Synthesis and structure determination of SCR7, a DNA ligase inhibitor

Article abstract of DOI:10.1016/j.tetlet.2016.06.037

In contrast to a published report, reaction of 4,5-diamino-6-hydroxy-2-mercaptopyrimidine with 2?equiv of aromatic aldehydes produces a mixture of 6,7-diaryl-2-thioxopteridine-4-one and 6,7-diaryl-2-thioxo-7,8-dihydropteridine-4-one rather than a diimine. These compounds represent the correct structure for SCR7, a substance reported to be an inhibitor of nonhomologous end-joining, a DNA repair pathway. The dihydropteridine can be isolated as a minor product, and it can be oxidized to the pteridine.

Full text of DOI:10.1016/j.tetlet.2016.06.037

Tetrahedron Letters 57 (2016) 3204–3207
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis and structure determination of SCR7, a DNA ligase
inhibitor
a
a
a
b
George E. Greco a, , Zane A. Conrad , Alycia M. Johnston , Qingyao Li , Alan E. Tomkinson
⇑
a Department of Chemistry, Goucher College, Baltimore, MD 21204, United States
b University of New Mexico Cancer Center, Albuquerque, NM 87131, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
In contrast to a published report, reaction of 4,5-diamino-6-hydroxy-2-mercaptopyrimidine with 2 equiv
of aromatic aldehydes produces a mixture of 6,7-diaryl-2-thioxopteridine-4-one and 6,7-diaryl-2-thioxo-
7,8-dihydropteridine-4-one rather than a diimine. These compounds represent the correct structure for
SCR7, a substance reported to be an inhibitor of nonhomologous end-joining, a DNA repair pathway.
The dihydropteridine can be isolated as a minor product, and it can be oxidized to the pteridine.
Ó 2016 Elsevier Ltd. All rights reserved.
Received 19 May 2016
Accepted 9 June 2016
Available online 11 June 2016
Keywords:
Pteridines
Dihydropteridines
DNA ligase inhibitor
Structure determination
NMR
Pericyclic reaction
Introduction
equivalents of benzaldehyde for 3 h under reflux (155 degrees) in
a 3:1 mixture of DMF and acetic acid (Eq. 1).
During the course of our work with human DNA ligase
inhibitors,1,2 we encountered a report describing SCR7, reported
to be a selective inhibitor of DNA Ligase IV.3 The proposed struc-
ture for SCR7 is a diimine with structure A (Fig. 1).3 Based on our
analysis of 1H and 13C NMR spectra, structure A is not isolated
when following the published procedure for the synthesis of
SCR7. Herein, we provide evidence that the major compound
obtained from the published procedure is pteridine 1a. Further-
more, based on our analysis of a sample originating in the lab of
the authors, the compound used for the reported biological studies
is 7,8-dihydropteridine 2a, a minor product of the reaction that
forms SCR7. Full details of our biological studies with these com-
pounds were reported elsewhere,4 but in our hands, both 1a and
2a are inhibitors of all three human DNA ligases. Pteridines 1a
and 1b are known compounds with biological activity. Both com-
pounds have been previously shown to be active as nematocides,5,6
and 1a has recently been shown to inhibit replication of HIV-1,
although it is also toxic to host cells.7
O
DMF
HOAc
(3:1)
Δ
H2N
H2N
N
SH
+
SCR7
2
H
ð1Þ
N
OH
Following this procedure, we obtained a yellow solid upon addi-
tion of water to the cooled reaction mixture. Based on NMR analy-
sis, we conclude that this crude material is approximately a 2:1
mixture of compound 1a and the previously published ligase inhi-
bitor L189, (which Raghavan refers to as SCR6). Consistent with the
published report, we obtained L189 as the sole product when we
reacted
4,5-diamino-6-hydroxy-2-mercaptopyrimidine
with
1 equiv of benzaldehyde overnight at room temperature. The
published purification of SCR7 consists of recrystallization from
DMF–ethanol.3 In our hands, an orange solid can be crystallized
in very low yields from DMF-ethanol, but it is still not pure. NMR
analysis showed it to be a 1:1.5 mixture of compound 1a and
L189 (enriched in L189 relative to the crude precipitate).
