Synthesis of ertugliflozin from D-glucose

Article abstract of DOI:10.1016/j.tetlet.2019.03.009

A new synthesis of ertugliflozin from D-glucose in 7% overall yield is reported. The reaction sequence involves conversion of D-glucose to a fully protected open chain D-glucose aldehyde in 4 steps, further installation of the aglycon group by a Grignard

Full text of DOI:10.1016/j.tetlet.2019.03.009

Accepted Manuscript
Synthesis of Ertugliflozin from D-Glucose
Virginia V. Triantakonstanti, Thanos Andreou, Theoharis V. Koftis, John K.
Gallos
PII:
S0040-4039(19)30210-2
https://doi.org/10.1016/j.tetlet.2019.03.009
TETL 50650
DOI:
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
18 January 2019
27 February 2019
4 March 2019
Please cite this article as: Triantakonstanti, V.V., Andreou, T., Koftis, T.V., Gallos, J.K., Synthesis of Ertugliflozin
from D-Glucose, Tetrahedron Letters (2019), doi: https://doi.org/10.1016/j.tetlet.2019.03.009
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. 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.

Graphical Abstract
1

Synthesis of Ertugliflozin from D-Glucose†
‡
Virginia V. Triantakonstanti,a Thanos Andreou,b Theoharis V. Koftisb and John K. Gallosa
aDepartment of Chemistry, Aristotle University of Thessaloniki, Thessaloniki 54124, Greece
bPharmathen Industrial S.A., 9th km Thermi-Thessaloniki, Thessaloniki 57001, Greece
Keywords: ertugliflozin; D-glucose; Grignard reaction; benzylic oxidation; SGLT2 inhibitors
Abstract: A new synthesis of ertugliflozin from D-glucose in 7% overall yield is reported. The reaction sequence
involves conversion of D-glucose to a fully protected open chain D-glucose aldehyde in 4 steps, further
installation of the aglycon group by a Grignard reaction, introduction of the hydroxymethyl group by a
sequential oxidation/aldol reaction/Canizzarro reduction, and finally benzylic oxidation and global
deprotection.
Diabetes is a group of metabolic diseases caused by high blood glucose concentrations either
because insulin production in the body is inadequate (type 1 diabetes) or because body cells are not
responding well to insulin (type 2 diabetes), or both. Type 2 diabetes is the most common glucose
homeostatic disorder that accounts for 90-95% of all cases of diabetes, and is a major risk factor for
the development of both micro- and macro-vascular complications.1 Glycemic control and prevention
of over-complications is a major health objective for a significant portion of humanity. The previously
applied therapies focus on targeting insulin resistance and insulin secretion, carbohydrate digestion,
glucagon production, and providing exogenous insulin.2 Treatment with traditional glucose lowering
therapies, including metformin, sulphonylureas and insulin, are being phased out due to
gastrointestinal side effects, weight gain and hypoglycaemia. This has led to research into newer
alternatives such as selective sodium-glucose transporter-2 (SGLT2) inhibitors.3
† Respectfully dedicated to Prof. K. C. Nicolaou for his outstanding contribution to Organic Synthesis
‡ Corresponding author. E-mail address: igallos@chem.auth.gr
2

SGLT2 are proteins in the human body with high glucose capacity, and are responsible for 90% of
glucose reabsorption in the kidney. SGLT2 inhibitors block glucose resorption in the kidney thereby
increasing its excretion and lowering glucose blood levels. The mechanism of action of this new class
of drugs also offers further glucose control allowing for increased insulin sensitivity and glucose
uptake in muscle cells, reduced gluconeogenesis and improved first phase insulin release from beta
cells. Thus, hyperglycemia is controlled in a glucose-dependent but insulin independent manner.
Inspired by the natural product phlorizin, a number of O- and C-glucosides were developed, and
among them dapagliflozin and canagliflozin are now on the market as SGLT2 inhibitors (Fig. 1). In
addition, researchers at Pfizer disclosed a new class of such C-glucosides with a unique dioxa-
bicyclo[3.2.1]octane structure and among them ertugliflozin (1) represents one of the most potent
and selective SGLT2 inhibitors.4 Recently, the US Food and Drug Administration (FDA) approved
ertugliflozin (Steglatro) for the treatment of glycemic control in patients with type 2 diabetes as a
drug taken on its own as a fixed-dose, or in combination with metformin and sitagliptin, which are
both oral antihyperglycemic agents.
Ertugliflozin (1) was first prepared from D-glucose in 13 steps4a but in very low overall yield (0.3%). In
the crucial step, an advanced intermediate Weinreb amide was treated with the lithiated aryl aglycon
to give the final precursor, which was then converted to ertugliflozin by deprotection, unfortunately
together with its epimer which needed to be separated by HPLC. In a substantially improved
synthesis of ertugliflozin (1) by the same research group, also suitable for the synthesis of any
compound of this class,4b commercially available di-O-isopropylidene mannofuranose was converted
into the target molecule in 5 steps and 25% overall yield. This approach, however, was found to be
unsuitable for large scale preparations, mainly due to the lack of crystalline intermediates and the
requirement for low temperature reactions.
3

