Scalable Asymmetric Syntheses of Foslevodopa and Foscarbidopa Drug Substances for the Treatment of Parkinson's Disease

Drug Substances for the Treatment of Parkinson ’s Disease

Article

Scalable Asymmetric Syntheses of Foslevodopa and Foscarbidopa

Alexander D. Huters, James Stambuli, Russell C. Klix, Mark A. Matulenko, Vincent S. Chan,

Justin Simanis, David R. Hill,* Rajarathnam E. Reddy, Timothy B. Towne, John R. Bellettini,

Brian J. Kotecki, Benoit Cardinal-David, Jianguo Ji, Eric A. Voight, Minshan Shou,

Selvakumar Balaraman, Abhishek Ashok, and Soma Ghosh

Cite This: https://doi.org/10.1021/acs.joc.1c00905

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: Foslevodopa (FLD, levodopa 4′-monophosphate, 3) and foscarbi-

dopa (FCD, carbidopa 4′-monophosphate, 4) were identiﬁed as water-soluble

prodrugs of levodopa (LD, 1) and carbidopa (CD, 2), respectively, which are

useful for the treatment of Parkinson’s disease. Herein, we describe asymmetric

syntheses of FLD (3) and FCD (4) drug substances and their manufacture at pilot

scale. The synthesis of FLD (3) employs a Horner−Wadsworth−Emmons

oleﬁnation reaction followed by enantioselective hydrogenation of the double

bond as key steps to introduce the α-amino acid moiety with the desired

stereochemistry. The synthesis of FCD (4) features a Mizoroki−Heck reaction

followed by enantioselective hydrazination to install the quaternary chiral center

bearing a hydrazine moiety.

INTRODUCTION

barrier. However, LD (1) is rapidly decarboxylated to

dopamine in extracerebral tissues, especially in the gastro-

intestinal tract, and activates peripheral dopamine receptors

and often causes severe side eﬀects, such as nausea, vomiting,

■

Parkinson’s disease (PD) is a progressive and debilitating

neurodegenerative disorder which is characterized by de-

creased dopaminergic activity in the brain and aﬀects over six

and dyskinesia. Therefore, levodopa (1) is often administered

million people worldwide. One of the standard treatments for

in combination with carbidopa (Figure 1, CD, 2), an inhibitor

of DOPA decarboxylase enzyme. Carbidopa (2) does not

eﬀectively cross the blood−brain barrier and therefore

selectively inhibits the conversion of LD (1) to dopamine

outside of the brain, which increases the LD (1) half-life and

early Parkinson’s disease involves oral administration of

levodopa (Figure 1, LD, 1) as a dopamine replacement

therapy. Levodopa (1) is converted in vivo to dopamine by

DOPA decarboxylase. The active metabolite of LD (1),

dopamine, controls the symptoms of Parkinson’s disease, and

unlike LD (1 ), dopamine does not cross the blood-brain

uptake in the brain while reducing its dosage and side eﬀects.

One challenge with LD (1) treatments is the relatively short

half-life in plasma, even with the coadministration of CD (2),

which makes it diﬃcult for patients to achieve and maintain

optimal levels of dopamine in the brain. Moreover, upon

progression of Parkinson’s disease, gastric motility is typically

impaired, which makes the absorption of LD (1) from tablet

form unpredictable. To circumvent these issues, a novel

formulation of LD (1) and CD (2) (Duopa) was developed as

a gel that enables continuous administration of the drug via

direct infusion into the intestine through a jejunal tube.

Special Issue: Excellence in Industrial Organic Syn-

thesis 2021

Received: April 19, 2021

Figure 1. Levodopa and carbidopa and their phosphate prodrugs.

XXXX American Chemical Society

The Journal of Organic Chemistry

J. Org. Chem. XXXX, XXX, XXX−XXX

Figure 2. Retrosynthesis of foslevodopa.

Article

Figure 3. Retrosynthesis of foscarbidopa.

Scheme 1. Synthesis of Foslevodopa (3)

Duopa gel formulation allows continuous dosing of LD (1)

and CD (2) with aid of a pump to prevent the peaks and

troughs associated with traditional LD (1) and CD (2) tablet

therapy. Despite the multiple beneﬁts of Duopa, drug

administration requires surgery at the beginning of the therapy

physicochemical properties and undergoes an in vivo chemical

or enzymatic transformation to deliver the active molecule at

eﬃcacious levels.

Levodopa (1) and carbidopa (2) have poor aqueous

solubility, approximately 4−5 mg/mL at physiological pH

(7.4), which limits options for solution dosage forms. Our

investigations found that phosphate derivatives of LD (1) and

CD (2) imparted higher aqueous solubility and enabled

to install the jejunal tube.

With the goal of overcoming the challenges associated with

LD (1) and CD (2) existing therapies, a prodrug strategy was

envisioned for the treatment of Parkinson’s disease. Prodrug

strategies can enable drug development endeavors when an

active ingredient presents challenges, such as poor solubility,

inadequate pharmacokinetic proﬁle, undesirable pill burden or

high injection volume, formulation diﬃculties, etc., and they

have been successfully applied to bring novel drugs to market

in scenarios where the parent compound would have likely

subcutaneous delivery of drug at adequate dosages.

Foslevodopa (3) and foscarbidopa (4) were identiﬁed as

phosphate prodrugs of levodopa (1) and carbidopa (2),

respectively, for treatment of Parkinson’s disease via con-

tinuous subcutaneous infusion.

No selective LD monophosphate synthesis has been

reported previously, and there was no precedent for CD

failed during clinical development. A prodrug is generally

phosphates; therefore, synthetic routes to access foslevodopa

described as a weakly active or inactive chemical entity capable

of converting to the parent drug, which achieves the desired

and foscarbidopa were developed. Here, we describe the

asymmetric synthesis of FLD (3) which employs a Horner−

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

Article

Table 1. Catalysts and Ligand Screening for Enantioselective Enamine (6) Reduction

Entry

Catalyst

Ligand

% Conversion

% ee (5)

[((R,R)-Et-DuPhos)Rh(cod)BF₄]

−

>99

<10

95.6

[((S,S)-Et-DuPhos)Rh(cod)BF ] (16)

97.8

68.8

31.0

[Rh(nbd) ]BF

Josiphos J001-2

−

(R)-Monophos

(R)-BINAP

−

((S)-Segphos)Ru(OAc₂)

[Ir(cod)Cl]₂

Not Determined

46.0

((R)-BINAP)Ru(OAc)₂

[((S)-BINAP)RuCl ]p-cymene

Not Determined

Not Applicable

Not Determined

[((R)-Segphos)RuCl ]p-cymene

[Rh(nbd) ]BF

(R)-BINAP

(S,S)-iPr-DuPhos

>99

[Rh(nbd) ]BF

[((S,S)-Et-DuPhos)Rh(cod)BF ] (16)

−

97.0

[((S,S)-Et-DuPhos)Rh(cod)OTf]

>99

Not Determined

[((S,S)-Et-DuPhos)Rh(cod)BF ] (16)

94.3

95.1

[((S,S)-Me-DuPhos)Rh(cod)BF₄]

e,f

[((S,S)-Et-DuPhos)Rh(cod)BF ] (16)

>99.5%

Notes: R-isomer. Reaction led to signiﬁcant debenzylation. Reaction run using 50 psig H . Reaction run at 30 °C. All reactions run in

dichloromethane except for Entry 16, which was run in THF. It was not clear if both E- and Z-isomers of 6 were directly reduced to the desired

(S)-5 or if one of the isomers of 6 was selectively reduced to (S)-5 and the other oleﬁn isomer of 6 was concurrently isomerized/equilibrated and

was then reduced to (S)-5.

Wadsworth−Emmons oleﬁnation followed by enantioselective

hydrogenation as key steps to introduce the α-amino acid

moiety. Also reported is the synthesis of FCD (4), which

employs a Mizoroki−Heck reaction, followed by enantiose-

lective hydrazination, for installation of the quaternary center

with the desired stereochemistry. In addition, optimization of

the FLD (3) and FCD (4) processes and manufacture of both

drug substances (3 and 4) on kilogram scale is reported.

THF, to provide 3′-protected benzyl ether 7 in 58% yield and

98.5% purity (Scheme 1). Under these conditions, the

corresponding 4′-isomer and 3′/4′-dibenzyl impurities were

also observed. These were rejected through recrystallization.

