Process development and scale-up of the PPAR agonist NNC 61-4655

Article abstract of DOI:10.1021/op034048j

A scalable synthetic route of the nonselective but PPARα-preferring potent PPAR agonist NNC 61-4655 aimed for treatment of type 2 diabetes was developed. The synthetic pathway comprises the convergent synthesis and coupling of the two key intermediates E-5-(chloropent-3-en-1-ynyl)benzene 8 (prepared in a five-step synthesis in 18% overall yield) and (S)-2-ethoxy-3-(4- hydroxyphenyl)propanoic acid isopropyl ester 9. The 2-aminoethanol salt of NNC 61-4655 was selected in a preclinical salt selection program as the appropriate salt form for further development. More than 900 g of NNC 61-4655, 2-aminoethanol was finally synthesized under GMP in 98.7% purity. In comparison to the original medicinal chemistry route, starting from phenylpropargyl aldehyde 1, the overall yield towards NNC 61-4655 could be enhanced from 24 to 37%. An improved scalable two-step synthesis for 8 was developed on a laboratory scale (≥33-35% overall yield) shortly after the GMP batch.

Full text of DOI:10.1021/op034048j

Organic Process Research & Development 2004, 8, 363−371
Process Development and Scale-Up of the PPAR Agonist NNC 61-4655
Heinz-Josef Deussen,*,† Lone Jeppesen,† Norbert Scha¨rer,‡ Finn Junager,† Bjørn Bentzen,† Beat Weber,‡ Volker Weil,†
Sandor Josef Mozer,† and Per Sauerberg†
NoVo Nordisk A/S, Research & DeVelopment, NoVo Nordisk Park, 2760 MaaloeV, Denmark, and Siegfried AG,
Chemical DeVelopment, Untere Bru¨hlstrasse 4, 4800 Zofingen, Switzerland
Abstract:
A scalable synthetic route of the nonselective but PPARr-
preferring potent PPAR agonist NNC 61-4655 aimed for
treatment of type 2 diabetes was developed. The synthetic
pathway comprises the convergent synthesis and coupling of
the two key intermediates E-5-(chloropent-3-en-1-ynyl)benzene
8 (prepared in a five-step synthesis in 18% overall yield) and
Figure 1. Structure of the PPAR-agonist NNC 61-4655.
(S)-2-ethoxy-3-(4-hydroxyphenyl)propanoic acid isopropyl ester
9. The 2-aminoethanol salt of NNC 61-4655 was selected in a
preclinical salt selection program as the appropriate salt form
for further development. More than 900 g of NNC 61-4655,
2-aminoethanol was finally synthesized under GMP in 98.7%
purity. In comparison to the original medicinal chemistry route,
starting from phenylpropargyl aldehyde 1, the overall yield
towards NNC 61-4655 could be enhanced from 24 to 37%. An
improved scalable two-step synthesis for 8 was developed on a
laboratory scale (g33-35% overall yield) shortly after the GMP
batch.
as rosiglitazon (PPARγ),6 pioglitazone (PPARγ),7 fenofibrate
(PPARR),8 and clofibrate (PPARR),8 have been shown to
have beneficial effects on the described characteristics of
type 2 diabetes.9 However, neither the fibrates nor the
glitazones both lower triglycerides and increase HDL-c as
well as lower blood glucose and improve insulin sensitivity.
To achieve this biological response both dual-acting PPARR,γ
agonists10 and triple-acting PPARR,γ,δ agonists11 have been
designed. Several dual-acting PPARR,γ agonists (e.g.,
ragaglitazar/NNC 61-0029,12 tesaglitazar/AZ242,13 and
LY46560814) are currently being investigated in preclinical
and clinical trials.
Introduction
NNC 61-4655 belongs to the class of nonselective but
PPARR-preferring potent PPAR agonists (R:EC50 ) 0.0067
µM; γ:EC50 ) 1.13 µM; δ:EC50 ) 6.90 µM). NNC 61-4655
