A practical synthesis of a potent δ-opioid antagonist: Use of a modified Knorr pyrrole synthesis

Article abstract of DOI:10.1021/op034198u

A modified Knorr pyrrole reaction is described which may have practical benefits for the large-scale synthesis of SB-342219. The use of zinc to reduce a phenylhydrazone is replaced by the hydrogenation of the corresponding oxime to provide a common aminoketone intermediate. The modifications have practical benefit with respect to carrying out the reaction and isolating the product. A modified N-demethylation procedure is also described. The use of zinc to reduce an intermediate trichloroethylcarbamate is replaced with the aqueous hydrolysis of the corresponding 1-chloroethylcarbamate.

Full text of DOI:10.1021/op034198u

Organic Process Research & Development 2004, 8, 279−282
A Practical Synthesis of a Potent δ-Opioid Antagonist: Use of a Modified Knorr
Pyrrole Synthesis
Richard K. Bellingham, John S. Carey,* Nigel Hussain, David O. Morgan, Paul Oxley, and Laurence C. Powling
Synthetic Chemistry, GlaxoSmithKline Pharmaceuticals, Old Powder Mills, Nr. Leigh, Tonbridge, Kent TN11 9AN, UK
Scheme 1
Abstract:
A modified Knorr pyrrole reaction is described which may have
practical benefits for the large-scale synthesis of SB-342219. The
use of zinc to reduce a phenylhydrazone is replaced by the
hydrogenation of the corresponding oxime to provide a common
aminoketone intermediate. The modifications have practical
benefit with respect to carrying out the reaction and isolating
the product. A modified N-demethylation procedure is also
described. The use of zinc to reduce an intermediate trichlo-
roethylcarbamate is replaced with the aqueous hydrolysis of
the corresponding 1-chloroethylcarbamate.
SB-342219 (4) is a selective δ-opioid receptor agonist
and as such has undergone preclinical evaluation for the
potential treatment of neuropathic pain.1 In this report we
describe some technical improvements to the synthesis of
SB-342219 (4) that are applicable to a large-scale synthesis.
The medicinal chemistry route2 is summarized in Scheme
1. The advanced intermediate ketone 1 was obtained from
oxycodone in accordance with the published procedures.3
The tetra-substituted pyrrole 2 was prepared using the Knorr
pyrrole synthesis4 where the R-aminoketone is formed in situ,
by reduction of the phenylhydrazone 7 (Scheme 2). N-
Demethylation of 2 was achieved via the intermediate tri-
chloroethylcarbamate 3, and then cleavage of the carbamate
with zinc gives the target compound SB-342219 (4).
Scheme 2
Although the Knorr pyrrole reaction was highly regiose-
lective (none of the alternate regioisomer was observed) and
high yielding, there were aspects of the process that we felt
would cause problems for scale-up, which were mainly
associated with the use of zinc. There was a need to add the
finely divided zinc powder in portions to the hot, flammable
solvent to maintain a low concentration of the intermediate
R-aminoketone, formed by reduction of the phenylhydrazone
7. The crude product 2 also required extensive purification
by chromatography primarily to remove the by-product
aniline. The problems associated with the N-demethylation
procedure were the following: (1) the use of chloroform for
Scheme 3
* Author for correspondence. Telephone: +44(0)1732 372245. Fax:
+(0)1732 372355. E-mail: John_S_Carey@gsk.com.
(1) Dondio, G.; Ronzoni, S.; Farina, C.; Graziani, D.; Parini, C.; Petrillo, P.;
Giardina, G. A. M. Farmaco 2001, 56, 117-119.
(2) Clarke, S. E.; Dondio, G.; Raveglia, L. G.; Ronzoni, S. WO 00/63210,
2000.
(3) Preparation of 14-Hydroxy-17-methylmorphinan-6-one 1. (a) Schmidham-
mer, H.; Jacobson, A. E.; Atwell, L.; Brossi, A. HelV. Chim. Acta 1981,
64, 2540-2543. (b) Brossi, A.; Atwell, L.; Jacobson, A. E.; Rozwadowska,
M. D.; Schmidhammer, H.; Flippen-Anderson, J. L.; Gilardi, R. HelV. Chim.
Acta 1982, 65, 2394-2404.
the formation of the intermediate carbamate, (2) the addition
of zinc powder to a hot, flammable solvent for the cleavage
of the carbamate, and (3) the requirement for chromatography
to ensure pure SB-342219 (4).