Use of THF as a cosolvent was necessary to avoid the potential

safety hazards of sodium hydride in DMSO, especially during

scale-up. The 4′-phenolic group in 3-(benzyloxy)-4-hydroxy-

benzaldehyde (7) was then converted to the corresponding

dibenzyl phosphate 15, in excellent yield (93%), by reaction

RESULTS AND DISCUSSION

with tetrabenzyl pyrophosphate (14, TBPP, 1.05 equiv) and

DBU (1.15 equiv) in acetonitrile. These conditions had not

been previously reported and were found to be uniquely

eﬀective and convenient for the preparation of sterically

■

(

Retrosynthesis of Foslevodopa (3) and Foscarbidopa

4). The foslevodopa (3) retrosynthetic analysis is shown in

Figure 2. We envisioned that FLD (3) could arise from (S)-

amino-ester 5, which could be accessed through an

hindered phenol phosphates. The remaining carbon frame-

work of foslevodopa (3) was installed through a Horner−

Wadsworth−Emmons oleﬁnation. Accordingly, the phosphate

15 was treated with 1.1 equiv of (±)-benzyloxycarbonyl-α-

phosphonoglycine-trimethyl ester (8), in the presence of

tetramethylguanidine (TMG), to furnish enamine 6 in 85%

enantioselective reduction of enamine 6 as the key step.

The enamine 6 could be derived by Horner−Wadsworth−

Emmons oleﬁnation of 3-(benzyloxy)-4-hydroxybenzaldehyde

(

7) and commercially available phosphonate ester 8. The

retrosynthetic analysis of foscarbidopa (4) is shown in Figure

, and it was envisioned from (S)-hydrazine derivative 9 via

yield, as a 10:1 mixture of E/Z isomers.

With enamine 6 in hand, our eﬀorts then turned toward

ﬁnding the best conditions for an enantioselective hydro-

genation to provide (S)-amino-ester 5. These eﬀorts are

summarized in Table 1. Use of 1 mol % rhodium catalyst 16

containing the Et-DuPhos ligand was shown to give excellent

conversion (>99%) and 97.8% ee (entry 2). A variety of

catalysts known for enantioselective reduction of enamines

were examined but gave poor reactivity and/or selectivity

(entries 3−10). Use of i-Pr-DuPhos or Me-DuPhos ligands

led to lower reactivity (entries 11 and 15). The counterion was

also shown to be critical for optimal reactivity (entry 13), with

triﬂate providing inferior results to those with tetraﬂuorobo-

oxidation of the aldehyde to a carboxylic acid and cleavage of

protective groups. The (S)-hydrazine 9 would be obtained

from racemic aldehyde 10 using an enantioselective α-

hydrazination to install the quaternary chiral center. Finally,

the (±)-aldehyde 10 would arise from 2-(benzyloxy)-4-

bromophenol (11) and 2-methylprop-2-en-1-ol (12) through

a Mizoroki−Heck coupling reaction.

Synthesis of Foslevodopa (3). The synthesis of

foslevodopa (3) began with selective protection of 3,4-

dihydroxybenzaldehyde (13) via alkylation of the more

nucleophilic alkoxide. This was performed with benzyl chloride

using 2.2 equiv of sodium hydride, in a mixture of DMSO and

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

Article

Scheme 2. Synthesis of Foscarbidopa (6)

Table 2. Optimization of α-Hydrazination of (± )-Aldehyde 10 to (S)-Hydrazine 9

Entry

Catalyst

Additive

None

TFA (20 mol %)

TCA (15 mol %)

TFA (15 mol %)

TFA (5 mol %)

Solvent

MeCN

Time (h)

Conversion (%)

(S)-9 ee (%)

(S)-20

(S)-23

(R,R)-24

(S)-20

(R)-20

−51

−45

−46

−50

−46

−40

−50

−60

CH Cl₂

PhMe

THF

EtOAc

MTBE

MeCN

(R)-20

Notes: 20 mol % catalyst used. 5 mol % catalyst used.

rate. A decrease of hydrogen pressure or temperature gave

slightly diminished enantioselectivity (entries 12 and 14). A

change to THF from dichloromethane as solvent allowed for

enamine 6 to be reduced to the (S)-amino-ester 5 in 94% yield

and >99.5% ee (entry 16). It was gratifying to note that both

E/Z-enamine isomers 6 converted to the desired product (S)-

(2S,5S)-16 and hydrogen at 100 psig to give the (S)-amino-

ester 5 in 94% yield and >99.5% ee.

Finally, to complete the synthesis of foslevodopa (3), the

(S)-amino ester 5 was subjected to hydrogenolysis using Pd/C

to remove the benzyl and Cbz groups. This was accomplished

in THF and water in the presence of sodium bicarbonate. After

the reaction was complete, pH adjustment with hydrochloric

acid followed by crystallization gave (S)-ester 17 in 85% yield

and 98.7% ee. The debenzylated product, (S)-ester 17, was

subjected to hydrolysis using sodium hydroxide. The resulting

. Further studies would be warranted to evaluate the

hydrogenation reaction mechanism. Thus, under optimized

conditions (entry 16), the enamine 6 was selectively reduced

in the presence of 1 mol % (0.01 equiv) of rhodium catalyst

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

Article

mixture was treated with hydrochloric acid to aﬀord

foslevodopa (3) in 87% yield and 99.6% ee. Thus, starting

from 3,4-dihydroxybenzaldehyde (13), the overall yield of

foslevodopa (3) drug substance was 31% for six steps, at lab

scale. This six-step synthesis was executed on pilot scale to

obtain 29.3 kg of foslevodopa (3) drug substance for clinical

studies, with the overall yield and quality comparable to those

of lab scale batches (99.7 area% purity and 99.2% ee).

2-(benzyloxy)-4-bromophenol (11), the overall yield of

foscarbidopa (4) drug substance was 25% over six steps. The

foscarbidopa (4) process, which consists of six steps starting

from 2-(benzyloxy)-4-bromophenol (11), was optimized and

several scale-up batches were manufactured on pilot scale to

obtain 14.3 kg of foscarbidopa (4) drug substance, with overall

yield and quality comparable to that of lab scale batches (99.7

area% purity and 99.9% ee).

Synthesis of Foscarbidopa (4). The synthesis of

foscarbidopa (4, Scheme 2) began with a Mizoroki−Heck

coupling of 2-(benzyloxy)-4-bromophenol (11) with 2-

methylprop-2-en-1-ol (12). The coupling reaction was

catalyzed by Pd (dba) /t-Bu P-HBF (1.5/3.0 mol %) in 1,4-

CONCLUSION

■

Synthetic routes were developed for foslevodopa (FLD, 3) and

foscarbidopa (FCD, 4), which are phosphate prodrugs of

levodopa (LD, 1) and carbidopa (CD, 2) drug substances. The

synthesis of FLD (3) featured an enantioselective hydro-

genation of enamine (6) in excellent selectivity as the key step.

The synthesis of FCD (4) employed an enantioselective α-

hydrazination of (±)-aldehyde (10) as the key step to install

the quaternary chiral center. The syntheses of FLD (3) and

FCD (4) were optimized to enable manufacturing at

multikilogram scale.

dioxane in the presence of N-cyclohexyl-N-methlcyclo-

hexanamine (Cy NMe) at 100 °C. After the reaction was

complete, puriﬁcation of the expected (±)-aldehyde product

18) via crystallization was complicated by the low melting

(

point of this intermediate. For this reason, the (±)-aldehyde

18) was derivatized as its bisulﬁte adduct (±)-19 and isolated

in a 2-step yield of 64%, without the need for chromatog-

(

raphy. In the next step, the bisulﬁte adduct (19) was cleaved

in situ with sodium bicarbonate, and the aldehyde intermediate

EXPERIMENTAL SECTION

■

(

18) was phosphorylated using tetrabenzyl pyrophosphate (14,

14 15

General Procedure. Unless otherwise noted, all solvents and

reagents were purchased from commercial sources and used without

additional puriﬁcation. The temperature listed for each experiment

refers to the internal reaction vessel temperature. Small-scale reactions

were performed using a heating mantle with a thermocouple/

temperature probe. Large scale reactions were performed using a

jacketed reactor using ethylene glycol-water as the heating media.

TBPP) and DBU (1.15 equiv) in acetonitrile, to provide

′-phosphorylated (±)-aldehyde 10 in 92% yield over the two

steps.

With (±)-aldehyde 10 in hand, our focus turned to

exploration of the key enantioselective α-hydrazination to

install the desired quaternary chiral center (Table 2).

