Synthesis of an adenine N-3 substituted CBI adduct by alkylation of adenine with a 1-iodomethylindoline seco-CBI precursor

Article abstract of DOI:10.1016/j.tet.2018.09.057

The synthesis of a 1,2,9,9a-tetrahydrocyclopropa[c]benz[e]indol-4-one (CBI)-adenine adduct via regioselective N-3 alkylation of adenine with a 1-(iodomethyl)-2,3-dihydro-1H-benzo[e]indol-5-ol (I-seco-CBI)-containing precursor is described. Spectroscopic analyses of the unlabeled and adenine-C8 carbon-13 labeled adducts utilizing ROESY NMR techniques allowed structural assignment of the alkylation product as the adenine N-3 substituted regioisomer. A stable-isotopically labeled version of the N-3 adduct incorporating six carbon-13 labels was also prepared by this method for use as an LC-MS internal standard. The cyclized CBI-containing compound was also found to alkylate adenine at elevated temperatures to produce the N-3 adduct albeit at a significantly slower rate than that observed for the I-seco-CBI precursor. Adenine alkylation with an I-seco-CBI precursor offers scalable and facile access to a CBI-adenine adduct facilitating its use as an efficacy marker for the development of duocarmycin-based ADCs.

Full text of DOI:10.1016/j.tet.2018.09.057

Tetrahedron xxx (2018) 1e9
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Synthesis of an adenine N-3 substituted CBI adduct by alkylation of
adenine with a 1-iodomethylindoline seco-CBI precursor
,
Alban J. Allentoff a *, Michael A. Wallace a, Sarah C. Traeger b, Naiyu Zheng c, Qian Zhang d,
Samuel J. Bonacorsi Jr. a
a Discovery Chemistry Platforms-Radiochemistry, Bristol-Myers Squibb, Route 206 and Province Line Road, Princeton, NJ 08543, USA
b Department of Discovery Analytical Sciences, Bristol-Myers Squibb, Route 206 and Province Line Road, Princeton, NJ 08543, USA
c Department of Bioanalytical Sciences, Bristol-Myers Squibb, Route 206 and Province Line Road, Princeton, NJ 08543, USA
d Discovery Chemistry Oncology, Bristol-Myers Squibb, 700 Bay Road, Redwood City, CA 94063, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of a 1,2,9,9a-tetrahydrocyclopropa[c]benz[e]indol-4-one (CBI)-adenine adduct via regio-
selective N-3 alkylation of adenine with a 1-(iodomethyl)-2,3-dihydro-1H-benzo[e]indol-5-ol (I-seco-
CBI)-containing precursor is described. Spectroscopic analyses of the unlabeled and adenine-C8 carbon-
13 labeled adducts utilizing ROESY NMR techniques allowed structural assignment of the alkylation
product as the adenine N-3 substituted regioisomer. A stable-isotopically labeled version of the N-3
adduct incorporating six carbon-13 labels was also prepared by this method for use as an LC-MS internal
standard. The cyclized CBI-containing compound was also found to alkylate adenine at elevated tem-
peratures to produce the N-3 adduct albeit at a significantly slower rate than that observed for the I-seco-
CBI precursor. Adenine alkylation with an I-seco-CBI precursor offers scalable and facile access to a CBI-
adenine adduct facilitating its use as an efficacy marker for the development of duocarmycin-based
ADCs.
Received 5 July 2018
Received in revised form
27 September 2018
Accepted 30 September 2018
Available online xxx
Keywords:
Duocarmycin
Adenine alkylation
CBI
Antibody-drug conjugate
Isotopic labeling
© 2018 Elsevier Ltd. All rights reserved.
1. Introduction
[23]. Subsequent degradation of modified DNA affords adducts
which confirm that the adenine base on DNA is modified selectively
Duocarmycins and related analogs are well known highly potent
antitumor agents [1e3]. The known myelotoxicity and hepatotox-
icity of these drug molecules significantly narrows the therapeutic
window for their use as chemotherapy drugs [4,5]. In an attempt to
overcome these therapeutic limitations, duocarmycin-related
compounds have recently served as payloads in antibody-drug
conjugates (ADCs) utilizing targeted delivery to enhance their tu-
mor killing ability while avoiding the undesired toxicity associated
with these compounds [6e13].
The mechanism of cytotoxicity of the duocarmycins and related
analogs has been extensively investigated through the work of
Boger [14e18] and others [19e22]. This putative mechanism in-
volves initial minor groove binding of duocarmycins to adenine-
rich DNA. Molecular shaping enhances the electrophilicity of the
cyclopropyl moiety allowing regioselective alkylation of the N-3 of
adenine on the DNA resulting in tumor cell death through apoptosis
at the N-3 position [15,24].
2. Results and discussion
2.1. Synthesis
As part of bioanalytical assays to support development of ADCs
possessing a 1-(chloromethyl)-2,3-dihydro-1H- benzo[e]indol-5-ol
(seco-CBI)-containing payload component, measuring levels of N-3
adenine adduct 1 (Fig. 1) could serve as a valuable tool to evaluate
drug efficacy in preclinical in vivo models, and as a potential clinical
biomarker [25]. However, execution of this assay required sub-
stantial amounts of 1 for use as a reference standard, as well as an
isotopically labeled analog of 1 for use as an LC-MS internal stan-
dard. Previously reported preparations of 1,2,9,9a-tetrahy-
drocyclopropa[c]benz[e]indol-4-one (CBI)-adenine N-3 adducts
have required lengthy incubations of CBI derivatives with adenine-
rich calf thymus DNA, followed by depurination, extraction and
preparative chromatography to isolate the desired adduct from the
* Corresponding author.
E-mail address: alban.allentoff@bms.com (A.J. Allentoff).
https://doi.org/10.1016/j.tet.2018.09.057
0040-4020/© 2018 Elsevier Ltd. All rights reserved.
Please cite this article in press as: A.J. Allentoff, et al., Synthesis of an adenine N-3 substituted CBI adduct by alkylation of adenine with a 1-
iodomethylindoline seco-CBI precursor, Tetrahedron (2018), https://doi.org/10.1016/j.tet.2018.09.057

2
A.J. Allentoff et al. / Tetrahedron xxx (2018) 1e9
(Scheme 1) using Finkelstein reaction conditions in butanone at
78 ꢀC. Iodide 6 was found to have sufficient stability from cycliza-
tion to the CBI system allowing workup and characterization by LC-
MS and NMR analyses. Difficulties in efficiently isolating pure 6
coupled with the low reactivity of chloride 5 prompted us to utilize
the crude iodide (containing approximately 10% of chloride 5)
directly in subsequent reactions. Treatment of iodide 6 with five
equivalents of adenine in DMA at 80 ꢀC for 18 h resulted in com-
plete consumption of the starting material. LC-MS analyses of the
resulting product mixture showed approximately 40e50% of
adenine-containing adducts along with 20e30% of the cyclized CBI
system 7 and 5e10% of unreacted 5 which had been carried through
from the previous step. Further HPLC and LC-MS evaluations of the
adenine-containing components established the presence of
approximately 70% of one major adenine isomer in the mixture
with the observed molecular ion of m/z 710 in the mass spectrum
which was later identified as the adenine N-3 adduct. The observed
regioselectivity towards this isomer was consistent with previously
observed N3-alkylations of adenine by alkyl halides [31]. Three
other adenine adducts with the same molecular ion of m/z 710 were
also observed as minor components including one adduct (pre-
sumably the N-9 adenine substituted adduct [31]) comprising
approximately 14% of the adduct mixture. Isolation of the major
adenine-adduct by semi-preparative HPLC followed by acidic
removal of the N-Boc protecting group afforded the pure adduct 1b
in 24% isolated yield in two steps from chloride 5. This isolated
compound was found to co-elute on HPLC with a reference sample
of the R-enantiomer of adduct 1b derived from incubations of the R,
S-enantiomer of 7 with calf thymus DNA using literature proced-
ures [15].