To alleviate some of the problems associated with the
Knorr pyrrole reaction using phenylhydrazone 7, the use of
the corresponding oxime 8 was investigated as the R-ami-
(4) Joule, J. A.; Mills. K. Heterocyclic Chemistry, 4th ed.; Blackwell Sciences
Ltd: Oxford, 2000.
10.1021/op034198u CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/05/2004
Vol. 8, No. 2, 2004 / Organic Process Research & Development
•
279

Figure 1.
Figure 2.
Scheme 4
Figure 3.
The absence of aniline from the reaction mixture made the
work-up and isolation of product pyrrole 2 simpler. The
product could be extracted into aqueous citric acid, and the
nonbasic pyrazine by-product 12, Figure 2, formed from
R-aminoketone 9 by dimerisation and subsequent oxidation,
could be removed by washing with toluene. Pyrrole 2 could
then be isolated by crystallisation from toluene/heptane and
then recrystallised from i-PrOH in 68% overall yield if a
further up-grade in purity was required.
Alternative methods for the cleavage of trichloroethyl-
carbamate 3 to SB-342219 (4) were evaluated; however, none
proved suitable for scale-up. A number of alternative
chloroformates (e.g., benzyl, allyl, and 1-chloroethyl) have
been reported for N-demethylation. Of these the use of
1-chloroethyl chloroformate appeared attractive since cleav-
age of the intermediate carbamate is reported to proceed with
methanol.6 Complete, clean formation of the 1-chloroethyl-
carbamate 11 only proceeded in a satisfactory manner in
chlorinated solvents; dichloromethane was the most ap-
propriate solvent. Attempts to replace the heterogeneous
inorganic base, potassium hydrogen carbonate, with an
organic base (e.g. imidazole, DBU or pyridine) were not
successful. The intermediate 1-chloroethylcarbamate 11 was
not isolated, but used directly in the hydrolysis reaction.
Treatment of the crude carbamate 11 with methanol resulted
in the formation of product 4, but contamination with
methylcarbamate 13 was noted (Figure 3). Hydrolysis of
carbamate 11 was shown to be possible using wet, water-
miscible solvents, and THF was found to work particularly
well. By minimising the amount of water used, it was
possible to isolate pure SB-342219 (4), as the hydrochloride
salt, in 74% yield, directly from the reaction mixture and
thus avoid the use of chromatography. This modified
demethylation procedure overcomes all the main problems
associated with the original procedure along with the
additional benefit that 1-chloroethyl chloroformate is con-
siderably cheaper than the corresponding trichloroethyl
variant.
noketone 9 precursor. The preparation of oxime 8 is shown
in Scheme 2. Condensation of tert-butyl acetoacetate 5 with
N-isopropylbenzylamine gave the amide 6 in 97% yield
which was then converted to the oxime 8 in 82% yield by
treatment with sodium nitrite in aqueous acetic acid.5
Although the in situ reduction of oxime 8 to R-aminoke-
tone 9 for the Knorr pyrrole reaction could not be achieved,
a very workable two-step procedure was developed. Catalytic
hydrogenation of oxime 8 over palladium on charcoal in
acetic acid, in the presence of 2.0 equiv of anhydrous HCl,
could be achieved at 60 °C and 60 psi of hydrogen pressure,
Scheme 3. The hydrogenation reaction could be performed
at lower hydrogen pressures by using 2.0 equiv of methane-
sulphonic acid; however, this resulted in the formation of
elevated levels of des-benzyl impurity 10, Figure 1. It was
feared that impurities derived from the reaction of 10 would
be very difficult to remove from subsequent products.
Likewise the use of concentrated hydrochloric acid in place
of anhydrous HCl also led to elevated levels of 10 being
formed. Although R-aminoketone 9 could not be isolated as
a solid, the reduction was clean and efficient. The filtered
acetic acid solution of 9 was stable and was used directly in
the modified Knorr pyrrole reaction.
The Knorr pyrrole reaction was carried out by the slow
addition of the acetic acid solution of 9 to a solution of ketone
1 and sodium acetate in acetic acid at 100 °C, Scheme 4.
Between 2.5 and 3.0 equiv of R-aminoketone 9 was required
to ensure complete consumption of the valuable ketone 1.
