Efficient synthesis of a galectin inhibitor clinical candidate (TD139) using a Payne rearrangement/azidation reaction cascade

Article abstract of DOI:10.1039/d0ob00910e

Selective galectin inhibitors are valuable research tools and could also be used as drug candidates. In that context, TD139, a thiodigalactoside galectin-3 inhibitor, is currently being evaluated clinically for the treatment of idiopathic pulmonary fibrosis. Herein, we describe a new strategy for the preparation of TD139. Starting from inexpensive levoglucosan, we used a rarely employed reaction cascade: Payne rearrangement/azidation process leading to 3-azido-galactopyranose. The latter intermediate was efficiently converted into TD139 in a few simple and practical steps.

Full text of DOI:10.1039/d0ob00910e

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Efficient synthesis of a galectin inhibitor clinical
candidate (TD139) using a Payne rearrangement/
azidation reaction cascade†
Cite this: DOI: 10.1039/d0ob00910e
Jacob St-Gelais, Vincent Denavit and Denis Giguère
*
Selective galectin inhibitors are valuable research tools and could also be used as drug candidates. In that
context, TD139, a thiodigalactoside galectin-3 inhibitor, is currently being evaluated clinically for the treat-
ment of idiopathic pulmonary fibrosis. Herein, we describe a new strategy for the preparation of TD139.
Starting from inexpensive levoglucosan, we used a rarely employed reaction cascade: Payne rearrange-
ment/azidation process leading to 3-azido-galactopyranose. The latter intermediate was efficiently con-
verted into TD139 in a few simple and practical steps.
Received 1st May 2020,
Accepted 7th May 2020
DOI: 10.1039/d0ob00910e
rsc.li/obc
feature of thiodigalactoside analogues is their resistance to gly-
cosidase in vivo. Combined with the long and difficult syn-
Introduction
Galectins are proteins that bind to galactoside residues and thetic route to access bis-(3-azido-3-deoxy-β-D-galactopyrano-
their natural ligands are any glycoconjugates with a non-redu- syl)-sulfane core,12 there is a major need for a fast and efficient
cing galactopyranoside terminus.1 Galectins have the ability to synthesis of TD139.
regulate numerous biological processes, including neoplastic
Fig. 1 shows, in retrosynthetic format, the known synthetic
transformation, tumor cell survival processes, angiogenesis, strategy leading to TD139 1, a 3,3′-bis-(4-aryltriazol-1-yl) thiodi-
tumor metastasis, and cell homeostasis.2 Among the 15 mam- galactoside. The C2 symmetry of 1 allowed for a general two-
malian members, galectins contain a conserved carbohydrate directional strategy to install triazole moieties through click
recognition domain (CRD) of about 135 amino acids. Sharing chemistry. Then, bis-(2,4,6-tri-O-acetyl-3-azido-3-deoxy-β-D-
a consensus amino acid sequence for the CRD makes it galactopyranosyl)-sulfane 2 is accessible from dimerization of
difficult to prepare selective galectin inhibitors for biological 3-azido-3-deoxy-galactopyranoside bromide 3. Currently, two
investigations. Over the years, the scientific community has distinct approaches were reported for the synthesis of 3-azido-
directed efforts in the synthesis of potent and optimized galec- 3-deoxy-galactose. The first one was initially reported by
tin inhibitors. These works have been summarized in reviews Lemieux13 and involved a nucleophilic displacement of a gulo-
by the groups of Pieters,3 Kiss,4 Nilsson,5 Mayo6 and ours.7 furanose triflate derivative 4. This approach began with expen-
Small molecular weight glycomimetics are rationally designed sive gulofuranose, but the latter compound can be accessed
to bind the CRD and inhibit the galectin target. As such, from inexpensive glucose diacetonide 5 in a five-step proto-
TD139 1 8 (Fig. 1) is a ditriazolylthiodigalactoside that was col.14 The second approach was described more recently by the
designed as one of the most potent antagonist of galectin-3.9 group of Nilsson and involved the introduction of the 3-azido-
Compound 1 showed a Kd of 14 nM to galectin-3, as deter- functionality via nucleophilic azidation at C-3 of a gulopyrano-
mined using a competitive fluorescence anisotropy assay.10 side triflate intermediate 6.15 This compound arises from C-3
Developed by Galecto Inc., TD139 passed a Ib/IIa phase clini- inversion and functionalization of galactose 7. Because of the
cal trial in idiopathic pulmonary fibrosis patients (IPF) and is instability of triflate intermediates, this method was later on
currently in a randomized, double-blind, placebo-controlled improved by using more stable imidazylate and tosylate inter-
