Process development of sotagliflozin, a dual inhibitor of sodium- Glucose cotransporter-1/2 for the treatment of diabetes

Glucose Cotransporter-1/2 for the Treatment of Diabetes

Article

Process Development of Sotagliﬂozin, a Dual Inhibitor of Sodium−

Matthew M. Zhao,* Haiming Zhang, Shinya Iimura, Mark S. Bednarz, Qiu-Ling Song, Ngiap-Kie Lim,

Jie Yan, Wenxue Wu,* Kuangchu Dai, Xiaodong Gu, and Youchu Wang

Cite This: https://dx.doi.org/10.1021/acs.oprd.0c00359

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: The development of an eﬃcient manufacturing process for sotagliﬂozin (LX4211), a dual inhibitor of sodium−

glucose cotransporter-1/2 (SGLT-1/2) for the treatment of diabetes, is described. Sotagliﬂozin features ﬁve contiguous chiral

centers on the carbohydrate core ﬂanked by a thioether group and a biaryl moiety. Three chiral centers are obtained from the

starting material L-xylose, while the other two were established (or modiﬁed) via three highly stereoselective transformations: Luche

reduction (dr: 97/3), dynamic kinetic resolution of anomeric hemiacetal (dr: 95/5), and Lewis acid-promoted thiolation (dr: 1000/

). Global deprotection of the resulting penultimate intermediate with catalytic sodium methoxide followed by recrystallization

furnishes sotagliﬂozin. The longest linear sequence consists of 10 steps from L-xylose with an overall yield of 40%. This process has

been performed on multi-hundred kilogram batches to satisfy the drug substance development demands.

KEYWORDS: SGLT, diabetes, stereoselective, Luche reduction, DKR, thiolation

INTRODUCTION

Inhibitors of sodium-dependent glucose cotransporters (SGLT)

for the treatment of diabetes mellitus have attracted

conditions gave tetraol 8 (1/1 anomers). The latter was

peracetylated to tetraacetate 9 and then treated sequentially with

HBr/HOAc and NaSMe to aﬀord triacetate 10R. Finally,

■

deprotection by treatment with ammonia in MeOH furnished

sotagliﬂozin. This synthesis is lengthy (15 steps) mainly due to

the protection/deprotection steps and adjustment of the

oxidation state. Additional drawbacks include the stability

concerns of aldehyde 3, cryogenic conditions required for

preparing and handling aryl lithium 4-Li, and the use of

corrosive HBr.

considerable attention. Several SGLT2-selective inhibitors,

which lower blood glucose levels by inhibiting reabsorption of

glucose in the kidneys, have received regulatory approval.

Clinical studies of Lexicon’s drug candidate sotagliﬂozin

LX4211, Figure 1 ), a dual inhibitor of SGLT1 (expressed

A more concise medicinal chemistry route (11 steps) was

later devised starting with the conversion of L-glucose to a bis-

acetonide, followed by selective deprotection of the six-

membered ring acetonide to give 11 (Scheme 2). Cleavage of

the vicinal diol with NaIO , followed by oxidation of the

Figure 1. Sotagliﬂozin (LX4211).

resulting aldehyde with bromine in methanol aﬀorded methyl

ester 12. Treatment of the latter with morpholine and catalytic

primarily in the intestine) and SGLT2 (expressed primarily in

the kidneys), have shown beneﬁts of reducing glucose

absorption in the GI tract for patients with renal function

n-BuLi gave morpholine amide 13, an economical substitute for

the Weinreb amide. Coupling of 13 with aryl lithium 4-Li gave

ketone 6 which was converted to tetraacetate 9 following the

same chemistry in the ﬁrst medicinal chemistry route. TMSOTf

promoted thiolation with thiourea, followed by treatment with

MeI and Hunig’s base aﬀorded triacetate 10R. Finally,

deprotection with sodium methoxide in MeOH furnished

sotagliﬂozin.

impairment.

Sotagliﬂozin features ﬁve contiguous chiral centers on the

carbohydrate core ﬂanked by a thioether group and a biaryl

moiety. The initial medicinal chemistry synthetic route started

with protection of L-xylose as a bis-acetonide, followed by

selective deprotection of the six-membered ring acetonide to

give monoacetonide 1 (Scheme 1). After the manipulation of

protecting groups, the primary alcohol was oxidized to aldehyde

Received: August 3, 2020

which was coupled with in situ-prepared aryl lithium 4-Li to

give benzylic alcohol 5. Because of poor diastereoselectivity, 5

was oxidized to ketone 6 and then predominantly reduced to the

desired diastereomer 7S (dr: 9/1) via Luche reduction.

Deprotection of the acetonide and ring expansion under acidic

XXXX American Chemical Society

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development

Article

Scheme 1. First Medicinal Chemistry Route to Sotagliﬂozin

Scheme 2. Second Medicinal Chemistry Route to Sotagliﬂozin

While the second medicinal route is more concise and gives a

higher yield (∼30 vs ∼12% overall), it also has several signiﬁcant

drawbacks. First, L-glucose is expensive ($1430/25 g, Sigma-

Aldrich) and not available in bulk quantities. Sodium periodate

is also expensive and poor in atom economy. Usage of hazardous

elemental bromine is undesirable. Furthermore, the free hydroxy

group on amide 13 consumes an equivalent of valuable aryl

lithium 4-Li. Therefore, a hybrid process chemistry route was

designed to circumvent most of these issues (Scheme 3).

Using L-xylose as the starting material obviated the need for

corresponding aryl Grignard reagent (4-MgX) should circum-

vent the need for cryogenic conditions. Stereoselective Luche

reduction of ketone 6 and Lewis acid-promoted thiolation of

tetraacetate 9 using thiourea were retained. Isolation of several

crystalline intermediates should facilitate impurity purging

without resorting to column chromatography.

RESULTS AND DISCUSSION

■

Preparation of Morpholine Amide 13. The synthesis of

morpholine amide 13 started with the global protection of L-

xylose to bis-acetonide 15, followed by selective deprotection of

costly L-glucose and NaIO . Substituting 4-Li with the

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development

Article

Scheme 3. Process Chemistry Synthetic Plan for Sotagliﬂozin

the six-membered ring acetonide to give monoacetonide 1

sensitivity to acidic conditions. The lack of a chromophore also

hampered the reaction optimization. The commonly used

bleach/TEMPO (cat)/KBr (cat) protocol gave unreprodu-

(Scheme 4). Oxidation of the exposed primary alcohol to

carboxylic acid 14, followed by amidation gave morpholine

amide 13.

cible results (20−70% yield). TEMPO-catalyzed oxidation with

PhI(OAc) in acetonitrile/water was ineﬀective. Trichloro-

cyanuric acid with catalytic TEMPO gave a cleaner reaction,

but removal of the cyanuric acid byproduct was problematic.

Scheme 4. Preparation of Carboxylic Acid 14

RuCl -catalyzed oxidation with bleach gave only 57% yield due

to competitive oxidation of the secondary alcohol. Finally,

oxidation with sodium chlorite (NaClO ) and catalytic TEMPO

and NaClO aﬀorded promising results. Solutions of NaClO₂

and NaClO/K HPO were prepared separately to minimize the

generation of chlorine dioxide (explosive at high concen-

trations). A small portion of each solution was added to a

phosphate-buﬀered (pH ∼5) aqueous acetonitrile of 1 to initiate

the reaction, and the remainders were then charged slowly to

minimize the accumulation of oxidants in the reactor. The

reactor headspace was swept with a gentle nitrogen ﬂow to

prevent chlorine/chlorine oxides from reaching high concen-

trations. The product (14) was extracted into 2-MeTHF after

acidiﬁcation to pH ∼2. The crude product solution was not very

stable due to cleavage of the acetonide catalyzed by the innate

carboxylic acid. To mitigate the risk, morpholine (used for the

next step) was added as a stabilizer. To our delight, the

morpholine salt of 14 crystallized almost instantly. This

discovery was utilized for isolating high purity 14 in 75% overall

yield (3-steps from L-xylose, 400 kg).

For the bis-acetonide formation step, CuSO (5 equiv) was

substituted with environmentally benign MgSO (2 equiv). A

strong acid (H SO cat.) was more eﬀective than milder

phosphoric acid, achieving complete conversion in ∼20 h at 20−

5 °C. After quenching the reaction with concentrated aqueous

ammonia, the inorganic precipitate was removed by ﬁltration,

and the ﬁltrate was concentrated and codistilled with water

until residual acetone <0.5% w/w) to give an aqueous solution

The conversion of carboxylic acid 14 to morpholine amide 13

using conventional coupling reagents was surprisingly challeng-

ing. With EDC/HOBT or TBTU, the yields ranged from 30 to

60%. The purity was also unsatisfactory even after puriﬁcation

by silica gel chromatography. Another major challenge was the

removal of reagent-related byproducts from highly water-soluble

13 (250 mg/mL). Therefore, alternative amidation methods

that do not require stoichiometric coupling agents were

investigated.

(

of bis-acetonide 15. For the selective deprotection of the six-

membered ring acetonide, it is crucial to control the pH,

temperature, and reaction time to minimize overhydrolysis. The

optimal pH (∼2) was readily achieved and maintained with 0.1

equiv of H PO . The reaction mixture was stirred at 20−25 °C

until bis-acetonide 15 decreased to ∼2% [gas chromatography

GC)]. Pushing the reaction further proved counterproductive

due to overhydrolysis. To obviate the multiple tetrahydrofuran

THF) extractions of highly water-soluble product (1), the

reaction mixture was neutralized with K HPO and carried

(

Yamamoto

reported that 3,5-bis(triﬂuoromethyl)-

phenylboronic acid was an eﬀective catalyst for the amidation

of carboxylic acids (Scheme 5). However, its high cost and

availability on scale prompted us to investigate cheaper boric

(

forward to the next step.

