Synthesis of stable isotope labeled PNU-95666

Article abstract of DOI:10.1002/jlcr.848

Studies towards the synthesis of stable isotope labeled PNU-95666, a CNS agent currently under development for the treatment of Parkinson's disease, is described. The synthesis of non-labeled PNU-95666 starts using a large excess of methylamine. While the

Full text of DOI:10.1002/jlcr.848

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2004; 47: 609–614.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.848
Research Article
Synthesis of stable isotope labeled PNU-95666
Zhigang Jian*
Pharmaceutical Science, Pfizer, 301 Henrietta St., 0216-300-365.2, Kalamazoo,
MI 49007-4940, USA
Summary
Studies towards the synthesis of stable isotope labeled PNU-95666, a CNS agent
currently under development for the treatment of Parkinson’s disease, is described.
The synthesis of non-labeled PNU-95666 starts using a large excess of methylamine.
While the non-labeled methylamine is inexpensive and in plentiful supply, the stable
isotope labeled form is expensive and limited availability. Different synthetic routes
and conditions, including the use of different aminating agents and various amounts
of methylamine were explored and new synthetic conditions were established.
Copyright # 2004 John Wiley & Sons, Ltd.
Key Words: stable isotope labeled PNU-95666; sumanirole; CNS agent; [13CD3]
methylamine
Introduction
PNU-95666 is a highly selective D2 agonist currently under development for
the treatment of Parkinson’s Disease.1 The stable isotope labeled form is
needed as an internal standard for LC/MS study, which requires it to be at
least three mass unit higher than the corresponding parent compound.
Many synthetic routes for the synthesis of PNU-95666 have been reported1–6
and in the most recent method the desired methylamino group was introduced
by reaction of the chiral naproxen ester 1 with methylamine (Scheme 1).6
This route looks attractive for the synthesis of stable isotope labeled PNU-
95666 since it allows us to use the commercially available 13CD3 methylamine
as the starting material. The drawback, however, is the use of a large excess of
13CD3 methylamine (22 equivalents),6 which is not only expensive but also of
limited availability. One possible solution is to use ammonia or benzylamine
instead of methylamine to form amine 4. The reaction of amine 4 with much
cheaper and widely available 13CD3 methyl iodide should afford the desired
*Correspondence to: Z. Jian, Radiosynthesis, Pfizer, 301 Henrietta St., 0216-300-365.2, Kalamazoo,
MI 49007, USA. E-mail: zhigang.jian@pfizer.com
Copyright # 2004 John Wiley & Sons, Ltd.
Received 2 April 2004
Revised 15 April 2004
Accepted 30 April 2004

610
Z. JIAN
OCH3
O
NHCH3
OH
O
N
NHCH3
Br
40% aq. MeNH2
N
N
HN
N
N
O
O
Bn
O
Bn
PNU-95666
1
2
Scheme 1. Synthetic route for non-labeled PNU-95666
OCH3
NHR
O
NH2
NH13CD3
OH
O
N
OH
OH
NH3 or
Benzylamine
Br
Selective
debenzylation
13CD3I
N
O
N
O
N
N
N
N
N
O
Bn
Bn
Bn
O
Bn
R = H or Benzyl
13CD3 2
1
3
4
Scheme 2. Proposed synthesis of stable isotope labeled PNU-95666
intermediate 2 (Scheme 2). The other solution would be to lower the amount
of 13CD3 methylamine used to convert 1 to 2.
Results and discussion
The reaction of naproxen ester 1 with ammonia was carried out with both
concentrated aqueous ammonia (22 equivalents) and with ammonia in
methanol (7 N, 22 equivalents). Only starting material was observed in both
cases after overnight reaction at room temperature. No product was formed
even under prolonged reflux. Apparently, ammonia itself is not nucleophilic
enough for the reaction to occur. While the reaction of benzylamine with
naproxen ester 1 looked attractive, we later learned that the selective
debenzylation of 3 to 4 had been tried unsuccessfully.7
As the plan to use either ammonia or bezylamine to replace methylamine
failed, we turned our attention to the use of less methylamine. Non-labeled
amine 2 was made by the reaction of 22 equivalents of aqueous methylamine
with naproxen ester 1 in acetonitrile at room temperature overnight in 83%
yield. At least three equivalents of methylamine is needed for this
transformation: one for amidation, one to open the epoxide, and one to
neutralize the liberated HBr (Scheme 3).
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 609–614

