Facile synthesis of the glucosylceramide synthase inhibitor GZ667161

Article abstract of DOI:10.1016/j.tetlet.2020.152352

GZ667161 (GZ-161) is a quinuclidine-based small molecule inhibitor of the lysosomal enzyme glucosylceramide synthase. It represents an important tool molecule for studying the contribution of glycosphingolipids to disease pathology in lysosomal storage di

Full text of DOI:10.1016/j.tetlet.2020.152352

Journal Pre-proofs
Facile synthesis of the glucosylceramide synthase inhibitor GZ667161
Elizabeth D. Hewlett, Edward Melenski, Frederick V. Qiu, Hiu T. Leung,
Marlene Jacobson, Feng Qiu, Magid Abou-Gharbia, Wayne Childers
PII:
S0040-4039(20)30830-3
DOI:
Reference:
https://doi.org/10.1016/j.tetlet.2020.152352
TETL 152352
To appear in:
Tetrahedron Letters
Received Date:
Revised Date:
Accepted Date:
12 March 2020
2 August 2020
7 August 2020
Please cite this article as: Hewlett, E.D., Melenski, E., Qiu, F.V., Leung, H.T., Jacobson, M., Qiu, F., Abou-
Gharbia, M., Childers, W., Facile synthesis of the glucosylceramide synthase inhibitor GZ667161, Tetrahedron
Letters (2020), doi: https://doi.org/10.1016/j.tetlet.2020.152352
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Elsevier Ltd. All rights reserved.

2
Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Facile synthesis of the glucosylceramide synthase Leave this area blank for abstract info.
inhibitor GZ667161
Elizabeth D. Hewlett,a,† Edward Melenski,a,† Frederick V. Qiu,b Hiu T. Leung,a Marlene Jacobson,a
Feng Qiu,c Magid Abou-Gharbiaa and Wayne Childersa,*

1
Tetrahedron Letters
Facile synthesis of the glucosylceramide synthase inhibitor GZ667161
Elizabeth D. Hewlett,a,† Edward Melenski,a,† Frederick V. Qiu,b Hiu T. Leung,a Marlene Jacobson,a Feng
Qiu,c Magid Abou-Gharbiaa and Wayne Childersa,*
aMouder Center for Drug Discovery Research, Temple University School of Pharmacy, 3307 North Broad Street, Philadelphia, Pennsylvania, 19140. USA
bDepartment of Computer Science, Princeton University, 35 Olden Street, Princeton, NJ, 08540. USA
cQualytics LLC, 1979 Stout Drive, Warminster, PA, 18974. USA
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
GZ667161 (GZ-161) is a quinuclidine-based small molecule inhibitor of the lysosomal
enzyme glucosylceramide synthase. It represents an important tool molecule for studying
the contribution of glycosphingolipids to disease pathology in lysosomal storage disorders
such as Gaucher disease and GBA1 Parkinson’s disease. GZ667161 is not commercially
available. The published synthesis involves 6 steps and proceeds in 18% overall reported
yield. As part of a drug discovery program targeting Type 2 Gaucher disease we required
quantities of GZ667161 that would support animal studies. To facilitate the project, we
devised and executed an efficient 4-step convergent synthesis of the compound.
Available online
Keywords:
GZ667161
GZ-161
Glucosylceramide synthase
Gaucher disease
GBA mutations
2009 Elsevier Ltd. All rights reserved.
Introduction
Gaucher disease is the most common liposomal storage
disorder, with a prevalence of 1 in 57,000 births.1 Patients
possess deleterious mutations in the GBA1 gene encoding the
lysosomal enzyme glucocerebrosidase (Gcase). Deficient or loss
of Gcase activity results in accumulation of its substrates
glucosylceramide and glucosylsphingosine in diseased tissues.
Type 2 Gaucher disease is a variant that affects the neurological
system, causing severe and irreversible brain damage during the
first years of life. Additionally, accumulation of Gcase substrates
in the brain stem may play a pathological role in the
neurodegeneration associated with Parkinson’s disease.2 With no
drugs approved for their treatment, Type 2 Gaucher disease and
GBA-Parkinson’s disease are highly unmet medical needs.
Figure 1. Structure of GZ667161 (compound 1).4
mouse model, GZ667161 reduced brain glucosylceramide and
glucosylsphingosine levels and significantly extended lifespan.4
One of the current avenues under investigation for the
treatment of Gaucher disease is preventing the synthesis of Gcase
substrates. To accomplish this, the enzyme that produces
glucosylceramide from ceramide, glucosylceramide synthase
(GCS), has been targeted for inhibition. The quinuclidine analog
GZ667161 (GZ-161, 1, Fig. 1)3 is an inhibitor of GCS. It has
become an important tool molecule that is used to study the
contribution of glycosphingolipids to disease pathology. The
compound acts to reduce substrate flux into biosynthetic
sphingolipid pathways. In a neuronopathic Gaucher disease
In
a
Gaucher-related synucleinopathy mouse model,
GZ667161 also decreased -synuclein and improved cognition in
the mice.3 Alpha-synuclein has also been implicated in the
disease pathology of Parkinson’s disease.5 Recently, 1 has been
shown to slow the accumulation of hippocampal aggregates of α-
synuclein, ubiquitin, and tau, and improved the associated
memory deficits.3
As part of a drug discovery program targeting Type 2 Gaucher
disease we required quantities of 1 to support in vitro and in vivo
studies. The lack of a commercial supply of the molecule
prompted us to synthesize it. The only published synthesis of
GZ667161 appears in the patent literature.6,7 However, in our
hands, the published procedure failed to produce adequate
quantities of 1 to meet our needs. We therefore designed and
_________
* Corresponding author.
E-mail address: wayne.childers@temple.edu
† Contributed equally to this manuscript.