Oxidation of the primary alcohol (1) to carboxylic acid 14 was

challenging due to the presence of a secondary alcohol and its

acid as an alternative. Gratifyingly, reﬂuxing a mixture of

carboxylic acid 14, morpholine, and 20 mol % boric acid in

toluene with azeotropic removal of water aﬀorded complete

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development

Article

Scheme 5. Amidation of 14 to Morpholine Amide 13

using NaBH and AlCl in THF was developed aﬀording

4 3

biaryl iodide 4 in 90−95% yield with >99.5% purity (250 kg

scale).

Preparation of Ketone 6. For the synthesis of ketone 6,

coupling of morpholine amide 13 with the aryl Grignard (4-

MgX) was used instead of aryl lithium (4-Li) to avoid the

required cryogenic conditions. The preparation of aryl Grignard

by treatment of aryl iodide 4 with i-PrMgCl in THF was facile

even at −30 °C (<0.5 h) but was conducted more economically

at −10 °C in the plant. To avoid consuming an equivalent of

valuable aryl Grignard, the hydroxy group in 13 was

deprotonated with t-BuMgCl (1.1 equiv) without any

epimerization of the α-chiral center.

The coupling reaction of the aryl Grignard and morpholine

amide 13 was facile at −10 to −20 °C aﬀording ketone 6 without

noticeable bis-addition or epimerization. However, signiﬁcant

yield variability (70−90%) was encountered without apparent

changes in the reaction proﬁle [high-performance liquid

chromatography (HPLC)]. The root cause was traced to the

formation of non-UV active i-propyl ketone 21a due to

inadvertent overcharge of i-PrMgCl. Therefore, i-PrMgCl was

slightly undercharged initially, and a second charge (if

necessary) was performed based on the amount of unreacted

biaryl iodide 4. This protocol avoided sampling/titration of

highly air-/moisture-sensitive i-PrMgCl while still compensated

for residual moisture in the reaction system (Scheme 7).

conversion to morpholine amide 13. As a bonus, boric acid

could be easily removed by ﬁltration. Excess morpholine could

be removed by codistillation with toluene to minimize yield loss

in the mother liquor to give 13 in excellent yield (90%)

(

toluene/n-heptane). On production scale, the codistillation

was deemed too time consuming and morpholine was removed

by aqueous acetic acid washes instead. The product was

crystallized from DCM/n-heptane in 90% yield (600 kg scale).

Preparation of Biaryl Iodide 4. The synthesis of biaryl

iodide 4 was based on the chemistry of the bromide analogue.

Treatment of 2-chloro-5-iodobenzoic acid (17) with oxalyl

chloride and catalytic dimethylformamide (DMF) followed by

AlCl -promoted Friedel−Crafts acylation of ethoxybenzene

The reaction was initially quenched with aq NH Cl and

gave biaryl ketone 18 (Scheme 6). The unusually large variations

acidiﬁed with aq HCl to achieve a clean phase separation as well

as to remove the morpholine byproduct. The latter was shown to

catalyze the epimerization of the α-chiral center of the ketone to

Scheme 6. Preparation of Aryl Iodide 4

1b (via an enamine intermediate) during processing. To

mitigate the risk of premature cleavage of the acetonide due to

the large pH swing during pH adjustment with HCl, the reaction

mixture was quenched with 18% aqueous citric acid (∼4 vol)

instead. However, magnesium citrate sometimes precipitated

which would not dissolve even at 40 °C. Fortunately, the

induction period for the precipitation could be prolonged to >24

h by lowering the workup temperature to 0 °C. The product was

crystallized from EtOAc/n-heptane in 85−90% yield and with

99% purity, and its absolute conﬁgurations were conﬁrmed by

single-crystal X-ray crystallography data (Figure 2).

Conversion of Ketone 6 to Tetraacetate 9. Conversion

of ketone 6 to tetraacetate 9 entailed stereoselective Luche

reduction (dr: 97/3), deprotection of the acetonide/con-

comitant ring expansion to tetraol 8 (1/1 anomers), followed

by DKR (dynamic kinetic resolution) via stereoselective

acetylation (dr: 96/4) (Scheme 8).

in regioselectivity (8/1 to 100/1) were traced to the moisture

levels in DCM (Table 1). By minimizing moisture in the

reaction system, consistently high regioselectivity (∼100/1) was

achieved to aﬀord 18 in 85% isolated yield and with 99% purity

The screened multiple reagents (NaBH(OAc) , Zn(BH ) ,

(

425 kg scale).

4 2

BH ·DEA, pinacolborane, 9-BBN etc.) gave poor results.

Exhaustive reduction of biaryl ketone 18 was accomplished by

Interestingly, catecholborane aﬀorded the undesired diastereo-

treatment with Et SiH and BF ·Et O in acetonitrile aﬀording

mer 7R exclusively (Table 2, entry 1). LiAlH(t-BuO) gave high

biaryl iodide 4 in 89% yield with >99.5% purity (450 kg scale).

stereoselectivity (dr: 92/8) at −65 °C, but its high cost and

cryogenic conditions rendered this option unattractive (entry

Curiously, the putative alcohol intermediate 20 was not

observed during the reaction, presumably due to its higher

reactivity. To further reduce the cost, an alternative process

⁾^.^T^r^aⁿ^s^f^e^r^h^y^d^r^o^g^eⁿ^a^tⁱ^oⁿ^wⁱ^t^h⁽⁽^R^,^R⁾^-^T^s^D^P^E^N^,⁽^C^p^*^I^r^C^l2⁾2^,

HCO K) was sluggish but gave high stereoselectivity (dr: 92/8)

(

entry 3). The best result was obtained using NaBH /CeCl ·

4 3

Table 1. Inﬂuence of Water on Regioselectivity

H O in methanol (Luche reduction) (dr: 90/10) (entry 4).

Surprisingly, decreasing the temperature was ineﬀective in

enhancing the stereoselectivity (entries 4−7). Initial solvent

screen (THF, ACN, THF/MeOH, and ACN/MeOH)

suggested that the polar solvent (MeOH) should give better

stereoselectivity. Unfortunately, highly polar solvent systems

regioisomer

moisture in DCM (w/w %)

18 (%)

19 (%)

.15

.07

.01

88.2

92.3

99.5

11.8

7.7

0.5

[

DMF, dimethyl sulfoxide (DMSO), formamide, ethylene

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development

Article

Scheme 7. Preparation of Ketone 6

Table 2. Stereoselective Reduction of Ketone 6

temp

entry

reagent

catecholborane

LiAlH(t-BuO)₃

solvent

THF

(°C)

(7S/7R)

−15

−65

0/100

92/8

(R,R)-TsDPEN, [Cp*IrCl ] ,

ACN/H O

HCO K

NaBH /CeCl ·7H O

MeOH

EtOH

IPA

−25

−65

90/10

91/9

92/8

94/6

95/5

90/10

NaBH /CeCl ·7H O

higher stereoselectivity (dr: 95/5) than MeOH (entry 8).

However, this trend did not hold for IPA (dr: 90/10) probably

because most of the CeCl ·7H O remained undissolved (entry

With this promising lead, the reaction conditions were further

optimized systematically. In contrast to methanol, decreasing

the temperature markedly increased the stereoselectivity (58/1

at −25 °C; 100/1 at −65 °C) (Table 3, entries 3−4). Reducing

Figure 2. Single-crystal X-ray crystallography of ketone 6. Note:

ORTEP drawing of molecule of the asymmetric unit with 30%

probability thermal ellipsoids.

the usage of CeCl from 1.0 to 0.5 equiv did not incur signiﬁcant

penalty in stereoselectivity (Table 3, entry 5). However, further

cuts did impart noticeable drops in selectivity (Table 3, entries

−8).

glycol, MeOH/water, THF/water, and ACN/water] aﬀorded

disappointing results. Surprisingly, EtOH oﬀered signiﬁcantly

With the low solubility of the starting material, a high solvent

volume (>50 vol) was required for its complete dissolution.

Scheme 8. Preparation of Tetraacetate 9R

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development

Article

Table 3. Optimization of Luche Reduction in EtOH

anomer (9R) encouraged us to introduce an isolation point for

better impurity control. To minimize the yield penalty of losing

the minor anomer (9S), conversion of the latter to the desired

anomer (9R) via a crystallization-driven DKR was attempted.

With mild Lewis acids (Ag CO and AgOTs), no conversion

entry CeCl ·7H O (equiv) temp (°C) EtOH (vol) 7S/7R 7R (%)

1.0

0.50

0.25

0.10

0.05

0.50

17/1

30/1

58/1

100/1

66/1

38/1

9.6/1

6.3/1

50/1

30/1

5.6

3.3

1.7

1.0

1.5

2.6

9.4

13.7

2.0

3.2

−25

−65

−25

−5

was observed in acetic anhydride at 15 °C. Stronger Lewis acid

(

BF ·Et O, 2 equiv) promoted the interconversion but in the

opposite direction. For example, the undesired isomer (9S)

increased from 40 to 80% in 4 h at 15 °C, indicating that the

desired anomer is thermodynamically unfavorable. Indeed,

under equilibration conditions (Ac O/AcOH/H SO = 10/4/

.1, 90 °C 24 h), 9S increased to 91%.

5.6

Another strategy pursued was to leverage the facile anomeric

equilibration of tetraol 8 to achieve DKR via stereoselective

acetylation. Initial screening of bases in acetonitrile with

catalytic DMAP gave disappointing results (dr: 1/1). Serendip-

itously, when DMAP was omitted, much higher selectivity was

Fortunately, the product was very soluble in the reaction

medium which permitted reduced solvent usage by allowing the

starting material to gradually dissolve as the reaction progressed.