Proton and carbon nuclear magnetic resonance spectra ( H NMR and

Following literature precedent on a related substrate, initial

data showed that (±)-aldehyde 10 was converted preferentially

to the undesired (R)-enantiomer of hydrazine 9, when

catalyzed by (S)-tetrazole 20 in the presence of dibenzyl

azodicarboxylate (DBAD), albeit with low conversion and

moderate enantiomeric excess (entry 1). A solvent screening

study showed no improvement to the enantioselectivity and

only a moderate increase in the conversion of 10 to 9 (entries

C NMR) were measured with either a Varian 400 MHz or Varian

Inova 500 MHz spectrometer. H NMR chemical shifts are expressed

in parts per million (δ) downﬁeld from tetramethylsilane (CHCl

standardized at 7.26 ppm). C NMR chemical shifts are expressed in

parts per million (δ) downﬁeld from tetramethylsilane (central peak

of CHCl standardized at 77.0 ppm). P NMR chemical shifts are

expressed in parts per million using the uniﬁed chemical shift scale

(Xi: 40.480742), with H PO as external reference. Progress of

−6). Use of L-proline (S)-23 as a catalyst gave lower

conversion and enantioselectivity (entry 7). Diamine catalyst

R,R)-24 with triﬂuoroacetic acid (TFA) as an additive

provided excellent conversion albeit with no improvement to

reactions were monitored using either HPLC or LC-MS. Analytical

LC-MS was performed on a Thermo MSQ-Plus mass spectrometer

and Agilent 1100/1200 HPLC system running Xcalibur 2.0.7, Open-

Access 1.4, and custom login software. The mass spectrometer was

operated under positive APCI or ESI ionization conditions dependent

on the system used. High resolution mass spectra were acquired using

(

the enantioselectivy. Performing the reaction with (S)-

tetrazole 20, in the presence of the additive trichloroacetic acid

time-of-ﬂight (TOF MS ES ). The HPLC system comprised an

(

TCA), provided product in 62% conversion and 60% ee

entry 9). Switching to triﬂuoroacetic acid (TFA) as an

Agilent Binary pump, degasser, column compartment, autosampler,

and diode-array detector, with a Polymer Laboratories ELS-2100

evaporative light-scattering detector. The column used was a

Phenomenex Kinetex C8, 2.6 μm 100 Å (2.1 mm × 30 mm) at a

temperature of 65 °C. Puriﬁcation of the crude reaction products,

when performed, was accomplished on a Grace Reveleris X2 normal-

phase chromatography system using silica gel cartridges purchased

from Grace or Silicycle. Speciﬁc parameters used in the separation of

individual compounds are detailed under each entry. Unless otherwise

noted, reactions were carried out under an atmosphere of nitrogen

and yields refer to isolated yields of analytically pure (>95%) material.

Chiral purities were determined by HPLC using a Chiralpak IC

column (4.6 mm × 250 mm, 5 μm) with mobile phase 90:10 hexane/

additive with (S)-tetrazole 20 increased the conversion to 93%

with 60% ee (entry 10). The use of (R)-tetrazole 20 with TFA

as an additive provided excellent conversion of 10 with 60% ee

favoring the desired enantiomer (S)-9 (entry 11). Adequate

catalyst activity and conversion were retained upon lowering

the catalyst and additive loading to 5 mol % (entry 12).

With the optimized conditions in hand, the α-hydrazination

of (±)-aldehyde 10 provided (S)-hydrazine 9 in 84% assay

yield (57% ee). Addition of water allowed crystallization of

(

S)-9 directly from the crude reaction mixture in 50% yield.

Following the installation of the quaternary chiral center, the

S)-hydrazine 9 was oxidized under Lindgren conditions using

:1 MeOH:EtOH + 0.1% H PO at 25 °C and a ﬂow rate of 1.8 mL/

3 4

min with detection at 210 nM.

(

3-(Benzyloxy)-4-hydroxybenzaldehyde (7). In a dry reactor

equipped with a temperature probe, 3,4-dihydroxybenzaldehyde (13,

sodium chlorite for conversion of the aldehyde to the

carboxylic acid (S)-21 in 75% yield. Finally, hydrogenolysis of

.2 kg, 8.68 mol, 1.0 equiv) was dissolved in DMSO (2.4 L) and THF

(S)-acid 21 was performed using Pd/C as catalyst in aqueous

(1.2 L) under nitrogen. The resulting solution was slowly added to a

THF. The resulting mixture was treated with hydrochloric acid

and isopropanol to enable isolation of foscarbidopa (4) in 97%

yield and >99% ee. Thus, starting from commercially available

stirred mixture of sodium hydride (60% w/w, 0.768 kg, 19.08 mol, 2.2

equiv) in DMSO (6 L) and THF (2.4L) at 25−30 °C over a period of

1 h. The reaction mixture was stirred for 1 h. Benzyl chloride (1.20 kg,

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

Article

.55 mol, 1.1 equiv) was slowly added to the reaction mixture at 25−

0 °C over a period of 1 h. The resulting mixture was stirred, and the

DMSO-d ): δ 9.0 (bs), 7.62 (t, J = 1.4 Hz, 1H), 7.50−7.18 (m, 23H),

13 1

5.25−5.02 (m, 8H), 3.73 (s, 3H); C{ H} NMR (100 MHz, DMSO-

d ): δ 166.1, 155.1, 149.7 (d, J = 5.2 Hz), 140.4 (d, J_C_P= 7.3 Hz),

progress of the reaction was monitored by HPLC. After about 30 min,

the reaction mixture was slowly quenched into precooled water (5−

137.1, 136.6, 136.1 (d, J_C_P= 7.1 Hz), 131.61, 131.60 (d, J_C_P= 1.7 Hz),

131.9, 131.63, 131.62, 128.90, 128.88, 128.8, 128.5, 128.39, 128.35,

0 °C) under a steady ﬂow of nitrogen, and the mixture was stirred

for an additional 10 min. The pH of the reaction mixture was adjusted

to 3−4 using 6 N HCl solution and extracted 3 times with ethyl

acetate (2 × 12 L, 1 × 6 L). The combined ethyl acetate solution was

washed 3 times with puriﬁed water (3 × 12 L) and aqueous sodium

chloride solution (3.6 L). The organic layer was dried over sodium

sulfate and concentrated in vacuo. The residue was cooled to 25−30

128.2, 128.1, 126.3, 123.6 (d, J_C_P= 1.6 Hz), 121.7 (d, J = 3.0 Hz),

116.3, 70.6, 69.8 (d, J_C_P= 5.9 Hz), 66.5, 60.2, 52.7; P{ H} NMR

(162 MHz, DMSO-d ): δ −6.17; LC-MS (APCI) m/z: [M + H]

694.2; HRMS (ESI) m/z: [M + H] calcd for C H NO P 694.2206;

found 694.2205.

Methyl (S)-2-Amino-3-(3-hydroxy-4-(phosphonooxy)-

phenyl)propanoate (17). The enamine (6, E/Z Ratio: 10:1,

446.31 g, 521 mmol, 1.0 equiv) and 3.6 L of THF were charged to a

7.5 L reactor, and the solution was sparged with nitrogen. To another

7.5 L reactor, 1,2-bis[(2S,5S)-2,5-diethylphospholano]benzene-(1,5-

cyclooctadiene)-rhodium(I) tetraﬂuoroborate (16, 3.44 g, 5.21 mmol,

0.01 equiv) was charged, and the system was purged with nitrogen.

THF (about 0.4 L) was charged, and the catalyst solution was sparged

with nitrogen. To this catalyst solution, the THF solution of 6 was

charged, and the reactor (including transfer lines) was purged with

hydrogen. The reaction was stirred at 35 °C under 100 psig of

hydrogen. After 20 h, HPLC revealed complete consumption of

starting material 6. The reaction mixture was transferred into a 12 L

extractor, and 3.6 L of ethyl acetate were added. The solution was

°C, MTBE (0.96 L) and THF (0.3 L) were added, and the mixture

was stirred for 10 min. The resulting slurry was heated to 40 °C,

mixed for 1 h, cooled to 10−15 °C, and stirred for additional 30 min.

The resulting slurry was ﬁltered to collect the solid, and the wet cake

was washed with MTBE (0.3 L). The solid was dried under reduced

pressure at 45−50 °C for 6 h to give 1.15 kg of 3-(benzyloxy)-4-

hydroxybenzaldehyde (7) as a pale brown solid in 58% yield. By

following this lab scale procedure, a pilot scale batch of 7 (17.7 kg)

was manufactured starting from 20.0 kg of 3,4-dihydroxybenzaldehyde

(

13) in 53% yield. Purity by HPLC: 98.5 area%; H NMR (400 MHz,

DMSO-d ): δ 10.34 (s, 1H), 9.74 (1 H), 7.50−7.31 (m, 7H), 7.01−

.99 (m, 1 H), 5.19 (s, 2 H); 4′-Regioisomer of 7 by GC: 0.08%, Loss

on Drying: 0.4%.

Dibenzyl (2-(Benzyloxy)-4-formylphenyl)phosphate (15).

To a solution of 3-(benzyloxy)-4-hydroxybenzaldehyde (7, 10.0 g,

washed with aqueous 5 wt % cysteine/8 wt % NaHCO solution (2 ×

3.7 L) followed by 3.6 L of 5 wt % aqueous NaCl solution. The

organic layer was separated and stirred overnight with 43.4 g of ENO-

PC activated carbon at room temperature under nitrogen. The

mixture was ﬁltered, and the ﬁltrate was concentrated to aﬀord methyl

(S)-3-(3-(benzyloxy)-4-((bis(benzyl-oxy)phosphoryl)oxy)phenyl)-2-

(((benzyloxy)-carbonyl)-amino) propanoate (5, 420.1 g, 94% yield

and >99.5% ee) as an oil, which was used directly in the next step.