2
Scheme 1. Published synthesis of GZ667161 (compound 1).7
in which the reaction mixture was first concentrated and the
residuals were re-dissolved in 1N aq. HCl to generate the water
soluble quinuclidium salt. The aqueous solution was then
washed with diethyl ether and ethyl acetate, and then
concentrated. The resulting product was desalted by dissolving
the residuals in isopropanol, followed by filtration. The filtrate
was concentrated to give the hydrochloride salt of the desired
product as a yellow powder.
successfully carried out a modified synthesis of 1. Our results
are presented herein.
Results and Discussion
The reported synthesis of 1 is shown in Scheme 1.6
Treatment of methyl 3-bromobenzoate (2) with two equivalents
of methyl magnesium iodide affords alcohol 3 in quantitative
yield without purification. Alcohol 3 is then converted to
chloroacetamide 4 via a Ritter reaction with chloroacetonitrile.
Deacetylation of chloroacetamide 4 gives amine 5 (isolated as the
hydrochloride salt). Oxalyl chloride is used to convert amine 5 to
isocyanate 6 which is immediately converted to carbamate 7
through treatment with 3-quinuclidinol. Carbamate 7 is finally
converted to 1 via a Suzuki reaction employing 4-fluorophenyl-
boronic acid.
Unfortunately, we were unable to obtain the required
quantities of the final target 1, using the Suzuki chemistry
methods described in the published method. Four different
reaction conditions were explored (Table 1), however, we could
not isolate the desired compound in appreciable quantities.
Table 1. Suzuki reaction conditionsa explored for the attempted
Catalyst
Additives
Solvent
In our hands the Grignard addition and chloroacetamide
formation/Ritter reaction proceeded smoothly and did not require
purification. However, our ability to obtain amine 5 was severely
hindered by difficulties in isolating the compound. Trituration as
described in the literature afforded the intermediate, but, in a low
yield that could not be improved (27% as opposed to 43% stated
in the published synthesis). Purification of the subsequent
carbamate 7 also proved to be problematic. The published
synthesis cited the use of normal phase chromatography on silica
gel with a mixture of chloroform and methanol with 2N
ammonium hydroxide as a modifier. In our hands, these
conditions (and similar conditions substituting chloroform with
dichloromethane) provided the quaternary salt 8 instead of the
desired free base 7 (Scheme 2). We were finally able to obtain
abPd(PPh3)4
(10 mol %)
Pd(PPh3)4
(10 mol %)
Pd2(dba)3
(10 mol %)
Pd2(dba)3
(10 mol %)
Na2CO3
(2 eq.)
K2CO3
(2 eq.)
K3PO4, P(cyclo-C6H11)3
(2 eq.; 3 eq.)
KF, P(t-Bu)3
(1 eq.; 3 eq.)
Water/DME
Water/DME
Water/toluene
THF
conversion of compound 7 to GZ667161 (1).
aReactions were heated under microwave irradiation at 150 oC for 25 minutes
bPreviously published conditions.6
We hypothesized that the extremely nucleophilic quinuclidine
nitrogen might be interfering with the reaction by associating
with the palladium catalyst in preference to the boronic acid
reagent (in a Buchwald/Hartwig fashion). In addition, the small
amounts of crude 1 that were obtained (based on LC/MS) could
not be purified adequately due to the co-elution of left over 4-
fluorophenylboronic acid and the product under both normal
phase and reversed-phase conditions. Reducing the amount of
the boronic acid used had no effect on the ratio of product to 4-
flurophenyl boronic acid at the completion of the reaction. We
therefore designed an efficient four-step convergent synthesis
that circumvented this problem by avoiding the use of
Scheme 2. Attempts to purify intermediate 7 via chromatography
resulting in the formation of quaternary by-product 8.
the hydrochloride salt of intermediate 7 in 74% yield through an
ion exchange procedure

