δ 57.2, 59.4, 62.9, 119.0, 126.6, 132.1,
133.2, 140.9, 149.2; HRMSꢀES (m/z) [M+H]+ calcd for C11H16BrNO3: 290.0378, found: 290.0386.
Step 3. 4ꢀBromoꢀN,Nꢀbis(2ꢀchloroethyl)ꢀ2ꢀ(chloromethyl)aniline (24) was prepared from compound 22
1
(11.5 g) in a similar procedure to the synthesis of 14 from 13. Oil (10.3 g, 76.5% for two steps); H NMR
(CDCl3, 300 MHz): δ 3.41ꢀ3.53 (m, 8H, H2ꢀC(10'), H2ꢀC(11'), H2ꢀC(13'), H2ꢀC(14')), 4.79 (s, 2H, H2ꢀC(8')),
7.16 (d, J = 8.7 Hz, 1H, HꢀC(2')), 7.47 (dd, J = 2.4 Hz, J = 8.4 Hz, 1H, HꢀC(2')), 7.67 (d, J = 2.4 Hz, 1H,
HꢀC(1')); 13C NMR (CDCl3, 75 MHz): δ 41.3, 41.6, 57.1, 119.4, 126.6, 132.7, 134.3, 138.3, 146.5;
HRMSꢀES (m/z) [M+H]+ calcd for C11H13BrNCl3: 343.9349, found: 343.9370, chlorine isotopic peak
[M+H+2]+ found 345.9114.
Step 4. BocCysOMe (2.37 g, 7.1 mmol) was dissolved in Methanol (50 mL), K2CO3 (4.5 g, 32 mmol) and
24 (2.2 g, 6.4 mmol) were added and the reaction flushed with nitrogen. The reaction was stirred at room
temperature for 2 hours and then evaporated under vacuum, the residue was extracted with EtOAc (200 mL)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 35
and washed with H2O (200 mL) followed by the same amount of brine. The organic layer was dried
(MgSO4), filtered, and concentrated under reduced pressure. The crude product was purified by column
1
2
3
4
5
6
chromatography
(Hexane/Ethyl
Acetate
=
5:1)
to
provide
7
8
9
3ꢀ((2ꢀ(bis(2ꢀchloroethyl)amino)ꢀ5ꢀbromobenzyl)thio)ꢀ2ꢀ((tertꢀbutoxycarbonyl)amino)propanoate (25) as a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
white liquid (2.6 g, 75%). H NMR (CDCl3, 300 MHz) δ 1.43 (s, 9H, H3ꢀC(25') , H3ꢀC(26') , H3ꢀC(27')),
2.87ꢀ3.04 (m, 2H, H2ꢀC(17')), 3.36 (t, J = 6.0 Hz, 4H, H2ꢀC(11'), H2ꢀC(13')), 3.45 (t, J = 6.0 Hz, 4H,
H2ꢀC(10'), H2ꢀC(14')), 3.73 (s, 3H, H3ꢀC(19')), 3.84 (d, J = 13.2 Hz, 1H, HꢀC(8')), 3.90 (d, J = 13.2 Hz, 1H,
HꢀC(8')), 4.53ꢀ4.55 (m, 1H, HꢀC(18')), 5.40 (d, J = 7.8 Hz, 1H, HꢀN(20')), 7.09 (d, J = 8.4 Hz, 1H, HꢀC(2')),
13
7.33 (dd, J = 2.1 Hz, J = 8.4 Hz, 1H, HꢀC(1')), 7.53 (d, J = 2.1 Hz, 1H, HꢀC(5')); C NMR (CDCl3, 75
MHz): δ 28.3, 31.7, 35.1, 41.6, 52.5, 52.6, 53.2, 56.9, 57.2, 80.1, 118.9, 126.6, 131.4, 133.8, 138.3, 146.7,
155.2, 171.5; HRMSꢀES (m/z) [M+H]+ calcd for C20H29BrCl2N2O4S: 543.0472, found: 543.0481, chlorine
isotopic peak [M+H+2]+ found 545.0459.