Compound 1a is best purified by column chromatography eluting
with 2:1 CH2Cl2/ethyl acetate. A typical yield of compound 1a by
this route is 39%. The synthesis of pteridines from the reaction of
4,5-diaminopyrimidines with aromatic aldehydes has been previ-
ously described,8 and Ochoa used the same method to synthesize
compound 1a.5 SCR7 is now commercially available from XCessBio
(with the published structure in their online catalog), however, the
material they sold to us was actually compound 1a.
Synthesis and structure determination of pteridines
The published procedure3 for the synthesis of SCR7 consists of
heating 4,5-diamino-6-hydroxy-2-mercaptopyrimidine with
2
⇑
Corresponding author.
http://dx.doi.org/10.1016/j.tetlet.2016.06.037
0040-4039/Ó 2016 Elsevier Ltd. All rights reserved.

G. E. Greco et al. / Tetrahedron Letters 57 (2016) 3204–3207
3205
l
a related compound (1b) using the same method as our synthesis
of 1a. In the 1H NMR spectrum of 1b, the signals for hydrogens i,
j, k, and l (Fig. 1) on the two inequivalent aromatic rings are all
well-resolved doublets, and our spectrum matches the data
reported in the literature.
R
R
j
H
N
e
d
f
N
N
S
N
N
N
SH
NH
c
N
O
i
Given the overlap of signals in the 1H NMR, and the fact that all
of the carbons in the pteridine core are quaternary, 13C NMR is a
better tool to support our contention that 1a is the structure of
the major product formed under the conditions reported for the
synthesis of SCR7. Raghavan et al. do not report any 13C NMR data
in their Article,3 but 13C data for compounds 1a5 and 1b6 have been
previously reported. Our 13C data for compound 1b matches the
data reported in the literature exactly. There are slight discrepan-
cies between our data and the literature data for compound 1a,
but through the use of DEPT, HMQC, and HMBC spectra, we have
unambiguously assigned all of the carbons in the pteridine core
of compounds 1a and 1b (Table 1).
k
OH
1a (R=H)
1b (R=OCH3)
A
l
R
R
j
H
N
s
f
H
N
H2N
N
N
SH
e
d
S
N
NH
c
N
OH
O
i
k
L189
2a (R=H)
2b (R=OCH3)
Figure 1. Structures of compounds produced in reaction of 4,5-diamino-6-
hydroxy-2-mercaptopyrimidine with benzaldehyde or p-anisaldehyde. A = struc-
ture proposed by Raghavan for product obtained when the pyrimidine is heated
with 2 equivalents of benzaldehyde,3 1a = structure of major isolated product under
these conditions; 2a = structure of minor isolated product; L189 = previously
reported product1 obtained through reaction of the pyrimidine with 1 equiv of
benzaldehyde at room temperature.
Synthesis and structure determination of 7,8-dihydropteridines
We analyzed a sample of SCR7 that originated in the authors’s
lab by 1H, 13C, DEPT, and COSY NMR spectroscopy, expecting it to
be identical to the material that we isolated from the reaction
(compound 1a). However, the NMR spectra are not consistent with
structure A, compound 1a, or L189. Based on a detailed analysis of
the spectra, we conclude that this compound is actually a novel
7,8-dihydropteridine with structure 2a. Specifically, the peak for
Hi at d 7.83 is separated from the rest of the aromatic hydrogens,
suggesting that these hydrogens are ortho to a ‘typical’ C@N double
bond rather than a pteridine ring system. The doublet at d 7.52 dis-
appears upon shaking the NMR sample with D2O, suggesting that it
is an NAH (Hq). The HAH COSY spectrum contains cross peaks
between the resonances at d 7.52 and d 6.02, indicating that the
peak at d 6.02 is He. The absence of this distinctive peak in the
NMR data reported for SCR7 suggests that compound 2a was not
a significant component of the NMR sample analyzed for the orig-
inal Article. The 13C NMR spectrum contains a peak at d 52.3 which
indicates an aliphatic carbon, and according to the DEPT spectra,
this carbon bears a single hydrogen (He). The peak assigned to car-
bon c is further upfield in 2a, as compared to 1a, which is consis-
tent with an ordinary pyrimidine ring system rather than a
pteridine ring system. TLC analysis indicates that compound 2a is
much more polar than compound 1a, also consistent with the
NAH bond in the proposed structure.