SYNTHESIS OF STABLE ISOTOPE LABELED
611
Scheme 3. Reaction mechanism
To find out how much of the reagent is practically needed for the reaction to
proceed, we started with three different experiments: stirring a solution of
naproxen ester 1 with: (1) 22 equivalents, (2) 10 equivalents, and (3) 6
equivalents of 40% methylamine in acetonitrile at room temperature. The
reaction started as a suspension (naproxen ester 1 is sparely soluble in the
reaction media) and ended up as a homogeneous solution, as the reaction
proceed to completion. All three reaction went to completion: while reaction
(1) and reaction (2) became homogeneous solutions within a reasonable 18 h
and a manageable 96 h respectively, reaction (3) took a prohibitive 216 h.
This work and the fact that there was no other good starting material
available (either labeled or non-labeled) led us to carry out the synthesis as
outlined in Scheme 4.
Commercially available 13CD3NH2 hydrochloride was first converted to the
aqueous free amine, which was then reacted with naproxen ester 1 in
acetonitrile to afford stable isotope labeled amino alcohol 2 in 89% yield after
flash chromatography purification. Amino alcohol 2 was treated with 1 equiv
of BuLi followed by benzenesulfonyl chloride, which selectively sulfonates the
hydroxyl, and upon base treatment, causes ring closure to the aziridine 6 in
55% yield. Reductive ring opening and debenzylation was achieved in one step
when aziridine 6 was treated with lithium in ammonia, and the final compound
was isolated as the maleic acid salt in 70% yield. The chemical purity of 13CD3
PNU-95666 was >97% when analyzed by HPLC.y The NMR spectrum of
13CD3 PNU-95666 showed no N-methyl CH3 signal and no non-labeled
molecular ion was observed when analyzed by mass spectrometry.
Experimental
13CD3 Methylamine hydrochloride (99% C-13 and 99.5 deuterium enrich-
1
ment) was purchased from Aldrich. H NMR spectra were recorded on a
y HPLC condition: column, Zorbax RX-C8 (5 m, 4.6 ꢀ 250 mm); mobile phase, methanol/water/triethyla-
mine (500/500/1); flow-rate, 1 ml/min; UV detector, 282 nm.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 609–614

612
Z. JIAN
13
.
CD3NH2 HCl
OCH3
NaOH/H2O
NH13CD3
OH
O
O
N
Aqueous 13CD3NH2
CH3CN
1. n-BuLi, -23oC, THF
2. BsCl
Br
N
O
N
N
Bn
O
Bn
13CD3 2
1
N13CD3
NH13CD3
1. Li, NH3, t-BuOH
2. Maleic acid
.
N
N
Maleic Acid
HN
N
O
O
Bn
13CD3 6
13CD3 PNU-95666E
Scheme 4. Synthetic of stable isotope labeled PNU-95666
Bruker 300 MHz spectrometer. CDCl3 was used as the NMR solvent unless
1
otherwise indicated. H NMR spectra are reported relative to TMS. Mass
spectra were obtained with a Finnigan TSQ-700 mass spectrometer. HPLC
analysis was performed on a Perkin Ekmer ISS 200 HPLC instrument. All
solvents and reagents were of analytical grade and were used without
purification unless otherwise indicated.
13CD3 (5R,6R)-1-Benzyl-5-hydroxy-6-(methylamino)-5,6-dihydro-4H-imida-
zo[4,5,1-ij]quinolin-2(1H)-one 2
A solution of 10 g of sodium hydroxide (250 mmol) in 10 ml of water was
added dropwise with stirring to a solution of 14 g of 13CD3NH2 hydrochloride
(197 mmol) in 10 ml of water at room temperature. After the reaction was
complete, free 13CD3NH2, along with water, was distilled under vacuum into a
solution of 10 g of naproxen ester 1 (17.6 mmol) in 150 ml of acetonitrile. The
resulting reaction mixture was stirred at room temperature for 72 h. TLC
analysis (silica gel plate, hexane/EtOAc/MeOH 10/10/1) found no remaining
starting material. Excess 13CD3NH2 and solvent was removed under reduced
pressure. The residue was suspended in 115 ml of EtOAc, treated dropwise
with 37.5 ml of 5% HCl, and stirred at room temperature for 10 min. The two
layers were separated and the EtOAc layer was washed with water
(2 ꢀ 37.5 ml). The combined aqueous phase was washed with EtOAc
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 609–614