Step 5. A mixture of 25 (542 mg, 1.0 mmol), bis(pinacolato)diboron (500 g, 2.0 mmol), KOAc (167 mg, 1.7
mmol), and PdCl2(dppf) (165 mg, 0.2 mmol) in dioxane (25 ml) was flushed with argon for 10 minutes and
heated to reflux for 5 h under argon. After cooling to room temperature, the mixture was extracted with ethyl
acetate and washed with brine. The organic layers was collected, dried over Na2SO4 and evaporated under
vacuum, the residue was purified by column chromatography (Hexane/Ethyl Acetate = 5:1) to afford 7 as a
1
yellow oil (460 mg, 78%). H NMR (CDCl3, 300 MHz): δ 1.31 (s, 12H, H3ꢀC(32'), H3ꢀC(33'), H3ꢀC(34'),
H3ꢀC(35')), 1.43 (s, 9H, H3ꢀC(25') , H3ꢀC(26') , H3ꢀC(27')), 2.99ꢀ3.00 (m, 2H, H2ꢀC(17')), 3.41ꢀ3.49 (m, 8H,
H2ꢀC(11'), H2ꢀC(13') , H2ꢀC(10'), H2ꢀC(14')), 3.74 (s, 3H, H3ꢀC(19')), 3.87 (d, J = 11.7 Hz, 1H, HꢀC(8')),
3.94 (d, J = 11.7 Hz, 1H, HꢀC(8')), 4.54ꢀ4.56 (m, 1H, HꢀC(18')), 5.43 (d, J = 7.8 Hz, 1H, HꢀN(20')), 7.19 (d,
13
J = 8.4 Hz, 1H, HꢀC(2')), 7.67 (d, J = 8.4 Hz, 1H, HꢀC(1')), 7.79 (s, 1H, HꢀC(5')); C NMR (CDCl3, 75
MHz): δ 25.0, 28.3, 32.5, 35.4, 41.5, 52.5, 53.3, 56.7, 80.0, 83.8, 123.7, 134.6, 134.9, 137.9, 150.3, 155.2,
171.6; HRMSꢀES (m/z) [M+H]+ calcd for C26H41BBrCl2N2O6S: 591.2218, found: 591.2233, chlorine
ACS Paragon Plus Environment

Page 25 of 35
Journal of Medicinal Chemistry
isotopic peak [M+H+2]+ found 593.2188.
1
2
3
4
5
6
7
8
(3ꢀ(((2ꢀAminoꢀ3ꢀmethoxyꢀ3ꢀoxopropyl)thio)methyl)ꢀ4ꢀ(bis(2ꢀchloroethyl)amino)phenyl) boronic acid
(10). 500 mg of compound 7 was dissolved in 5 mL of DCM, 0.6 mL of trifluoroacetic acid was added via a
syringe within 10 min. The mixture was allowed to stir overnight and evaporated under vacuum. The residue
was dissolved in a solution of DCM (20 mL) and H2O (20 mL) and neutralized with K2CO3 (1 M) to pH 6ꢀ7.