Fig. 1. N-3 Adduct 1 with numbered adenine positions.
degradation mixtures [15,24]. This approach while well refined, is
not amenable to scale-up, or to modification for the preparation of
isotopically labeled analogs. No previous synthetic approach to
these adducts using direct alkylation of adenine with a CBI or seco-
CBI related precursor has been reported. We now describe the
synthesis of the CBI-adenine adduct 1 and its isotopically labeled
analog via N-3 alkylation of adenine with an iodo-substituted seco-
CBI precursor.
The S-enantiomer of 2 (Fig. 2) was recently developed as the
seco-CBI-containing payload component of a CD70 targeted ADC
[26,27] and was previously shown to have enhanced cytotoxicity in
comparison to naturally occurring duocarmycins [28]. Compound 2
is comprised of a seco-CBI alkylating subunit linked to a DNA
binding minor-groove component.
To investigate conditions allowing reaction of the alkylating
subunit in 2 with adenine in solution to form the desired adduct 1,
we prepared the racemic seco-CBI precursor
5 (Scheme 1).
Coupling of the known racemic chloride 3 [22] with N-Boc-pro-
tected 4-aminobenzoic acid using HATU activation readily afforded
the known seco-CBI precursor 5 [28] in high yield. The racemic
material was used due to the ready availability of racemic starting
material 3 and because all planned bioanalytical assays measuring
adduct levels utilized non-chiral LC-MS assays.
While CBI-containing compounds rapidly and regioselectively
react with the N-3 position of adenine in DNA [15], CBI and closely
related ring systems undergo slow solvolysis [29] and are unreac-
tive to adenine in solution [30]. Therefore, we investigated alky-
lating adenine regioselectively with a potentially more reactive
seco-CBI precursor using halogen substitution chemistry. Adenine
can behave as an ambident nucleophile and regioselective N-3
alkylation by alkyl halides has been shown by Leonard to require
elevated temperatures and polar solvents such as DMF or DMA
favoring ionic character [31]. These polar, hydrophilic solvents are
also required for appropriate solubility of adenine in reaction
mixtures.
2.2. Structural identification
Spectroscopic analyses utilizing 2D NMR techniques provided
further confirmation of the structural assignment of 1b as the
adenine N-3-substituted isomer. Boger has described 2D 1H-1H
NOESY NMR experiments in which diagnostic NOEs between both
the H-13 protons of the side-chain and the H-2 proton of the
adenine were observed exclusively in adenine N-3 substituted ad-
ducts (Fig. 3) [15]. In our hands, strong NOE crosspeaks were
observed in the 2D 1H-1H ROESY between the upfield adenine
proton H-2 at 7.68 ppm, and the CH2-13 protons (5.05 and
4.75 ppm) confirming the N3-substituted adenine structure 1b.
While the assignment of the upfield singlet (relative to the other
adenine proton at 8.11 ppm) to the adenine C2-H position was
consistent with spectra from related adenine adducts [15], the key
chemical shift differed from the literature values, most likely due to
pH differences in the sample and the effects of modified sub-
stituents. In order to unambiguously confirm the adenine proton
assignment and the isomeric structure, we synthesized the carbon-
13 labeled analog 1d (Scheme 2). This was accomplished via a re-
action of iodide 6 with adenine labeled with carbon-13 at the C8
position followed by acidic deprotection of the resulting adenine N-
Boc protected adduct 1c. It was expected that the large coupling
constant between the carbon-13 labeled adenine C8 and the H-8
proton would readily differentiate this proton from the H-2 proton
in the spectra. As displayed in Fig. 3 the ROESY spectrum of 1d
shows the absorbance associated with adenine H-8 as the expected
doublet possessing a large 13C-H coupling (8.3, 7.90 ppm, d,
J ¼ 220.8 Hz). The other proton associated with adenine (H-2) re-
mains as a singlet at 7.68 ppm. Furthermore, the downfield
component of the doublet assigned to H-8 at 8.32 ppm does not
overlap with any other absorbances in the spectrum, and clearly
shows no observable NOE correlations with other protons in the
spectrum. This lack of NOE is consistent with the distal positioning
Treatment of chloride 5 with excess adenine in DMA at 80 ꢀC
gave only trace amounts of adenine-containing adducts (as iden-
tified by LC-MS) even after extended heating (72 h). To enhance
reactivity, chloride 5 was converted to the corresponding iodide 6
Fig. 2. Seco-CBI-containing payload 2.
Please cite this article in press as: A.J. Allentoff, et al., Synthesis of an adenine N-3 substituted CBI adduct by alkylation of adenine with a 1-
iodomethylindoline seco-CBI precursor, Tetrahedron (2018), https://doi.org/10.1016/j.tet.2018.09.057

A.J. Allentoff et al. / Tetrahedron xxx (2018) 1e9
3
Scheme 1. Synthesis of iodide 6 and its reactions with adenine.
of H-8 to the H-13 methylene protons and confirms the N3-adenine
adduct structure assigned as 1d.
associated with the adenine alkylation conditions of iodide 6 may
enable 7 to react with adenine. In order to investigate this possi-
bility, we prepared the known CBI-containing compound 7 [32]
(Scheme 4) via treatment of chloride 5 with DBU under established
conditions [15]. Treatment of 7 with excess adenine at 82 ꢀC
showed slow formation of the N-3 adduct 1a with side-products.
After 72 h reaction time, HPLC-UV and LC-MS analyses showed a
product mixture consisting of 59% adenine-containing adducts as
an approximate 4:1 mixture of N-3 substituted adduct and a single
additional adduct (presumably the adenine N-9 substituted). Also
observed as 26% of the mixture was the hydrolyzed product 8 from
adventitious water in the reaction, and 14% of unreacted starting
material 7. Unlike the profile from adenine alkylation with iodide 6
in which four adenine-related isomers were observed, only two
adenine-related adducts were seen in the mixture. This difference
in isomeric mixture suggests sluggish reactivity of 7 with weakly
basic adenine compared to the seco-CBI iodide 6. While this result
demonstrates that at appropriately elevated temperatures, free CBI
systems can alkylate adenine in solution, the observed differences
in alkylation rates (18 h versus 72 h) and product profiles, provide
evidence that the primary pathway of adenine alkylation by iodide
6 does not include initial cyclization of 6 to the CBI 7.
2.3. Synthesis of stable-isotopically labeled 1 for use as an LC-MS
internal standard
With confirmation that the isolated product was in fact the
desired N-3 adenine adduct, we applied the identical sequence to
the synthesis of the C6 carbon-13 labeled analog 1f (Scheme 3) for
use as an LC-MS internal standard in bioanalytical assays. This
approach allowed facile isotope labeling in the benzoic ring system
of 1. Commercially available N-Boc-p-amino benzoic acid 4b con-
taining six carbon-13 isotopic ring labels was coupled to indole
amine 3 to give 5b in 48% yield. Following conversion to the iodide
(6b) and reaction with adenine, 5b was converted to the stable-
labeled adenine adduct 1e in 28% yield from chloride 5b. The N-
Boc group was then removed under standard conditions to afford
1f. With possession of six isotopic mass units, adduct 1f was well
suited for use as an LC-MS internal standard for adduct bio-
analytical assays with minimal interference or cross signal contri-
bution with unlabeled analyte during MS based quantifications.