To a 150 mL Parr hydrogenator were added 10 wt % (on a dry

3.8 mmol, 1.0 equiv) in acetonitrile (100 mL), tetrabenzyl

pyrophosphate (14, TBPP, 24.8 g, 46.0 mmol, 1.05 equiv) was

added at room temperature under nitrogen. The mixture was cooled

to about 4 °C and treated with 1,8-diazabicyclo(5,4,0)undec-7-ene

(

DBU, 7.67 g, 50.4 mmol, 1.15 equiv). After the addition was

complete, the reaction mixture was warmed to room temperature and

mixed for 1 h. The mixture was quenched with water (400 mL) and

extracted with MTBE (3 × 100 mL). The combined organic layers

basis) of 5% Pd/C (1.33 g, catalyst contains 63.6% H

O) and a 2.9 wt

were washed with saturated aqueous NaHCO solution (150 mL),

% aqueous NaHCO solution (20.7 g). A portion of the above

water (150 mL), and saturated aqueous NaCl solution (150 mL). The

organic phase was concentrated to aﬀord 20.7 g of dibenzyl (2-

prepared (S)-amine 5 (5.70 g) was dissolved in THF (48.5 mL) and

transferred to the reactor. The hydrogenator was pressurized with

argon to 60 psig, and then the pressure was vented to 10 psig (× 6).

In a similar fashion, the hydrogenator was then pressure purged with

hydrogen (× 3) and a ﬁnal ﬁll of hydrogen to 50 psig followed by

agitation at rt for at least 2 h. Following the completion of reaction as

determined by HPLC, the mixture was ﬁltered to remove the catalyst.

The wet cake was rinsed with water (4.1 mL, 2 mL/g relative to

theoretical yield of product). The biphasic reaction mixture was

washed with MTBE (2 × 16 mL; discarded), and the aqueous layer

was treated with 6 M HCl to adjust the pH to 1.8. To this slurry, i-

PrOH (73 mL) was added to bring the ﬁnal solvent composition to

(

benzyloxy)-4-formylphenyl)phosphate (15) in 93% yield as a pale-

yellow oil. By following this lab scale procedure, a pilot scale batch of

5 (49.1 kg) was manufactured starting from 25.6 kg of 3-

benzyloxy)-4-hydroxybenzaldehyde (7) in 89.6% yield. Purity by

(

HPLC: 97.0 area%; H NMR (400 MHz, DMSO-d ): δ 9.93 (s, 1H),

.69 (dd, J = 1.8, 0.9 Hz, 1H), 7.56 (dd, J = 8.1, 1.8 Hz, 1H), 7.49−

.41 (m, 3H), 7.38−7.24 (m, 13H), 5.22 (s, 2H), 5.11 (dd, J = 8.2,

.1 Hz, 4H); ¹³C{ H} NMR (100 MHz, DMSO-d ): δ 192.3, 150.6,

50.5, 144.5, 144.4, 136.5, 136.0, 135.9, 134.6, 129.0, 128.9, 128.9,

28.8, 128.6, 128.4, 128.3, 124.5, 124.5, 122.2, 122.2, 113.9, 70.8,

0.1, 70.0, 69.1, 69.0; P{ H} NMR (162 MHz, DMSO-d ): δ

3:1 i-PrOH/H O and stirring was continued overnight. The

−

6.63; LC-MS (ESI) m/z: [M + H] 489.0; HRMS (ESI) m/z: [M +

crystallization slurry was ﬁltered, and the wet cake solids were washed

with i-PrOH. The white solid was dried in the vacuum oven at 50 °C

to aﬀord 1.72 g of methyl (S)-2-amino-3-(3-hydroxy-4-(phosphono-

oxy)phenyl)propanoate (17) in 85% yield and 98.7% ee. By following

this lab procedure, a pilot scale batch of (S)-ester 17 (23.5 kg) was

manufactured starting from 40.0 kg of enamine 6 in 74.3% yield for

two steps. Purity by HPLC: 99.7 area%; Assay: 87.7% (w/w);

H] calcd for C H O P 489.1467; found 489.1481.

Methyl (Z)-3-(3-(Benzyloxy)-4-((bis(benzyloxy)phosphoryl)-

oxy)phenyl)-2-(((benzyloxy)carbon-yl)amino)acrylate (6). To a

cooled solution of (±)-benzyloxycarbonyl-α-phosphonoglycine-

trimethyl ester (8, 31.1 g, 94 mmol, 1.1 equiv) and dibenzyl (2-

(

benzyloxy)-4-formylphenyl)phosphate (15, 44.3 g, 85 mmol, 1.0

equiv) in dichloromethane (443 mL) was added 1,1,3,3-tetramethyl-

guanidine (TMG, 11.78 g, 102 mmol, 1.2 equiv) at about 2 °C under

nitrogen. The reaction mixture was allowed to warm to room

temperature. After stirring overnight, the mixture was washed with

water (3 × 222 mL), and the organic phase was concentrated to

aﬀord 68.9 g of crude product 6. The residue was stirred with 40.5 g

of silica gel 60 in 689 mL of ethyl acetate for 1 h and ﬁltered to

remove polar impurities. The ﬁltrate was concentrated and suspended

in MTBE (350 mL) at 4 °C and stirred for 1 h. The resulting slurry

was ﬁltered, and the wet cake was washed with cold MTBE. The solid

was dried in the vacuum oven at 40 °C overnight to aﬀord 50.4 g

Residual Solvents: 0.5% (w/w); Mp = 210.4 °C (dec); H NMR (400

MHz, D O): δ 7.23 (dd, J = 8.2, 1.3 Hz, 1H), 6.85 (d, J = 2.1 Hz,

1H), 6.78 (dd, J = 8.3, 2.2 Hz, 1H), 4.39 (dd, J = 7.9, 5.5 Hz, 1H),

3.85 (s, 3H), 3.28 (dd, J = 14.7, 5.5 Hz, 1H), 3.13 (dd, J = 14.6, 7.9

Hz, 1H); C{ H} NMR (100 MHz, D O): δ 170.0, 147.2 (d, J

5.5 Hz), 139.7 (d, J_C_P= 6.6 Hz), 130.3 (d, J_C_P= 1.6 Hz), 121.6 (d, J_C_P

= 7.5 Hz), 121.5 (d, J = 3.3 Hz), 117.7 (d, J_C_P= 0.8 Hz), 53.9, 53.5,

34.9; P{ H} NMR (162 MHz, D O) δ −2.95; LC-MS (ESI) m/z:

[M + H] 292.00. Anal. Calcd for C H NO P·1.0 HCl·1.0 H O: C,

34.93; H, 4.92; N, 4.07. Found: C, 34.93; H, 4.75; N, 4.21.

Foslevodopa (3). To a solution of (S)-methyl 2-amino-3-(3-

hydroxy-4-(phosphonooxy)phenyl)-propanoate (17, 10.0 g, 34.3

mmol) in 40 mL of water was added 22.9 mL (4.0 equiv) of 6 N

NaOH solution at 15−20 °C. When the pH reached 7−8, the

solution was passed through a ﬁlter for clariﬁcation, and the remaining

(

85% yield) of enamine (6, E/Z Ratio: 10:1) as a white solid. By

following this lab scale procedure, the pilot scale batch of enamine 6

49.92 kg) was manufactured starting from 48.0 kg of phosphate ester

(

5 in 63.0% yield. Purity by HPLC: 98.0 area%; H NMR (400 MHz,

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

Article

NaOH solution was added. After the addition was complete, the

reaction mixture was stirred at rt for 60 min (pH = 12.06). The

mixture was acidiﬁed with 6 N HCl (137 mmol, 22.89 mL) to a ﬁnal

pH of 1.8. After 10 min, the reaction mixture became cloudy and 200

mL of i-PrOH were added. The resulting slurry was stirred for 30 min,

and the solid was ﬁltered and washed with i-PrOH. The wet cake was

dried under vacuum at 40 °C overnight to aﬀord foslevodopa [(S)-2-

amino-3-(3-hydroxy-4-(phosphonooxy)phenyl)propanoic acid] (3.

diastereomer was not determined.) By following this lab scale

procedure, the pilot scale batch was manufactured using the 3-(3-

(benzyloxy)-4-hydroxyphenyl)-2-methylpropanal (18) solution pre-

pared from 75.1 kg of 2-(benzyloxy)-4-bromophenol (11), and 103.1

kg of adduct 19 was obtained in 64% yield. Potency: 60.0% by Q-

NMR; Mp > 166 °C (dec); H NMR (400 MHz, D O): δ 7.50−7.34

(m, 5H; A/B), 6.92 (m, 1H; A/B), 6.86 (dd, J = 8.0, 4.0 Hz, 1H; A/

B), 6.76 (dd, J = 8.0, 4.0 Hz, 1H; A/B), 5.20 (s, 0.67H; B), 5.19 (s,

1.33H; A), 4.29−4.24 (m, 1H; A/B), 3.11−3.05 (m, 0.33H; B),

2.70−2.62 (m, 0.67H; A), 2.54−2.46 (m, 0.67H; A), 2.40−2.24 (m,

1H; A/B), 2.24−2.12 (m, 0.33H; B), 0.95 (d, J = 8.0 Hz, 2H; A), 0.85

.85 g) in 87% yield. By following this procedure, a pilot scale

batch of FLD drug substance (3, 29.3 kg) was manufactured starting

from 39.9 kg of (S)-ester 17 in 77.3% yield. Chiral purity: 99.6% ee;