The mixture was extracted with DCM and washed with brine. The organic phase was collected and
concentrated. The crude products were purified by column chromatography (Hexane/Ethyl Acetate = 1:1
then DCM/MeOH = 20:1) to afford a white foam 10 (90 mg, yield = 26%). 1H NMR (DMSOꢀd6/D2O = 9/1,
300 MHz): δ 2.73ꢀ2.78 (m, 2H, H2ꢀC(17')), 3.37 (t = 6.3 Hz, 4H, H2ꢀC(11'), H2ꢀC(13')), 3.54 (t = 6.3 Hz, 4H,
H2ꢀC(10'), H2ꢀC(14')), 3.59 (s, 3H H3ꢀC(19')), 3.87 (s, 2H, H2ꢀC(8')), 7.24 (d, J = 8.1 Hz, 1H, HꢀC(2')), 7.63
(d, J = 8.1 Hz, 1H, HꢀC(1')), 7.76 (s, 1H, HꢀC(5')); 13C NMR (DMSOꢀd6/D2O = 9/1, 75 MHz): δ 32.1, 37.1,
42.5, 52.3, 54.2, 55.9, 123.9, 134.3, 134.5, 137.3, 150.3, 174.7; HRMSꢀES (m/z) [M+H]+ calcd for
C15H24BCl2N2O4S: 409.0921, found: 409.0925, chlorine isotopic peak [M+H+2]+ found 411.0915.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(R)ꢀ2ꢀAminoꢀ3ꢀ(1ꢀ(2ꢀ(bis(2ꢀchloroethyl)amino)ꢀ5ꢀboronobenzyl)ꢀ1Hꢀindolꢀ3ꢀyl)propanoic acid (11).
Step 1. NaH (75 mg, 60% in mineral oil) was added to a solution of 27 (560 mg, 1.56 mmol) in anhydrous
o
DMF (10 mL) at 0 C. After the mixture was stirred for 30 min at the same temperature, 24 (532 mg, 1.56
mmol) in DMF (2 mL) was added slowly. The reaction was allowed to stir at room temperature overnight.
The mixture was extracted with DCM and washed with brine. The organic phase was collected and
concentrated to get the crude product which was further purified by column chromatography (Hexane/Ethyl
Acetate
=
5:1),
yielding
(R)ꢀtertꢀButyl
3ꢀ(1ꢀ(2ꢀ(bis(2ꢀchloroethyl)amino)ꢀ5ꢀbromobenzyl)ꢀ1Hꢀindolꢀ3ꢀyl)ꢀ2ꢀ((tertꢀbutoxycarbonyl)amino)propanoat
1
e (28) as a white solid (700 mg, 65%). mp 69ꢀ71
℃
; H NMR (DMSOꢀd6, 300 MHz): δ 1.33 (s, 9H, Boc
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 35
t
(27')), 1.34 (s, 9H, Bu (28')), 2.98ꢀ3.16 (m, 2H, H2ꢀC(25')), 3.44 (t, J = 6.6 Hz, 4H, H2ꢀC(10'), H2ꢀC(11')),
3.67 (t, J = 6.6 Hz, 4H, H2ꢀC(9'), H2ꢀC(12')), 4.14ꢀ4.17 (m, 1H, HꢀC(26')), 5.51 (s, 2H, H2ꢀC(14')), 6.54 (s,
1H, HꢀC(20')), 7.03ꢀ7.13 (m, 3H, HꢀC(23') , HꢀC(1') , HꢀC(2')), 7.21 (s, 1H, HꢀC(5')), 7.28 (d, J = 7.8 Hz,
1H, HꢀC(22')), 7.35 (d, J = 8.4 Hz, 1H, HꢀC(21')), 7.42 (dd, J = 2.4 Hz, J = 8.7 Hz, 1H, HꢀC(24')), 7.57ꢀ7.60
(m, 1H, HꢀNBoc); 13C NMR (CDCl3, 75 MHz): δ 28.0, 28.4, 41.5, 45.5, 54.8, 56.5, 79.5, 81.8, 109.4, 110.4,
119.4, 119.6, 122.1, 126.0, 126.5, 128.7, 131.2, 131.6, 136.6, 138.2, 145.4, 171.2; HRMSꢀES (m/z) [M+H]+
calcd for C31H40N3O4Cl2: 668.1652, found: 668.1616, chlorine isotopic peak [M+H+2]+ found 670.1605.