In an attempt to block undesired cyclization of iodide to the CBI
during adenine alkylation, we protected the hydroxyl group in 6
with several protecting strategies. Initial attempts protected the 4-
hydroxyl in 6 with bulky silyl protecting groups. However, silyl
ethers were found to be unstable to the alkylation conditions. We
then turned to methylpiperazine carbamate as a known protecting
group for seco-CBI systems [26,32]. Carbamate protected iodide 10
was readily prepared via acylation of 6 with methylpiperazine
acetyl chloride affording 10 in 40% yield after purification (Scheme
5). After heating iodide 10 with 5 equivalents of adenine at 80 ꢀC for
24 h, approximately 35% of the starting material remained and the
product mixture was comprised of approximately 14% adenine-
2.4. Further investigation of the adenine alkylation
During our investigation of adenine alkylation with iodide 6,
formation of the ring-closed CBI 7 was consistently observed as
20e30% of the product mixture. The possibility therefore exists that
the observed alkylation of adenine is proceeding via initial cycli-
zation of iodide 6 to the CBI 7 which could then alkylate adenine in
a similar manner to the mechanism seen in DNA alkylation. Free
CBIs are known as very weak electrophiles in solution [29] and
structurally similar cyclopropylpyrroloindoles have been reported
to have no reaction with excess adenine in DMF for extended
heating at 45 ꢀC [30]. However, the elevated temperature (80 ꢀC)
Please cite this article in press as: A.J. Allentoff, et al., Synthesis of an adenine N-3 substituted CBI adduct by alkylation of adenine with a 1-
iodomethylindoline seco-CBI precursor, Tetrahedron (2018), https://doi.org/10.1016/j.tet.2018.09.057

4
A.J. Allentoff et al. / Tetrahedron xxx (2018) 1e9
Fig. 3. ROESY NMR analysis of adenine adduct 1b and 13C labeled adduct 1d.
Scheme 2. Synthesis of adenine C8-carbon-13 labeled adduct 1.
adducts 11a by HPLC analysis. These adducts showed a similar ratio
of the four isomers as observed previously in reactions of iodide 6
with adenine. Also identified in the product mixture were sub-
stantial amounts of compound 11b (presumably resulting from
hydrolysis with adventitious water), and a non-adenine containing
product 12. Continued heating for a total of 48 h showed 10% of
iodide 10 still remaining in the product mixture while levels of
adenine-containing adducts (11a) rose only slightly. However, the
hydrolysis product 11b was no longer observed after 48 h and 12
had become the major product comprising approximately 40% of
the mixture. Semi-preparative HPLC afforded pure product 12 in
30% yield which was structurally identified by MS and NMR ana-
lyses as 12: resulting from elimination and internalization of the
resulting double bond. Surprisingly, the carbamide modification
appeared to diminish observed iodide substitution with adenine,
and allowed side-reactions such as iodide substitution with water
to become substantial reaction pathways. The observed initial
accumulation of the hydrolysis product 11b followed by its disap-
pearance and corresponding increase of 12 suggests that it is in fact
the hydroxyl compound 11b (not the iodide 10) which primarily
undergoes elimination and double bond migration to produce 12.
While the scope of this study focused on reaction conditions pre-
cedented to favor N-3 adenine alkylation [31], future work could
explore alternative conditions using base or other additives to
enhance alkylation at lower temperatures and suppress formation
of the cyclized CBI side product.
Please cite this article in press as: A.J. Allentoff, et al., Synthesis of an adenine N-3 substituted CBI adduct by alkylation of adenine with a 1-
iodomethylindoline seco-CBI precursor, Tetrahedron (2018), https://doi.org/10.1016/j.tet.2018.09.057

A.J. Allentoff et al. / Tetrahedron xxx (2018) 1e9
5
Scheme 3. Synthesis of the 13C6 labeled internal standard 1f.
Scheme 4. Preparation of CBI-containing 7 and its reaction with adenine at elevated temperatures.
Scheme 5. Preparation of carbamate 10 and its reaction with adenine.
3. Conclusion
internal standards for mass spectrometry based efficacy and
biomarker assays in the development of duocarmycins and related
ADCs.
A procedurally facile method for synthesizing CBI-adenine ad-
ducts via alkylation of adenine in solvent with iodo-substituted
seco-CBI precursors has been developed. These alkylation condi-
tions should be applicable to other CBI and related ADC payload
systems in which DNA alkylation is a therapeutic target. Further-
more, this method is applicable to the synthesis of stable-
isotopically labeled analogs of these adducts which serve as
4. Experimental
4.1. General information and methods
All unlabeled reagents were of ACS grade or better. Racemic (5-
Please cite this article in press as: A.J. Allentoff, et al., Synthesis of an adenine N-3 substituted CBI adduct by alkylation of adenine with a 1-
iodomethylindoline seco-CBI precursor, Tetrahedron (2018), https://doi.org/10.1016/j.tet.2018.09.057

6
A.J. Allentoff et al. / Tetrahedron xxx (2018) 1e9
amino-1H-indol-2-yl)(1-(chloromethyl)-5-hydroxy-1H-benzo[e]
indol-3(2H)-yl)methanone (3), TFA salt was provided by Bristol-
Myers Squibb Chemical Development. Adenine (8e13C) was pur-
chased from Cambridge Isotope laboratories Inc. (Tewksbury, MA).
Method 7: Column: Phenomenex Luna C18 (2), 21.2 ꢁ 250 mm,
5
m
m particle size; Mobile Phase A: aqueous 0.1% TFA; Mobile Phase
B: Acetonitrile; Gradient: 30e100 over 20 min, flow rate: 10 mL/
min; UV Detection:
¼ 254 nm.
l
[
13C6]4-((tert-Butoxycarbonyl)amino)benzoic acid (4b) was pur-
chased from IsoSciences Inc. (King of Prussia, PA). Samples of the N-
3 adenine adduct 1b standard prepared from incubations of the
(R,S)-enantiomer of 7 with calf thymus DNA [15] were provided by
Bristol-Myers Squibb Biologics Discovery California (Redwood City,
CA). Reactions were run under an inert atmosphere of nitrogen and
magnetically stirred at a constant rate. Semi-preparative HPLC
purifications were performed using a Varian Chromatography
system. Analytical HPLC analyses were performed on a Shimadzu
model SCL-10A system with an SPD-10AV UVevis detector. LC-MS
analyses were performed on an Agilent 1100 series HPLC and a
Finnigan LXQ ion trap mass spectrometer (Desolvation Gas: Nitro-
gen; Capillary Temp: 270 ꢀC; Positive Electrospray). High resolution
mass spectra (HRMS) were obtained using a Q Exactive™ HF Hybrid
Quadrupole-Orbitrap™ Mass Spectrometer (with electrospray
ionization (ESI) source) that was connected to a Shimadzu Nexera
4.2. tert-Butyl(4-((2-(1-(chloromethyl)-5-hydroxy-2,3-dihydro-
1H-benzo[e]indole-3-carbonyl)-1H-indol-5-yl)carbamoyl)phenyl)
carbamate (5)
To a solution of 4-((tert-butoxycarbonyl)amino)benzoic acid (4,
196 mg, 0.825 mmol) in DMF (5 mL) was added 2-(3H-[1,2,3]tri-
azolo[4,5-b]pyridin-3-yl)-1,1,3,3-tetramethylisouronium
hexa-
fluorophosphate(V) (HATU, 314 mg, 0.825 mmol) followed by N,N-
diisopropylethylamine (0.432 mL, 2.48 mmol). The resulting solu-
tion was stirred for 45 min at ambient temp. The solution was then
added to another flask containing a solution of (5-amino-1H-indol-
2-yl)
(1-(chloromethyl)-5-hydroxy-1H-benzo[e]indol-3(2H)-yl)
methanone [22], TFA salt (3, 500 mg, 0.990 mmol) dissolved in 5 mL
of DMF. The flask containing the solution of activated 4 was rinsed
with 200 mL of DMF which was added to the reaction. To the flask
UPLC system (Cortecs UPLC C18 column, 1.6
m
(2.1 ꢁ 50 mm)). Re-
was then added Hünig's base (0.451 mL, 2.6 mmol) and the
resulting dark, uniform solution was allowed to stir at ambient
temp. After 45 min, a sample was dissolved in acetonitrile and
analyzed by LC-MS (method 1). This analysis showed the formation
of a new product (Tr ¼ 7.0 min) with m/z ¼ 611 which is the ex-
pected ion for the product. To the solution was added 0.5 N HCl
(55 mL) causing a precipitate to form. After stirring for 1 h, the
precipitate was collected by vacuum filtration on a fritted funnel.