Mp = 215.5 °C (dec); H NMR (400 MHz, D O): δ 7.23 (dd, J = 8.3,

(d, J = 8.0 Hz, 1H; B); C{ H} NMR (100 MHz, D O): δ 145.5,

.3 Hz, 1H), 6.90 (dd, J = 2.3, 0.8 Hz, 1H), 6.82 (dd, J = 8.3, 2.2 Hz,

H), 4.23 (dd, J = 8.3, 5.3 Hz, 1H), 3.28 (dd, J = 14.7, 5.2 Hz, 1H),

.10 (dd, J = 14.6, 8.1 Hz, 1H); C{ H} NMR (100 MHz, D O): δ

71.9, 147.2, 139.5 (d, J_C_P= 6.7 Hz), 130.9 (d, J_C_P= 1.5 Hz), 121.56

145.3, 143.6, 143.5, 136.5, 133.1, 132.8, 128.7, 128.6, 128.3, 128.2,

128.1, 122.8, 122.7, 116.5, 116.1, 115.6, 115.5, 87.6, 85.7, 71.0, 70.9,

39.3, 37.1, 36.2, 36.0, 15.4, 12.3; LC-MS (APCI) m/z: [M + H −

SO Na − H O] 253.0. Anal. Calcd for C H NaO S·1.5 NaHSO : C,

17 19

(

d, J_C_P= 1.1 Hz), 121.53 (d, J = 5.2 Hz), 117.8 (d, J_C_P= 1.0 Hz),

38.49; H, 3.90. Found: C, 38.27; H, 3.50.

4.4, 35.1; P{ H} NMR (162 MHz, D O): δ − 3.10; LC-MS

Dibenzyl (2-(Benzyloxy)-4-(2-methyl-3-oxopropyl)phenyl)-

phosphate (10). A 500 mL three-neck round-bottom ﬂask equipped

with a thermocouple and an overhead stirrer was charged with sodium

3-(3-(benzyloxy)-4-hydroxyphenyl)-1-hydroxy-2-methylpropane-1-

sulfonate solution (19, 15.1 g, 63.3 w/w%, 23.2 mol, 1.0 equiv),

APCI) m/z: [M + H] 278.04, [M + H − H O] 260.10; Anal.

Calcd for C H NO P·0.4 H O: C, 37.89; H, 4.56; N, 4.91. Found: C,

7.77; H, 4.29; N, 4.63.

-(3-(Benzyloxy)-4-hydroxyphenyl)-2-methylpropanal (18).

A 500 mL three-neck round-bottom ﬂask equipped with a

thermocouple, stir bar, and reﬂux condenser was charged with 2-

NaHCO (16.97 g, 202 mmol, 8.70 equiv), water (155 mL), and ethyl

acetate (140 mL). The resulting biphasic suspension was stirred

vigorously at room temperature. After the complete consumption of

starting material, the reaction was transferred to a separatory funnel,

and the layers were separated. The organic layer was washed with

brine (75 mL), dried (sodium sulfate), ﬁltered, and concentrated in

vacuo to provide 6.29 g of 3-(3-(benzyloxy)-4-hydroxyphenyl)-2-

methylpropanal (18). This intermediate (18) was taken into a 250

mL three-neck ﬂask, equipped with a thermocouple and an overhead

stirrer, acetonitrile (63 mL) and tetrabenzyl pyrophosphate (14, 13.54

g, 24.38 mmol) were added, and the mixture was cooled to about 2

°C. To this mixture, DBU (4.55 mL, 30.2 mmol) was added dropwise,

and the resulting solution was stirred for 1 h. The reaction was diluted

with water (65 mL) and extracted twice with MTBE (130 mL). The

combined organic layers were washed with water (65 mL), 5% NaCl

solution (30 mL), dried (sodium sulfate), ﬁltered, and concentrated in

vacuo to provide crude dibenzyl (2-(benzyloxy)-4-(2-methyl-3-

oxopropyl)phenyl)phosphate (10, 11.38 g) in 92% yield as yellow

oil. By following this lab scale procedure, a large pilot scale batch was

manufactured starting from 97.3 kg of bisulﬁte adduct 19, and

dibenzyl (2-(benzyloxy)-4-(2-methyl-3-oxopropyl)phenyl)phosphate

(10) was obtained as a solution in acetonitrile (HPLC assay: 22 wt

(

benzyloxy)-4-bromophenol (11, 25.04 g, 89.71 mmol, 1.0 equiv),

tris(dibenzylideneacetone) palladium (1.23 g, 1.34 mmol, 1.5 mol %),

and tri-tert-butylphosphonium tetraﬂuoroborate (0.88 g, 3.02 mmol,

.36 mol %). The ﬂask was purged with nitrogen for 1 h. During this

time, a second ﬂask was charged with 1,4-dioxane (200.0 mL), 2-

methylprop-2-en-1-ol (12, 6.95g, 8.30 mL, 96.34 mmol, 1.1 equiv),

and N-cyclohexyl-N-methylcyclohexanamine (Cy NMe, 30.0 mL, 140

mmol, 1.55 equiv), and the solution was sparged with nitrogen for 1 h.

The dioxane solution was transferred via cannula to the ﬂask

containing bromide 11, palladium catalyst, and ligand, and the

reaction mixture was heated at 100 °C for 1 h. The reaction mixture

was cooled to 35 °C and diluted with ethyl acetate (250 mL) and 1.0

M HCl (250 mL). The biphasic mixture was stirred for 10 min, and

the layers were separated. The organic phase was removed from the

reactor, and the aqueous phase was returned. Ethyl acetate (150 mL)

was added to the aqueous phase, and the mixture was agitated for 10

min. The aqueous layer was drained, and the combined organic

phases were washed with a 5% N-acetylcysteine/8% NaHCO mixture

(

250 g). The organic solution was ﬁltered through Celite, and the

solution containing 3-(3-(benzyloxy)-4-hydroxyphenyl)-2-methyl-

propanal (18) was used directly in the next reaction. A small portion

of this solution was dried to aﬀord 18 as an oﬀ-white solid. By

following this lab scale procedure, a pilot scale batch starting from

% and 85.3% yield), which was used directly in the next step. H

NMR (400 MHz, CDCl ): δ 9.68 (d, J = 1.2 Hz, 1H), 7.43−7.38 (m,

2H), 7.34−7.19 (m, 13H), 7.14 (dd, J = 8.0, 1.2 Hz, 1H), 6.78 (dd, J

= 2.0, 1.2 Hz, 1H), 6.68 (dd, J = 8.0, 2.0 Hz, 1H), 5.08 (s, 2H), 5.06

(s, 2H), 5.03 (s, 2H), 3.02 (dd, J = 13.6, 5.6, Hz, 1H), 2.65−2.55 (m,

1H), 2.53 (dd, J = 13.6, 8.0 Hz, 1H), 1.06 (d, J = 7.2 Hz, 3H);

5.1 kg of 2-(benzyloxy)-4-bromophenol (11) was performed and the

-(3-(benzyloxy)-4-hydroxyphenyl)-2-methylpropanal (18) solution

was directly used in the next step pilot scale batch. Residual Water by

Karl Fischer: 3.3 wt %. Mp 45.0−47.5 °C; H NMR (400 MHz,

C{ H} NMR (100 MHz, CDCl ): δ 204.1, 148.80, 149.75, 138.7,

DMSO-d ): δ 9.59 (d, J = 1.5 Hz, 1H), 8.79 (s, 1H), 7.47−7.39 (m,

136.9, 136.8, 136.4, 135.8, 135.7, 128.57, 128.55, 128.52, 128.50,

128.4, 128.1, 128.0, 127.81, 127.77, 121.59, 121.57, 115.22, 115.21,

H), 7.38−7.22 (m, 3H), 6.80 (d, J = 2.1 Hz, 1H), 6.67 (d, J = 8.0

Hz, 1H), 6.55 (dd, J = 8.0, 2.0 Hz, 1H), 5.03 (s, 2H), 2.85 (dd, J =

71.0, 69.84, 69.78, 47.97, 47.96, 36.3, 13.3; P{ H} NMR (162 MHz,

3.7, 6.1 Hz, 1H), 2.64−2.53 (m, 1H), 2.41 (dd, J = 13.7, 8.0 Hz,

CDCl ): δ −5.88; LC-MS (APCI) m/z: [M + H] 531.2. Anal. Calcd

H), 0.86 (d, J = 6.9 Hz, 3H); C{ H} NMR (100 MHz, DMSO-d ):

for C H O P·1.0 H O: C, 67.87; H, 6.06 Found: C, 67.68; H, 5.75.

