Discovery and optimization of aspartate aminotransferase 1 inhibitors to target redox balance in pancreatic ductal adenocarcinoma

Article abstract of DOI:10.1016/j.bmcl.2018.04.061

Pancreatic ductal adenocarcinoma (PDAC) is a lethal malignancy that is extremely refractory to the therapeutic approaches that have been evaluated to date. Recently, it has been demonstrated that PDAC tumors are dependent upon a metabolic pathway involving aspartate aminotransferase 1, also known as glutamate-oxaloacetate transaminase 1 (GOT1), for the maintenance of redox homeostasis and sustained proliferation. As such, small molecule inhibitors targeting this metabolic pathway may provide a novel therapeutic approach for the treatment of this devastating disease. To this end, from a high throughput screen of ～800,000 molecules, 4-(1H-indol-4-yl)-N-phenylpiperazine-1-carboxamide was identified as an inhibitor of GOT1. Mouse pharmacokinetic studies revealed that potency, rather than inherent metabolic instability, would limit immediate cell- and rodent xenograft-based experiments aimed at validating this potential cancer metabolism-related target. Medicinal chemistry-based optimization resulted in the identification of multiple derivatives with >10-fold improvements in potency, as well as the identification of a tryptamine-based series of GOT1 inhibitors.

Full text of DOI:10.1016/j.bmcl.2018.04.061

Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery and optimization of aspartate aminotransferase 1 inhibitors to
target redox balance in pancreatic ductal adenocarcinoma
Justin Anglin a, Reza Beheshti Zavareh a,b, Philipp N. Sander b, Daniel Haldar c, Edouard Mullarky d,
Lewis C. Cantley d, Alec C. Kimmelman e, Costas A. Lyssiotis f,g, Luke L. Lairson a,b,
⇑
a California Institute for Biomedical Research, 11119 North Torrey Pines Road, La Jolla, CA 92037, USA
b Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA
c Department of Systems Biology, Harvard Medical School, Boston, MA 02215, USA
d Meyer Cancer Center, Department of Medicine, Weill Cornell Medical College, New York, NY 10065, USA
e Department of Radiation Oncology, Perlmutter Cancer Center, NYU Langone Medical Center, New York, NY 10016, USA
f Department of Molecular and Integrative Physiology, University of Michigan Medical School, Ann Arbor, MI 48109, USA
g Department of Internal Medicine, Division of Gastroenterology, University of Michigan Medical School, Ann Arbor, MI 48109, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Pancreatic ductal adenocarcinoma (PDAC) is a lethal malignancy that is extremely refractory to the ther-
apeutic approaches that have been evaluated to date. Recently, it has been demonstrated that PDAC
tumors are dependent upon a metabolic pathway involving aspartate aminotransferase 1, also known
as glutamate-oxaloacetate transaminase 1 (GOT1), for the maintenance of redox homeostasis and sus-
tained proliferation. As such, small molecule inhibitors targeting this metabolic pathway may provide
a novel therapeutic approach for the treatment of this devastating disease. To this end, from a high
throughput screen of ꢀ800,000 molecules, 4-(1H-indol-4-yl)-N-phenylpiperazine-1-carboxamide was
identified as an inhibitor of GOT1. Mouse pharmacokinetic studies revealed that potency, rather than
inherent metabolic instability, would limit immediate cell- and rodent xenograft-based experiments
aimed at validating this potential cancer metabolism-related target. Medicinal chemistry-based opti-
mization resulted in the identification of multiple derivatives with >10-fold improvements in potency,
as well as the identification of a tryptamine-based series of GOT1 inhibitors.
Received 11 March 2018
Revised 24 April 2018
Accepted 25 April 2018
Available online xxxx
Keywords:
Aspartate aminotransferase-1
Glutamic-oxaloacetic transaminase-1
Inhibitor
Cancer metabolism
Pancreatic ductal adenocarcinoma
Ó 2018 Published by Elsevier Ltd.