After air drying for 10 min, the funnel was dried under vacuum at
ambient temp for 12 h to afford crude 5 [22] as a greenish-grey
actions were monitored by HPLC and LC-MS with comparisons
made to authentic unlabeled materials when available. Melting
points were obtained using an EZ-Melt Automated melting point
apparatus (Stanford Research Systems).
NMR. Proton NMR spectra were recorded on a Bruker 300 MHz
Avance II NMR spectrometer with a 5 mm 3H/1H/13C QNP probe at
27 ꢀC. One and two-dimensional NMR spectra for structure eluci-
dation purposes were collected on either a Bruker 500 MHz Avance
III NMR Spectrometer with a 5 mm Bruker BBFOþ 500 probe at
27 ꢀC, a Bruker 500 MHz Avance III NMR Spectrometer with a 5 mm
Bruker BBO Prodigy 500 cyroprobe at 27 ꢀC, or a Bruker 600 MHz
NEO NMR Spectrometer with a 5 mm Bruker TCI cryo probe at
27 ꢀC.
powder: 1H NMR (DMSO‑d6):
d 11.69 (s, 1 H), 10.42 (s, 1 H), 10.02
(s, 1 H), 9.66 (br s, 1 H), 8.17 (br s, 1 H), 8.14 (d, 1 H, J ¼ 8.36 Hz), 7.99
(br s, 1 H), 7.95 (d, 2 H, J ¼ 8.8 Hz), 7.87 (d, 1 H, J ¼ 8.36 Hz), 7.61 (d,
2 H, J ¼ 8.8 Hz), 7.56 (dd,1 H, J ¼ 8.8,1.9 Hz), 7.53 (d,1 H, J ¼ 7.04 Hz),
7.47 (d, 1 H, J ¼ 8.8 Hz), 7.39 (br t, 2 H, J ¼ 7.04 Hz), 7.21 (s, 1H), 4.82
(br t, 1 H, J ¼ 11 Hz), 4.58 (br dd, 1 H, J ¼ 11, 1.8 Hz), 4.25 (m, 1 H),
4.04 (m, 1 H), 3.89 (dd, 1 H, J ¼ 11, 7.3 Hz), 1.51 (s, 9 H); mp
131e135 ꢀC (dec).
LCMS. LC-MS methods described below were used for in-
process and final product analyses.
Method 1: Column: Luna C18(2), 3.0 ꢁ 100 mm, 3
mm particle
size. Flow rate: 0.5 mL/min. Mobile phase A: water, 0.1% formic
acid; Mobile phase B: acetonitrile, 0.1% formic acid, linear gradient
of 20% B to 80% B over 13 min.
4.3. tert-Butyl(4-((2-(1-((6-amino-3H-purin-3-yl)methyl)-5-
hydroxy-2,3-dihydro-1H-benzo[e]indole-3-carbonyl)-1H-indol-5-
yl)carbamoyl)phenyl)carbamate (1a)
Method 2: Column: Luna C18(2), 3.0 ꢁ 100 mm, 3
mm particle
size. Flow rate: 0.5 mL/min. Mobile phase A: water, 0.1% formic
acid; Mobile phase B: acetonitrile, 0.1% formic acid, linear gradient
of 20% B to 95% B over 7 min.
To a solution of 5 (150 mg, 0.245 mmol) in 2-butanone (10 mL)
was added sodium iodide (184 mg, 1.23 mmol) and the resulting
mixture was stirred at 78 ꢀC under nitrogen and monitored by HPLC
(method 3, compound 6, Tr ¼ 15.1 min, chloride 5, Tr ¼ 14.8 min).
After heating for 72 h conversion of 5 to the corresponding iodide 6
was approximately 90% complete. After cooling, the mixture was
evaporated to dryness in vacuo, and the residue partitioned be-
tween water and ethyl acetate. The layers were separated, the
aqueous layer extracted twice with ethyl acetate, and the combined
organic layers were washed with brine and then dried over mag-
nesium sulfate to give 170 mg of crude 6: LC-MS analysis (method
2) of the solution (compound 6, Tr ¼ 5.6 min) showed m/z [MþH]þ
calcd for C34H31IN4O5 703.13, found 703.12. This material was
stored as a solution in EtOAc at 4 ꢀC and was evaporated to give a
solid residue as required for the next step.
HPLC. HPLC methods described below were used for in-process
and final product analyses.
Method 3: Column: Luna, 4.6 ꢁ 150 mm, 5
mm particle size.
Mobile phase A: water with 0.1% TFA. Mobile phase B: CH3CN.
Linear gradient of 30% B to 100% B over 20 min; Flow rate: 1 mL/
min, injection volume: 10
¼ 220 nm.
Semi-preparative HPLC. HPLC methods described below were
mL, diluent: CH3CN; UV Detection:
l
used for purifications.
Method 4: Column: Phenomenex Luna C18 (2), 21.2 ꢁ 250 mm,
5 mm particle size; Mobile Phase A: aqueous 0.1% TFA; Mobile Phase
B: Acetonitrile; Gradient: 55e85 over 30 min, flow rate: 10 mL/min;
UV Detection:
l
¼ 254 nm.
Method 5: Column: Phenomenex Luna C18 (2), 21.2 ꢁ 250 mm,
5
m
m particle size; Mobile Phase A: aqueous 0.1% TFA; Mobile Phase
A 12 mg portion of crude 6 was subjected to semi-preparative
HPLC (method 4) which afforded purified 6 (6 mg, 0.01 mmol)
used for characterization: HRMS [MþH]þ calcd for C34H31IN4O5
B: Acetonitrile; Gradient: 20e80 over 20 min, flow rate: 10 mL/min;
UV Detection:
l
¼ 254 nm.
Method 6: Column: Eclipse XDB-C8 column, 21.2 ꢁ 250 mm,
703.1412, found 703.1405; 1H NMR (500 MHz, DMSO‑d6)
d 11.67 (br
5
m
m particle size; Mobile Phase A: aqueous 0.1% TFA; Mobile Phase
d, J ¼ 1.3 Hz, 1H), 10.02 (s, 1H), 9.66 (s, 2H), 8.14 (d, J ¼ 1.3 Hz, 1H),
8.10 (d, J ¼ 8.5 Hz, 1H), 7.98 (br s, 1H), 7.92 (br d, J ¼ 8.8 Hz, 2H), 7.80
(d, J ¼ 8.2 Hz, 1H), 7.59 (br d, J ¼ 8.8 Hz, 2H), 7.55 (br dd, J ¼ 9.0,
B: Acetonitrile; Gradient: 20e100% over 30 min, flow rate: 10 mL/
min; UV Detection:
¼ 254 nm.
l
Please cite this article in press as: A.J. Allentoff, et al., Synthesis of an adenine N-3 substituted CBI adduct by alkylation of adenine with a 1-
iodomethylindoline seco-CBI precursor, Tetrahedron (2018), https://doi.org/10.1016/j.tet.2018.09.057

A.J. Allentoff et al. / Tetrahedron xxx (2018) 1e9
7
2.0 Hz, 1H), 7.50 (td, J ¼ 7.6, 1.1 Hz, 1H), 7.46 (d, J ¼ 8.8 Hz, 1H),
7.36e7.30 (m, 1H), 7.20 (d, J ¼ 1.6 Hz, 1H), 4.72e4.65 (m, 1H),
4.65e4.60 (m, 1H), 3.87e3.80 (m, 1H), 3.77 (dd, J ¼ 10.9, 3.9 Hz, 1H),
3.46e3.45 (m, 1H), 1.50 (m, 9H). 13C NMR (126 MHz, DMSO‑d6)
product Tr ¼ 12.2 min). After cooling, the resulting crude mixture
was directly purified by semi-preparative HPLC (method 6) which
afforded 16 mg of purified 1c (0.023 mmol, 25% in two steps from
chloride 5) as a light-yellow solid: LC-MS m/z [MþH]þ calcd for
C1338CH35N9O5: 711.28, found: 711.28.
d
164.7, 160.2, 153.4, 152.6, 142.4, 141.7, 133.2, 132.0, 131.4, 130.1,
128.4 (2C), 127.0, 126.9, 123.2, 123.0, 122.1, 119.5, 117.1 (s, 2C), 116.6,
113.1, 112.0, 105.6, 100.5, 79.5, 62.9, 54.6, 42.5, 28.1 (3C), 32/34
carbons identified. Purified iodide 6 was found to convert to CBI 7
and hydrolysis product 8 on extended exposure to the acidic HPLC
mobile phase.