31 31

δ 205.7, 146.8, 145.8, 137.8, 130.1, 128.7, 128.14, 128.09, 122.1,

16.1, 115.7, 70.4, 47.7, 35.9, 13.2; LC-MS (APCI) m/z: [M + H −

Dibenzyl (S)-1-(1-(3-(Benzyloxy)-4-((bis(benzyloxy)-

phosphoryl)oxy)phenyl)-2-methyl-3-oxo-propan-2-yl)-

hydrazine-1,2-dicarboxylate (9). A 500 mL three-neck round-

bottom ﬂask equipped with a thermocouple were charged with (R)-5-

(pyrrolidin-2-yl)-1H-tetrazole (20, 0.15 g, 1.07 mmol, 4.97 mol %),

acetonitrile (40 mL), and triﬂuoroacetic acid (0.084 mL, 1.07 mmol)

at room temperature under nitrogen. To this mixture was added (E)-

dibenzyl diazene-1,2-dicarboxylate (8.25 g, 27.7 mmol, 1.29 equiv)

followed by a solution of dibenzyl (2-(benzyloxy)-4-(2-methyl-3-

oxopropyl)phenyl) phosphate (10, 11.4 g, 21.49 mmol, 1.0 equiv) in

acetonitrile (70 mL) added via cannula. The resulting solution was

stirred at room temperature for 15 h. After the complete consumption

of starting material, the reaction mixture was diluted with acetonitrile

(88 mL), and water (58 mL) was added to precipitate the product;

H O] 253.0. Anal. Calcd for C H O ·0.15 H O: C, 74.79; H, 6.76.

Found: C, 74.82; H, 6.68.

17 18

Sodium 3-(3-(Benzyloxy)-4-hydroxyphenyl)-1-hydroxy-2-

methylpropane-1-sulfonate (19). To the solution of crude (3-

(

benzyloxy)-4-hydroxyphenyl)-2-methylpropanal (18) prepared

above sodium bisulﬁte (18.7 g, 179 mmol) was added, and the

mixture was heated to 40 °C for 13 h. The resulting precipitate was

ﬁltered, and the solid was washed with ethyl acetate (3 × 100 mL) to

give 36.0 g (64%) of sodium 3-(3-(benzyloxy)-4-hydroxyphenyl)-1-

hydroxy-2-methylpropane-1-sulfonate (19) as a colorless solid as a

5:35 mixture of diastereomers (A and B), assigned based on

NMR chemical shifts. (The relative stereochemistry of each

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

https://pubs.acs.org/doi/10.1021/acs.joc.1c00905.

Article

the resulting slurry was stirred overnight at room temperature. The

solid was collected by ﬁltration, and the wet cake was washed with 28

wt % water in acetonitrile (30 mL) to provide dibenzyl (S)-1-(1-(3-

biphasic ﬁltrate was diluted with MTBE (370 mL) and stirred for 10

min. The layers were separated (organic discarded), and the aqueous

layer was washed with MTBE (370 mL) in portions. The pH of the

aqueous layer was adjusted with 6 M HCl to 1.9, followed by seeding

with 0.1 wt % of foscarbidopa (4) to induce nucleation. Isopropanol

(1326 mL) was added to the slurry and mixed for 5 h at room

temperature. The slurry was ﬁltered, and the wet cake was washed

with isopropanol (370 mL) followed by air-drying on the funnel for 2

h to provide 38.5 g (97%) of foscarbidopa (4, (S)-2-hydrazinyl-3-(3-

hydroxy-4-(phosphonooxy)phenyl)-2-methylpropanoic acid) as a

trihydrate and white solid. By following this lab scale procedure, a

pilot scale batch was manufactured starting with 40.0 kg of (S)-acid

(21), and 14.3 kg of FCD (4) were obtained in 86% yield as the

(

benzyloxy)-4-((bis(benzyloxy)-phosphoryl)oxy)phenyl)-2-methyl-3-

oxopropan-2-yl)hydrazine-1,2-dicarboxylate (9, 8.9 g, 50% yield,

99% ee) as a white solid. By following this lab scale procedure, a

large pilot scale batch was manufactured by using the acetonitrile

solution of dibenzyl (2-(benzyloxy)-4-(2-methyl-3-oxopropyl)phenyl)

phosphate (10, HPLC assay: 22 wt %) and 63.0 kg of (S)-hydrazine

(

9) was obtained in 29.2% overall yield for two steps. Mp: 114.0−

14.6 °C; [α]

−71.7° (c, 1.0; CHCl ); H NMR (400 MHz,

DMSO-d ): δ 9.82 (s, 1H), 9.55 (s, 1H), 7.46−7.22 (m, 25H), 7.11−

.95 (m, 2H), 6.76−6.67 (m, 1H), 5.27−4.96 (m, 10H), 3.20−2.87

13 1

(

m, 2H), 1.14−0.98 (br s, 3H); C{ H} NMR (100 MHz, DMSO-

trihydrate. Purity by HPLC: 99.7 area%; Assay: 99.6 wt %; ee >

d ): δ 198.7, 157.5, 149.43, 149.38, 138.5, 138.4, 137.0, 136.7, 136.6,

99.9%; Mp 167.2 °C (dec); H NMR (400 MHz, D

O): δ 7.21 (d, J =

36.2, 136.1, 128.87, 128.86, 128.81, 128.79, 128.78, 128.54, 128.46,

28.4, 128.3, 128.23, 128.21, 128.17, 128.16, 128.1, 123.5, 121.1,

8.0 Hz), 6.84 (d, J = 2.0 Hz, 1H), 6.77 (dd, J = 8.0, 2.0 Hz, 1H), 3.18

(d, J = 16.0 Hz, 1H), 3.00 (d, J = 16.0 Hz, 1H), 1.53 (s, 3H); C{ H}

NMR (100 MHz, D O): δ 175.1, 146.8 (d, JCP = 5.2 Hz), 139.5 (d,

JCP = 6.7 Hz), 130.4 (d, JCP = 1.2 Hz), 122.3 (d, JCP = 1.5 Hz), 121.2

⁽^d^,^JCP = 3.0 Hz), 118.5, 66.3, 41.3, 18.9; P{ H} NMR (162 MHz,

D O): δ −3.13; LC-MS (APCI) m/z: [M + H] 307.1. Anal. Calcd

for C H N O P·3.2 H O: C, 33.01; H, 5.93; N, 7.70. Found: C,

17.3, 70.5, 70.4, 69.67, 69.65, 69.64, 69.61, 69.59, 67.02, 66.95;

P{ H} NMR (162 MHz, DMSO-d ) δ −5.98, −6.04 (2.6:1.0

−

mixture of rotamers); LC-MS (ESI negative mode) m/z: [M − H]

27.0. Anal. Calcd for C H N O P·3.3 H O: C, 63.55; H, 5.85; N,

.15. Found: C, 63.34; H, 5.61; N, 3.22.

S)-3-(3-(Benzyloxy)-4-((bis(benzyloxy)phosphoryl)oxy)-

10 15

33.38; H, 5.68; N, 7.33.

(

phenyl)-2-(1,2-bis((benzyloxy)carbon-yl)-hydrazinyl)-2-

methylpropanoic Acid (21). A 100 mL three-neck round-bottom

ﬂask equipped with a thermocouple was charged with dibenzyl (S)-1-

ASSOCIATED CONTENT

sı Supporting Information

The Supporting Information is available free of charge at

■

(

1-(3-(benzyloxy)-4-((bis(benzyloxy)-phosphoryl)oxy)phenyl)-2-

methyl-3-oxopropan-2-yl)hydrazine-1,2-dicarboxylate (9, 5.10 g, 6.15

mmol, 1.0 equiv), acetonitrile (50 mL), and dimethyl sulfoxide

(

DMSO, 1.00 mL, 14.1 mmol, 2.29 equiv) under nitrogen. The white

Copies of NMR spectra, HPLC and MS data

descriptions (PDF )

suspension was stirred, and a mixture of water (2.0 mL) and sodium

dihydrogen phosphate monohydrate (1.78 g, 12.90 mmol, 2.09 equiv)

was added to the reaction. Following this addition, a solution of

sodium chlorite (2.88 g (80 wt %), 25.5 mmol, 4.15 equiv) in water

AUTHOR INFORMATION

(

2.0 mL) was added dropwise. The resulting cloudy reaction turned

■

yellow and became clearer as the reaction proceeded. After 90 min,

Corresponding Author

the reaction was quenched with a solution of sodium sulﬁte (1.60 g,

David R. Hill − Process Research & Development, AbbVie Inc.,

North Chicago, Illinois 60064, United States; orcid.org/

2.7 mmol) in water (6.0 mL). The reaction was stirred for 20 min

and transferred into a separatory funnel with isopropyl acetate (50

mL) and water (50 mL). The organic layer was washed with brine (70

mL) and concentrated to a volume of about 10 mL. The reaction ﬂask

was held at 2−8 °C for 16 h. The solid that formed was collected,

washed with 20 mL of isopropyl acetate, and dried in vacuo to give

0000-0002-4766-7211 ; Email: david.hill@abbvie.com

Authors

Alexander D. Huters − Process Research & Development,

AbbVie Inc., North Chicago, Illinois 60064, United States

James Stambuli − Process Research & Development, AbbVie

Inc., North Chicago, Illinois 60064, United States

.92 g of (S)-acid (21) in 75% yield as white solid. By following this

lab scale procedure, a large pilot scale batch was manufactured by

using dibenzyl (S)-1-(1-(3-(benzyloxy)-4-((bis(benzyloxy)-

phosphoryl)oxy)phenyl)-2-methyl-3-oxopropan-2-yl)hydrazine-1,2-

dicarboxylate (9, 49.0 kg) and 42.8 kg of (S)-acid (21) was obtained

Russell C. Klix − Process Research & Development, AbbVie

Inc., North Chicago, Illinois 60064, United States

in 85.7% yield. Mp 120.3 °C−120.9 °C; [α]