Step 2. A mixture of 28 (500 mg, 0.75 mmol), bis(pinacolato)diboron (380 g, 1.5 mmol), KOAc (150 mg,
1.5 mmol), and PdCl2(dppf) (55 mg, 0.075 mmol) in dioxane (25 ml) was flushed with argon for 10 minutes
and heated to reflux for 24 h under argon. After cooling to room temperature, the mixture was extracted with
ethyl acetate and washed with brine. The organic layers was collected, dried over Na2SO4 and evaporated
under vacuum, the residue was purified by column chromatography (Hexane/Ethyl Acetate = 5:1) to afford
(R)ꢀtertꢀButyl
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3ꢀ(1ꢀ(2ꢀ(bis(2ꢀchloroethyl)amino)ꢀ5ꢀ(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)benzyl)ꢀ1Hꢀindolꢀ3ꢀyl)ꢀ2ꢀ
1
((tertꢀbutoxycarbonyl)amino)propanoate (8) as a white solid (350 mg, 66%). mp 69ꢀ71
℃
; H NMR
(DMSOꢀd6, 300 MHz): δ 1.19 (s, 12H, H3ꢀC(33'), H3ꢀC(34'), H3ꢀC(35') , H3ꢀC(36')), 1.31 (s, 9H, Boc (29')),
1.34 (s, 9H, tBu (27')), 2.97ꢀ3.15 (m, 2H, H2ꢀC(25')), 3.44 (t, J = 11.7 Hz, 4H, H2ꢀC(10'), H2ꢀC(11')), 3.62 (t,
J = 11.7 Hz, 4H, H2ꢀC(9'), H2ꢀC(12')), 4.06ꢀ4.11 (m, 1H, HꢀC(26')), 5.46 (s, 2H, H2ꢀC(14')), 6.98 (d, J = 7.8
Hz, 1H, HꢀC(2')), 7.02ꢀ7.13 (m, 4H, HꢀC(22') , HꢀC(23') , HꢀC(25') , HꢀC(5')), 7.30 (d, J = 7.8 Hz, 1H,
13
HꢀC(21')), 7.38 (d, J = 7.8 Hz, 1H, HꢀC(24')), 7.55ꢀ7.58 (m, 2H, HꢀNBoc, HꢀC(1')); C NMR (DMSOꢀd6,
75 MHz): δ 25.0, 28.6, 28.0, 40.5, 45.3, 55.3, 55.8, 78.6, 80.7, 84.0, 110.4, 110.6, 119.4, 119.2, 121.8, 123.8,
128.0, 128.2, 134.3, 134.5, 135.0, 136.9, 150.5, 155.8, 171.8; HRMSꢀES (m/z) [M+H]+ calcd for
C37H52N3O6BCl2: 716.3405, found: 716.3390, chlorine isotopic peak [M+H+2]+ found 718.3359.
Step 3. 200 mg of compound 8 was dissolved in 4 mL of DCM, 0.6 mL of trifluoroacetic acid was added via
ACS Paragon Plus Environment

Page 27 of 35
Journal of Medicinal Chemistry
a syringe within 10 min. The mixture was allowed to stir overnight and evaporated under vacuum. The
residue was dissolved in a solution of DCM (20 mL) and H2O (20 mL) and neutralized with K2CO3 (1 M) to
pH 6.2. The mixture was extracted with DCM and washed with brine. The organic phase was collected and
concentrated. The crude products were purified by column chromatography (Hexane/Ethyl Acetate = 1:1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
then DCM/MeOH = 20:1) to afford 11 as a white foam (67 mg, 50%). H NMR (DMSOꢀd6/D2O = 9/1, 300
MHz): δ 3.05 (dd, J = 9.0 Hz, J = 15.0 Hz, 1H, HꢀC(25')), 3.05 (dd, J = 5.0 Hz, J = 15.0 Hz, 1H, HꢀC(25')),
3.43 (t = 6.5 Hz, 4H, H2ꢀC(10'), H2ꢀC(11')), 3.54ꢀ3.57 (m, 1H, HꢀC(26')), 3.64 (t = 6.5 Hz, 4H, H2ꢀC(9'),
H2ꢀC(12')), 5.42 (d, J = 16.5 Hz, 1H, HꢀC(14')), 5.48 (d, J = 16.5 Hz, 1H, HꢀC(14')), 6.98ꢀ7.04 (m, 2H,