3
S
Scheme 3. Modified 4-step convergent synthesis of GZ667161 (1).
the palladium catalyst in the presence of the nucleophilic
quinuclidine nitrogen. Our modified synthesis is depicted in
Scheme 3. The key to the modified synthetic sequence was to
relocate the Suzuki step to the beginning of the synthesis and
reserve insertion of the troublesome quinuclidine moiety until the
end. Due to the previously describe purification challenges with
chromatography on silica gel. We did not require the use of
microwave irradiation to obtain high yields of the product. The
carbonylbenzyloxy (Cbz) group was then removed via catalytic
hydrogenation at one atmosphere over 10% palladium on carbon
to afford biphenyl amine 11 in 82% yield.
amine
5,
commercially
available
benzyl
was
(2-(3-
employed.
Difficulties obtaining an isocyanate derivate similar to
intermediate 6 prompted the use of a convergent synthesis. 3-
Quinuclidinol was first treated with diphosgene to provide
quinuclidin-3-yl carbonochloridate hydrochloride (prepared as
described in the literature and isolated as a white solid).8
Treatment of biphenyl amine 11 with the carbonochloridate in
bromophenyl)propan-2-yl)carbamate
9
Treatment of carbamate 9 with 4-fluorophenylboronic acid under
typical Suzuki chemistry conditions afforded the desired 4-
fluorophenyl derivative 10 in 84% yield following normal phase
Table 2. H- 13C- and 19F-NMR chemical shifts for compound 1.
1
o
pyridine at 90 C provided 1 as a white solid in 44% yield.
Optimum purification conditions were found to be a basic amine
column using a gradient of ethyl acetate in hexanes. Attempts to
improve the yield of the final step of the synthesis proved
unsuccessful. Nevertheless, we were able to reproducibly obtain
1 in 30% overall yield, which is higher than the yield reported for
the published synthesis (18%).
Chemical Shifts (ppm)
Atom
1
2
3
4
5/8
6/7
9
10
11
12
13
14/15
16
17
18
19
20
21
1H
13C
1H (NH)
19F
1
The H NMR spectra and ESIMS were consistent with that
previously published.7 Because of the broadness in the aliphatic
3.02 4.42
4.43
1.84
1.42 1.32 0.90
2.74 2.62 2.54
55.8
70.5
25.8
1
1H NMR region and some protons in the aromatic rings, the H
NMR was run at different temperatures and in different solvents
to enhance the resolution of the peaks, with the best results
24.7 19.7
47.4 46.4
o
obtained in DMSO-d6 at 27 C. 2D-COSY and Heteronuclear
1
Multiple Quantum Coherence (HMQC) were run to obtain H-1H
155.1
correlation and 1H-13C one-bond correlation to assist in the
proton and carbon assignments. Finally, 2D-Heteronuclear
3.54
55.0
30.3 29.8
149.5
123.6
139.3
124.8
129.1
124.4
1
Multiple Bond Correlation (HMBC) was run to obtain H-13C
1.60 1.56
7.56
multi-bond correlations (two to three bonds) which provided the
information needed to unequivocally assign most of the proton
and carbon resonances.
Furthermore, covalent bonding
7.45 dt
7.39 t
7.35
information between the functional groups could be deduced
1
from the multi-bond H-13C correlations, which unambiguously
22
137.4
129.1
116.2
162.3
confirmed the desired structure. The proton, carbon and fluorine
assignments are given in Table 2 and details are provided in the
ESI.
23/27
24/26
25
7.64 dd
7.30 t
28
-115.8
Conclusion