has excellent pharmacokinetic properties and was shown to
have more efficacious in vivo effects in male db/db mice
than seen with both rosiglitazone and pioglitazone.15 Due to
the unique PPAR receptor subtype profile together with the
impressive pharmacological properties, NNC 61-4655 was
chosen as a promising antidiabetic drug candidate (see Figure
1).
Type 2 diabetes is a polygenic and progressive metabolic
disorder characterized by insulin resistance, hyperglycaemia,
hypertriglyceridaemia, and low plasma HDL-cholesterol.
Untreated type 2 diabetes leads to several chronic diseases
such as retinopathy, nephropathy, neuropathy, and cardio-
vascular diseases such as atherosclerosis, the latter leading
to increased mortality.1 Due to the forecasted epidemic in
type 2 diabetes, the increasing financial and social costs, and
the complicated pathology of the disease, new therapies are
needed which address both the insulin resistance and
dyslipidemic components of the disease.2-4
Peroxisome proliferator-activated receptors (PPARs) are
transcription factors that belong to the nuclear hormone
receptor superfamily. PPAR has three isoforms designated
PPARR, PPARγ, and PPARδ, which differ in ligand
selectivity and biological action.5 A number of PPAR
agonists exhibiting different receptor subtype profiles, such
(6) Chilcott, J.; Tappenden, P.; Jones, M. L.; Wright, J. P. Clin. Ther. 2001,
23, 1792-1823.
(7) Boyle, P. J.; King, A. B.; Olansky, L.; Marchetti, A.; Lau, H.; Magar, R.;
Martin, J. Clin. Ther. 2002, 24, 378-396.
(8) Staels, B.; Dallongeville, J.; Auwerx, J.; Schoonjans, K.; Leitersdorf, E.;
Fruchart, J.-C. Circulation 1998, 98, 2088-2093.
(9) Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke, B. R. J. Med. Chem.
2000, 43, 527-550.
(10) Sauerberg, P.; Pettersson, I.; Jeppesen, L.; Bury, P. S.; Mogensen, J. P.;
Wassermann, K.; Brand, C. L.; Sturis, J.; Wo¨ldike, H. F.; Fleckner, J.;
Andersen, A.-S. T.; Mortensen, S. B.; Svensson, L. A.; Rasmussen, H. B.;
Lehmann, S. V.; Polivka, Z.; Sindelar, K.; Panajotova, V.; Ynddal, L.; Wulff,
E. M. J. Med. Chem. 2002, 45, 789-804.
(11) Mogensen, J. P.; Jeppesen, L.; Bury, P. S.; Pettersson, I.; Fleckner, J.; Nehlin,
J.; Frederiksen, K. S.; Albrektsen, T.; Din, N.; Mortensen, S. B.; Svensson,
L. A.; Wassermann, K.; Wulff, E. M.; Ynddal, L.; Sauerberg, P. Bioorg.
Med. Chem. Lett. 2003, 13, 257-260.
* Corresponding author. Current address: Leo Pharma A/S, Industriparken
55, DK-2750 Ballerup. Fax: +45 4466 0012. Phone: + 45 7226 2453. E-mail:
heinz-josef.deussen@leo-pharma.com.
† Novo Nordisk A/S.
‡ Siegfried AG.
(1) Porte, D., Jr.; Schwartz, M. W. Science 1996, 27, 699-700.
(2) Cobb, J.; Dukes, I. Ann. Rep. Med. Chem. 1998, 33, 213-221.
(3) Grover, S. A.; Coupal, L.; Zowall, H.; Dorais, M. Circulation 2000, 102,
722-727.
(12) Ebdrup, S.; Pettersson, I.; Rasmussen, H. B.; Deussen, H.-J.; Jensen, A. F.;
Mortensen, S. B.; Fleckner, J.; Pridal, L.; Nygaard, L.; Sauerberg, P. J.
Med. Chem. 2003, 46, 1306-1317.
(4) Bailey, C. J. TiPS 2000, 21, 259-265
(5) Guan, G.; Zhang, Y.; Breyer, M. D. Drug News Perspect. 2002, 15(3), 147-
154.
(13) Cronet, P.; Petersen, J. F. W.; Folmer, R.; Blomberg, N.; Sjo¨blom, K.;
Karlsson, U.; Lindstedt, E.-L.; Bamberg, K. Structure 2001, 9, 699-706.
10.1021/op034048j CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/22/2004
Vol. 8, No. 3, 2004 / Organic Process Research & Development
•
363