In conclusion we have developed an alternative procedure
for the Knorr pyrrole reaction that does not use zinc metal
in the preparation of the R-aminoketone 9. Likewise we have
(5) Paine, J. B.; Brough, J. R.; Buller, K. K.; Erikson, E. E.; Dolphin, D. J.
Org. Chem. 1987, 52, 3993-3997.
(6) Hengeveld, J. E.; Gupta, A. K.; Kemp, A. H.; Thomas, A. V. Tetrahedron
Lett. 1999, 40, 2497-2500.
280
•
Vol. 8, No. 2, 2004 / Organic Process Research & Development

developed a very mild method for N-demethylation; both of
these procedures worked efficiently and cleanly, and we
believe they would be suitable for the large-scale preparation
of a potential pharmaceutical ingredient.
2. To a suitable pressure vessel were charged oxime 8 (16.0
g, 61.0 mmol), 10% palladium on charcoal (4.0 g, Johnson
Matthey type 87L, 50% water wet), and acetic acid (80 mL).
The system was purged with nitrogen, then HCl gas (5.0 g,
137 mmol) added, and the solution stirred for 5 min. The
vessel was purged with hydrogen and pressurised to 60 psi
with hydrogen. The stirrer was started, and the vessel was
heated to 60 °C. After 90 min the vessel was cooled,
depressurised, and then purged with nitrogen. The palladium
on charcoal was removed by filtration through Celite, and
the filtercake was washed with acetic acid (20 mL) to leave
a solution of R-aminoketone 9 as the hydrochloride salt in
100 mL of acetic acid.
Experimental Section
N-(1-Methylethyl)-3-oxo-N-(phenylmethyl)butana-
mide 6. To N-isopropylbenzylamine (25.00 g, 168 mmol)
at 100 °C was added tert-butyl acetoacetate (24.76 g, 157
mmol) over 10 min, maintaining the temperature at 100 °C.
The reaction was heated to 115 °C and a vacuum cautiously
applied (100 mmHg). Once the collection of distillate had
stopped, the reaction was cooled to 30 °C and EtOAc (75
mL) added. The solution was washed with 0.6 M aqueous
HCl (25 mL), brine (25 mL), saturated aqueous NaHCO3
(25 mL), brine (25 mL), and then the solvent was removed
by vacuum distillation to leave acetoacetamide 6 (35.47 g,
To a solution of morphinanone 1 (6.21 g, 22.9 mmol)
and sodium acetate (14.1 g, 172 mmol) in acetic acid (55
mL) at 100 °C was added the acetic acid solution of
R-aminoketone 9 (61 mmol, 100 mL) over 2 h, maintaining
the temperature at 100 °C. Once the reaction was complete,
the acetic acid was removed by distillation under reduced
pressure. Toluene (120 mL) was added and the solvent
removed by distillation under reduced pressure. Toluene (145
mL) was added followed by water (120 mL) and the pH
adjusted to pH ) 9 by the addition of 14.8 M ammonia (40
mL). The layers were separated, and the aqueous layer was
discarded. The toluene solution was extracted with 5% v/v
aqueous citric acid (140 mL) and the toluene discarded.
Toluene 145 (mL) was added and the pH adjusted to pH )
6.8 by the addition of 14.8 M ammonia (8 mL). The layers
were separated, and the aqueous phase was discarded. The
toluene solution was washed with water (25 mL) and then
concentrated by vacuum distillation to leave a residual
volume of 35 mL. The solution was warmed to 45-50 °C
to induce crystallisation. Once nucleation had occurred, a
mixture of toluene (12 mL) and heptane (12 mL) was added
at 50 °C. The slurry was aged at 50 °C for 1 h then at 0 °C
for 30 min prior to isolation by filtration. The filtercake was
washed with a mixture of toluene (12 mL) and heptane (6
mL) and then partially dried under under vacuum. The crude
2 was then recrystallised from refluxing i-PrOH (60 mL).