Deregulation of metabolism is a hallmark feature of cancer.1–5
The rewiring of metabolic programs in cancer cells can serve a
range of functions. For example, cancer cells alter metabolism in
order to facilitate bioenergetics and biosynthesis, to manage cellu-
lar stressors and challenges to redox balance and/or to regulate
gene expression.5 The role of metabolic deregulation in cancer sig-
nificantly depends on context. Tissue of origin, tissue architecture,
degree of oxygenation, as well as other factors influence the meta-
bolic programs of a particular cancer. Thus, small molecule phar-
macological probes that target defined components of metabolic
pathways/processes have the potential to address specific cancer
metabolism-related hypotheses that are challenging to query using
genetic methods. Further, such probe molecules may serve as start-
ing points for drug discovery programs aimed at targeting cancer
metabolism, as part of drug combination therapeutic approaches
for the treatment of refractory tumor types.6–8 As cancer metabo-
lism targets are frequently found to be overexpressed in tumor
cells, which therefore results in the need for higher levels of target
engagement in tumor versus non-diseased cell types, a challenge
to this approach is the identification of cancer metabolism targets
that selectively sensitize cancer cells without impacting non-dis-
eased cell types.
PDAC is a lethal cancer type that is projected to become the sec-
ond leading cause of cancer-related deaths in the United States by
2020 and, at present, has a 5-year survival rate of ꢀ6%.9,10 Recently,
several PDAC cell lines were demonstrated to be reliant on a KRAS-
regulated non-canonical glutamine (Gln) metabolism pathway that
enables proliferation and tumor growth.11 Specifically, KRAS-
dependent down-regulation of glutamate dehydrogenase (GLUD1)
and upregulation of GOT1 in PDAC results in Gln-derived aspartate
being converted to oxaloacetate (OAA) by GOT1 in the cytoplasm,
which is subsequently converted to pyruvate through a series of
reactions that ultimately lead to the generation of NADPH, which
is used to maintain redox balance.11 GOT1 knockdown in PDAC
⇑
Corresponding author at: Department of Chemistry, The Scripps Research
Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA.
E-mail address: llairson@scripps.edu (L.L. Lairson).
https://doi.org/10.1016/j.bmcl.2018.04.061
0960-894X/Ó 2018 Published by Elsevier Ltd.
Please cite this article in press as: Anglin J., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.04.061

2
J. Anglin et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
Table 1
was demonstrated to stunt growth in vitro and in vivo, whereas
GOT1 knockdown in normal cells had no effect on proliferation.11
As such, GOT1 inhibitors may provide a much-needed therapeutic
approach for selectively targeting PDAC.
GOT1 inhibition activity of compounds 1a–n.
A biochemical assay-based screen of ꢀ800,000 compounds at
the California Institute for Biomedical Research resulted in the
identification 4-(1H-indol-4-yl)-N-phenylpiperazine-1-carboxam-
ide (1a) as an inhibitor of GOT1 enzyme activity (Fig. 1).12
The synthesis of the urea-based compounds described below
was readily accomplished by addition of aryl- or alkyl-isocyanates
to 1-(aryl)piperazines or tryptamine in the presence of triethy-
lamine (Schemes 1 and 2). The structural integrity of the identified
screening hit was corroborated by re-synthesis of 1a, which was
demonstrated to possess a confirmed IC50 of 85 mM in an MDH cou-
pled GOT1 enzymatic assay.13
a
Compound
R
IC50 (mM)
1a
85
1b
1c
>100
>100
Evaluation of 1a in preliminary ADME PK studies indicated that
the parent compound of this series is stable in mouse plasma, with
>95% remaining following 4 h of incubation. In addition, 1a was
found to have reasonable bioavailability and exposure properties
in the mouse following oral administration of a 20 mg/kg dose
(t1/2 = 0.7 h, Cmax = 4133 ng/mL, AUC(0–24 hr) = 11,734 h * ng/mL).14
Based on these findings, it was rationalized that potency, rather
than inherent exposure or metabolic stability properties, would
limit sufficient target engagement by 1a to enable its use as an
in vivo probe for the evaluation of GOT1 as a potential PDAC drug
target. This limitation was the primary focus of an initial medicinal
chemistry effort aimed at optimizing potency and establishing a
structure-activity relationship for 1a.