It was important to keep the addition rate of NaBH below the

observed (dr: 8/1 with Et N), albeit at a slower reaction rate.

dissolution rate of ketone 6 as NaBH decomposed rapidly (<5

Solvent screening results showed that highly polar solvents

min) in the presence of CeCl in protic solvents. Thus, a solution

(

DMSO, DMF, DMAc, and NMP) gave the lowest selectivity

of NaBH (0.4−0.5 equiv) in 1 N aq NaOH (to stabilize

while chlorinated solvents (DCM, DCE, and chloroform) the

highest (dr: 15/1 in chloroform). The stereoselectivity also

NaBH ) was slowly (5 h) added to a suspension of 6 in 5−6

volumes of ethanol at −5 to −10 °C to aﬀord the product with

improved with increasing Et N charge, plateauing at 6 equiv.

high stereoselectivity (3% 7R) (Table 3, entry 10). The amount

Surprisingly, decreasing the temperature from 20 to 0 °C led to

decreased stereoselectivity (from 9/1 to 5/1). This unusual

behavior was attributed to the signiﬁcantly higher activation

(

0.1 vol) of 1 N NaOH was kept to a minimum to avoid

signiﬁcant deterioration in stereoselectivity. Residual boric acid

must be completely removed by washing with dilute NaOH in

the workup to prevent stalling of the acetylation of tetraol 8 due

to the formation of boric acid−tetraol complex 32 (Scheme

energy for the anomeric equilibration than that of

acetylation. Consequently, decreasing the temperature

imparts greater rate reduction of anomeric equilibration than

that of the acetylation, leading to a deﬁciency of anomer 8S and

thereby decreased selectivity. Conversely, lowering the concen-

tration of acetic anhydride should decrease the acetylation rate

without impacting the rate of anomeric equilibration. Indeed,

excellent stereoselectivity (dr: 25/1) was achieved by slowly

). The crude product solution in acetonitrile was carried

forward to the next step.

For the deprotection and concomitant ring expansion of diol

S to tetraol 8, the aqueous acetonitrile solution of 7S was

treated with 0.2 equiv of HCl at 75 °C for 2−3 h to give a 1/1

anomeric mixture of tetraol 8. Quenching the reaction with

aqueous K CO and extraction with MTBE led to signiﬁcant

adding acetic anhydride (5 h) to a mixture of tetraol 8 and Et N

(

6 equiv) in acetonitrile at 30−40 °C. The product (9R) was

emulsiﬁcation during the brine wash as well as noticeable

product degradation via the Cannizzaro reaction. Therefore, the

reaction mixture was simply saturated with NaCl, and the

resulting organic layer was carried forward to the next step after

azeotropic distillation with acetonitrile (until KF <1.0%).

crystallized from IPA in 80% overall yield (from ketone 6, 3-

steps) with 99% purity.

Stereoselective Thiolation of Tetraacetate 9 to

Triacetate 10R. Thiolation was achieved by treatment of the

tetraacetate with TMSOTf and thiourea followed by MeI and

Hunig’s base to give predominantly the desired triacetate 10R

(dr: 94/6, Scheme 10). The high stereoselectivity was attributed

to nucleophilic attack of thiourea from underneath the oxonium

ion 22 stabilized by the neighboring acetate group. The resulting

Peracetylation of tetraol 8 (Ac O, cat. DMAP) typically gave a

/1 to 3/1 anomeric mixture of tetraacetate 9. The stereo-

selectivity was considered inconsequential as both anomers were

converted to the desired product in the TMSOTf-catalyzed

thiolation. The discovery of a crystalline form of the major

Scheme 9. Eﬀect of Boric Acid in the Acetylation Step

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development

Article

Scheme 10. Stereoselective Thiolation of Tetraacetate 9

Scheme 11. Stereoselective Thiolation with BF ·Et O

isothiourea intermediate (23) proved quite stable under acidic

conditions and could be isolated as its tetraﬂuoroborate salt. For

the methylation stage, it is important to add MeI prior to

Hunig’s base to minimize the formation of disulﬁde dimer 25.

As TMSOTf was a main cost driver for the thiolation step,

(0.1 equiv) was required as the base was regenerated in MeOH.

Interestingly, no partial deprotection intermediates were

observed during the reaction. This was likely due to their higher

reactivity because of shuttling of NaOMe by neighboring

hydroxy groups. Adding water to the reaction mixture as an

antisolvent often gave various sotagliﬂozin hydrates that were

diﬃcult to ﬁlter and dry. Subsequently discovered anhydrous

form I sotagliﬂozin not only ﬁltered and dried rapidly but also

provided better impurity purging. It was consistently produced

from MeOH/water (>70/30). The wet cake of “form I”

sotagliﬂozin turned out to be a methanol solvate which readily

desolvated upon drying. Single-crystal X-ray crystallography

data conﬁrmed the absolute conﬁgurations of sotagliﬂozin as

well as the presence of methanol (Figure 3). Form I sotagliﬂozin

was converted to the thermodynamically most stable polymorph

more economical BF ·Et O was evaluated (Scheme 11). With

.5 equiv BF ·Et O, only 3% isothiourea 23 was observed after

3 2

4 h at rt in 1,4-dioxane. The reaction was much faster in DCM,

reaching 90% conversion under similar conditions. The

thiolation reached 99% conversion in 3 h at 40 °C by increasing

BF ·Et O to 2.5 equiv. As a bonus, the stereoselectivity was

much higher due to milder reaction conditions (1000/1).

Interestingly, the minor anomer of the tetraacetate (9S) was

essentially inert under milder reaction conditions. For the

methylation step, a notable impurity (∼4%) was identiﬁed as

guanidine 35 (HRMS). It could be converted to triacetate 10R

but only slowly. The conversion was greatly accelerated by

adding the cosolvent MeOH (2 vol).

(

form II) by recrystallization in MEK/n-heptane in 95% yield.

CONCLUSIONS

■

Although the reaction was generally very clean, the formation

of small amounts of side products derived from thiol 24 and

DCM prompted a search for an alternative solvent. Toluene

gave an intractable gel, while acetonitrile oﬀered a less clean

reaction. On the other hand, EtOAc and IPAc performed well,

aﬀording isothiourea 23 cleanly (99% conversion in 3−4 h, 55

An eﬃcient manufacturing process for Lexicon’s SGLT-1/2 dual

inhibitor sotagliﬂozin was developed. The synthesis starts with

protection of L-xylose as a bis-acetonide (Scheme 12). Selective

deprotection of the 6-membered ring acetonide, followed by

oxidation of the unveiled primary alcohol using sodium chlorite

with catalytic TEMPO and bleach gave carboxylic acid 14, which

was isolated as a crystalline morpholine salt. Boric acid-catalyzed

amidation of 14 with morpholine aﬀorded morpholine amide 13

(an economical substitute for the Weinreb amide) in 90% yield.

Coupling reaction of 13 with in situ-prepared aryl Grignard (4-

MgX) aﬀorded ketone 6 in a high yield (85−90%). The latter

C). EtOAc was selected for a more eﬃcient solvent exchange

during workup and isolation. The product was isolated in 95%

yield and >99% purity from IPA/water (300 kg scale).

Global Deprotection to Sotagliﬂozin. With penultimate

triacetate 10R in hand, a simple deprotection by treatment with

NaOMe in MeOH aﬀorded sotagliﬂozin. Only catalytic NaOMe

Organic Process Research & Development

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Article

(

95%) and purity (>99.5%). Global deprotection by treatment

with catalytic sodium methoxide in MeOH, followed by

recrystallization from MEK/n-heptane furnished sotagliﬂozin

drug substance (form II). The longest linear sequence consisted

of 10 chemical steps from L-xylose with ∼40% overall yield

(

average 91% per step). This process has been utilized for the

manufacturing of thousands of kilograms of sotagliﬂozin to

support the clinical studies for commercialization.

EXPERIMENTAL SECTION

■

General. All reagents and materials were used as received,

unless otherwise speciﬁed. Reactions were monitored by

reverse-phase HPLC, using a C18 or phenyl-hexyl column

with water/ACN or water/MeOH as a mobile phase and TFA as

a modiﬁer. Melting point information was acquired using a

melting-point apparatus or based on the peak temperature on a

diﬀerential scanning calorimeter (DSC). NMR spectra were

acquired in deuterated solvents on 400 MHz ( H) spectrom-

eters. Mass spectrometry data were obtained during LC−MS

analysis. Compound purity data were determined by reverse

phase HPLC and/or H NMR with an internal standard.

(

3 a S , 5 S , 6 R , 6 a S ) - 5 - ( H y d r o x y m e t h y l ) - 2 , 2 -

dimethyltetrahydrofuro[2,3-d][1,3]dioxol-6-ol (1). A mixture

of L-xylose (280.0 kg), acetone (2300 kg), and anhydrous

MgSO₄(505 kg, 2.25 equiv) was stirred for 30 min.

Concentrated H SO (53 kg, 0.3 equiv) was added at 20−25

C and the reaction mixture stirred until reaction completion

Figure 3. Single-crystal X-ray crystallography of the sotagliﬂozin

methanol solvate.