Figure 1. Structures of selected gliflozins
In two more recent approaches, D-glucose derivatives were used by Pfizer researchers as starting
materials. Firstly, persilylated D-gluconolactone was converted into ertugliflozin in several steps
involving inter alia Grignard addition, selective desilylation, oxidation and aldol-crossed-Cannizzaro
reactions.5 Although this process could be scaled up in very good overall yields, it was not
commercially viable. However, it enabled toxicological and clinical tests for the development of
ertugliflozin as a medicine. Finally, Pfizer laboratories developed a practical and commercially
acceptable process for the preparation of ertugliflozin in a 12-step sequence, starting from 2,3,4,6-
tetra-O-benzyl-D-glucose,6 which involved nucleophilic hydroxymethylation of a ketogluconamide
intermediate, and a highly efficient arylation of the protected diol thus obtained.
Recently, we reported a novel synthesis of ertuglifozin, starting from a known protected
hydroxymethyl L-erythrose derivative, which was easily accessible from L-arabinoze.7 Standard
manipulations afforded an aldehyde intermediate, which after aldol condensation with 1-(4-chloro-3-
(4-ethoxybenzyl)phenyl)ethanone led to formation of the carbon skeleton of the target molecule.
Dihydroxylation of the formed double bond and global deprotection gave ertugliflozin in eight steps
with 4% overall yield from known protected hydroxymethyl-L-erythrose. However, the undesired
diastereoselectivity in the dihydroxylation step represents a drawback of this approach, compared to
the existing synthesis methods. With the aim to improve the synthesis of ertugliflozin, we examined
a different approach to ertugliflozin (1) using much cheaper D-glucose as the starting material.
4

The designed new synthesis of ertugliflozin is depicted in Scheme 1. Ertugliflozin (1) and other
compounds of this class varying in the aglycon part (Ar), could be prepared by benzylic oxidation of 2,
followed by complete TFA deprotection. Compound 2 is the product of sequential aldol – Cannizzaro
reactions of aldehyde 3 with formaldehyde, which in turn could be derived from the known fully
protected D-glucose 48 by a Grignard addition and further standard selective protection/deprotection
manipulations. Compared to the first reported synthesis,4a where epimerization was observed in this
step, the Grignard addition introducing the aglycon part of the glycoside takes place on a more
reactive aldehydic carbonyl, expecting thus to avoid epimerization of the neighboring chiral carbon.
Scheme 1. Retrosynthetic analysis of ertugliflozin (1)
In our hands, known fully protected D-glucose 4 was prepared in 4 steps from D-glucose, using
modified literature procedures9 (Scheme 2). This conversion involved diethyl dithiacetal formation,
benzoylation of the primary hydroxyl group, protection of the secondary hydroxyl groups as
acetonides and finally deprotection of the aldehyde group (for details, see ESI). From a number of
successful deprotection procedures applied, the use of NBS in acetone gave the best yields. Grignard
addition of the aglycon part led to the formation of a ca. 1:1 inseparable diastereomeric mixture of
benzylic alcohols 6 and to our delight no epimerization was observed, as shown by HPLC/MS analysis,
where the two peaks appeared gave the correct molecular ion.
In the next step, we needed to protect the free benzylic hydroxyl group before hydrolysis of the
benzoic ester. This step proved laborious, presumably due to steric hindrance, and the introduction
of the TBS group using literature procedures10 was very slow affording low yields of 7. After
experimentation, we found that high yields of the silylated product 7 were obtained (92%) when
compound 6 was treated with TBS triflate (1.5 equiv.), di(tert-butylsilyl) ditriflate (0.25 equiv.), 2,6-
5

ο
lutidine (2 equiv.) and a catalytic amount of DMAP for 1 h at 0 C and then 0.5 h at rt. Although the
role of di(tert-butylsilyl) ditriflate is not obvious, it is possible that di(tert-butylsilyl) ditriflate acting as
a Lewis acid coordinates with an oxygen of the triflate group of TBS triflate, thus activating the later
in its reaction with the alcohol. Nevertheless, this is an interesting point, which needs further
detailed examination.
The benzoate was then cleaved with K2CO3 in MeOH and the resulting primary hydroxyl group was
oxidized to give aldehyde 3 in high yield, which upon treatment with formaldehyde and K2CO3 in
MeOH at reflux afforded the desired branched diol 9 in 72% yield, evidently via a sequential aldol
reaction/Cannizzaro reduction. This substrate has the required carbon skeleton with the correct
stereochemistry and what remained was achieved in three simple steps: the TBS group was removed
using TBAF and the product was subjected to benzylic oxidation with MnO2 to give ketone 10, which
was easily converted to ertugliflozin upon global deprotection with TFA, with physical and spectral
data identical to those reported in the literature.
Scheme 2. Synthesis of ertugliflozin (1)
6