1d
1e
41
46
1f
>100
Initial attempts to modify the structure of 1a were focused on
replacement of the indole ring, which could represent a liability
as a result of potential oxidization by cytochrome P450 enzymes
to oxindole and hydroxyindole metabolites.15 Representative
derivatives that consist of simple substitution on the indole ring
to replacement with alternative aryl, heteroaryl or alkyl groups
are shown in Table 1. A complete list of derivatives that were syn-
thesized and tested are shown in Supplementary Table 1. The
observed GOT1 inhibitory activity of these analogs, as determined
using the MDH coupled GOT1 enzymatic assay, indicate that
replacement of indole by phenyl, heteroaryl or simple alkyl group
substituents ablates activity (Table 1). These findings, combined
with the lack of observed activity for compounds 1j and 1k, indi-
cate that the nature and relative geometry of the hydrogen bond
donor of the indole ring system is essential for GOT1 inhibition
activity.
1g
1h
>100
>100
1i
1j
>100
>100
1k
>100
1l
>100
>100
>100
1m
1n
a
Data are reported as mean of n = 3 determinations.
Fig. 1. HTS GOT1 inhibitor hit 1a.
We next focused on evaluation of the unsubstituted phenyl
amide region of 1a which, as a result of the potential for release
of aniline containing metabolites, represents a genotoxicity liabil-
ity. Analogs containing mono- or di-substitution of the phenyl
group were synthesized (Scheme 1) and evaluated using the
MDH-coupled GOT1 enzymatic assay (Table 2 and Supplementary
Table 2). In addition, analogs in which phenyl is replaced by alkyl,
aryl, or heteroaryl substituents were similarly prepared and evalu-
ated. In general, introduction of electron-withdrawing groups on
the phenyl ring resulted in improved observed potency against
GOT1, as exemplified by the relative activities of 2b as compared
to 2c. Encouragingly, as demonstrated by the observed activities
of 2c, 2d, and 2g, ꢀ10-fold enhancement in potency was achieved
Scheme 1. Synthesis of 1a, 1g–1i, 1l–1w, 2a–2j, 2l, 2n–2p, 2t–2as, 2aw, 2ba, 4c and
4f. Reagents and conditions: (i) 1.1 eq. TEA, DCM, R.T., 15 m. 20–80%.
Scheme 2. Synthesis of 3a–j. Regents and conditions: (i) 1.1 eq. TEA, DCM, R.T., 30
min, 10–50%.
Please cite this article in press as: Anglin J., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.04.061

J. Anglin et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
3
Table 2
Table 2 (continued)
GOT1 inhibition activity of compounds 2a–t.
a
Compound
R
IC50 (mM)
2t
76
a
Data are reported as mean of n = 3 determinations.
a
Compound
R
IC50 (mM)
2a
>100
Table 3
GOT1 inhibition activity of compounds 3a–f.
2b
2c
2d
>100
8.2
16
a
2e
2f
>100
52
Compound
R
IC50 (mM)
3a
>100
36
3b
3c
3d
39
62
2g
14
2h
2i
>100
>100
3e
3f
32
34
2j
>100
2 k
2l
>100
43
a
Data are reported as mean of n = 3 determinations.
by introducing an electron withdrawing substituent at either the
meta- or para-position of the N-phenyl group. Indeed, 2c and 2d
represent molecules that are currently being evaluated in cell-
based metabolomics, cell-based selective toxicity and protein
co-crystallization experiments. As illustrated by the observed
potencies of 2o and 2p, bi-substitution failed to significantly
improve activity over the corresponding mono-substituted
derivative. ortho-Substitutions were not found to be tolerated,
demonstrated by 2h, 2i and 2j. Further, as demonstrated by the
observed activities of 2l and 2n, the wide variety of substituents
tolerated at the distal position of this series suggests that the
binding pocket may accommodate additional functionality at this
position, which may afford additional beneficial binding modes
and improved affinity.
2m
2n
>100
39
2o
2p
56
22
To evaluate if the piperazine linkage to the indole moiety is
essential for activity, a series of alternative linkage strategies,
including tryptamine derivatives, were designed, synthesized
(Scheme 2) and tested using the MDH coupled GOT1 enzymatic
assay (Table 3, Supplementary Tables 3 and 4). While the unsubsti-
tuted phenyl derivative 3a showed no significant activity, inclusion
of electron-withdrawing groups on phenyl in the context of the
tryptamine core, again improved activity and resulted in the iden-
tification of active analogs within this context. Interestingly, in
contrast to the relative observed activities of 2c and 2o, the
2q
2r
>100
>100
2s
>100
Please cite this article in press as: Anglin J., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.04.061

4
J. Anglin et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
observed equipotent activities of 3b and 3e demonstrates that
ortho-substitution is tolerable within the tryptamine series.