HꢀC(22') , HꢀC(23')), 7.09 (s, 1H, HꢀC(20')), 7.18 (d, J = 7.5 Hz, 1H, HꢀC(2')), 7.26 (s, 1H, HꢀC(5')), 7.29 (d,
J = 7.5 Hz, 1H, HꢀC(1')), 7.34ꢀ7.66 (m, 2H, HꢀC(21') , HꢀC(24')); 13C NMR (DMSOꢀd6/D2O = 9/1, 75 MHz):
δ 25.9, 41.4, 44.5, 53.8, 54.2, 88.8, 107.8, 108.9, 117.9, 120.4, 121.9, 126.9, 127.6, 128.1, 132.4, 132.8,
133.2, 135.5, 147.8, 169.8. HRMSꢀES (m/z) [M+H]+ calcd for C22H27BCl2N3O4: 478.1466, found: 478.1465,
chlorine isotopic peak [M+H+2]+ found 478.1488.
Detection of DNA crossꢀlinking. ICL formation and crossꢀlinking yields were analyzed via denaturing
polyacrylamide gel electrophoresis (PAGE) with phosphorimager analysis. The DNAꢀDNA crossꢀlinking
abilities of these compounds were investigated by reacting with a 32Pꢀlabelled 49 mer oligonucleotide
(Figure 1 and Figure S2) then subjected to 20% denaturing PAGE analysis. The 32Pꢀlabelled oligonucleotide
(1.0 ꢁM) was annealed with 1.5 equiv of the complementary strand by heating to 65 oC for 3 min in a buffer
of 10 mM potassium phosphate (pH 7) and 100 mM NaCl, followed by slowꢀcooling to room temperature
overnight. The 32Pꢀlabeled duplex DNA (2 ꢁL, 1.0 ꢁM) was mixed with 1.0 M NaCl (2 ꢁL), 100 mM
potassium phosphate (2 ꢁL, pH 8), 10 ꢁM to 50 mM H2O2 (2 ꢁL) and 10 mM compound (resulted in a
concentration of 1 mM) and appropriate amount of autoclaved distilled water was added to give a final
volume of 20 ꢁL. The reaction mixture was incubated at room temperature for 16 h and then quenched by an
equal volume of 90% formamide loading buffer. Finally it was subjected to 20% denaturing polyacrylamide
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 35
gel analysis.
1
2
3
4
5
6
7
8
9
MDAꢀMBꢀ468 cell line. The human tumor cell line MDAꢀMBꢀ468 was cultured in the media (media
component: 500 mL of Lꢀ15 Leibovitz media (HyClone no: SH30525.01), 50 mL of Fetal bovine serum
(MIDSCI no: S01520HI), 5 mL of NEAA (HyClone no: SH30238.01), 5 mL of Penicillin (HyClone no:
SV30010). Cells are inoculated into 384ꢀwell microtiter plates in 20 ꢁL at plating densities ranging from
5,000 to 10,000 cells/well. After cell inoculation, the microtiter plates were incubated at 37 ° C and 100%
relative humidity for 2ꢀ3 h prior to the addition of reagents. Reagents are solubilized in dimethyl sulfoxide at
20 mM concentration and serially diluted ten times each time 50% decrease in DMSO in a 384ꢀwell plate.
Then 200 nL of the serially diluted compounds were added to the cell plate (1:100 dilution) by using
Freedom EVOware. The plates were incubated for an additional 48 h at 37 °C and 100% relative humidity.