Structure A is not consistent with either the published NMR
data3 or the NMR for the compound that we isolated. Specifically,
the two imine substituents in a hypothetical structure A would
be inequivalent due to the asymmetry of the pyrimidine ring, so
the spectrum for structure A should contain two singlets (not a
multiplet) for the two inequivalent imine hydrogens around
9.6 ppm. In addition, the spectrum for structure A should contain
two inequivalent doublets between 7.8 and 7.9 ppm integrating
to 4 hydrogens (ortho hydrogens on the phenyl rings), a series of
peaks between 7.3 and 7.4 ppm integrating to 6 hydrogens (meta
and para hydrogens on phenyl rings), and no peaks around
8.1 ppm. The published NMR data were most likely obtained from
a mixture of compound 1 and L189, which is the result of following
the published procedure exactly. The reported peak at d 12.80 and
part of the reported multiplet between d 7.53 and d 7.36 are con-
sistent with our spectrum for compound 1a. The reported peaks
at d 11.97, 9.64, 7.88–7.86, and 7.53–7.36 are present in L189.
The only peak that does not belong to either compound is the peak
from 8.11 to 8.08, so it must be due to another impurity.
The more established route to pteridines involves the reaction
of
a-diketones with 4,5-diaminopyrimidines. For example, reac-
tion of the 4,5-diamino-2-mercapto-4-hydroxypyrimidine with
benzil is reported to produce compound 1a in 40% yield (Eq. 2).9
We were unable to isolate any of compound 2a using the pub-
lished purification procedure, but we were able to observe it as a
minor product in the NMR spectra of the crude product. In order
H
N
H2N
H2N
N
SH
O
O
N
N
S
+
N
ð2Þ
NH
OH
O
1a
Table 1
When we carried out the synthesis of compound 1a starting
13C NMR assignments of selected carbons in 1a and 1b
from these reagents but using a more current procedure,10 the
material we obtained was physically and spectroscopically identi-
cal to the sample that we purified from the benzaldehyde reaction
as described above. The mass spectrum of compound 1a indicates
that the molecular mass of the compound is 332, 2 mass units less
than that of structure A suggesting loss of 2 hydrogen atoms.
Furthermore, the high resolution mass spectrum of compound 1a
confirms a molecular formula of C18H12N4OS.
l
l
O
n
n
j
h
j
h
p
H
N
H
N
a
N
N
a
S
N
N
S
e
d
f
e
d
f
l
l
j
j
i
i
NH
NH
k
k
c
c
g
i
g
i
b
O
b
O
o
m
m
O
k
k
1a carbons
d
1b carbons
d
The 1H NMR spectrum of compound 1a is difficult to interpret
because all of the aromatic peaks overlap, and the only other
hydrogens in the molecule are the downfield amide hydrogens.
There is a single recent report containing 1H NMR data for com-
pound 1a in CDCl3, however that spectrum is also different from
our spectrum, and the literature report suggests their route pro-
duces a different tautomer. As a result, we decided to synthesize
a
b
c
d
e
f
175.7
a
b
c
d
e
f
m/n
o/p
175.5
158.6
126.4
146.6
155.2
148.7
160.6, 159.7
55.35, 55.25
158.5
127.2
147.0
155.9
149.1
137.7
137.1
g
h

3206
G. E. Greco et al. / Tetrahedron Letters 57 (2016) 3204–3207
to increase the yield of compound 2a relative to compound 1a, we
modified the reaction conditions by omitting acetic acid, and carry-
ing out the reaction under nitrogen. Furthermore, we developed a
different workup procedure to take advantage of the polarity dif-
ference. Even under these improved conditions, pure compound
2a can only be isolated in 22% yield (Eq. 3).
for similar compounds with an oxygen at the 2-position.11 With
the ability to synthesize and purify compound 2a, we were able
to record HMQC and HMBC spectra for the compound. Resonances
that we were able to assign based on these 2D spectra can be found
in Table 2. The presence of an HMBC cross peak between He
(6.02 ppm) and Cf (143.2 ppm) confirms the 7,8-dihydropteridine
structure. If the 5,6-dihydropteridine structure were correct, and
the hydrogen were on carbon d, there would be a cross peak to car-
bon c instead.