(∼24 h, residual L-xylose by HPLC with ELSD). The reaction

was quenched with conc. aq ammonia (84 kg, ∼0.7 equiv) until

the pH reached 7.5−8.5. The reaction mixture was ﬁltered and

the ﬁlter cake washed with acetone (970 kg). The combined

ﬁltrate was concentrated to a low volume under reduced

pressure (0.08 MPa) at < 45 °C. Water (1680 kg) was added and

the mixture was concentrated again until residual acetone was

≤0.5 wt %. The aqueous solution of bis-acetonide 15 was used

for the next step. An analytical sample was obtained by

was reduced to 7S via a highly stereoselective Luche reduction

(

dr: 97/3) in ethanol. Cleavage of the acetonide and the

concomitant ring expansion by treatment with aqueous acid

aﬀorded tetraol 8 as a 1/1 anomeric mixture. DKR of tetraol 8

via stereoselective acetylation (dr: 96/4) aﬀorded high purity

(

>99%) tetraacetate 9R in 80% overall yield from ketone 6 (3-

steps). By taking advantage of neighboring group assistance, a

concentrating the acetone solution to an oil. H NMR

highly stereoselective (dr: 1000/1) Lewis acid (BF ·Et O)-

promoted thiolation of tetraacetate 9R, followed by methylation

with methyl iodide aﬀorded triacetate 10R in excellent yield

(DMSO-d ): δ ppm 1.23 (s, 3H), 1.25 (s, 3H), 1.386 (s, 3H),

1.390 (s, 3H), 2.82 (d, J = 13.5 Hz, 1H), 3.91 (d, J = 1.0 Hz, 1H),

4.06 (dd, J = 13.5, 2.4 Hz, 1H), 4.27 (d, J = 2.2 Hz, 1H), 4.46 (d,

Scheme 12. Sotagliﬂozin Manufacturing Process Summary

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development

Article

J = 3.8 Hz, 1H), 5.87 (d, J = 3.8 Hz, 1H); C NMR (100 MHz,

equiv), boric acid (26 kg, 0.2 equiv), and toluene (2210 kg) was

reﬂuxed for 14−18 h with a Dean−Stark trap to remove water.

After conﬁrming completion of the reaction, the reaction

mixture was diluted with toluene (580 kg), stirred at 30−50 °C,

and ﬁltered to remove solids. The ﬁlter cake was washed with

toluene (230 kg), and water (1390 kg) was added to the

combined ﬁltrate. The mixture was acidiﬁed with AcOH (180

kg) to pH = 6−7 and the organic layer was washed with dilute

AcOH (23 kg + 580 kg water) (aq pH = 6−7). The aqueous

layer was extracted four times with DCM (4870, 4870, 3450, and

2300 kg) and the combined organic layer was concentrated

under reduced pressure below 45 °C until residual DCM was

≤0.5%. The distillation residue was heated to 70−90 °C, diluted

with heptane (3944 kg), cooled to 0−5 °C (6 h), and agitated for

10 h. The solid was ﬁltered and dried under reduced pressure at

20−30 °C (24 h) to aﬀord morpholine amide 13 as a yellow

solid (520 kg, 99% wt, 90% yield). mp 137.2 °C (DSC peak);

DMSO-d ): δ ppm 18.74, 26.03, 26.47, 28.80, 59.42, 71.10,

2.65, 84.06, 96.80, 104.50, 110.62.

The aq solution of bis-acetonide 15 was acidiﬁed to pH 1.8−

.2 with H PO (16.6 kg, 0.1 equiv) and stirred at 20−25 °C

until residual bis-acetonide 15 was ≤2.5% (∼24 h; product 1

and residual 15 were monitored by GC; over-hydrolysis product

L-xylose by HPLC with ELSD). The reaction mixture was

neutralized with K HPO ·3H O (50.8 kg) to pH 6.0−7.0 and

used for the next step. An analytical sample of diol 1

hygroscopic and low-melting solid) was obtained by extraction

with EtOAc, passing through a silica gel plug, concentrating to

an oil, and crystallization from 2/1 n-heptane/EtOAc. GC−MS

M + H] calcd, 191.1; found, 191.2; H NMR (DMSO-d ): δ

ppm 1.22 (s, 3H), 1.37 (s, 3H), 3.51 (dd, J = 11.1, 5.8 Hz, 1H),

.61 (dd, J = 11.1, 5.1 Hz, 1H), 3.93−4.00 (m, 1H), 3.96 (s, 1H),

.36 (d, J = 3.8 Hz, 1H), 4.86 (br s, 2H), 5.79 (d, J = 3.5 Hz, 1H);

(

[

C NMR (DMSO-d ): δ ppm 26.19, 26.73, 59.01, 73.59, 81.42,

LC−MS [M + H] calcd, 274.1; found, 274.1; H NMR

5.19, 104.40, 110.44.

(CDCl

): δ ppm 1.29 (s, 3H), 1.45 (s, 3H), 3.49−3.38 (m, 2H),

Morpholin-4-ium (3aS,5R,6S,6aS)-6-Hydroxy-2,2-

3.80−3.55 (m, 6H), 4.42 (d, J = 2.2 Hz, 1H), 4.55 (d, J = 3.7 Hz,

dimethyltetrahydrofuro[2,3-d][1,3]dioxole-5-carboxylate

14). NaClO (353 kg) was dissolved in water (810 kg) at 20−25

C. Separately, 10% NaClO (540 kg) was added to a mixture of

K HPO (437 kg) in water (345 kg) at 20−25 °C with stirring to

give a solution. K HPO (111 kg), KH PO (300 kg),

1H), 4.57 (d, J = 2.5 Hz, 1H), 5.06 (br s, 1H), 5.96 (d, J = 3.7 Hz,

(

1H); C NMR (CDCl

66.49, 66.83, 75.59, 76.12, 83.64, 105.41, 111.77, 167.63; Anal.

Calcd for C12 : C, 52.74; H, 7.01; N, 5.13. Found: C,

): δ ppm 25.94, 26.86, 42.17, 46.30,

19NO

52.71; H, 7.04; N, 5.14.

acetonitrile (840 kg), and TEMPO (3.5 kg) were added to the

aqueous solution of diol 1 obtained above. A portion of the

(2-Chloro-5-iodophenyl)(4-ethoxyphenyl)methanone

(18). To a mixture of 2-chloro-5-iodobenzoic acid (17) (426 kg,

1.0 equiv), DMF (1.5 kg, 1.4 mol %), and methylene chloride

(2820 kg) was slowly charged oxalyl chloride (230 kg, 1.20

equiv) at 15−25 °C. The mixture was aged at 20−30 °C for 2 h,

concentrated to 500−600 L, and then diluted with methylene

chloride (620 kg) to give a solution of acyl chloride. The latter

was transferred into a mixture of ethoxybenzene (212 kg, 1.16

NaClO solution (227 kg) and the NaClO solution (60 kg)

above were charged over 1 h at 15−25 °C. A gentle nitrogen

sweep of the reactor headspace (>3 m /h) was maintained

throughout the reaction to prevent accumulation of chlorine

oxide in the headspace. A second portion of the NaClO solution

(

305 kg) and NaClO solution (42 kg) were charged over ∼2 h at

5−25 °C. The remainder of the NaClO solution (620 kg) was

equiv) and AlCl (247 kg, 1.24 equiv) in methylene chloride

added over ∼3 h at 15−25 °C. After 1−2 h of aging, a portion of

the NaClO solution (480 kg) was slowly (6−8 h) charged.

Additional TEMPO (1.8 kg) was added followed by slow

addition (6−8 h) of the remaining NaClO solution (560 kg).

The reaction mixture was agitated at 20−25 °C until reaction

completion (5−8 h) and then quenched with Na SO (115 kg)

(1610 kg) at 5−15 °C. The reaction mixture was aged at 5−15

°C until reaction completion (2 h) and then quenched with 2 N

aq HCl (1394 kg) at <30 °C. The organic layer was separated

and washed sequentially with 2 N HCl (1170 kg), 10% sodium

bicarbonate (2127 kg), and brine (1080 kg). The washed

organic layer was concentrated to a low volume, diluted with

ethanol (2655 kg) and water (425 kg), heated to 70−75 °C to

dissolve the solids, then slowly cooled to 15−20 °C, and aged for

16 h to crystallize the product. The suspension was ﬁltered and

the ﬁlter cake was dried under reduced pressure at 40−50 °C to

in portions at 5−25 °C. The inorganic precipitate was ﬁltered oﬀ

and the ﬁlter cake washed with water (100 kg). The pH of the

combined ﬁltrate was adjusted to 7−8 with 30% NaOH (∼280

kg) and then washed with 2-MeTHF twice (100 kg × 2) to

remove TEMPO-related neutral impurities. The aqueous layer

was acidiﬁed with H PO (600 kg) to pH = 2.0−2.2 and

aﬀord 511 kg of biaryl ketone 18 (88% yield, 99.9% purity by

HPLC). mp 101−103 °C; H NMR (CDCl ): δ ppm 1.44 (t, J =

extracted with 2-MeTHF (1400 kg × 5). The combined organic

layer was concentrated to 4000−5000 L under reduced pressure

below 25 °C. Morpholine (198 kg) was charged in portions at

7.1 Hz, 3H), 4.10 (q, J = 7.1 Hz, 2H), 6.92 (d, J = 9.1 Hz, 2H),

7.17 (d, J = 8.6 Hz, 1H), 7.66 (d, J = 2.1 Hz, 1H), 7.70 (dd, J =

8.4, 2.1 Hz, 2H), 7.76 (d, J = 8.8 Hz, 2H); C NMR (CDCl ): δ

15−30 °C and the suspension aged at 20−25 °C for 3−4 h,

ppm 14.7, 64.0, 91.4, 114.5, 128.6, 131.0, 131.6, 132.6, 137.3,

139.6, 141.0, 163.8, 191.9; Anal. Calcd for C H ClIO : C,

slowly (5 h) cooled from −5 to 0 °C, and aged for 6−8 h. The

solids were collected by ﬁltration and dried under reduced

pressure (0.08 MPa) at 15−30 °C for 12−16 h to aﬀord the

morpholine salt of the carboxylic acid 14 as a white solid (398 kg,

46.60; H, 3.13. Found: C, 46.80; H, 3.28.