phase IIb trial in subjects with IPF investigating its efficacy mediates.16 Nevertheless, a more convenient preparation of
and safety (see http://www.clinicaltrials.gov).11 A particular 3-azido-3-deoxy-galactose is much needed for enabling large
scale synthesis of thiodigalactoside galectin inhibitors and for
a rapid access to active pharmaceutical ingredients. We aimed
to develop a new synthetic route to 1,2,4,6-tetra-O-acetyl-3-
azido-3-deoxy-D-galactopyranose that could operate on large
scale and be initiated from inexpensive starting material. The
Département de Chimie, 1045 av. De la Médecine, Université Laval, GlycoNet,
Québec City, Qc, Canada G1V 0A6. E-mail: denis.giguere@chm.ulaval.ca
†Electronic supplementary information (ESI) available. CCDC 1956891. For ESI
and crystallographic data in CIF or other electronic format see DOI: 10.1039/
intended synthetic pathway was designed to develop a practical
d0ob00910e
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 1 Retrosynthetic analysis of TD139 1.
alternative to known reported methods.13,15,16 We explored the dro-β-D-gulopyranose 12, an ideal precursor for azidation reac-
possibility of generating crucial intermediate 3 via cleavage of tions. Efforts towards this end are presented in Table 1.
the 1,6-anhydro core from 1,6-anhydro-3-azido-3-deoxy-galacto- Accordingly, compound 11 was treated with sodium hydride in
pyranose derivative 8. The requisite azide moiety was intro- N,N-dimethylformamide and allowed to come to equilibrium
duced using
a rarely employed reaction cascade: Payne for 6 h prior addition of sodium azide followed by heating at
rearrangement/azidation process.17 Levoglucosan 9 is an ideal 120 °C for 24 h (entry 1). As propose by the group of Fraser-
starting material since the 1,6-anhydro core avoided the pre- Reid, in an aprotic solvent, the gulose form should predomi-
liminary protection of O-6 and anomeric positions, and can nate because of a possible chelation between a sodium cation,
easily lead to scalable 3-azido-3-deoxy-galactose derivatives via the O-3 alkoxide, the endocyclic oxygen, and the 1,6-anhydro
simple experimental protocols.
bridge.25 The 1H NMR of the crude reaction mixture after treat-
Cascade processes generate in situ a series of reactive inter- ment with sodium hydride revealed formation of compound
mediates that undergo consecutive transformations. They can 12 as major isomer (11/12 ≈ 1 : 9). Unexpectedly, we only iso-
offer cost saving in terms of reagents and solvents, as well as lated an unprecedented dimeric by-product 15 in 72% yield. In
time and effort. Cascade processes can drastically shorten a order to improve the azidation step, we added 5 equivalents of
synthetic route and thus reducing the number of manipula- ammonium chloride (entry 2) and compound 13 was isolated
tions. Because of their many advantages, these reactions have in 65% yield, along with 7% of compound 14. According to the
found applications in the synthesis of useful pharmaceuticals Fürst–Plattner rule, nucleophilic attack occurs on the C-3
and other fine chemicals.18
oxirane carbon of compound 12 (leading to 13) and on the C-4
Finally, this novel synthetic process could provide a rapid carbon of compound 11 (leading to 14).26 The obtained selecti-
access to 3-azido-3-deoxy-galactose analogues, crucial com- vity can be explain using steric and electronic reasons. Hence,
ponent to probe the active site of glycosyltransferases,19 as an incoming nucleophile suffered from 1,3-diaxial interaction
inhibitors of bacterial adhesins,20 and to generate novel ami- with the C-2 hydroxyl group for compound 11 and the partial
noglycoside antibiotics.21
positive charge at C-3 of the oxirane carbon was stabilized by
nearby acetal for compound 12.27 Further optimisation using
other additives (Na2SO4: entry 3; Na2HPO4: entry 4; NH4Br:
entry 5) failed, but lowering the temperature to 100 °C for
66 hours (entry 6) provided the desired 1,6-anhydro-3-azido-3-
Results and discussion
The synthesis of 1,6-anhydro-3-azido-3-deoxy-β-D-galactopyra- deoxy-β-D-galactopyranose 13 in 81% yield, along with 9% of
nose 13 from levoglucosan 9 is summarized in Scheme 1. compound 14 as inseparable mixture using standard flash
Monotoluenesulfonylation of levoglucosan 9 afforded known column chromatography. Finally, the latter reaction was even
1,6-anhydro-4-O-p-tolylsulfonyl-β-D-glucopyranose 10 22 and the perform on gram scale with no apparent formation of by-
latter compound was treated under basic condition to generate product 15.