4
A robust, reproducible 4-step convergent synthesis of the
brain penetrant GCS inhibitor GZ667161 is presented that
overcomes issues experienced with the previously reported
method and provides the desired material in less steps and higher
overall yield. The keys to the modified synthetic sequence were
to move the Suzuki reaction to the beginning of the reaction
sequence and incorporating the troublesome quinuclidine moiety
in the last step. We have subsequently used this sequence to
produce multi-gram quantities of GZ667161. The availability of
this improved, efficient synthesis should facilitate the use of this
important pharmacological tool molecule by researchers in the
field of lysosomal storage disorders.
Supplementary Material
Supplementary data to this article (experimental procedures and
spectroscopic data) can be found online at:
References and notes
1. P.J. Meikle, J.J. Hopwild, W.F. Clauge, W.F. Carery, JAMA 281
(1999) 249.
2. E. Sidransky, G. Lopez, Lancet Neurol. 11 (2012) 986.
3. S.P. Sardi, C. Veil, J. Clarke, C. M. Treleaven, A. M. Richards, H.
Park, M.A. Olszewski, J.C. Dodge, J. Marshall, E. Makino, B.
Want, RL. Sidman, S.H. Cheng, L.S. Shihabuddin, PNAS 114
(2017) 2699.
Declaration of Competing Interest
4. M.A. Cabrera-Salazar, M. DeRiso, S.D. Bercury, L. Li, J.T.
Lydon, W. Weber, N. Pande, M.A. Cromwell, D. Copeland, J.
Leonard, S.H. Cheng, R.K. Scheule, PlosOne 7 (2012) e43310.
5. L. Stefanis, Cold Spring Harb. Perspect. Med. 4 (2012) a009399.
6. E. Bourque, C. Celatka, B. Hirth. M. Metz, Z. Zhao, R. Skerlj, Y.
Xiang, K. Hancisics, J. Marshall, S. Cheng, R. Scheule, M.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to
influence the work reported in this paper.
Acknowledgments
Cabrera-Salazar,
A
Good,, World Patent Application
WO2012129084A2, 2012; Chem. Abstr. 2012, CAN157:549491.
7. S.H. Cheng, L. Shihabuddin, S.P. Sardi, World Patent Application
WO2016145046A1, 2016; Chem. Abstr. 2016, CAN165:406076.
8. M. Turconi, M. Nicola, M.G. Quintino, L. Maiocchi, R.
Micheletti, E. Giraldo, A. Donetti, J Med Chem. 33 (1990) 2101.
The authors would like to thank Dr. Furong Sun and the team at
the Mass Spectrometry Lab, School of Chemical Sciences,
University of Illinois at Urbana-Champaign for providing the
high resolution mass spectrometry data. The authors also thank
Qualytics LLC, Warminster, PA, for providing detailed
spectroscopic data.