−84.5° (c, 1.0;

Mark A. Matulenko − Drug Discovery Science & Technology,

AbbVie Inc., North Chicago, Illinois 60064, United States

Vincent S. Chan − Process Research & Development, AbbVie

Inc., North Chicago, Illinois 60064, United States; Present

Address: Merck & Co., Inc., 213 East Grand Avenue,

South San Francisco, California 94080

CHCl ); H NMR (400 MHz, DMSO-d ): δ 7.43−7.21 (m, 25H),

.11−6.90 (m, 2H), 6.80−6.64 (m, 1H), 5.18−4.99 (m, 10H), 3.28−

13 1

.92 (m, 2H), 1.26 (br s, 3 H); C{ H} NMR (175 MHz, DMSO-

d₆): δ 175.8, 160.2, 159.8, 158.2, 157.7, 152.1, 152.0, 151.9, 141.5,

41.4, 141.3, 141.2, 139.8, 139.7, 139.5, 139.0, 138.96, 138.94, 138.92,

36.8, 131.7, 131.63, 131.61, 131.59, 131.58, 131.53, 131.3, 131.17,

31.15, 131.1, 131.0, 130.98, 130.97, 130.96, 130.8, 130.5, 126.4,

26.0, 123.8, 123.6, 120.3, 119.9, 73.3, 73.2, 72.43, 72.42, 72.40,

2.39, 72.37, 70.4, 70.2, 69.9, 69.3, 24.8; LC-MS (ESI) m/z: [M +

Justin Simanis − Process Research & Development, AbbVie

Inc., North Chicago, Illinois 60064, United States

Rajarathnam E. Reddy − Operations Science & Technology,

AbbVie Inc., North Chicago, Illinois 60064, United States

Timothy B. Towne − Process Research & Development,

AbbVie Inc., North Chicago, Illinois 60064, United States

John R. Bellettini − Process Research & Development, AbbVie

Inc., North Chicago, Illinois 60064, United States

Brian J. Kotecki − Process Research & Development, AbbVie

Inc., North Chicago, Illinois 60064, United States

Benoit Cardinal-David − Process Research & Development,

AbbVie Inc., North Chicago, Illinois 60064, United States

Jianguo Ji − Process Research & Development, AbbVie Inc.,

North Chicago, Illinois 60064, United States

H] 845.1, [M + NH ] 862.0. Anal. Calcd for C H N O P·0.9

H O: C, 65.56; H, 5.48; N, 3.25. Found: C, 65.57; H, 5.42; N, 3.11.

47 45

Foscarbidopa (4). A 3.7 L Parr reactor was charged with 5 wt %

(

dry basis) of 5% Pd/C (63.6% H O, 15.0 g), water (182 mL), and 5

wt % aqueous NaHCO (215 mL). To this catalyst slurry was added a

THF solution (1090 mL) of (S)-acid 21 (109 g, 129.1 mmol, 1.0

equiv, Potency: 85%), and the reactor was purged with nitrogen

followed by hydrogen. The reactor was then repressurized to 30 psig

with hydrogen and agitated at room temperature for 1 h. Upon

completion of the reaction, hydrogen was vented, and the reactor was

purged with nitrogen. The biphasic reaction mixture was ﬁltered to

remove the catalyst, followed by rinsing with water (93 mL). The

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

orcid.org/0000-0002-9542-5356

Article

Eric A. Voight − Drug Discovery Science & Technology,

AbbVie Inc., North Chicago, Illinois 60064, United States;

Intravenously Administered Levodopa: The Mechanism of Action in

the Treatment of Parkinsonism. Ann. Neurol. 1985 , 18 , 537 −543.

(4) For Duopa treatment, see: (a) Nyholm, D. Enteral Levodopa/

Carbidopa Gel Infusion for the Treatment of Motor Fluctuations and

Dyskinesias in Advanced Parkinson ’s Disease. Expert Rev. Neurother.

Minshan Shou − Analytical Research & Development, AbbVie

Inc., North Chicago, Illinois 60064, United States

Selvakumar Balaraman − Anthem Biosciences, Bangalore 560

006 , 6 , 1403 −1411. (b) Nyholm, D. Duopa Treatment for

Advanced Parkinson ’s Disease: A Review of Efficacy and Safety.

Parkinsonism Relat. Disord. 2012 , 18 , 916 −929. (c) Wirdefeldt, K.;

Odin, P.; Nyholm. Levodopa-Carbidopa Intestinal Gel in Patients

with Parkinson ’s Disease: A Systematic Review. CNS Drugs 2016, 30,

381−404.

99 Karnataka, India

Abhishek Ashok − Anthem Biosciences, Bangalore 560 099

Karnataka, India

Soma Ghosh − Anthem Biosciences, Bangalore 560 099

Karnataka, India

(5) For prodrug design, see: (a) Stella, V. J.; Nti-Addae, K. W.

Prodrug Strategies to Overcome Poor Water Solubility. Adv. Drug

Delivery Rev. 2007, 59, 677−694. (b) Rautio, J.; Kumpulainen, H.;

Heimbach, T.; Oliyai, R.; Oh, D.; Järvinen, T.; Savolainen, J.

Prodrugs: Design and Clinical Applications. Nat. Rev. Drug Discovery

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.joc.1c00905

Notes

2008 , 7 , 255 −270. (c) Huttunen, K.; Raunio, H.; Rautio, J. Prodrugs:

The authors declare the following competing ﬁnancial

interest(s): AbbVie sponsored and funded the study,

contributed to the design, participated in collection, analysis,

and interpretation of data, as well as in writing, reviewing, and

approval of the ﬁnal publication. A.D.H., J.S., R.C.K., M.A.M.,

V.S.C., J.S., D.R.H., R.E.R., T.B.T., J.R.B., B.J.K., B.C.D., J.J.,

E.V., and M.S. are, or were, employees of AbbVie and may own

AbbVie stock. V.S.C. is currently an employee of Merck and

had no additional conﬂicts to disclose. S.B., A.A., and S.G. are

employees of Anthem Biosciences and have no additional

conﬂicts to disclose.

From Serendipity to Rational Design. Pharmacol. Rev. 2011, 63, 750−

771. (d) Jornada, D. H.; Fernandes, G.; F, D.; Chiba, D. E.; de Melo,

T. R. F.; dos Santos, J. L.; Chung, M. C. The Prodrug Approach: A

Successful Tool for Improving Drug Solubility. Molecules 2016, 21 , 42.

(e) Rautio, J.; Meanwell, N. A.; Di, L.; Hageman, M. J. The

Expanding Role of Prodrugs in Contemporary Drug Design and

Development. Nat. Rev. Drug Discovery 2018, 17, 559−587.

(6) For phosphate prodrugs, see: (a) Wiemer, A. J.; Wiemer, D. F.

Prodrugs of Phosphonates and Phosphates: Crossing the Membrane

Barrier. Top. Curr. Chem. 2014 , 360 , 115. (b) Hecker, S. J.; Erion, M.

D. Prodrugs of Phosphates and Phosphonates. J. Med. Chem. 2008 ,

51 , 2328 −2345. (c) Schultz, C. Prodrugs of Biologically Active

Phosphate Esters. Bioorg. Med. Chem. 2003 , 11 , 885 −898.

(d) Sanches, B. M. A.; Ferreira, E. I. Is Prodrug Design an Approach

to increase water Solubility? Int. J. Pharm. 2019, 568, 118498−

ACKNOWLEDGMENTS

■

The authors wish to acknowledge and thank Jeﬀrey Bien

AbbVie, Process R&D) for excellent support in procuring

materials and valuable suggestions.

118510.