Then, 20 ꢁL of celltiterꢀGlo Luminescent solutions were added to the cell plates. The plates were further
incubated at room temperature for 10 min before the luminescent was measured with Infinite M1000.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NCI in vitro antitumor screen (IVCLSP). The IVCLSP of the NCI Developmental Therapeutics Program
screens up to 3,000 compounds annually for potential anticancer activity. This program uses 60 human
tumor cell lines representing leukemia, melanoma, and cancers of the lung, colon, brain, ovary, breast,
prostate, and kidney. In a twoꢀstage screening process, compounds are tested for growthꢀinhibitory activity
in all 60 cell lines at a single dose of 10 µmol/L, and compounds that achieve a threshold activity are
retested in all 60 cell lines in a fiveꢀdose screen. The growthꢀinhibitory activity of a tested compound is
expressed as log (GI50, TGI, or LC50), where GI50 is the concentration required to inhibit tumor cell
growth by 50%, TGI is the concentration causing total growth inhibition, and LC50 is the lethal
concentration at which 50% of cells are killed. For each compound tested, 60 activity values (one for each
cell line) make up the activity pattern, or fingerprint, of the compound.
CLL cells and normal lymphocytes. Leukemic lymphocytes were isolated from fresh peripheral blood
sample obtained from patients with CLL. Separate laboratory protocols were used to obtain blood samples
ACS Paragon Plus Environment

Page 29 of 35
Journal of Medicinal Chemistry
from patients with CLL and healthy donors. All individuals signed written informed consent forms in
accordance with the Declaration of Helsinki, and the laboratory protocols approved by the institutional
review board at the University of Texas MD Anderson Cancer Center. (A) Isolation of CLL and normal
lymphocytes: Whole blood was collected in heparinized tubes and diluted 1:3 with cold PBS (0.135 M NaCl,
2.7 mM KCl, 1.5 mM KH2PO4, 8 mM Na2HPO4 [pH 7.4]) and layered onto FicollꢀHypaque (specificgravity,
1.086; Life Technologies, Grand Island, NY). The blood was then centrifuged at 433 g for 20 minutes, and
mononuclear cells were removed from the interphase. Cells were washed twice with cold PBS and
resuspended in 10 mL RPMI 1640, supplemented with 10% autologous plasma. A Coulter channelyzer
(Coulter Electronics, Hialeah, FL) was used to determine cell number and the mean cell volume. The CLL
or normal lymphocytes were suspended in medium at a concentration of 1 × 107 cells/mL and fresh cells
were used for all experiments. (B) Measurement of apoptosis: Cell death is evaluated by flow cytometry
analysis with the use of annexin V–PI double staining. CLL or normal lymphocytes in suspension are
incubated with 10 µM of compounds and the cell death was measured by annexin V binding assay. Time
matched control samples with no compound are also maintained side by side. At the end of incubation time,
cells are washed with PBS and resuspended in 200 µL of 1× annexin binding buffer (BD Biosciences), at a
concentration of 1 × 106 cells/mL. Annexin V–FITC (5 ꢁL) is added, and the cells are incubated in the dark for
15 minutes at room temperature. A total of 10 ꢁL of PI (50 ꢁg/mL) is added to the labeled cells and analyzed
immediately with a FACSCALIBUR cytometer (Becton Dickinson). Data from at least 10000 events per
sample are recorded and processed using Cell Quest software (Becton Dickinson).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Solubility Assay. Calibration plots were prepared for each compound for the relationship between solute
concentration and absorbance using backgroundꢀsubtracted values: In a 384 UV plate (Greiner no. 781801),
compounds were serially diluted in quadruplicate starting from a 10 mM compound stock solution in DMSO
(final concentrations: 0 ꢁM to 300 ꢁM). Buffer (90 mM ethanolamine, 90 mM KH2PO4, 90 mM potassium
acetate, and 30mM KCl (pH 7.4)) containing 20% acetonitrile was used as the dilution solution. The plate
ACS Paragon Plus Environment