1,6:3,4-dianhydro-β-D-galactopyranose 11 in quantitative
A proposed mechanism for the formation compound 15 is
yield.23 An epoxide migration (known as Payne rearrange- shown in Fig. 2 and presumably involves an intermolecular
ment)24 of compound 11 lead to the requisite 1,6:2,3-dianhy- epoxide opening followed by an intramolecular attack from the
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 1 Synthesis of 1,6-anhydro-3-azido-3-deoxy-β-D-galactopyranose 13 from levoglucosan 9.
Table 1 Synthesis of 1,6-anhydro-3-azido-3-deoxy-β-D-galactopyranose 13 from 1,6:3,4-dianhydro-β-D-galactopyranose 11
Yielda (%)
Entry
Conditions
13
14
15
1
2
3
4
NaN3, DMF, 120 °C, 24 h
—
65
—
44
51
81
—
7
—
11
13
9
72
—
50
30
—
—
NaN3, NH4Cl, DMF, 120 °C, 24 h
NaN3, Na2SO4, DMF, 120 °C, 24 h
NaN3, Na2HPO4, DMF, 120 °C, 24 h
NaN3, NH4Br, DMF, 120 °C, 24 h
NaN3, NH4Cl, DMF, 100 °C, 66 h
5
6b
a Yields refer to isolated pure products after flash column chromatography. b The reaction was perform on gram scale.
Scheme 2 Development of a green reaction sequence: epoxide for-
mation/Payne rearrangement/azidation, leading to 1,6-anhydro-3-
azido-3-deoxy-β-D-galactopyranose 13.
Fig. 2 Proposed mechanism for the formation of dimeric by-product
15 and X-ray derived ORTEP of compound 15 showing 50% thermal
ellipsoid probability, carbon (gray), oxygen (red), hydrogen (white).
epoxide formation/Payne rearrangement/azidation reaction
sequence.
Encouraged by the successful synthesis of 3-azido-3-deoxy-
cis-vicinal alkoxide. Also, an X-ray crystallographic analysis of β-D-galactopyranose derivative from levoglucosan, we main-
15 confirmed its dimeric nature unambiguously (see X-ray tained our efforts towards the preparation of TD139
derived ORTEP in Fig. 2).28
(Scheme 3). Thus, triethylsilyl triflate-catalyzed acetolysis of a
1
With the ongoing objective of shortening the number of mixture of 13 and 14 furnished a separable mixture of 3-azido-
steps to prepare valuable 3-azido-3-deoxy-galactose, we per- galactose 16 (77%) and 4-azido-glucose 17 (19%). Then, the
formed our key step starting from mono-tosylate 10 using galactosyl bromide 3 was slowly generated in 74% yield using
aqueous sodium hydroxide as base (Scheme 2).29 To our TiBr4 from intermediate 16. Alternatively, intermediate 3 was
delight, compound 13 and 14 were isolated in 79% yield (13/14 generated from compound 11. Accordingly, acetylation of the
= 4 : 1, ∼93% per step). Presumably, this green process avoided crude reaction mixture of the key step allowed the preparation
organic solvent and involved the formation of the 3,4-anhydro of a separable mixture of 18 and 8 (1 : 10) in 81% yield over 2
analogue followed by Payne rearrangement and azidation. To steps. With compound 8 in hand, a direct bromolysis with
the best of our knowledge, this is the first example of an TiBr4 afforded in 97% yield a mixture of galactosyl bromide 3
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 3 Rapid synthesis of TD139 1 from 1,6:3,4-dianhydro-β-D-galactopyranose 11.