ð3Þ
Due to the overlap of most aromatic resonances in the 1H NMR
spectrum of 2a, we were unable to assign all of the protons and
carbons even using 2D techniques. To confirm the assignments,
we synthesized compound 2b in 8% isolated yield by the same
method. All of the protons and carbons can be unambiguously
assigned for compound 2b (Table 2). The specific HMBC cross
peaks that confirm the 7,8-dihydropteridine structure are between
He (5.91 ppm), and Cf (143.0 ppm), between Hi (7.76 ppm) and Cd
(146.8 ppm), and between Hj (7.19 ppm) and Ce (51.8 ppm).
To provide further evidence in support of our proposed struc-
ture for compounds 2a and 2b, we carried out an oxidation of each
compound in refluxing nitrobenzene, conditions which are known
to promote dehydrogenation to produce aromatic systems12 (Eq.
4). Compound 1a was isolated in 44% yield, and compound 1b
was isolated in 49% yield with no evidence of any other by-
products.
The 1H, 13C, DEPT, and COSY spectra provided evidence that
compound 2a is a dihydropteridine, but the question remained of
whether the correct structure is a 5,6-dihydropteridine or a
7,8-dihydropteridine (Fig. 2). Both structures have been reported
H
N
H
N
H
N
e
f
e
f
R
R
N
S
R
R
S
NH
NH
N
H
d
N
c
c
d
O
O
5,6-dihydropteridine
7,8-dihydropteridine
Figure 2. 5,6-Dihydropteridine versus 7,8-dihydropteridine.
H
N
H
N
H
N
Table 2
S
N
N
S
PhNO2
1H and 13C NMR assignments for 2a and 2b
ð4Þ
NH
NH
Δ
N
l
l
O
O
O
n
n
O
2a
1a
j
j
h
p
H
N
H
N
H
N
H
N
q
q
r
r
a
S
h
a
S
e
d
f
e
d
f
l
l
The synthesis of similar dihydropteridines from
a
-hydroxyke-
j
j
tones has been reported.11,13 In analogous fashion, reaction of
4,5-diamino-6-hydroxy-2-mercaptopyrimidine with benzoin in
acetic acid affords compound 2a in 42% yield (Eq. 5).
i
i
NH
NH
k
s
k
s
N
N
c
c
g
i
g
i
b
O
b
O
o
m
m
k
k
2a
1H
13C
2b
1H
13C
H
N
H
N
H2N
H2N
N
SH
a
b
c
d
e
f
g
h
i
j
k
l
m
n
—
—
—
—
6.02
—
173.1
157.6
104.3
146.7
52.3
143.2
136.1
140.1
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
p
q
—
—
—
—
5.91
—
—
172.8
157.6
104.2
146.8
51.8
143.0
128.8
132.4
128.1
128.3
114.0
114.5
160.6
159.3
55.3
OH
O
S
HOAc
+
ð5Þ
N
NH
Δ
N
OH
O
2a
Figure 3 outlines the most likely mechanism of the reaction
—
—
between 4,5-diamino-6-hydroxy-2-mercaptopyrimidine and aro-
matic aldehydes. Under mild conditions, only one amino group
reacts with benzaldehyde to form an imine and give L189. Under
more forcing conditions, the second amino group presumably
reacts the same way as the first to give the bis-imine (structure
A). Not only did we not isolate a compound with structure A, there
was no evidence of such a structure in crude reaction mixtures. It is
not surprising that structure A cannot be isolated, since in the lit-
erature, analogous compounds are only postulated as unstable
intermediates. For example, reaction of o-phenylenediamine with
benzaldehyde produces a substituted benzimidazole.14 Structure
—
7.83
7.27
Unassigned
Unassigned
Unassigned
Unassigned
126.4
7.76
7.19
6.92
6.88
—
128.56
Unassigned
Unassigned
Unassigned
Unassigned
—
3.75
3.68
7.33
11.95
11.88
55.2
—
—
—
q
7.52
12.07
12.00
—
—
—
r or s
r or s
r or s
r or s
O
O
H
H2N
N
N
SH
H
N
N
SH
SH
H2N
H2N
N
SH
N
N
N
N
A
OH
L189
OH
OH
H
H
H
N
H
H
N
N
S
N
N
N
N
S
H
NH
N
NH
N
N
O
OH
2a
O
1a
Figure 3. Possible mechanism of formation of compounds 2a and 1a through L189 and structure A.