The resulting slurry was cooled to 0 °C and then isolated by
filtration. The filtercake was washed with i-PrOH (12 mL)
and then dried under vacuum at 50 °C to afford pyrrole 2
(7.53 g, 68%) as a white solid: mp ) 232-233 °C. 1H NMR
(400 MHz, CDCl3) δ 1.13 (d, J ) 6.4 Hz, 3 H), 1.17 (d, J
) 6.6 Hz, 3 H), 1.19-1.27 (m, 1 H), 1.93 (s, 3 H), 2.08-
2.23 (m, 2 H), 2.40 (s, 3 H), 2.32-2.46 (m, 3 H), 2.85-
2.93 (m, 2 H), 2.96-3.05 (m, 1 H), 3.12 (d, J ) 16.2 Hz, 1
H), 3.24 (d, J ) 18.6 Hz, 1 H), 4.32 (d, J ) 15.5 Hz, 1 H),
4.43 (hept, J ) 6.5 Hz, 1 H), 4.58 (s, 1 H), 4.77 (d, J )
15.4 Hz, 1 H), 7.01-7.12 (m, 3 H), 7.12-7.28 (m, 6 H),
8.26 (s, 1 H); 13C NMR (100 MHz, CDCl3) δ 10.5, 20.9,
21.0, 24.5, 28.5, 29.4, 36.7, 40.4, 42.5, 44.5, 45.1, 50.2, 62.0,
69.8, 114.1, 118.4, 120.8, 124.4, 125.4, 125.9, 126.1, 126.8,
126.8, 127.2, 127.8, 135.3, 139.4, 139.6, 166.5; IR (neat,
cm-1) 1614 and 1324; HRMS (CI +ve) m/z Calcd for
C31H38N3O2 (M+ + H): 484.2959. Found: 484.2954; Anal.
Calcd for C31H37N3O2: C, 77.0; H, 7.7; N, 8.7. Found: C,
77.0; H, 7.7; N, 8.75.
1
97%) as a yellow oil. H NMR (400 MHz, DMSO-d6) δ
1.04-1.09 (m, 6 H), 1.80 (s, 0.4 H), 1.95 (s, 0.3 H), 2.09 (s,
0.9 H), 2.21 (s, 1.4 H), 3.32 (s, 0.4 H), 3.42 (s, 0.6 H), 3.78
(s, 1.0 H), 4.03 (hept, J ) 6.6 Hz, 0.5 H), 4.38 (m, 0.1 H),
4.47 (s, 1.4 H), 4.52-4.74 (m, 0.8 H), 5.11 (s, 0.1 H), 5.58
(s, 0.1 H), 7.16-7.41 (m, 5 H); 13C NMR (100 MHz, CDCl3)
δ 20.1, 20.1, 20.3, 21.2, 21.4, 21.4, 21.9, 30.2, 30.4, 43.5,
43.9, 44.8, 45.7, 46.1, 46.5, 49.9, 50.3, 50.5, 86.9, 88.3,
125.7, 125.9, 126.6, 126.7, 126.8, 127.0, 127.3, 128.3, 128.6,
128.8, 138.0, 138.3, 139.0, 166.6, 167.6, 175.1, 202.5, 202.6;
IR (neat, cm-1) 1721, 1631; LRMS (CI +ve) m/z 234 (M+
+ H).
(2EZ)-2-(Hydroxyimino)-N-(1-methylethyl)-3-oxo-N-
(phenylmethyl)butanamide 8. To a solution of acetoaceta-
mide 6 (15.0 g, 64.3 mmol) and acetic acid (15 mL) in THF
(30 mL) at 15 °C was added a solution of sodium nitrite
(6.21 g, 90 mmol) in water (9.3 mL) over 2 h, maintaining
the temperature <20 °C. The reaction mixture was then
stirred at 20 °C for a further 1 h before being added to water
(150 mL). The resulting slurry was stirred at 20 °C for a
further 30 min before the solid was isolated by filtration.
The wet solid was dissolved in methanol (50 mL) and then
heated to reflux. Water (50 mL) was added, maintaining the
solution at reflux. The resulting slurry was cooled to 10 °C
and then aged at 10 °C for 30 min prior to isolation by
filtration. The filtercake was washed with water (3.75 mL)
and then dried under vacuum at 80 °C to give oxime 8 (13.8
g, 82%) as a white solid: mp ) 158-160 °C. 1H NMR (400
MHz, CDCl3) δ 1.07 (d, J ) 6.5 Hz, 4.5 H), 1.22 (d, J )
6.8 Hz, 1.5 H), 2.17 (s, 0.75 H), 2.44 (s, 2.25 H), 3.70 (hept,
J ) 6.5 Hz, 0.75 H), 4.27 (s, 0.5 H), 4.55 (hept, J ) 6.7 Hz,
0.25 H), 4.72 (s, 1.5 H), 7.19-7.41 (m, 5 H); 13C NMR (100
MHz, DMSO-d6) δ 19.7, 20.9, 21.4, 25.2, 25.4, 40.1, 42.3,
46.6, 48.0, 49.9, 126.3, 126.5, 127.1, 127.6, 127.9, 128.2,
137.9, 138.8, 153.4, 162.6, 163.0, 194.3, 194.9; IR (neat,
cm-1) 1682, 1591 and 1582; HRMS (CI +ve) m/z Calcd for
C14H18N2O3Na: (M+ + Na): 285.1209. Found: 285.1208;
Anal. Calcd for C14H18N2O3: C, 64.1; H, 6.9; N, 10.7.