and 3-azido-galactose 16 (3/16 = 3 : 2). Subsequently, we used a dimeric by-product 15. We are currently exploring the chem-
similar strategy to the group of Nilsson for the preparation of istry and biology of this novel C2-symmetrical compound.
compound 1.30 Briefly, a base promoted SN2 substitution of Finally, the strategy described herein allowed the preparation
galactosyl bromide 3 with triisopropylsilanethiol afforded of galactose derivatives functionalized at C-3 that could
triisopropylsilyl β-thio-galactoside 19 in 76% yield.31 The provide useful options for the synthesis of other galectin
dimeric nature of the thiodigalactoside core was achieved by inhibitors.
treating compounds 3 and 19 under tetrabutylammonium flu-
oride to generate bis-(2,4,6-tri-O-acetyl-3-azido-3-deoxy-β-D-
galactopyranosyl)-sulfane 2 in 75% yield. Finally, triazole
installation with known alkyne 20 preceded global de-
Conflicts of interest
protection, allowing the preparation of TD139 1 in 65% over 2
steps. The synthesis of compound 1 was described by the
group of Nilsson in 11 steps (6% global yield) starting from
known intermediate phenyl 4,6-O-benzylidene-1-thio-β-D-
glucopyranoside.9,12 Using the present strategy, compound 1
was prepared in only 8 steps and 21% yield from known tosy-
late 10. It is also important to point out that our strategy
avoided the use of triflate intermediates,9,12 unsuitable in
large scale synthesis of active pharmaceutical ingredients.
A related provisional patent has been filed with the title
“Synthesis of 3-azido-3-deoxy-D-galactopyranose” by Jacob
St-Gelais, Vincent Denavit and Denis Giguère (USPTO
62/861,476).
Acknowledgements
This work was supported by the Natural Sciences and
Engineering Research Council of Canada (NSERC) and
Université Laval. J. St-G. thanks the Fonds de Recherche du
Québec-Nature et Technologies for a postgraduate fellowship.
Conclusions
TD139 is a galectin-3 inhibitor and is currently being evaluated
clinically for the treatment of idiopathic pulmonary fibrosis.
An efficient synthesis of TD139 1 from levoglucosan was
described. To the best of our knowledge, this is the shortest
route to 3-azido-3-deoxy-galactose, a crucial intermediate for
the preparation of bis-(2,4,6-tri-O-acetyl-3-azido-3-deoxy-β-D-
galactopyranosyl)-sulfane. In that manner, we used a rarely
employed reaction cascade: Payne rearrangement/azidation,
leading to 1,6-anhydro-3-azido-3-deoxy-β-D-galactopyranose. In
the course of this study, we also isolated an unprecedented
References
1 (a) S. H. Barondes, V. Gastronovo, D. N. W. Cooper,
R. D. Cummings, K. Drickamer, T. Feizi, M. A. Gitt,
J. Hirabayashi, C. Hughes, K.-I. Kasai, H. Leffler, F.-T. Liu,
R. Lotan, A. M. Mercurio, M. Monsigny, S. Pillai, F. Poirier,
A. Raz, P. W. J. Rigby, J. M. Rini and J. L. Wang, Cell, 1994,
76, 597–598; (b) S. H. Barondes, D. N. W. Cooper, M. A. Gitt
and H. Leffler, J. Biol. Chem., 1994, 269, 20807–20810.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
2 (a) G. A. Rabinovich, M. Toscano, D. A. Jackson and 16 K. Peterson, A. Wymouth-Wilson and U. J. Nilsson,