The mixture was placed under nitrogen and heated at 75 ꢀC for
18 h with monitoring of the reaction progress by HPLC (method 3,
product Tr ¼ 12.2 min). After cooling, the resulting crude mixture
was directly purified by semi-preparative HPLC (method 6) and
afforded 16 mg (0.023 mmol, 25% in two steps from chloride 5) of
purified 1c as a light-yellow solid: LC-MS m/z [MþH]þ calcd for
C1338CH35N9O5: 711.28, found: 711.28.
To a solution of crude 6 (85 mg, 0.196 mmol) in N,N-dimethy-
lacetamide (6.76 mL) was added 9H-purin-6-amine (82 mg,
0.605 mmol) and the resulting partial slurry was stirred under ni-
trogen at 80 ꢀC for 42 h while monitored by HPLC (method 3)
showing a mixture of adenine-containing adducts Tr ¼ 11.1 min
(15%), 11.8 min (N-3 substituted 1a, 70%), 12.2 min (7%) and
12.4 min (8%) as confirmed by LC-MS analyses (method 2). Also
observed by HPLC analysis (method 3) was the cyclized CBI 7
(Tr ¼ 15.7 min, 20e30%), and unreacted chloride 5 (Tr ¼ 14.8 min,
5e10%) which was carried over from the crude iodide starting
material. The resulting dark-red solution was cooled and subjected
directly to semi-preparative HPLC (method 6) which afforded
20 mg (0.028 mmol, 24% in two steps from 5) of 1a as a light-yellow
powder: HRMS (ESI) [MþH]þ calcd for C39H35N9O5: 710.2834,
found: 710.2827.
4.6. 4-((2-(1-((6-amino-8-[13C]-3H-purin-3-yl)methyl)-5-hydroxy-
2,3-dihydro-1H-benzo[e]indole-3-carbonyl)-1H-indol-5-yl)
carbamoyl)benzenaminium TFA salt (1d)
To a partial slurry of 1c (16 mg, in 3 mL of methylene chloride
was added 750 mL of TFA. The resulting solution was stirred at
ambient temp and reaction progress was monitored by HPLC
(method 3, 1d Tr ¼ 8.3 min). After 1.5 h the mixture was concen-
trated and the resulting solid washed once with methylene chlo-
ride. The solid was then dissolve into
a solution of 1:1
acetonitrile:0.1%TFA (2 mL) and the solution was subjected to semi-
preparative HPLC (method 6) affording 9.1 mg (0.012 mmol, 53%) of
pure 1d as a light-yellow solid: 1H NMR (500 MHz, Methanol-d4)
4.4. 4-Amino-N-(2-(1-((6-amino-3H-purin-3-yl)methyl)-5-
hydroxy-2,3-dihydro-1H-benzo[e]indole-3-carbonyl)-1H-indol-5-
yl)benzamide TFA salt (1b)
d
8.25 (d, J ¼ 8.5 Hz, 1H), 8.01 (d, J ¼ 1.3 Hz, 1H), 8.12 (d, J ¼ 207.8 Hz,
1H, H-13C), 7.88 (d, J ¼ 7.6 Hz, 1H), 7.79 (d, J ¼ 8.8 Hz, 2H), 7.68 (s,
1H), 7.59e7.54 (m, 1H), 7.59e7.54 (m, 1H), 7.51e7.48 (m, 1H),
7.48e7.45 (m, 1H), 7.45e7.43 (m, 1H), 7.43e7.39 (m, 1H), 6.97 (s,
1H), 6.80e6.76 (m, 2H), 5.03 (dd, J ¼ 13.9, 3.8 Hz, 1H), 4.91 (br d,
J ¼ 11.0 Hz, 1H), 4.72e4.60 (m, 2H), 4.49e4.39 (m, 1H) (20/27 pro-
Adduct 1a (20.0 mg, 0.028 mmol) was partially dissolved into
3 mL of methylene chloride. To the mixture was added 750 mL of
TFA and the resulting slurry was stirred at ambient temp for 1.5 h.
After the mixture was concentrated to dryness with a stream of
nitrogen, the residue was washed twice with methylene chloride.
The resulting residue was dissolved in acetonitrile (1 mL) and to
this solution was added 0.1% TFA in water (1 mL). This solution was
then subjected to semi-preparative HPLC (method 6) to afford
11.7 mg (0.019 mmol, 67% yield) of 1b as a yellow solid: HRMS m/z
[MþH]þ calcd C34H27N9O3 610.2310, found 610.2301; 1H NMR
tons observed); 13C NMR (126 MHz, methanol-d4)
d 169.4, 162.2,
156.5, 155. 1 (m), 153.4, 150.3, 149.0, 146.7, 146.4, 146.2 (13C label),
135.6, 133.1, 132.2, 131.8, 130.5 (2C), 129.1, 129.0, 125.0, 124.9, 124.5,
123.9, 123.6, 122.2, 116.1, 116.0, 115.1 (2C), 113.2, 107.7, 101.4, 57.0,
13
53.4, 41.2; HRMS (ESI) m/z [MþH]þ calcd for
C
CH35N9O5:
38
611.2344, found: 611.2342.
4.7. tert-Butyl(4-((2-(1-(chloromethyl)-5-hydroxy-2,3-dihydro-
1H-benzo[e]indole-3-carbonyl)-1H-indol-5-yl)carbamoyl)[ phenyl-
1,2,3,4,5,6e13C6)carbamate (5b)
(500 MHz, methanol-d4)
d
8.24 (d, J ¼ 8.4 Hz, 1H), 8.09 (s, 1H), 8.01
(s, 1H), 7.87 (d, J ¼ 8.4 Hz, 1H), 7.81 (d, J ¼ 8.5 Hz, 2H), 7.68 (s, 1H),
7.61 (d, J ¼ 8.7 Hz, 1H), 7.56 (t, J ¼ 7.6 Hz, 2H), 7.47 (d, J ¼ 11.7 Hz,
4H), 7.43e7.39 (m, 2H), 6.96 (br. s., 1H), 6.82 (d, J ¼ 8.5 Hz, 2H), 5.03
(dd, J ¼ 14.0, 4.0 Hz, 1H), 4.90 (d, J ¼ 11.1 Hz, 1H), 4.71e4.65 (m, 1H),
4.65e4.59 (m,1H), 4.44 (br s,1H); This spectrum was comparable to
that observed from an authentic standard of the (R)-enantiomer of
1b obtained from incubation of the (R,S)-enantiomer of 7 with calf
To a solution of [13C6]4-((tert-butoxycarbonyl)amino)benzoic
acid (4b, 145 mg, 0.611 mmol) in DMF (5 mL) was added 2-(3H-
[1,2,3]triazolo[4,5-b]pyridin-3-yl)-1,1,3,3-tetramethylisouronium
hexafluorophosphate(V) (HATU, 232 mg, 0.611 mmol) followed by
Hünig's Base (0.291 mL, 1.666 mmol). The resulting solution was
stirred for 1 h at ambient temp with monitoring by LC-MS (method
1, Tr ¼ 5.5, m/z ¼ 362 for the 13C6 labeled activated ester). The so-
lution was then added to another flask containing (5-amino-1H-
indol-2-yl)(1-(chloromethyl)-5-hydroxy-1H-benzo[e]indol-3(2H)-
yl)methanone [22], TFA salt (3, 281 mg, 0.555 mmol). The flask was
thymus DNA.15 13C NMR (126 MHz, methanol-d4)
d 167.8, 160.9,
155.1, 154.9, 152.0, 151.4, 149.7, 143.9, 142.0, 134.1, 131.5, 130.7, 130.4,
128.9 (s, 2C), 127.5, 127.3, 123.3, 123.2, 122.9, 122.3, 121.9, 120.4,
119.6, 115.1, 114.4, 113.4 (s, 2C), 111.6, 106.0, 99.9, 55.5, 52.3, 39.0.