Found: C, 64.15; H, 6.9; N, 10.8.
(6R,6aS,11aR) 5,6,6a,7,10,11-Hexahydro-6a-hydroxy-
8,14-dimethyl-N-(1-methylethyl)-N-(phenylmethyl)-6,11a-
(iminoethano)-11aH-naphth[2,1-f]indole-9-carboxamide
Vol. 8, No. 2, 2004 / Organic Process Research & Development
•
281

(6R,6aS,11aR)-5,6,6a,7,10,11-Hexahydro-6a-hydroxy-
8-methyl-N-(1-methylethyl)-N-(phenylmethyl)-6,11a-(imi-
noethano)-11aH-naphth[2,1-f]indole-9-carboxamide mono-
hydrochloride 4. To a solution of N-methylamine 2 (3.65
g, 7.55 mmol) in CH2Cl2 (55 mL) at room temperature was
added potassium hydrogen carbonate (9.07 g, 90.6 mmol)
and then 1-chloroethyl chloroformate (4.85 g, 33.9 mmol).
Once the addition was complete, the reaction was heated at
40 °C for 16 h. The reaction mixture was cooled to room
temperature, the inorganic salts were removed by filtration,
and the filtercake was washed with CH2Cl2 (20 mL). The
combined filtrate was concentrated by vacuum distillation
to leave a residual volume ) 10 mL. THF (73 mL) was
added and a further 15 mL of distillate collected by vacuum
distillation. Water (3.6 mL) was added and the reaction
mixture heated at 50 °C for 5 h. The resulting slurry was
then cooled to 20 °C and then aged for 1 h prior to isolation
by filtration. The filtercake was washed with THF (20 mL)
and then dried under vacuum at 50 °C to afford amine
hydrochloride 4 (2.83 g, 74%) as a white solid: mp > 250
°C; H NMR (400 MHz, CD3OD) δ 1.15 (d, J ) 6.6 Hz, 6
1
H), 1.46 (d, J ) 13.0 Hz, 1 H), 1.87 (s, 3 H), 2.42-2.58 (m,
3 H), 2.81 (td, J ) 13.7, 3.2 Hz, 1 H), 2.94 (d, J ) 16.5 Hz,
1 H), 3.09 (dd, J ) 13.0, 3.9 Hz, 1 H), 3.24 (d, J ) 19.3
Hz, 1 H), 3.40 (d, J ) 16.4 Hz, 1 H), 3.64-3.78 (m, 2 H),
4.37 (hept, J ) 6.9 Hz, 1 H), 4.59 (d, J ) 16.6 Hz, 1 H),
4.62 (d, J ) 16.6 Hz, 1 H), 7.13-7.24 (m, 8 H), 7.37 (d,
J ) 7.4 Hz, 1 H); 13C NMR (100 MHz, CD3OD) δ 8.3,
19.5, 27.4, 28.8, 29.5, 32.4, 36.0, 39.9, 49.7, 55.7,
67.7, 111.9, 118.0, 120.4, 124.5, 125.8, 126.0, 126.3, 126.6,
127.2, 127.4, 132.7, 137.7, 138.6, 167.1; IR (neat, cm-1)
1714 and 1573; HRMS (CI +ve) m/z Calcd for C30H36N3O2
(M+ + H): 470.2802. Found: 470.2798; Anal. Calcd for
C30H35N3O2.HCl: C, 71.2; H, 7.2; N, 8.3. Found: C, 71.1;
H, 7.15; N, 8.4.
Received for review December 22, 2003.
OP034198U
282
•
Vol. 8, No. 2, 2004 / Organic Process Research & Development