G. Vasta, Curr. Opin. Struct. Biol., 2007, 17, 513–520; J. Carbohydr. Chem., 2015, 34, 490–499.
(b) F.-T. Liu and G. A. Rabinovich, Nat. Rev. Cancer, 2005, 5, 17 (a) T. Yoshino, Y. Nagata, E. Itoh, M. Hashimoto, T. Katoh
29–41; (c) S. Califice, V. Castronovo and F. van den Brûle,
Int. J. Oncol., 2004, 25, 983–992; (d) F.-T. Liu and
G. A. Rabinovich, Nat. Rev. Cancer, 2005, 5, 29–41;
and S. Terashima, Tetrahedron Lett., 1996, 37, 3475–3478;
(b) T. Yoshino, Y. Nagata, E. Itoh, M. Hashimoto, T. Katoh
and S. Terashima, Tetrahedron Lett., 1997, 53, 10239–10252.
(e) S. Califice, V. Castronovo and F. van den Brûle, 18 (a) E. A. Anderson, Org. Biomol. Chem., 2001, 9, 3997–4006;
Int. J. Oncol., 2004, 25, 983–992; (f) G. A. Rabinovich and
M. Toscano, Nat. Rev. Immunol., 2009, 9, 338–352; (g) S. Di
Lella, V. Sundblad, J. P. Cerliani, C. M. Guardia,
(b) K. C. Nicolaou and J. S. Chen, Chem. Soc. Rev., 2009, 38,
2993–3009; (c) K. C. Nicolaou, D. J. Edmonds and
P. G. Bulger, Angew. Chem., Int. Ed., 2006, 45, 7134–7186.
D. A. Estrin, G. R. Vasta and G. A. Rabinovich, Biochemistry, 19 (a) H. P. Nguyen, N. O. L. Seto, Y. Cai, E. K. Leinala,
2011, 50, 7842–7857; (h) G. A. Rabinovich, M. A. Toscano,
J. M. Ilarregui and N. Rubinstein, Glycoconjugate J., 2004,
19, 565–573.
3 R. J. Pieters, ChemBioChem, 2006, 7, 721–728.
4 L. Ingrassia, I. Camby, F. Lefranc, V. Mathieu,
P. Nshimyumukiza, F. Darro and R. Kiss, Curr. Med. Chem.,
2006, 13, 3513–3527.
5 (a) C. T. Öberg, H. Leffler and U. J. Nilsson, Chimia, 2011,
65, 18–23; (b) H. Leffler and U. J. Nilsson, Low-molecular
weight inhibitors of galectins, in ACS Symposium Series.
Galectins and disease implications for targeted therapeutics,
S. N. Borisova, M. M. Palcic and S. V. Evans, J. Biol. Chem.,
2003, 278, 49191–49195; (b) N. Soya, G. K. Shoemaker,
M. M. Palcic and J. Klassen, Glycobiology, 2009, 19, 1224–
1234; (c) S. Laferté, N. W. C. Chan, K. Sujino, T. L. Lowary
and M. M. Palcic, Eur. J. Biochem., 2000, 267, 4840–4849;
(d) J. A. L. M. van Dorst, J. M. Tikkanen, C. H. Krezdorn,
M. B. Streiff, E. G. Berger, J. A. van Kuik, J. P. Kamerling
and J. F. G. Vliegenthart, Eur. J. Biochem., 1996, 242, 674–
681; (e) T. L. Lowary and O. Hindsgaul, Carbohydr. Res.,
1994, 251, 33–67; (f) T. L. Lowary, S. J. Swiedler and
O. Hindsgaul, Carbohydr. Res., 1994, 256, 257–273.
ed. A. A. Klyosov and P. G. Traber, 2012, vol. 115, 20 (a) S. Haataja, P. Verma, O. Fu, A. C. Papageorgiou,
pp. 47–59.
S. Poysti, R. J. Pieters, U. J. Nilsson and J. Finne, Chem. –
Eur. J., 2018, 24, 1905–1912; (b) J. Ohlsson, A. Larsson,
S. Haataja, J. Alajaaski, P. Stenlund, J. S. Pinkner,
S. J. Hultgren, J. Kihlberg and U. J. Nilsson, Org. Biomol.
Chem., 2005, 3, 886–900.
6 K. H. Mayo, From Carbohydrate to peptidomimetic
inhibitors of galectins, in ACS Symposium Series. Galectins
and disease implications for targeted therapeutics, ed.