Compound 1b was found to co-elute on HPLC (method 5,
Tr ¼ 12.7 min) and had identical LC-MS properties (method 1) with
an authentic standard of the (R)-enantiomer of 1b.
rinsed with 200 mL of DMF which was added to the reaction. To the
flask was then added Hünig's base (0.291 mL, 1.666 mmol) and the
resulting dark, uniform solution was allowed to stir at ambient
temp while monitored by LC-MS (method 1, 5b Tr ¼ 7.0 min, m/
z ¼ 617) The mixture was then concentrated for 2 h in vacuo on a
rotary evaporator. Water (50 mL) was added to the residue causing
a precipitate to form. After stirring for 15 min, the precipitate was
collected by vacuum filtration on a fritted funnel. After air drying
for 10 min, the funnel was dried under vacuum at ambient temp for
12 h to afford crude 5b as a greenish-grey powder. Purification by
flash chromatography (gradient 20:80 to 70:30 ethyl acetate:hex-
ane [v:v]) afforded 181 mg (0.30 mmol, 48%) of pure 5b: 1H NMR
4.5. tert-butyl(4-((2-(1-((6-amino-8[13C]-3H-purin-3-yl)methyl)-
5-hydroxy-2,3-dihydro-1H-benzo[e]indole-3-carbonyl)-1H-indol-
5-yl)carbamoyl)phenyl)carbamate (1c)
To a solution of crude iodide 6 (85 mg, 0.121 mmol, containing
approximately 10% of chloride 5 as an impurity) in dimethylace-
tamide (5 mL) was added 8-[13C]-9H-purin-6-amine (82 mg, 5 eq).
The mixture was placed under nitrogen and heated at 75 ꢀC for 18 h
with monitoring of the reaction progress by HPLC (method 3,
Please cite this article in press as: A.J. Allentoff, et al., Synthesis of an adenine N-3 substituted CBI adduct by alkylation of adenine with a 1-
iodomethylindoline seco-CBI precursor, Tetrahedron (2018), https://doi.org/10.1016/j.tet.2018.09.057

8
A.J. Allentoff et al. / Tetrahedron xxx (2018) 1e9
(600 MHz, Methanol-d4)
d
8.24 (d, J ¼ 8.4 Hz, 1H), 8.08 (s, 1H), 8.00
150.3149.0, 146.4, 143.6, 135.6, 133.2, 132.2, 131.8, 130.5 (td, J ¼ 59.7,
6.0 Hz, 2C, 13C label position), 129.2, 129.0, 125.0, 125.0, 124.5, 123.9,
123.7123.7 (td, J ¼ 58.9, 7.6 Hz, 1C, 13C label position), 122.2, 116.2,
116.0, 114.9 (td, J ¼ 59.9, 5.4 Hz, 2C, 13C label position), 107.7, 101.4,
(s, 1H), 7.86 (br d, J ¼ 8.1 Hz, 1H), 7.67 (s, 1H), 7.76 (dq, J ¼ 159.2,
8.3 Hz, 2H, H-13C), 7.55 (br t, J ¼ 7.5 Hz, 1H), 7.51e7.47 (m, 1H),
7.47e7.44 (m, 1H), 7.44e7.41 (m, 1H), 7.41e7.38 (m, 1H), 6.95 (br s,
1H), 6.74 (dq, J ¼ 156.4, 6.8 Hz, 2H, H-13C), 5.02 (br dd, J ¼ 14.0,
3.4 Hz,1H), 4.90 (br d, J ¼ 11.0 Hz,1H), 4.70e4.58 (m, 2H), 4.47e4.39
57.0, 53.4, 41.2 (32/34 carbons observed); LC-MS (ESI) m/z [MþH]þ
13
calcd for C C6H28N9O3: 616.25, found: 616.24; HRMS (ESI) m/z
28
13
(m, 1H) (20/27 protons observed); LC-MS analysis: m/z [MþH]þ
[MþH]þ calcd for C C6H28N9O3: 616.2511, found: 616.2493; mp
28
13
calcd for C C6H31ClN4O5 617.2, 619.2, found 617.5, 619.5; HRMS
220e224 ꢀC (dec).
28
13
(ESI) m/z [MþH]þ calcd for
C
CH35N9O5: 617.2257, found:
38
617.2244; mp 130e134 ꢀC (dec).
4.10. 3-(5-(4-((tert-Butoxycarbonyl)amino)benzamido)-1H-indole-
2-carbonyl)-1-(iodomethyl)-2,3-dihydro-1H-benzo[e]indol-5-yl 4-
methylpiperazine-1-carboxylate (10)
4.8. tert-Butyl(4-((2-(1-((6-amino-3H-purin-3-yl)methyl)-5-
hydroxy-2,3-dihydro-1H-benzo[e]indole-3-carbonyl)-1H-indol-5-
yl)carbamoyl)phenyl-1,2,3,4,5,6e13C6)carbamate (1e)
In
0.085 mmol) was dissolved into 8 mL of a 3% solution of allyl
alcohol in methylene chloride, and pyridine (41.4 L, 0.512 mmol)
a 25 mL round bottom flask, crude iodide 6 (60 mg,
To a solution of 5b (120 mg, 0.196 mmol) in 2-butanone (5 mL)
was added sodium iodide (147 mg, 0.982 mmol) and the resulting
mixture was stirred at 80 ꢀC under nitrogen and monitored by LC-
MS (method 1, iodide 6b Tr ¼ 5.0 min, m/z ¼ 709.12). After heating
for 72 h conversion of 5b to the corresponding iodide 6b was
approximately 90% complete. After cooling, the mixture was
evaporated to dryness in vacuo, and the residue partitioned be-
tween water and ethyl acetate. The layers were separated, the
aqueous layer extracted twice with ethyl acetate, and the combined
organic layers dried over sodium sulfate and filtered to give a crude
solution of 6b. LC-MS analysis of the solution (method 1), m/z
m
was added followed by 4-methylpiperazine-1-carbonyl chloride (9,
27.8 mg, 0.171 mmol). The resulting solution was stirred at ambient
temp for 72 h while monitoring product formation by HPLC
(method 3, 10 Tr ¼ 13.0 min. After evaporation of the solvent, the
resulting crude was subjected to semi-preparative HPLC purifica-
tion (method 7) affording 28.1 mg (0.034 mmol, 40%) of pure 10 as a
white solid: 1H NMR (DMSO‑d6):
d 11.68 (s, 1 H, N-H), 10.01 (s, 1 H,
NH), 9.65 (s, 1 H, CH), 8.21 (s, 1 H, CH), 8.18 (br s, 1 H, CH), 8.00 (d,
J ¼ 8.36 Hz, CH), 7.92 (d, 2 H, J ¼ 8.8 Hz, CH), 7.85 (d, 1 H, J ¼ 8.36 Hz,
CH), 7.45e7.64 (m, 4 H), 7.59 (d, 2 H, J ¼ 8.8 Hz, CH), 7.23 (br s, 1 H,
CH), 4.88 (dd, 1 H, J ¼ 9.7, 11 Hz, CH2), 4.47 (dd, 1 H, J ¼ 11, 2.2 Hz,
CH2), 4.27 (m, 1 H, CH), 3.68e3.85 (m, 4 H), 3.49 (m, 2 H), 2.41 (m,
4 H), 2.26 (s, 3 H, CH3), 1.50 (s, 9 H, C(CH3)3); HRMS (ESI) m/z
[MþH]þ calcd for C40H41IN6O6: 829.2205, found: 829.2181.