1-Chloro-2-(4-ethoxybenzyl)-4-iodobenzene (4). Option

(a) reduction with Et SiH/BF ·OEt . To a mixture of biaryl

5% wt, 73.3% overall yield from L-xylose). mp 166 °C (DSC

ketone 18 (450 kg) and triethylsilane (455 kg) in acetonitrile

(1450 kg) was charged BF ·OEt (66 kg) at <30 °C and the

peak); H NMR (D O): δ ppm 1.25 (s, 1H), 1.40 (s, 1H), 3.18

(

m, 2H), 3.84 (m, 2H), 4.30 (d, J = 3.2 Hz, 1H), 4.53 (d, J = 3.2

mixture was aged for 0.5 h. Additional BF ·OEt (331 kg) was

3 2

Hz, 1H), 4.58 (d, J = 3.6 Hz, 1H), 5.96 (d, J = 3.6 Hz, 1H); C

slowly added below 30 °C and the mixture stirred at 20−30 °C

until reaction completion (2 h). The reaction mixture was

concentrated to 300−400 L (−0.09 MPa, <55 °C), diluted with

water (2000 kg), cooled to 0−5 °C, aged for 2 h, and ﬁltered.

The wet cake was dissolved in acetonitrile (650 kg) at 45−50 °C,

cooled to 0−5 °C, aged for 3−4 h, and ﬁltered. The ﬁlter cake

NMR (D O): δ 25.1, 25.6, 43.1, 63.6, 75.0, 81.7, 84.2, 104.6,

12.5, 174.5.

(3aS,5R,6S,6aS)-6-Hydroxy-2,2-dimethyltetrahydrofuro-

2,3-d][1,3]dioxol-5-yl)(morpholino)methanone (13). A mix-

ture of morpholine salt of 14 (650 kg), morpholine (260 kg, 1.3

(

[

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development

Article

was dried under reduced pressure (−0.08 MPa) at 45−55 °C to

aﬀord 384 kg of biaryl iodide 4 (89% yield, 99.9% HPLC purity).

Option (b) reduction with AlCl /NaBH . To precooled THF

87% yield). mp 121.6 °C (DSC peak); LC−MS [M + H] calcd,

433.1; found, 433.1; H NMR (CDCl ): δ ppm 1.36 (s, 3H),

1.41 (t, J = 7.0 Hz, 3H), 1.53 (s, 3H), 2.98 (d, J = 4.6 Hz, 1H),

4.01 (q, J = 6.8 Hz, 3H), 4.07 (d, J = 12 Hz, 1H), 4.11 (d, J = 12

Hz, 1H), 4.54 (dd, J = 3.9, 3.2 Hz, 1H), 4.57 (d, J = 3.8 Hz, 1H),

5.2 (d, J = 2.8 Hz, 1H), 6.05 (d, J = 3.5 Hz, 1H), 6.84 (d, J = 8.6

Hz, 2H), 7.10 (d, J = 8.6 Hz, 2H), 7.48 (d, J = 8.3 Hz, 1H), 7.80

(d, J = 2.0 Hz, 1H), 7.86 (dd, J = 8.2, 2.2 Hz, 1H); C NMR

(CDCl ): δ ppm 14.84, 26.16, 26.89, 38.3, 63.34, 76.42, 82.09,

84.22, 105.31, 112.15, 114.67, 114.67, 128.3, 129.89, 129.95,

129.95, 130.22, 131.31, 133.92, 140.01, 140.52, 157.67, 195.84.

Anal. Calcd for C H ClO : C, 63.81; H, 5.82; Cl, 8.19. Found:

(

444 kg, 0−5 °C) was added AlCl (129 kg, 1.5 equiv) in

portions below 35 °C (highly exothermic!). The mixture was

stirred at 20−25 °C until all the solid dissolved (2 h) and then

cooled to 0−5 °C. NaBH (29.4 kg, 1.2 equiv) was added and

the mixture aged for 1 h at 0−5 °C before warming to 50 °C. A

solution of ketone 18 (250 kg, 1.0 equiv) in THF (250 kg) was

slowly added (5 h) and the reaction mixture aged at 65−70 °C

until reaction completion (12 h). The reaction mixture was

cooled to 15−20 °C and then slowly quenched into precooled

(

5−10 °C) 2 N HCl (1000 kg) below 25 °C (caution: gas

C, 64.17; H, 5.77; Cl, 8.43.

evolution!). After aging for 0.5 h, MTBE (401 kg) was added

and the mixture stirred for 0.5 h. The organic layer was separated

and washed sequentially with 2 N HCl (520 kg), 7% aq

(3aS,5S,6R,6aS)-5-((S)-(4-Chloro-3-(4-ethoxybenzyl)-

phenyl)(hydroxy)methyl)-2,2-dimethyltetrahydrofuro[2,3-d]-

[1,3]dioxol-6-ol (7S). A solution of NaBH (30.0 kg, 0.50 equiv)

NaHCO (500 kg) (caution: gas evolution!), and 5% aq.

in 4.5% aq NaOH (89.3 kg) was slowly (5 h) added to a mixture

Na SO (516 kg). The washed organic layer was concentrated to

of ketone 6 (680 kg, 1.00 equiv) and CeCl ·7H O (293 kg, 0.50

75−500 L below 40 °C and ﬂushed with EtOH (480 kg × 2)

equiv) in EtOH (3600 L) at −10 °C. The mixture was aged for

0.5 h and HPLC showed complete conversion (>99%)

(stereoselectivity: 7S/7R ∼97:3). The reaction mixture was

quenched with water (1360 L) at <20 °C, aged for 1 h,

concentrated to ∼1700 L, and then diluted with MTBE (4080

L) and water (2040 L). The mixture was acidiﬁed with 19% aq

HCl (∼200 kg) to pH 2−3. The layers were separated and the

organic layer was sequentially washed with 2% NaOH (1635 kg)

and 10% NaCl (1600 kg × 3), concentrated to ∼2400 L, and

codistilled with ACN (3800 L) to give a solution of crude 7S in

ACN (∼2400 L). A sample of a hydrate of 7S was isolated from

wet MTBE/heptane (1/5). X-ray diﬀraction conﬁrmed its

crystallinity, DSC showed a broad peak at ∼50 °C, and

thermogravimetric analysis showed 3.8% weight loss between 30

and 60 °C. A sample of anhydrous 7S was obtained as a white

until residual THF was <1.0%. The distillation residue was

diluted with EtOH (750 kg) and heated to 50−55 °C to dissolve

the solid. Water (376 kg) was slowly (3−4 h) added and the

resulting suspension slowly (3−4 h) cooled to 15−20 °C, aged

for 2−3 h, and ﬁltered. The ﬁlter cake was washed with water

(

180 kg) and recrystallized from acetone/water by dissolving it

in acetone (524 kg) at 20−25 °C and adding antisolvent water

(

1000 kg) over 2−3 h. After aging for 2−3 h, the suspension was

ﬁltered and the ﬁlter cake was washed with water (230 kg).

Drying at 40−50 °C in a vacuum oven aﬀorded 224 kg of biaryl

iodide 4 (92% yield, 99.7% purity). mp 65.8 °C (DSC peak);

HRMS, calcd for C H ClIO [M + H] , 372.9856; found,

72.9851; H NMR (CDCl ): δ ppm 1.45 (t, J = 7.0 Hz, 3H),

.00 (s, 2H), 4.05 (q, J = 7.0 Hz, 2H), 6.88 (d, J = 8.8 Hz, 2H),

.08−7.13 (m, 3H), 7.46−7.51 (m, 2H); C NMR (CDCl ): δ

solid by crystallization from MTBE/heptane (1/5, 35−0 °C).

ppm 14.97, 38.12, 63.43, 91.79, 114.62, 129.96, 130.47, 131.23,

34.23, 136.58, 139.55, 141.60, 157.65; Anal. Calcd for

C H ClIO: C, 48.35; H, 3.79; Cl, 9.51; I, 34.06. Found: C,

mp 86 °C (DSC peak); H NMR (CDCl ): δ ppm 1.30 (s, 3H),

1.40 (t, J = 7.0 Hz, 3H), 1.45 (s, 3H), 3.17 (d, J = 3.79 Hz, 1H),

3.97−4.01 (m, 3H), 4.03−4.05 (m, 2H), 4.09−4.12 (m, 2H),

4.48 (d, J = 3.5 Hz, 1H), 5.15 (t, J = 3.8 Hz, 1H), 6.00 (d, J = 3.8

Hz, 1H), 6.79−6.86 (m, 2H), 7.09 (d, J = 8.6 Hz, 2H), 7.19 (d, J

= 2.0 Hz, 1H), 7.26 (dd, J = 8.0, 2.0 Hz, 1H), 7.39 (d, J = 8.3 Hz,

8.61; H, 3.75; Cl, 9.71; I, 33.77.

4-Chloro-3-(4-ethoxybenzyl)phenyl)((3aS,5R,6S,6aS)-6-

(

hydroxy-2,2-dimethyltetrahydrofuro[2,3-d][1,3]dioxol-5-yl)-

methanone (6). To a solution of aryl iodide 4 (637 kg, 1.06

equiv) in THF (1730 L, 3.9 vol) was added i-PrMgCl (18%, 871

kg, 157 kg active, 0.95 equiv) over 5 h at 0 °C. The mixture was

aged for 1 h at 0 °C to give the aryl Grignard 4-MgX (additional

i-PrMgCl may be added based on the IPC result). Meanwhile, to

a solution of morpholine amide 13 (440 kg active, 1.00 equiv) in

THF (1487 L) was added t-BuMgCl (20.3%, 1001 kg, 203 kg

active, 1.08 equiv) over 5 h at −5 °C. The mixture was cooled to

1H); C NMR (CDCl ): δ ppm 15.04, 26.29, 26.91, 38.57,

63.62, 73.04, 75.57, 82.24, 85.34, 105.20, 111.97, 114.80,

125.30, 128.73, 130.07, 131.31, 134.12, 138.25, 139.79, 157.71.