In conclusion, we have achieved a new synthesis of ertugliflozin (1) from D-glucose in 7% overall yield
in 12 steps (16.2% from known compound 5 in 9 steps). Regarding the number of steps and yields,
this synthesis is comparable with above mentioned syntheses. However, our synthesis starts from
inexpensive D-glucose and, in addition, utilizes simple methods and cheap materials and reagents.
SUPPORTING INFORMATION
1
Experimental procedure and copies of H and 13C NMR spectra of compounds reported are available
in the Supporting Information.
ACKNOWLEDGMENT
This research has been co‐financed by the European Union and Greek national funds through the
Operational Program Competitiveness, Entrepreneurship and Innovation, under the call RESEARCH –
CREATE – INNOVATE (project code:T1EDK-01161).
REFERENCES
1. IDF Diabetes Atlas, 8th Ed., International Diabetes Federation: Montreal, Canada, 2017.
2. Stumvoll, M.; Goldstein, B. J.; Van Haeften, T. W. Lancet 2005, 365, 1333–1346.
3. (a) Neal, B.; Perkovic.; Matthews, D. R.; Mahaffey, K. W.; Fulcher, G.; Meininger, G.; Erondu,
N.; Desai, M.; Shaw, W.; Vercruysse, F.; Yee, J.; Deng, H.; de Zeeuw, D. Diabetes Obes. Metab.
2017, 19, 387–393; (b) Sonesson, C.; Johansson, P. A.; Johnsson, E.; Gause-Nilsson, I.
Cardiovasc. Diabetol. 2016, 15:37; (c) Tinahones, F. J.; Gallwitz, B.; Nordaby, M.; Götz, S.;
Maldonado-Lutomirsky, M.; Woerle, H. J.; Broedl, U. C. Diabetes Obes. Metab. 2017, 19, 266–
274. (d) Aguillón, A. R.; Mascarello, A.; Segretti, N. D.; de Azevedo, H. F. Z.; Guimaraes, C. R.
7

W.; Miranda, L. S. M.; Rodrigo O. M. A. de Souza, R. O. M. A. Org. Process Res. Dev. 2018, 22,
467–488.
4. (a) Mascitti, V.; Collman, B. M. WO10023594, 2010. (b) Mascitti, V.; Préville, C.
Stereoselective Org. Lett. 2010, 12, 2940–2943.
5. Bernhardson, D.; Brandt, T. A.; Hulford, C. A.; Lehner, R. S.; Preston, B. R.; Price, K.; Sagal, J.
F.; St. Pierre, M. J.; Thompson, P. H.; Thuma, B. Org. Process Res. Dev. 2014, 18, 57−65.
6. o les,, . renek, . . aron, . o, . . re ler, . T. uan, . ub , P.; Hansen, E. C.;
Jones, B. P.; Jones, K. N.; Ljubicic, T. A.; Makowski, T. W.; Mustakis, J.; Nelson, J. D.; Olivier,
M.; Peng, Z.; Perfect, H. H.; Place, D. W.; Ragan, J. A.; Salisbury, J. J.; Stanchina, C. L.;
Vanderplas, B. C.; Webster, M. E.; R. Matt Weekly, R. M. Org. Process Res. Dev. 2014, 18,
66−81.
7. Triantakonstanti, V. V.; Mountanea, O. G.; Papoulidou, K.-E.; Andreou, T.; Koftis, T. V.; Gallos
J. K. Tetrahedron 2018, 74, 5700–5708.
8. Herczegh, P.; Zsély, M.; Szilágyi, L.; Dinya Z.; Bognár, R. Tetrahedron 1989, 45, 5995−6002.
9. (a) Wong, M. Y. H.; Gray, G. R. Carbohydr. Res. 1980, 80, 87−98. (b) hadbhade, . raig, .
C.; Ng, C. J.; Odier, L.; Stevens, J. D. Carbohydr. Res. 2012, 353, 86−91. (c) Tsutsui, N.; Tanabe,
G.; Gotoh, G.; Morita, N.; Nomura, N.; Kita, A.; Sugiura, R.; Muraoka, O. Bioorg. Med. Chem.
2014, 22, 945−959. (d) Papadakis, P. J. Am. Chem. Soc. 1937, 59, 841-843.
10. (a) Bols, M.; Pedersen, C. M. Beilstein J. Org. Chem. 2017, 13, 93–105. (b) Patschinski, P.;
Zhang, C.; Zipse, H. J. Org. Chem. 2014, 79, 8348−8357. (c) Patschinski, P.; Zipse, H. Org. Lett.
2015, 17, 3318−3321. (d) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, 3rd Ed., Wiley, New York 1999. (e) Kociénski, P. J. Protecting Groups, Thieme,
Stuttgart 1994.
11. Fisher, E. Chem. Ber. 1894, 27, 673–679.
8

Highlights:



A new synthesis of ertugliflozin from D-glucose in 7% overall yield
A new method for silylation of a hindered alcohol was developed
Successful sequential aldol reaction/Cannizzaro reduction
9