[MþH]þ calcd for C C6H31IN4O5 709.15, found 709.12.
13
28
This material was stored as a crude solution at 4 ꢀC and was
evaporated to give a solid residue as required for the next step.
To a solution of crude 6b (0.138 g, 0.196 mmol) in DMA (6.76 mL)
was added 9H-purin-6-amine (0.132 g, 0.980 mmol) and the
resulting partial slurry was stirred under nitrogen at 80 ꢀC for 18 h
while monitored by LC-MS (method 1, 1e Tr ¼ 5.0 min). The
resulting dark-red solution was cooled and subjected directly to
semi-preparative HPLC (method 6) which afforded 39.2 mg (28% in
4.11. 3-(5-(4-((tert-butoxycarbonyl)amino)benzamido)-1H-indole-
2-carbonyl)-1-methyl-3H-benzo[e]indol-5-yl 4-methylpiperazine-
1-carboxylate (12)
two steps from 4b) of 1e as a light-yellow powder: LC-MS (ESI) m/z
A screw cap test tube was charged with a solution of iodide 10
(10.8 mg, 0.013 mmol) and adenine (8.81 mg, 0.065 mmol) in 1 mL
of DMA. The tube was closed and heated at 80 ꢀC with stirring for
48 h. The progress of the reaction was monitored by HPLC (method
3) and confirmed by LC-MS (method 2) which showed adenine
adducts (m/z ¼ 836, 5:1 ratio presumed N-3 isomer 11a [HPLC
Tr ¼ 10.15 min] to other adenine-containing isomers [HPLC
Tr ¼ 9.6 min, 9.9 min and 10.4 min]), hydrolysis product 11b (m/
z ¼ 761, HPLC Tr ¼ 12.3 min), starting iodide 10 (m/z ¼ 829, HPLC
Tr ¼ 13.0 min) elimination product 12 (m/z ¼ 701, HPLC
Tr ¼ 13.2 min). After 48 h the reaction showed approximately 10%
of the starting iodide 10 remaining and the mixture was allowed to
13
[MþH]þ calcd for
C C6H35N9O5: 716.30, found: 716.26; mp
33
260e263 ꢀC (dec).
4.9. 4-((2-(1-((6-amino-3H-purin-3-yl)methyl)-5-hydroxy-2,3-
dihydro-1H-benzo[e]indole-3-carbonyl)-1H-indol-5-yl)carbamoyl)
benzenaminium- 13C6 TFA salt (1f)
To a partial slurry of 1e (36.2 mg, in 1 mL of methylene chloride
To the slightly cloudy mixture was added trifluoroacetic acid
(0.4 mL, 5.19 mmol) and the resulting uniform, light yellow solution
was stirred at ambient temp for 1.5 h with monitoring by LC-MS
(method 1, 1f Tr ¼ 2.5 min). The solution was evaporated to dry-
ness on a rotary evaporator to afford a yellowish oil. This oil was
dissolved in methanol (3 mL) and evaporated to dryness to afford 1f
cool and to the solution was added 0.1% aqueous TFA (100
acidified mixture was subjected to semi-preparative HPLC to isolate
12 (3.6 mg, 40%) as a white solid: 1H NMR (DMSO‑d6):
12.12 (s,1 H,
mL). This
d
(29 mg, 0.039 mmol, 76%) as a tan solid: 1H NMR (DMSO‑d6):
d
11.52
N-H), 10.07 (s, 1 H, NH), 9.66 (s, 1 H, N-H), 8.57 (d, 1 H, J ¼ 8.6 Hz,
CH), 8.32 (s, 1 H, CH), 8.24 (br s, 1 H, CH), 8.05 (s 1 H, CH), 8.03 (br d,
1 H, J ¼ 8.36 Hz, CH), 7.92 (d, 2 H, J ¼ 8.8 Hz, CH), 7.74 (dd, 1 H, J ¼ 8,
8 Hz, CH), 7.59 (d, 2 H, J ¼ 9 Hz, CH), 7.63 (m, 2 H), 7.52 (d, 1 H,
J ¼ 8.8 Hz, CH), 7.36 (br d, 1 H, J ¼ 1.5 Hz, CH), 6.50 (br s, 2 H, CH2),
4.56 (Br s, 1 H, CH2), 4.88 (dd, 1 H, J ¼ 9.7, 11 Hz, CH2), 4.19 (br s, 1 H,
CH2), 3.54 (br s, 4 H), 2.90 (br s, 3 H, CH3), 2.73 (s, 3 H, CH3), 1.50 (s,
9 H, C(CH3)3); LC-MS (ESI) m/z [MþH]þ calcd for C40H40N6O6: 701.3,
found: 701.48; HRMS (ESI) m/z [MþH]þ calcd for C40H40N6O6:
701.3082, found: 701.3065.
(s, 1 H), 10.51 (s, 1 H), 9.70 (s, 1 H), 9.29 (s, 1 H), 8.75 (s, 1 H), 8.59 (s,
2 H), 8.17 (d, 1 H, J ¼ 8.3 Hz), 8.11 (s, 1 H), 8.05 (d, 1 H, J ¼ 8.3 Hz),
7.99 (br s, 1 H), 7.74 (br d, 2 H, J ¼ 166 Hz, aryl 13C-H), 7.55 (ddd, 1 H,
J ¼ 1.1, 8, 8 Hz), 7.52 (dd, 1 H, J ¼ 1.9, 8.8 Hz), 7.49 (d, J ¼ 8.8 Hz), 7.41
(ddd, 1 H, J ¼ 1.1, 8, 8 Hz), 6.98 (s, 1 H), 6.64 (br d, 2 H, J ¼ 157 Hz, aryl
13C-H), 4.78 (br d, 1 H, J ¼ 11.4 Hz), 4.45e4.69 (m, 4 H); 13C NMR
(126 MHz, DMSO‑d6)
d
160.2, 154.4, 151.8 (td, J ¼ 59.3, 7.7 Hz, 1C, 13
C
label position), 148.0, 145.3 (br s, 1C), 142.2, 133.1, 132.5, 132.5, 131.1,
130.0, 129.1 (td, J ¼ 59.0, 5.4 Hz, 2C, 13C label position), 127.5, 127.0,
127.0 (br s, 1C), 123.3, 123.2 (br s, 1C), 122.7, 122.1, 121.4 (td, J ¼ 58.6,
7.3 Hz, 1C,13C label position), 119.6, 114.3, 112.5 (td, J ¼ 59.7, 5.9 Hz,
2C, 13C label position), 110.5 (br s, 1C), 105.6, 100.2, 54.4, 51.8 (br s,
Acknowledgements
1C), 39.01e38.95 (m, 1C); 13C NMR (151 MHz, Methanol-d4)
d
162.3,
The authors wish to thank Bilal Sufi, and Lourdes Thevanayagam
of Bristol-Myers Squibb Biologics Discovery California for providing
156.5, 155.2, 153.6 (td, J ¼ 59.9, 7.6 Hz, 1C, 13C label positon),
Please cite this article in press as: A.J. Allentoff, et al., Synthesis of an adenine N-3 substituted CBI adduct by alkylation of adenine with a 1-
iodomethylindoline seco-CBI precursor, Tetrahedron (2018), https://doi.org/10.1016/j.tet.2018.09.057

A.J. Allentoff et al. / Tetrahedron xxx (2018) 1e9
9
2862.
samples of the (R)-enantiomer of adduct 1b prepared from in-
cubations of the (R,S)-enantiomer of 7 with calf thymus DNA.