SYNTHESIS OF STABLE ISOTOPE LABELED
613
(2 ꢀ 37.5 ml), and then, basified with dropwise addition of 19 ml of 10 M
NaOH at 08C. Free amino alcohol 2 was extracted with EtOAc (2 ꢀ 75 ml),
washed with water (2 ꢀ 35 ml), and dried over anhydrous sodium sulfate.
Filtration and solvent removal left an oily residue. The residue was then
purified by flash chromatography (silica gel column, eluted with hexane/
EtOAc/MeOH 3/3/1) to afford 4.88 g (89%) of 13CD3 labeled amino alcohol 2,
which was found to be identical with an authentic sample when analyzed by
TLC (silica gel plate, hexane/EtOAc/MeOH 1/1/1, Rf 0.41). Mass spectrum
(ESI): m=z 314 (MH+). 1H NMR (DMSO-d6, 300 MHz), d: 7.35 (m, 5H), 7.0
(m, 3H), 5.0–5.15 (m, 3H), 4.25, (m, 1H), 3.78 (m, 2H), 3.50 (m, 1H).
13CD3 (7aS,8aR)-4-Benzyl-8-methyl-7,7a,8,8a-tetrahydroazireno [2,3-c] imi-
dazo [4,5,1-ij] quinolin-5(4H)-one 6
n-Butyl lithium in hexane (2.5 M, 6.6 ml, 16.5 mmol) was added dropwise to a
solution of 4.88 g of 13CD3 labeled amino alcohol 2 (15.1 mmol) in 25 ml of
THF kept at ꢁ258C. The resulting reaction mixture was stirred at ꢁ258C for
15 min and then 2.116 ml of benzensulfonyl chloride (16.5 mmol) was added
slowly, keeping the temperature below ꢁ238C. The solution was stirred at
ꢁ258C for 1 h, warmed up to room temperature over 60 min, and stirred for
10 min. Saturated aqueous NaHCO3 solution (11 ml) was added, followed by
75 ml of water. The solid precipitated was collected by filtration and washed
with water (2 ꢀ 5 ml) to afford 2.50 g (55%) of 13CD3 labeled aziridine 6 as a
colorless solid after drying. Aziridine 6 was used directly in next reaction
without further purification and identification.
13CD3 PNU-95666E
13CD3 Aziridine 6 (500 mg, 1.7 mmol) was charged into a 3-necked flask, which
was then attached to a dewar condenser. The condenser and flask were cooled
in dry ice/acetone bath, and NH3 gas was let in to the flask until about 10 ml of
liquid ammonia had been condensed. t-Butyl alcohol (0.5 ml) was added to the
solution, followed by 97 mg of lithium (13.8 mmol) in small pieces. The
reaction mixture was refluxed for 90 min at ꢁ338C and then quenched by the
slow addition of 1 ml of water. After quench, the slurry was stirred for 15 min
at reflux and warmed up to room temperature to remove ammonia. More
water (4.5 ml) and hexane (10 ml) was added and the resulting mixture was
stirred until all the solid dissolved. The phases were allowed to separate and
the aqueous phase was washed with more hexane (10 ml). Most of the water
was removed by concentration and the desired product was extracted by
stirring the aqueous residue with EtOAc overnight (2 ꢀ 150 ml). Ethyl acetate
extracts were combined and the solvent was removed to afford a gummy
residue. The residue was dissolved in 3 ml of ethanol. The ethanol solution was
brought to gentle reflux and a warm solution of 200 mg of maleic acid in 3 ml
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 609–614

614
Z. JIAN
of ethanol was added slowly over 30 min. Crystals started to appear while the
addition was proceeding. The resulting suspension was cooled to 08C and
filtered. The solid collected was again taken up in 7 ml of ethanol. The ethanol
suspension was heated to reflux, cooled to room temperature, and kept in a
refrigerator overnight. The solid was collected by filtration and dried under
vacuum to afford 382 mg (70%) of 13CD3 PNU-95666 E. The chemical purity
was 97.2% when analyzed by HPLC.y Mass spectrum (ESI): m=z 208 (MH+).
No non-labeled PNU-95666 (m/z 204) was observed. NMR (DMSO-d6,
300 MHz): d 3.04 (dd, 1H, J = 16.5, 6), 3.23 (dd, 2H), 3.9–4.06 (m, 3H), 6.02
(s, 2H, maleic acid CH=CH), 6.85–6.97 (m, 3H), 8.55 (br s, 2H), 10.83 (s, 1H).
Acknowledgements
The author would like to express his appreciation to Dr. Peter G. M. Wuts, Jill
M. Northuis, Rui Lin Gu and Tricia A. Kwan of Pfizer CRD for the chemistry
support and to John Easter of Pfizer Radiosynthesis for the mass spectral
measurements.
References
1. Heier RF, Dolak LA, Duncan JN, Hyslop DK, Lipton MF, Martin IJ, Mauragis
MA, Piercey MF, Nichols NF, Schreur PJKD, Smith MW, Moon WM. J Med
Chem 1997; 40: 639–646.
2. Romero AG, Darlington WH, McMillan MW. J Org Chem 1997; 62: 6582–6587.
3. Wuts PGM. Curr Opin Drug Discov Devel 1999; 2: 557–564.
4. Wuts PGM. WO 00/46226, August 2000.
5. Acker BA, Heier RF, Jin AQ, Moon MW. WO 01/83483, November 2001.
6. Wuts PGM, Gu RL, Northuis JM, Kwan TA, Beck DM, White MJ. Pure Appl
Chem 2002; 74: 1359–1368.
7. Peter Wuts, private communication.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 609–614