(3S,4R,5R,6S)-6-(4-Chloro-3-(4-ethoxybenzyl)phenyl)-

tetrahydro-2H-pyran-2,3,4,5-tetraol (8). Water (1360 L, 2.0

vol) and 32% aq HCl (31 L, 0.20 equiv) were added to the

solution of crude 7S in ACN obtained above and the mixture

was aged at 75 °C until reaction completion (2 h). The reaction

mixture was cooled to <40 °C and salted with solid NaCl (272

kg). The resulting organic layer was separated, washed with 25%

aq NaCl (1620 kg × 2), concentrated to 2.5 vol, and dried by

azeotropic distillation with ACN (2280 L × 3) until KF <0.3%.

The distillation residue (∼1700 L) was diluted with ACN (700

L), heated to 40 °C, and ﬁltered to remove the small amount of

NaCl precipitate to give a solution of tetraol 8 as a 1/1 anomer in

−

10 °C and then the aryl Grignard prepared above was added

over 5 h. After stirring for 1.5 h, IPC showed complete

conversion. The reaction mixture was quenched into 18% aq

citric acid (357 kg citric acid monohydrate + 1493 kg water) at

−5 °C and stirred for 1 h. The organic layer was separated,

washed with 25% aq NaCl (880 kg × 2), and concentrated to

1500 L under reduced pressure. The distillation residue was

codistilled with EtOAc (1450 L × 3) and diluted with EtOAc

1450 L). It was passed through activated carbon cartridges,

concentrated to ∼1350 L, and heated to 60 °C. n-Heptane

1221 kg) was added over 2 h, followed by seeds of ketone 6 (1.3

∼

ACN. LC−MS [M + NH ] calcd, 412.2; found, 412.0; H

(

NMR (DMSO-d ): δ ppm 1.29 (t, J = 7.0 Hz, 3H), 3.03−3.10

(m, 1H), 3.14 (t, J = 9.1 Hz, ∼0.5H), 3.25 (t, J = 8.9 Hz, ∼0.5H),

3.31 (dd, J = 9.5, 3.7 Hz, ∼0.5H), 3.54 (t, J = 9.2 Hz, ∼0.5H),

3.93−4.00 (m, 4H), 4.05 (d, J = 9.4 Hz, ∼0.5H), 4.46 (d, J = 7.7

Hz, ∼0.5H), 4.51 (d, J = 9.7 Hz, ∼0.5H), 5.01 (d, J = 3.5 Hz,

∼0.5H) 6.83 (d, J = 8.7 Hz, 2H), 7.08 (d, J = 8.6 Hz, 2H), 7.21

(dd, J = 8.3, 2.0 Hz, 1H), 7.30 (t, J = 2.4 Hz, 1H) 7.36 (t, J = 7.9

(

kg). After aging for 0.5 h, more n-heptane (2716 kg) was added

over 2 h and the mixture aged for 1 h. The suspension was slowly

cooled to 10 °C, aged, and ﬁltered. The ﬁlter cake was washed

with n-heptane (448 L) and dried to give ketone 6 (607.5 kg,

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development

https://pubs.acs.org/doi/10.1021/acs.oprd.0c00359.

Article

Hz, 1H); ¹³C NMR (DMSO-d ): δ ppm 14.71, 37.63, 62.89,

19.99, 20.24, 20.41, 37.37, 62.85, 68.72, 72.25, 72.92, 77.51,

80.99, 114.27, 126.57, 129.32, 129.63, 130.11, 130.87, 132.99,

136.05, 138.56, 156.95, 168.42, 169.13, 169.43.

(2S,3R,4R,5S,6R)-2-(4-Chloro-3-(4-ethoxybenzyl)phenyl)-

6-(methylthio)tetrahydro-2H-pyran-3,4,5-triol (Sotagliﬂozin

Form I). NaOMe (30 wt % in MeOH, 16 kg, 0.08 equiv) was

added to a suspension of triacetate 10R (597 kg, 1.0 equiv) in

MeOH (2474 L, 4.2 vol) and the mixture aged at 45 °C until

reaction completion (∼4 h). Water (961 L, 1.61 vol) was added

and the mixture was seeded with LX4211 (2.4 kg form I) at 40

2.36, 74.86, 75.45, 76.73, 76.89, 92.81, 97.36, 114.30, 127.37,

27.50, 128.65, 129.59, 131.24, 131.78, 131.94, 137.94, 138.01,

39.31, 139.86, 156.92.

(

2R,3S,4R,5S,6S)-6-(4-Chloro-3-(4-ethoxybenzyl)phenyl)-

tetrahydro-2H-pyran-2,3,4,5-tetrayl Tetraacetate (9R). To

the above-obtained solution of tetraol 8 in ACN was added Et N

(

1325 L, 6.0 equiv) followed by slow addition (5 h) of Ac O

887 L, 6.0 equiv) at 30−35 °C. After aging for 3 h, the reaction

mixture was cooled to 5−10 °C, quenched with IPA (580 L, 0.85

vol), aged at 15−20 °C for 1 h, concentrated to ∼2040 L, diluted

with IPA (2040 L), aged at 55 °C for 2 h, cooled to −5 °C over 3

h, and aged for 2 h. The suspension was ﬁltered and the ﬁlter

cake washed with precooled IPA (1600 L, −5 °C) and dried at

C, aged (6 h), cooled to −7 °C (5 h), aged (6 h), and ﬁltered.

The ﬁlter cake was washed with cold (−10 °C) MeOH/water

416/179 L) and dried under reduced pressure to give 448 kg of

(

sotagliﬂozin (form I) as a white solid, 97% yield, 99.9% purity.

50 °C under reduced pressure to give 690 kg of 9R as an oﬀ-

LC−MS [M + NH ] calcd, 442.2; found, 442.0; mp 123 °C

white solid, with 79% overall yield from ketone 6, 99.5% purity.

mp 143.5 °C (DSC peak); LC−MS [M + NH ] calcd, 580.2;

found, 580.3; H NMR (DMSO-d ): δ ppm 1.29 (t, J = 7.0 Hz,

(

DSC peak); H NMR (DMSO-d ): δ ppm 1.30 (t, J = 6.9 Hz,

H), 2.04 (s, 3H), 3.17 (ddd, J = 9.5, 8.6, 5.7 Hz, 1H), 3.19 (ddd,

J = 9.5, 8.6, 5.7 Hz, 1H), 3.27 (dt, J = 8.6, 5.0 Hz, 1H), 3.97 (q, J

7.0 Hz, 2H), 3.95−4.02 (m, 2H), 4.10 (d, J = 9.5 Hz, 1H), 4.35

d, J = 9.5 Hz, 1H), 4.95 (d, J = 5.7 Hz, 1H), 5.13 (d, J = 5.0 Hz,

H), 5.22 (d, J = 5.7 Hz, 1H), 6.84 (d, J = 8.8 Hz, 2H), 7.10 (d, J

8.8 Hz, 2H), 7.21 (dd, J = 8.1, 2.1 Hz, 1H), 7.27 (d, J = 2.1 Hz,

H), 1.69 (s, 3H), 1.94 (s, 3H), 2.02 (s, 3H), 2.06 (s, 3H), 3.92−

.03 (m, 4H), 4.89 (d, J = 9.60 Hz, 1H), 5.08 (t, J = 9.6 Hz, 1H),

.17 (dd, J = 9.9, 8.3 Hz, 1H), 5.48 (t, J = 9.6 Hz, 1H), 6.02 (d, J

(

8.3 Hz, 1H), 6.83 (d, J = 8.6 Hz, 2H), 7.06 (d, J = 8.6 Hz, 2H),

.27 (s, 1H), 7.28 (d, J = 9.5 Hz, 1H), 7.42 (d, J = 7.8 Hz, 1H);

1H), 7.39 (d, J = 8.1 Hz, 1H); C NMR (DMSO-d

): δ ppm

C NMR (DMSO-d ): δ ppm 14.49, 19.84, 20.09, 20.14, 20.31,

0.95, 14.65, 37.56, 62.83, 72.07, 74.32, 77.89, 80.58, 85.35,

7.21, 62.70, 69.89, 71.56, 71.84, 74.41, 90.93, 114.13, 126.59,

29.27, 129.42, 130.14, 130.71, 133.06, 135.09, 138.45, 156.79,

68.26, 168.67, 168.95, 169.29.

14.25, 127.06, 128.73, 129.63, 130.42, 131.05, 132.01, 138.03,

39.02, 156.89. Anal. Calcd for C H ClO S: C, 59.35; H, 5.93;

Cl, 8.34; O, 18.83; S, 7.55. Found: C, 59.29; H, 5.95; Cl, 8.45; O,

(

2S,3S,4R,5S,6R)-2-(4-Chloro-3-(4-ethoxybenzyl)phenyl)-

8.88; S, 7.48.