(

(7) For Foslevodopa (3) and Foscarbidopa (4) Prodrug discovery

and development, see: (a) Cardinal-David, B.; Chan, V. S.; Dempah,

K. E.; Enright, B. P.; Henry, R. F.; Ho, R.; Huang, Ye.; Huters, A. D.;

Klix, R. C.; Krabbe, S. W.; Kym, P. R.; Lao, Y.; Lou, X.; Mackey, S. E.;

Matulenko, M. A.; Mayer, P. T.; Miller, C. P.; Stambuli, J.; Voight, E.

A.; Wang, Z.; Zhang, G. G.; Stella, V. J. Carbidopa and L-Dopa

Prodrugs and Methods of Use. US 2016/9446059B2, 2016 and

references cited therein. (b) Rosebraugh, M.; Kym, P.; Liu, W.;

Facheris, M.; Benesh, J. A Novel Levodopa/Carbidopa Prodrug

REFERENCES

■

(

1) For reviews on Parkinson’s disease, see: (a) Tanner, C. M.;

Goldman, S. M. Epidemiology of Parkinson’s Disease. Neurol. Clin.

992 , 14 , 317 −335. (b) Tysnes, O.; Storstein, A. Epidemiology of

Parkinson ’s Disease. J. Neural Transm. 2017, 124, 901−905. (c) Fahn,

S. Int. Congress of Parkinson’s Disease and Movement Disorders , 1

November 2006 . (d) Goole, J.; Amighi, K. Levodopa Delivery

Systems for the Treatment of Parkinson ’s Disease: An Overview. Int.

J. Pharm. 2009, 380, 1−15.

(ABBV-951) 24-h Continuous Subcutaneous Infusion Treatment for

Parkinson’s Disease (P3.8−037). Neurology 2019, 92 (15 Supple-

ment), P3.8−037 Apr. (c) Rosebraugh, M.; Kym, P.; Liu, W.;

Facheris, M.; Benesh, J. Pharmacokinetics of ABBV-951 Following 24-

h Continuous Subcutaneous Infusion to Arm, Abdomen and Thigh

(

2) For treatment of Parkinson’s disease, see: (a) Sandler, M.;

Johnson, R. D.; Ruthven, C. R. J.; Reid, J. L.; Calne, D. B.

Transaminase is a Major Path of L-Dopa Metabolism Following

Peripheral Decarboxylase Inhibition. Nature 1974, 247, 364−366.

(Abstract). Mov. Disord. 2019, 34 (Suppl 2). (d) Rosebraugh, M.;

Voight, E.; Matulenko, M.; Moussa, E.; Jameel, F.; Lou, X.; Zhang, G.;

Mayer, P.; Stolarik, D.; Carr, R.; Enright, B.; Liu, W.; Facheris, M.;

Kym, P. Foslevodopa/Foscarbidopa: ANew Subcutaneous Treatment

for Parkinson ’s Disease. Ann. Neurol. 2021, 90, 52−61.

(8) Tang, P.; Wang, H.; Zhang, W.; Chen, F.-E. Asymmetric

Catalytic Hydrogenation of Imines and Enamines in Natural Product

Synthesis. Green Synthesis Catalysis 2020 , 1 , 26 −41.

b) Hughes, R. C.; Polgar, J. G.; Weightman, D.; Walton, J. N. L-

Dopa in Parkinsonism and the Influence of Previous Thalamotomy.

Br. Med. J. 1971, 1, 7−13. (c) Porras, G.; De Deurwaerdere, P.; Li, Q.;

Marti, M.; Morgenstern, R.; Sohr, R.; Bezard, E.; Morari, M.;

Meissner, W. G. L-Dopa-induced Dyskinesia: Beyond an Excessive

Dopamine Tone in the Striatum. Sci. Rep. 2015, 4, 3730 −3075.

d) Tolosa, E.; Martí, M. J.; Valldeoriola, F.; Molinuevo, J. L. History

(9) Bisceglia, J.; Orelle, L. R. Recent Advances in Horner-

Wadsworth-Emmons Reaction. Curr. Org. Chem. 2015 , 19 , 744 −775.

(10) Chowdari, N. S.; Barbas III, C. F. Total Synthesis of LFA-1

Antagonist BIRT-377 via Organocatalytic Asymmetric Construction

of a Quaternary Stereocenter. Org. Lett. 2005, 7, 867−870.

of Levodopa and Dopamine Agonists in Parkinson ’s Disease

Treatment. Neurology 1998 , 50 , S2 −S10. (e) Armstrong, M. J.;

Okun, M. S. Diagnosis and Treatment of Parkinson ’s Disease: A

Review. JAMA 2020, 323, 548−560.

(

3) For the levodopa (1) and carbidopa (2 ) mechanism of action,

(11) Scrivanti, A.; Bertoldini, M.; Beghetto, V.; Matteoli, U.

see: (a) Khor, S.-P.; Hsu, A. The Pharmacokinetics and Pharmacody-

namics of Levodopa in the Treatment of Parkinson ’s Disease. Curr.

Clin. Pharmacol. 2007, 2, 234−243. (b) Mena, M. A.; Casarejos, M. J.;

Solano, R. M.; García de Yébenes, J. Half a Century of L-DOPA. Curr.

Top. Med. Chem. 2009 , 9 , 880 −93. (c) Juncos, J. L. Levodopa:

Pharmacology, Pharmacokinetics, and Pharmacodynamics. Neurol.

Clin. 1992, 10 , 487 −509. (d) Nutt, J. G.; Woodward, W. R.;

Anderson, J. L. The effect of Carbidopa on the Pharmacokinetics of

Arylation of S-Methallyl alcohol Catalyzed by Pd(OAc)

Combination with P(t -Bu) : Application to Fragrance Synthesis.

Tetrahedron 2008, 64, 543−548.

(12) Taratula, O.; Hill, P. A.; Bai, Y.; Khan, N. S.; Dmochowski, I. J.

Shorter Synthesis of Trifunctionalized Cryptophane-A Derivatives.

Org. Lett. 2011, 13, 1414−1417.

(13) (a) Yang, Q.; Sheng, M.; Henkelis, J. J.; Tu, S.; Wiensch, E.;

Zhang, H.; Zhang, Y.; Tucker, C.; Ejeh, D. E. Explosion Hazards of

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

Sodium Hydride in Dimethyl Sulfoxide, N,N -Dimethylformamide,

Article

and N,N -Dimethylacetamide. Org. Process Res. Dev. 2019, 23, 2210−

217. (b) Wang, Z.; Richter, S. M.; Bellettini, J. R.; Pu, Y.-M.; Hill, D.

R. Safe Scale-up of Pharmaceutical Manufacturing Processes with

Dimethylsulfoxide as the Solvent and Reactant or a Byproduct. Org.

Process Res. Dev. 2014, 18, 1836−1842.

(

14) Nelson, T. D.; Rosen, J. D.; Bhupathy, M.; McNamara, J.;

Sowa, M. J.; Rush, C.; Crocker, L. S. Tetrabenzyl Pyrophosphate. Org.

Syntheses 2003, 80, 219−226.

(

15) Stronger bases such as n-BuLi, NaH, LHMDS, or KOt-Bu have

been reported for this reaction, but these were deemed problematic

for some of our substrates. See for example: (a) Nicolaou, M. G.;

Yuan, C.-S.; Borchardt, R. T. Phosphate Prodrugs for Amines

Utilizing a Fast Intramolecular Hydroxy Amide Lactonization. J. Org.

Chem. 1996, 61, 8636−8641. For Lewis acid catalyzed phenol

phosphate formation using N-ethyl-N,N-diisopropylamine as the base,

see: (b) Fenton, O. S.; Allen, E. E.; Pedretty, K. P.; Till, S. D.; Todaro,

J. E.; Sculimbrene, B. R. Tetrahedron 2012, 68, 9023−9028.

(

16) Endo, A.; Yanagisawa, A.; Abe, M.; Tohma, S.; Kan, T.;

Fukuyama, T. Total Synthesis of Ecteinacidin 743. J. Am. Chem. Soc.

002, 124, 6552−6554.

17) Burk, M. J. Modular Phospholane Ligands in Asymmetric

Catalysis. Acc. Chem. Res. 2000, 33 , 363 −372.

18) Buntin, S. A.; Heck, R. F. 2-Methyl-3-phenylpropanal. Org.

Synth. 1983, 61, 82−83.

19) Lindgren, B. O.; Nilsson, T. Preparation of Carboxylic Acids

from Aldehydes (Including Hydroxylated benzaldehydes) by

Oxidation with Chlorite. Acta Chem. Scand. 1973 , 27 , 888 −890.

20) Planchenault, D.; Dhal, R.; Robin, J.-P. New Agents of Biaryl

Oxidative Coupling in Fluoro Acid Medium VI. Application to the

Synthesis of Phenolic Bisbenzocyclooctadiene Lignans. Tetrahedron

1995, 51, 1395−1404.