[12] T.K. Owonikoko, A. Hussain, W.M. Stadler, D.C. Smith, H. Kluger, A.M. Molina,
P. Gulati, A. Shah, C.M. Ahlers, P.M. Cardarelli, L.J. Cohen, Cancer Chemother.
Pharmacol. 77 (2016) 155e162.
[13] R.Y. Zhao, H.K. Erickson, B.A. Leece, E.E. Reid, V.S. Goldmacher, J.M. Lambert,
R.V.J. Chari, J. Med. Chem. 55 (2012) 766e782.
Appendix A. Supplementary data
[14] D.L. Boger, Pure Appl. Chem. 66 (4) (1994) 837e844.
[15] D.L. Boger, T. Ishizaki, H. Zarrinmayeh, J. Am. Chem. Soc. 113 (1991)
6645e6649.
[16] D.L. Boger, S.L. Brunette, R.M. Garbaccio, J. Org. Chem. 66 (2001) 5163e5173.
[17] D.L. Boger, B. Bollinger, D.L. Hertzog, D.S. Johnson, H. Cai, P. Mesini,
R.M. Garbaccio, Q. Jin, P.A. Kitos, J. Am. Chem. Soc. 119 (1997) 4987e4998.
[18] P.S. Eis, J.A. Smith, J.M. Rydzewski, J.A. Case, D.L. Boger, W.J. Chazin, J. Mol. Biol.
272 (1997) 237e252.
[19] A.V. Vargiu, P. Ruggerone, A. Magistrato, P. Carloni, Biophys. J. 94 (2) (2008)
550e561.
[20] M. Ichimura, T. Ogawa, K. Takahashi, A. Mihara, I. Takahashib, H. Nakano,
Oncol. Res. 5 (1993) 165e171.
[21] K. Kiakos, A. Sato, T. Asao, P.J. McHugh, M. Lee, J.A. Hartley, Mol. Cancer
Therapeut. 6 (2007) 2708e2718.
[22] M. Minoshima, T. Bando, S. Sasaki, K. Shinohara, T. Shimizu, J. Fujimoto,
H. Sugiyama, J. Am. Chem. Soc. 129 (2007) 5384e5390.
[23] W. Wrasidlo, D.S. Johnson, D.L. Boger, Bioorg. Med. Chem. Lett. 4 (4) (1994)
631e636.
[24] A. Asai, S. Nagamura, H. Saito, J. Am. Chem. Soc. 116 (1994) 4171e4177.
[25] L. Thevanayagam, A. Bell, I. Chakraborty, B. Sufi, S. Gangwar, A. Zang,
V. Rangan, C. Rao, Z. Wang, C. Pan, C. Chong, P. Cardarelli, S. Deshpande,
S. Srinivasan, Bioanalysis 5 (9) (2013) 1073e1081.
[26] H. Wang, V.S. Rangan, M.-C. Sung, D. Passmore, T. Kempe, X. Wang,
L. Thevanayagam, C. Pan, C. Rao, M. Srinivasan, Q. Zhang, S. Gangwar,
S. Deshpande, P. Cardarelli, P. Marathe, Z. Yanga, Biopharm Drug Dispos. 37
(2016) 93e106.
[27] T.K. Owonikoko, A. Hussain, W.M. Stadler, D.C. Smith, H. Kluger, A.M. Molina,
P. Gulati, A. Shah, C.M. Ahlers, P.M. Cardarelli, L.J. Cohen, Cancer Chemother.
Pharmacol. 77 (2016) 155e162.
[28] D.L. Boger, H.W. Schmitt, B.E. Fink, M.P. Hedrick, J. Org. Chem. 66 (2001)
6654e6661.
[29] D.L. Boger, T. Ishizaki, P.A. Kitos, O. Suntornwat, J. Org. Chem. 55 (1990)
5823e5832.
[30] M.A. Warpehoski, D.E. Harper, J. Am. Chem. Soc. 117 (1995) 2951e2952.
[31] T. Fujii, G.C. Walker, N.J. Leonard, J. Med. Chem. 22 (1979) 125e129.
[32] Zhang, Q.; Gangwar, S.; Pan, C.; Derwin, D.W. PCT WO 2012162482 A1
20121129 Int. Appl. 2012.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tet.2018.09.057.
References
[1] D.L. Boger, D.S. Johnson, Angew. Chem., Int. Ed. Engl. 35 (1996) 1438e1474.
[2] K.S. MacMillan, D.L. Boger, J. Med. Chem. 52 (2009) 5771e5780.
[3] P.C. Patila, V. Satamb, M. Lee, Anti Cancer Agents Med. Chem. 15 (2015)
616e630.
[4] P.G. Baraldi, A. Bovero, F. Fruttarolo, D. Preti, M.A. Tabrizi, M.G. Pavani,
R. Romagnoli, Med. Res. Rev. 24 (4) (2004) 475e528.
[5] G.A. Vielhauer, M. Swink, N.K. Parelkar, J.P. Lajiness, A.L. Wolfe, D. Boger,
Cancer Biol. Ther. 14 (6) (2013) 527e536.
[6] G. Menderes, E. Bonazzoli, S. Bellone, J. Black, G. Altwerger, A. Masserdotti,
F. Pettinella, L. Zammataro, N. Buza, P. Hui, S. Wong, B. Litkouhi, E. Ratner, D.-
A. Silasi, G.S. Huang, M. Azodi, P.E. Schwartz, A.D. Santin, Gynecol. Oncol. 146
(2017) 179e186.
[7] J. Black, G. Menderes, S. Bellone, Schwab, E. Bonazzoli, F. Ferrari, F. Predolini,
C. De Haydu, E. Cocco, N. Buza, P. Hui, S. Wong, S. Lopez, E. Ratner, D.-A. Silasi,
M. Azodi1, B. Litkouhi, P.E. Schwartz, P. Goedings, P.H. Beusker, M.C. van der
Lee, C.M. Timmers, W.H.A. Dokter, A.D. Santin, Mol. Cancer Therapeut. 15
(2016) 1900e1909.
[8] B.E. de Goeij, J.M. Lambert, Curr. Opin. Immunol. 40 (2016) 14e23.
[9] M.M.C. van der Lee, P.G. Groothuis, R. Ubink, M.A.J. van der Vleuten, T.A. van
Achterberg, E.A. Loosveld, D. Damming, D.C.H. Jacobs, D. Rouwette,
D.F. Egging, D. van den Dobbelsteen, P.H. Beusker, P. Goedings,
G.F.M. Verheijden, J.M. Lemmens, M. Timmers, W.H.A. Dokter, Mol. Cancer
Therapeut. 14 (2015) 692e703.
[10] R.C. Elgersma, R.G.E. Coumans, T. Huijbregts, W.M.P.B. Menge, J.A.F. Joosten,
H.J. Spijker, F.M.H. de Groot, M.M.C. van der Lee, R. Ubink, D.J. van den Dob-
belsteen, D.F. Egging, W.H.A. Dokter, G.F.M. Verheijden, J.M. Lemmens,
C.M. Timmers, P.H. Beusker, Mol. Pharm. 12 (2015) 1813e1835.
[11] G. Menderes, E. Bonazzoli, S. Bellone, J. Black, F. Predolini, F. Pettinella,
A. Masserdotti, L. Zammataro1, G. Altwerger, N. Buza, P. Hui, S. Wong,
B. Litkouhi1, E. Ratner, D.N. Silasi, M. Azodi, P.E. Schwartz, A.D. Santin, Clin.
Canc. Res. 2017 (July 5 2017), https://doi.org/10.1158/1078-0432.CCR-16-
Please cite this article in press as: A.J. Allentoff, et al., Synthesis of an adenine N-3 substituted CBI adduct by alkylation of adenine with a 1-
iodomethylindoline seco-CBI precursor, Tetrahedron (2018), https://doi.org/10.1016/j.tet.2018.09.057