Indeed, even phenyl substitution at the ortho position (3f) was well
tolerated, consistent with the existence of an expansive, hydropho-
bic binding pocket. Overall, the observed activities within the tryp-
tamine series strongly suggests that the linkage between the
indole and phenyl moieties is not constrained and that other link-
ing chemistries may further enhance GOT1 inhibitory and physic-
ochemical properties.
In summary, we have developed two novel series of pheny-
lurea-based compounds that inhibit GOT1 enzymatic activity with
potencies in the low micromolar IC50 range. Future directions will
include further exploration of the piperazine series at the meta or
para positions of the phenyl moiety, in order to explore the binding
pocket around this region in more detail, which may improve
observed potencies. In addition, optimization of the linkage
between the indole and phenyl moieties will be prioritized, given
the structural differences between tryptamine and indole-4-piper-
azine and the observed relative activities. Finally, effort will be
required to improve the overall metabolic stability of these com-
pound series, in order to mitigate potential Cmax-driven toxicities
that might be predicted for a metabolic target. Overall, these com-
pounds represent first-in-class non-covalent inhibitors of GOT1
and potent derivatives may serve as leads for the development of
potential therapeutics for the treatment of PDAC.
A. Supplementary data
Supplementary data (synthetic methodology, compound char-
acterization and additional data tables) associated with this article
can be found, in the online version, at https://doi.org/10.1016/j.
bmcl.2018.04.061.
References
1. Warburg O, Wind F, Negelein E. The metabolism of tumors in the body. The
Journal of General Physiology. 1927;8:519–530.
2. Warburg O. On the origin of cancer cells. Science (New York, N.Y.).
1956;123:309–314.
3. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell.
2011;144:646–674.
4. Vander Heiden MG, Cantley LC, Thompson CB. Understanding the Warburg
effect: the metabolic requirements of cell proliferation. Science.
2009;324:1029–1033.
5. Pavlova NN, Thompson CB. The emerging hallmarks of cancer metabolism. Cell
Metabolism.. 2016;23:27–47.
6. Vander Heiden MG. Targeting cancer metabolism: a therapeutic window opens.
Nat Rev Drug Disc. 2011;10:671–684.
7. Halbrook CJ, Lyssiotis CA. Employing metabolism to improve the diagnosis and
treatment of pancreatic cancer. Cancer Cell. 2017;31:5–19.
8. Chakrabarti G, Moore ZR, Luo X, et al. Targeting glutamine metabolism
sensitizes pancreatic cancer to PARP-driven metabolic catastrophe induced by
ss-lapachone. Cancer & Metabolism. 2015;3:12.
9. Rahib L, Smith BD, Aizenberg R, Rosenzweig AB, Fleshman JM, Matrisian LM.
Projecting cancer incidence and deaths to 2030: the unexpected burden of
thyroid, liver, and pancreas cancers in the United States. Cancer Research.
2014;74:2913–2921.
10. Bailey P, Chang DK, Nones K, et al. Genomic analyses identify molecular
subtypes of pancreatic cancer. Nature. 2016;531:47–52.
Acknowledgments
11. Son J, Lyssiotis CA, Ying H, et al. Glutamine supports pancreatic cancer growth
through a KRAS-regulated metabolic pathway. Nature. 2013;496:101–105.
12. GLOX-coupled GOT1 assay: GOT1 enzymatic activity was measured utilizing two
coupling reactions: GOT1/GLOX and GOT1/MDH1. In the first method, GOT1
reaction was coupled to GLOX and HRP. In this reaction, aspartate and alpha-
ketoglutarate are provided as substrates to GOT1 at KM concentrations (4 mM
and 0.5 mM, respectively), which are then converted to glutamate and
This work was supported by a Lustgarten Foundation grant to A.