-(methylthio)tetrahydro-2H-pyran-3,4,5-triyl Triacetate

Sotagliﬂozin Form II. Sotagliﬂozin (form I, 193 kg) was

3 2

dissolved in methyl ethyl ketone (MEK) (763 L, 4.0 vol) at 55−

0 °C and ﬁltered through a polishing ﬁlter with a small rinse of

MEK (153 L, 0.8 vol). n-Heptane (preheated to 65−70 °C)

1435 L, 7.4 vol) was added over ∼2 h followed by sotagliﬂozin

of tetraacetate 9R (325 kg, 1.0 equiv), thiourea (49.3 kg, 1.12

equiv), and EtOAc (1174 L, 3.6 vol) at 55 °C and the mixture

was aged at 55 °C for 3 h to give a thick suspension of isothiourea

(

3. An analytical sample of 23·HBF salt was obtained by

form II seeds (2.1 kg). The mixture was aged at 65−70 °C for 6 h

and more n-heptane (preheated to 65−70 °C) (1111 L, 4.9 vol)

was added over 7 h. The mixture was aged at 75−80 °C for 7 h,

cooled to 20 °C over 4 h, and aged. The suspension was ﬁltered

and the ﬁlter cake was washed with n-heptane (4.8 vol). The wet

cake was dried under reduced pressure furnishing sotagliﬂozin

form II (DSC peak 134 °C) as the ﬁnal drug substance (183 kg,

95% yield).

ﬁltering a small sample of the suspension and washing the ﬁlter

cake with water, followed by drying. mp 202.8 °C (DSC peak);

LC−MS [M + H] calcd, 579.2; found, 579.2; H NMR

(

DMSO-d ): δ ppm 1.30 (t, J = 7.0 Hz, 3H), 1.71 (s, 3H), 1.97

s, 3H), 2.08 (s, 3H), 3.92−4.03 (m, 4H), 4.79 (d, J = 9.4 Hz,

H), 5.28−5.42 (m, 3H), 5.70 (d, J = 9.3 Hz, 1H), 6.83 (d, J =

.6 Hz, 2H), 7.07 (d, J = 8.6 Hz, 2H), 7.29−7.38 (m, 2H), 7.44

(

d J = 8.1 Hz, 1H), 9.10 (br s, 4H); C NMR (DMSO-d ): δ

ppm 14.61, 19.94, 20.16, 20.27, 37.44, 62.92, 68.87, 71.32,

2.60, 78.56, 80.26, 114.35, 126.93, 129.43, 129.56, 130.46,

30.84, 133.42, 135.03, 138.58, 156.99, 166.17, 168.38, 169.32,

69.42. After cooling the reaction mixture to −2 °C, MeI (107

ASSOCIATED CONTENT

■

sı Supporting Information

The Supporting Information is available free of charge at

kg, 1.30 equiv) was added, followed by MeOH (652 L, 2.0 vol)

and slow addition of i-Pr NEt (449 kg, 6.0 equiv). The reaction

mixture was aged at 20 °C for 1 h and then at 35 °C until reaction

completion (∼3 h). It was concentrated to 976 L (3.0 vol),

codistilled with IPA (1305 L × 2), diluted with IPA (1305 L, 4.0

vol) and water (976 L, 3.0 vol), and aged at 20 °C for 1 h. The

suspension was ﬁltered, the ﬁlter cake washed sequentially with

Copies of H and C NMR spectra for new compounds

PDF )

AUTHOR INFORMATION

■

/1 IPA/water (1958 L, 6.0 vol) and IPA (1057 L, 3.3 vol), and

dried to give 298 kg of triacetate 10R as a white solid, 94% yield,

Corresponding Authors

99.7% purity. mp 157.6 °C (DSC peak); LC−MS [M + NH ]

Matthew M. Zhao − Chemical Development, Lexicon

Pharmaceuticals Incorporation, Basking Ridge, New Jersey

07920, United States; orcid.org/0000-0002-4163-3555;

Email: mzhao@lexpharma.com

Wenxue Wu − Chemical Development, Lexicon Pharmaceuticals

Incorporation, Basking Ridge, New Jersey 07920, United States;

orcid.org/0000-0002-5979-9350; Email: wwu@

lexpharma.com

calcd, 568.2; found, 568.3; H NMR (DMSO-d ): δ ppm 1.29 (t,

J = 6.95 Hz, 3H), 1.70 (s, 3H), 1.94 (s, 3H), 2.04 (s, 3H), 2.09 (s,

H), 3.92−4.03 (m, 4H), 4.70 (d, J = 9.9 Hz, 1H), 4.88 (d, J =

.9 Hz, 1H), 5.05 (t, J = 9.8 Hz, 1H), 5.14 (t, J = 9.8 Hz, 1H),

.36 (t, J = 9.5 Hz, 1H), 6.80−6.86 (m, 2H), 7.04−7.11 (m,

H), 7.25 (d, J = 2.0 Hz, 1H), 7.30 (dd, J = 8.2, 2.2 Hz, 1H), 7.42

(

d, J = 8.1 Hz, 1H); C NMR (DMSO-d ): δ ppm 10.25, 14.63,

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development

Kimball, S. D.; Powell, D. R.; Rawlins, D. B. Novel L-xylose derivatives

Article

Authors

Mascarello, A.; Segretti, N. D.; de Azevedo, H. F. Z.; Guimaraes, C. R.

W.; Miranda, L. S. M.; de Souza, R. O. M. A. Synthetic strategies toward

SGLT2 inhibitors. Org. Process Res. Dev. 2018, 22, 467−488. and

references therein

Haiming Zhang − Chemical Development, Lexicon

Pharmaceuticals Incorporation, Basking Ridge, New Jersey

7920, United States; orcid.org/0000-0002-2139-2598

(3) (a) Goodwin, N. C.; Mabon, R.; Harrison, B. A.; Shadoan, M. K.;

Shinya Iimura − Chemical Development, Lexicon

Pharmaceuticals Incorporation, Basking Ridge, New Jersey

Almstead, Z. Y.; Xie, Y.; Healy, J.; Buhring, L. M.; DaCosta, C. M.;

Bardenhagen, J.; Mseeh, F.; Liu, Q.; Nouraldeen, A.; Wilson, A. G. E.;

as selective sodium-dependent glucose cotransporter 2 (SGLT2)

inhibitors for the treatment of type 2 diabetes. J. Med. Chem. 2009, 52,

6201−6204. (b) Powell, D. R.; Smith, M.; Greer, J.; Harris, A.; Zhao, S.;

DaCosta, C.; Mseeh, F.; Shadoan, M. K.; Sands, A.; Zambrowicz, B.;

Ding, Z.-M. LX4211 Increases serum glucagon-like peptide 1 and

peptide YY Levels by reducing sodium/glucose cotransporter 1

7920, United States

Mark S. Bednarz − Chemical Development, Lexicon

Pharmaceuticals Incorporation, Basking Ridge, New Jersey

7920, United States

Qiu-Ling Song − Chemical Development, Lexicon

Pharmaceuticals Incorporation, Basking Ridge, New Jersey

7920, United States; orcid.org/0000-0002-9836-8860

Strumph, P.; Sands, A. Development of sotagliflozin, a dual sodium-

SGLT1)−mediated absorption of intestinal glucose. J. Pharmacol.

Ngiap-Kie Lim − Chemical Development, Lexicon

Pharmaceuticals Incorporation, Basking Ridge, New Jersey

Exp. Ther. 2013, 345, 250 −259. (c) Lapuerta, P.; Zambrowicz, B.;

dependent glucose transporter 1/2 inhibitor. Diabetes Vasc. Dis. Res.

7920, United States; orcid.org/0000-0002-3760-4874

Jie Yan − Chemical Development, Lexicon Pharmaceuticals

Incorporation, Basking Ridge, New Jersey 07920, United States

Kuangchu Dai − Process R&D, WuXi Apptec, Shanghai 200131,

China

Xiaodong Gu − Process R&D, WuXi Apptec, Shanghai 200131,

China

015 , 12 , 101 −110. (d) Sims, H.; Smith, K. H.; Bramlage, P.; Minguet,

J. Sotagliflozin: a dual sodium-glucose co-transporter-1 and -2 inhibitor

for the management of type 1 and type 2 diabetes mellitus. Diabetic

Med. 2018, 35, 1037−1048.

(4) Unpublished internal communications at Lexicon Pharmaceutic-

als.

Youchu Wang − Process R&D, WuXi Apptec, Shanghai 200131,

(5) Gelas, J.; Horton, D. Acetonation of D-ribose and D-arabinose

with alkyl isopropenyl ethers. Carbohydr. Res. 1975 , 45 , 181 −195.

China

6) (a) Luche, J. L. Lanthanides in organic chemistry. 1. selective 1,2

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.oprd.0c00359

reductions of conjugated ketones. J. Am. Chem. Soc. 1978, 100, 2226−

227. (b) Stastna, E.; Cerny, I.; Pouzar, V.; Chodounska, H.

́ ́

̌ ̌

Stereoselectivity of sodium borohydride reduction of saturated

steroidal ketones utilizing conditions of Luche reduction. Steroids

2010, 75, 721−725. (c) Fruhmann, P.; Hametner, C.; Mikula, H.;

Notes

The authors declare no competing ﬁnancial interest.

Adam, G.; Krska, R.; Fro

deoxynivalenol and three of its acetylated derivatives at C8. Toxins

014, 6, 325−336.

7) Ruttens, B.; Kovac

colitose thioglycosides. Synthesis 2004, 2505−2508.

8) Williams, D. R.; Klingler, F. D.; Allen, E. E.; Lichtenthaler, F. W.

hlich, J. Stereoselective Luche reduction of

ACKNOWLEDGMENTS

■

, P. A facile synthesis of armed and disarmed

The authors thank Leonard Hargiss and Vincent Caruso for

analytical support, particularly for assistance with HPLC

method development and LC−MS analysis. The authors also

thank Philip Keyes for NMR structural assignments. They

gratefully acknowledge the collaboration with discovery

chemists, Bryce Harrison, Nicole Goodwin, Ross Mabon,

David R. Rawlins, and David Kimball.

́ ̌

Bromine as an oxidant for direct conversion of aldehydes to esters.

Tetrahedron Lett. 1988, 29, 5087−5090.