A. A. Klyosov and P. G. Traber, 2012, vol. 115, pp. 61–77.
7 V. Denavit, D. Lainé, T. Temblay, J. St-Gelais and 21 (a) J. Wang, J. Li, H.-N. Chen, H. Chang, C. T. Tanifum,
D. Giguère, Trends Glycosci. Glycotechnol., 2018, 30, SE21–
SE40.
8 TD139 is also known as GB0139.
H.-H. Liu, P. G. Czyryca and C.-W. T. Chang, J. Med. Chem.,
2005, 48, 6271–6285; (b) R. Albert, K. Dax and A. E. Stutz,
Carbohydr. Res., 1984, 132, 162–167.
9 T. Delaine, P. Collins, A. MacKinnon, G. Sharma, 22 B. T. Grindley and R. Thangarasa, Carbohydr. Res., 1988,
J. Stegmayr, V. Rajput, S. Mandal, I. Cumpstey, A. Larumbe, 172, 311–318.
B. A. Salameh, B. Kahl-Knutsson, H. van Hattum, M. van 23 M. Cerny, I. Buban and J. Pacak, Collect. Czech. Chem.
Scherpenzeel, R. J. Pieters, T. Sethi, H. Schambye, Commun., 1963, 28, 1569–1578.
S. Oredsson, H. Leffler, H. Blanchard and U. J. Nilsson, 24 (a) G. B. Payne, J. Org. Chem., 1962, 27, 3819–3822;
ChemBioChem, 2016, 17, 1759–1770.
10 (a) P. Sörme, B. Kahl-nutsson, U. Wellmar, U. J. Nilsson
(b) C. H. Behrens, S. Y. Ko, B. K. Sharpless and F. J. Walker,
J. Org. Chem., 1985, 50, 5687–5696.
and H. Leffler, Methods Enzymol., 2003, 362, 504–512; 25 A. Mubarak and B. Fraser-Reid, J. Org. Chem., 1982, 47,
(b) P. Sörme, B. Kahl-nutsson, U. Wellmar, 4265–4268.
B.-G. Magnusson, H. Leffler and U. J. Nilsson, Methods 26 A. Fürst and P. A. Plattner, Helv. Chim. Acta, 1949, 32, 275–
Enzymol., 2003, 363, 157–169; (c) P. Sörme, B. Kahl- 283.
nutsson, M. Huflejt, U. J. Nilsson and H. Leffler, Anal. 27 P. Crotti, V. Di Bussolo, L. Favero, F. Macchia and
Biochem., 2004, 334, 36–47. M. Pineschi, Tetrahedron, 2002, 58, 6069–6091.
11 A. Girard and J. L. Magnani, Trends Glycosci. Glycotechnol., 28 CCDC 1956891 contains the supplementary crystallo-
2018, 30, SE211–SE220. graphic data for compound 15.†
12 U. J. Nilsson, H. Leffler, N. Henderson, T. Sethi and 29 M. Dzoganova, M. Cerny, M. Budesinsky, M. Dracinsky and
A. Mackinnon, WO2014067986A1, 2014.
T. Trnka, Collect. Czech. Chem. Commun., 2006, 71, 1497–
1515.
30 K. Peterson, R. Kumar, O. Stenström, P. Verma,
P. R. Verma, M. Hakansson, B. Kahl-Knutsson,
F. Zetterberg, H. Leffler, M. Akke, D. T. Logan and
U. J. Nilsson, J. Med. Chem., 2018, 61, 1164–1175.
13 R. U. Lemieux, R. Sweda, E. Paszkiewicz-Hnatiw and
U. Spohr, Carbohydr. Res., 1990, 205, C12–C17.
14 A. Hoffmann-Röder and M. Johannes, Chem. Commun.,
2011, 47, 9903–9905.
15 C. T. Oberg, A.-L. Noresson, T. Delaine, A. Larumbe,
J. Tejler, H. von Wachenfeldt and U. J. Nilsson, Carbohydr. 31 S. Mandal and U. J. Nilsson, Org. Biomol. Chem., 2014, 12,
Res., 2009, 344, 1282–1284.
4816–4819.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.