G. E. Greco et al. / Tetrahedron Letters 57 (2016) 3204–3207
3207
A then undergoes a 6
p
electrocyclic ring closure to produce
article can be found, in the online version, at http://dx.doi.org/10.
1016/j.tetlet.2016.06.037. These data include MOL files and
InChiKeys of the most important compounds described in this
article.
another intermediate which has not been identified or isolated.
This intermediate undergoes a [1,5] sigmatropic rearrangement
to give dihydropteridines 2a or 2b. The dihydropteridine is oxi-
dized under the reaction conditions to give the fully aromatic pter-
idine 1a or 1b.
References and notes
We have demonstrated that the reaction conditions reported to
produce SCR7 actually produce a mixture of 1a and 2a, with 2a
being a novel compound. The scope of this dihydropteridine syn-
thesis and the optical spectroscopic properties of the dihy-
dropteridines are currently under investigation.
1. Chen, X.; Zhong, S.; Zhu, X.; Dziegielewska, B.; Ellenberger, T.; Wilson, G. M.;
MacKerell, A. D., Jr.; Tomkinson, A. E. Cancer Res. 2008, 68, 3169–3177.
2. Tomkinson, A. E.; Howes, T. R.; Wiest, N. E. Transl. Cancer Res. 2013, 2, 1219.
3. Srivastava, M.; Nambiar, M.; Sharma, S.; Karki, S. S.; Goldsmith, G.; Hegde, M.;
Kumar, S.; Pandey, M.; Singh, R. K.; Ray, P.; Natarajan, R.; Kelkar, M.; De, A.;
Choudhary, B.; Raghavan, S. C. Cell (Cambridge, MA, U.S.) 2012, 151, 1474–1487.
4. Greco, G. E.; Matsumoto, Y.; Brooks, R. C.; Lu, Z.; Lieber, M. R.; Tomkinson, A. E.
DNA Repair 2016, 43, 18–23.
Acknowledgements
5. Ochoa, C.; Rodriguez, J.; Lopez Garcia, M. L.; Martinez, A. R.; Martinez, M. M.
Arzneim-Forsch. 1996, 46, 643–648.
We thank Goucher College for financial support of this project
including the Lewent and Claasen Funds for summer research
opportunities. We thank Michael Lieber (Keck School of Medicine,
University of Southern California) for a sample of SCR7. We also
thank Phil Mortimer in the Johns Hopkins University Chemistry
Department Mass Spectrometry Facility for acquiring high resolu-
tion mass spectrometry data.
6. Ochoa, C.; Rodriguez, J.; Rodriguez, M.; Chana, A.; Stud, M.; Alonso-Villalobos,
P.; Martinez-Grueiro, M. M. Med. Chem. Res. 1997, 7, 530–545.
7. Bhatt, H.; Patel, P.; Pannecouque, C. Chem. Biol. Drug Des. 2014, 83, 154–166.
8. Pfleiderer, W.; Blank, H. U. Angew. Chem., Int. Ed. Engl. 1968, 7, 535–536.
9. Schneider, H. J.; Pfleiderer, W. Chem. Ber. 1974, 107, 3377–3394.
10. Wrasidlo, W.; Doukas, J.; Royston, I.; Noronha, G.; Hood, J. D.; Dneprovskaia, E.;
Gong, X.; Splittgerber, U.; Zhao, N. US Patent Appl., 2005; pp 95.
11. Viscontini, M.; Leidner, H. Helv. Chim. Acta 1968, 51, 1030–1037.
12. Dubey, P. K.; Chowdary, K. S.; Ramesh, B.; Reddy, P. V. V. P. Synth. Commun.
2010, 40, 697–708.
13. Viscontini, M.; Huwyler, S. Helv. Chim. Acta 1965, 48, 764–768.
14. Chebolu, R.; Kommi, D. N.; Kumar, D.; Bollineni, N.; Chakraborti, A. K. J. Org.
Chem. 2012, 77, 10158–10167.
Supplementary data
Supplementary data (experimental procedures (synthetic pro-
cedures, and compound characterization)) associated with this