(9) (a) Kopach, M. E.; Singh, U. K.; Kobierski, M. E.; Trankle, W. G.;

Murray, M. M.; Pietz, M. A.; Forst, M. B.; Stephenson, G. A.; Mancuso,

V.; Giard, T.; Vanmarsenille, M.; DeFrance, T. Practical synthesis of

chiral 2-morpholine: (4-benzylmorpholin-2-(S)-yl)-(tetrahydropyran-

4-yl)methanone mesylate, a useful pharmaceutical intermediate. Org.

Process Res. Dev. 2009, 13, 209−224. (b) Badioli, M.; Ballini, R.;

Bartolacci, M.; Bosica, G.; Torregiani, E.; Marcantoni, E. Addition of

organocerium reagents to morpholine amides: synthesis of important

pheromone components of Achaea janata. J. Org. Chem. 2002, 67,

ADDITIONAL NOTES

Satisfactory safety evaluation results were obtained before

being implemented in the plant.

Satisfactory safety evaluation data were obtained before scale-

■

up.

10) (a) Anelli, P. L.; Biffi, C.; Montanari, F.; Quici, S. Fast and

938−8942.

REFERENCES

■

selective oxidation of primary alcohols to aldehydes or to carboxylic

1) (a) Tahrani, A. A.; Barnett, A. H.; Bailey, C. J. SGLT Inhibitors in

acids and of secondary alcohols to ketones mediated by oxoammonium

salts under two-phase conditions. J. Org. Chem. 1987 , 52 , 2559 −2562.

(b) Davis, N. J.; Flitsch, S. L. Selective oxidation of monosaccharide

derivatives to uronic acids. Tetrahedron Lett. 1993, 34, 1181−1184.

(11) (a) De Mico, A.; Margarita, R.; Parlanti, L.; Vescovi, A.;

Piancatelli, G. A Versatile and highly selective hypervalent iodine (III)/

2,2,6,6-tetramethyl-1-piperidinyloxyl-mediated oxidation of alcohols to

carbonyl compounds. J. Org. Chem. 1997 , 62 , 6974 −6977. (b) Epp, J.

B.; Widlanski, T. S. Facile preparation of nucleoside-5 ’-carboxylic

Acids. J. Org. Chem. 1999, 64, 293−295.

(12) De Luca, L.; Giacomelli, G.; Masala, S.; Porcheddu, A.

Trichloroisocyanuric/TEMPO oxidation of alcohols under mild

conditions: a close investigation. J. Org. Chem. 2003, 68, 4999−5001.

(13) Zhao, M.; Li, J.; Mano, E.; Song, Z.; Tschaen, D. M.; Grabowski,

E. J. J.; Reider, P. J. Oxidation of primary alcohols to carboxylic acids

with sodium chlorite catalyzed by TEMPO and bleach. J. Org. Chem.

1999, 64, 2564−2566.

management of diabetes. Lancet Diabetes Endocrinol. 2013, 1, 140−151.

(

b) Sabino-Silva, R.; Mori, R. C.; David-Silva, A.; Okamoto, M. M.;

Freitas, H. S.; Machado, U. F. The Na /Glucose cotransporters: from

genes to therapy. Braz. J. Med. Biol. Res. 2010, 43 , 1019 −1026.

and cardiovascular considerations of SGLT2 inhibition. Nat. Rev.

Nephrol. 2017, 13 , 11 −26. (d) Mudaliar, S.; Polidori, D.; Zambrowicz,

B.; Henry, R. R. Sodium-glucose cotransporter inhibitors: effects on

renal and intestinal glucose transport: from bench to bedside. Diabetes

Care 2015, 38, 2344−2353.

(2) For reviews and syntheses, see: (a) He, Z.-X.; Zhou, Z.-W.; Yang,

Y.; Yang, T.; Pan, S.-Y.; Qiu, J.-X.; Zhou, S.-F. Overview of clinically

approved oral antidiabetic agents for the treatment of type 2 diabetes

mellitus. Clin. Exp. Pharmacol. Physiol. 2015 , 42 , 125 −138. (b) Cai, W.;

Jiang, L.; Xie, Y.; Liu, Y.; Liu, W.; Zhao, G. Design of SGLT2 inhibitors

for the treatment of type 2 diabetes: a history driven by biology to

chemistry. Med. Chem. 2015, 11, 317−328. (c) Aguillon, A. R.;

Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development

VCH: Weinheim, 2010. (b) Jin, R.-Y.; Hu, S.-Q.; Zhang, Y.-G.; Bo, T.

Article

(14) (a) Vogt, H.; Balej, J.; Bennett, J. E.; Wintzer, P.; Sheikh, S. A.;

Gallone, P.; Vasudevan, S.; Pelin, K. Chlorine Oxides and Chlorine

Oxygen Acids. Ullmann’s Encyclopedia of Industrial Chemistry; Wiley-

Concentration-dependence of the explosion characteristics of chlorine

dioxide gas. J. Hazard. Mater. 2009, 166, 842−847.

15) (a) Ishihara, K.; Ohara, S.; Yamamoto, H. 3,4,5-Trifluorobenze-

neboronic acid as an extremely active amidation catalyst. J. Org. Chem.

996 , 61 , 4196 −4197. (b) Maki, T.; Ishihara, K.; Yamamoto, H. New

boron (III)-catalyzed amide and ester condensation reactions.

Tetrahedron 2007 , 63 , 8645 −8657. (c) Al-Zoubi, R. M.; Marion, O.;

Hall, D. G. Direct and waste-free amidations and cycloadditions by

organocatalytic activation of carboxylic acids at room temperature.

Angew. Chem., Int. Ed. 2008, 47, 2876−2879.

(

16) (a) Mylavarapu, R. K.; GCM, K.; Kolla, N.; Veeramalla, R.;

Koilkonda, P.; Bhattacharya, A.; Bandichhor, R. Boric acid catalyzed

amidation in the synthesis of active pharmaceutical ingredients. Org.

Process Res. Dev. 2007 , 11 , 1065 −1068. (b) Tang, P. Boric acid

catalyzed amide formation from carboxylic acids and amines: N -

benzylbutyramide. Org. Synth. 2005, 81, 262−272.

(17) (a) Meng, W.; Ellsworth, B. A.; Nirschl, A. A.; McCann, P. J.;

Patel, M.; Girotra, R. N.; Wu, G.; Sher, P. M.; Morrison, E. P.; Biller, S.

A.; Zahler, R.; Deshpande, P. P.; Pullockaran, A.; Hagan, D. L.; Morgan,

N.; Taylor, J. R.; Obermeier, M. T.; Humphreys, W. G.; Khanna, A.;

Discenza, L.; Robertson, J. G.; Wang, A.; Han, S.; Wetterau, J. R.;

Janovitz, E. B.; Flint, O. P.; Whaley, J. M.; Washburn, W. N. Discovery

of dapagliflozin: a potent, selective renal sodium-dependent glucose

cotransporter 2 (SGLT2) inhibitor for the treatment of type 2 diabetes.

J. Med. Chem. 2008 , 51 , 1145 −1149.

18) Smonou, I. One step reduction of diaryl ketones to hydrocarbons

by etherated boron trifluoride-triethylsilane system. Synth. Commun.

994, 24, 1999−2002.

19) (a) Ono, A.; Maruyama, T.; Suzuki, N. Hydrogenation of diaryl

and arylalkyl ketones by sodium borohydride and anhydrous aluminum

chloride. Synth. Commun. 1987 , 17 , 1001 −1005. (b) Ono, A.; Suzuki,

N.; Kamimura, J. Hydrogenolysis of diaryl and aryl alkyl ketones and

carbinols by sodium borohydride and anhydrous aluminum (III)

chloride. Synthesis 1987, 736−738.

(20) Minkina, V. G.; Shabunya, S. I.; Kalinin, V. I.; Martynenko, V. V.;

Smirnova, A. L. Stability of alkaline aqueous solutions of sodium

borohydride. Int. J. Hydrogen Energy 2012, 37 , 3313 −3318.

(21) (a) Rietjens, M.; Steenbergen, P. A. Crosslinking mechanism of

boric acid with diols revisited. Eur. J. Inorg. Chem. 2005, 1162−1174.

(

b) Boron-carbohydrate interactions. In CarbohydratesComprehen-

sive Studies on Glycobiology and Glycotechnology; Pappin, B., Kiefel, M. J.,

Houston, T. A., Chang, C.-F., Eds.; IntechOpen, 2012; Chapter 3.

23) (a) Kruger, H. G. Ab initio mechanistic study of the protection of

22) Lewis, B. E.; Choytun, N.; Schramm, V. L.; Bennet, A. J.

Transition states for glucopyranose interconversion. J. Am. Chem. Soc.

006, 128, 5049−5058.

alcohols and amines with anhydrides. J. Mol. Struct.: THEOCHEM

M.; Honarparvar, B.; Kruger, H. G. Mechanistic investigation of the

002, 577, 281−285. (b) Lawal, M. M.; Govender, T.; Maguire, G. E.

uncatalyzed esterification reaction of acetic acid and acid halides with

methanol: a DFT study. J. Mol. Model. 2016, 22, 235.

(24) (a) Hanaya, T.; Baba, H.; Toyota, H.; Yamamoto, H. Synthetic

studies on pterin glycosides: the first synthesis of 2 ’-O-(α -D-

glucopyranosyl)biopterin. Tetrahedron 2009, 65, 7989−7997.

(

b) Luber, T.; Niemietz, M.; Karagiannis, T.; Monnich, M.; Ott, D.;

Perkams, L.; Walcher, J.; Berger, L.; Pischl, M.; Weishaupt, M.; Eller, S.;

Hoffman, J.; Unverzagt, C. A single route to mammalian N-glycans

substituted with core fucose and bisecting GlcNAc. Angew. Chem., Int.

Ed. 2018, 57, 14543−14549.