C.K., L.C.C., L.L.L. C.A.L. was supported by a Pancreatic Cancer Action
Network/AACR Pathway to Leadership award (13-70-25-LYSS);
Dale F. Frey Award for Breakthrough Scientists from the Damon
Runyon Cancer Research Foundation (DFS-09-14); Junior Scholar
Award from The V Foundation for Cancer Research (V2016-009);
Kimmel Scholar Award from the Sidney Kimmel Foundation for
Cancer Research (SKF-16-005); and a 2017 AACR NextGen Grant
for Transformative Cancer Research (17-20-01-LYSS). This work
was partially supported by National Cancer Institute Grants
R01CA157490, R01CA188048 and a Pancreatic Cancer P01 Grant
(P01CA117969) to LCC and ACK; ACS Research Scholar Grant
RSG-13-298-01-TBG and NIH grant R01GM095567; to A.C.K. This
work was partially supported by a R35-CA197588 to LCC. C.A.L.
and A.C.K. are inventors on patents pertaining to Kras regulated
metabolic pathways, redox control pathways in pancreatic cancer,
and targeting GOT1 as a therapeutic approach. A.C.K. also holds a
patent on the autophagic control of iron metabolism and is on
the SAB and has ownership interests in Cornerstone Pharmaceuti-
cals and Vescor Therapeutics. LCC is a founder and part owner of
Petra Pharmaceuticals, a company developing cancer drugs. Petra
provides support to the laboratory of LCC for research unrelated
to this manuscript. L.C.C. owns equity in, receives compensation
from, and serves on the Board of Directors and Scientific Advisory
Board of Agios Pharmaceuticals. Agios Pharmaceuticals is identify-
ing metabolic pathways of cancer cells and developing drugs to
inhibit such enzymes in order to disrupt tumour cell growth and
survival.
oxaloacetate. The glutamate serves as
a substrate for GLOX which is
converted to alpha-ketoglutarate and H2O2. In the presence of HRP, H2O2
oxidizes non-fluorescent Amplex red producing Resorfin which is readily
measured by fluorescence (530 nm Ex/550 nm Em). The final concentration of
the components of the assay are as follows: HEPES buffer 100 mM, KCL 100
mM, GLOX 80 nM, HRP 0.05 U/mL, Amplex Red (15
lg/mL), GOT1 (ꢀ1 nM,
adjusted from each purification to produce 20 min linear reaction). The
reaction components are mixed except for GOT1, which is prepared in a
separate buffer; all buffers are kept on ice prior to dispensing. For measuring
the reaction in the HTS format, first the compounds are transferred to
microplates (384 or 1536) then the two components of the assay are added.
The plates are incubated for 20 min, which is during the linear part of the
reaction at 37 °C prior to measurement of fluorescence.
13. MDH1-coupled GOT1 assay: In the second method, the GOT1 reaction is coupled
to MDH1. MDH1 catalyzes the conversion of oxaloacetate to malate by using
NADH as a source of proton. Therefore, for monitoring the enzyme activity, loss
of NADH fluorescence (340 nm Ex/445 nm Em) is measured. For the GOT1/
MDH1 coupling, NADH and MDH1 are added at final concentrations of 250
and 0.3 g/mL, respectively. The other details are same as described for GOT1/
GLOX assay.
lM
l
14. Pharmacokinetic studies: Female CD1 mice (9 weeks old, Charles River) were
dosed by oral gavage at a dose of 20 mg/kg (5 mL/kg). At 0.5-, 1-, 4-, 8- and 24-h
time points (n = 3 for each time point) 250 mL of whole blood was collected via
retro-orbital sinus into a heparinized capillary tube, centrifuged at 12,000 rpm
for 3 min and the resulting plasma was stored frozen at À80 °C prior to
analysis. Mice were deprived of food overnight and had food returned after the
4-h bleed time point. Drug levels in plasma were determined using an AB Sciex
API 3000 LC/MS/MS system, based on observed analyte area.
15. Gillam EM, Notley LM, Cai H, De Voss JJ, Guengerich FP. Oxidation of indole by
cytochrome P450 enzymes. Biochemistry. 2000;39:13817–13824.
Please cite this article in press as: Anglin J., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.04.061