Scheme 1. Medicinal chemistry synthesis of the
PPAR-agonist NNC 61-4655
carbonyl)dipiperidine (ADDP), followed by chromatography,
gave the ester 5 in 80% yield. Hydrolysis of 5 to the carbox-
ylic acid gave the desired product NNC 61-4655 in 24%
overall yield23 starting from phenylpropargyl aldehyde 1.
The existing route had a number of obstacles that had to
be solved before scaling. The starting phenylpropargyl
aldehyde 1 was not commercially available in larger quanti-
ties and therefore had to be synthesized. At some stage in
the Wittig-Horner reaction, foaming was observed during
the addition of 1 to the deprotonated triethyl phosphono-
acetate. Running the DIBAL-H reduction of 2 at such a low
temperature as -70 °C was not possible with our regular
equipment. The exothermic quenching of the DIBAL-H
reaction mixture by addition of water would be a safety issue
on large scale. The Mitsunobu coupling of 3 and 4 uses
expensive reagents and requires the cumbersome removal
of tri-n-butylphosphine oxide and the reduced ADDP from
the product. All reactions were run rather diluted, which
would lead to unacceptable large reaction volumes on large
scale. Furthermore, the numerous chromatographic steps
were unsuitable for scaling.
The Scaled GMP-Kilo-Laboratory Synthesis. To pro-
vide material for phase I studies as rapidly as possible,24 we
decided to retain and to optimize the synthetic procedures
provided by Medicinal Chemistry for scaling, except for the
Mitsunobu coupling, which was judged to be nonscalable.
The acetylenic compounds were considered to be a
potential safety risk during scale-up, particularly as some are
being distilled.25,26 Therefore, before scaling the synthesis
in the GMP-kilo laboratory, we investigated the thermal
properties of intermediates by DSC to identify our safety
window towards exotherms which were found to be in the
range of 95-127 kJ/mol.27
The synthesis of 1 in a two-step synthesis from com-
mercially available phenylacetylene 6 had been described
in the literature.28,29 The intermediate phenylpropargyl al-
dehyde diethyl acetal 7 was then prepared, closely following
the original literature procedure (Scheme 2).28 The subse-
quent procedure for the preparation of 129 had to be optimized
prior to scaling. The acetal cleavage was run more concen-
trated, and the formed ethanol was removed from the reaction
mixture by distillation to shift the equilibrium of the reaction
towards 1. It was important to control the distillation of
ethanol accurately as otherwise we observed losses of product
1 that was removed azeotropically together with the ethanol
during distillation. The diethyl ether used in the literature
synthesis was replaced by MTBE for extraction. 1 (2.1 kg,
93% pure) was finally synthesized in 46% overall yield.30
Thus, NNC 61-4655 was needed in amounts of several
hundred grams for preclinical development. A GMP batch,
preferably as a pharmaceutical acceptable salt, to be used as
first human dose material in phase I studies, had to be
produced on a multi-100-g scale.16 It therefore became
necessary to develop a scalable synthesis that would provide
the quantities needed for further development and to identify
a pharmaceutically acceptable salt of NNC 61-4655.
Results and Discussion
The Medicinal Chemistry Synthesis. The original
medicinal chemistry route is depicted in Scheme 1. Com-
mercially available phenylpropargyl aldehyde 1 (Aldrich) was
reacted with triethyl phosphonoacetate in a Horner-Emmons
reaction to give the R,â-unsaturated ester 2. Following a
general procedure for Wittig-Horner olefinations, the condi-
tions used were slightly different from those of the procedure
earlier described by Krause (e.g. THF was replaced by
toluene).17 The isolated yield was significantly lower than
that reported earlier,18 apparently due to significant losses
during the chromatography purification step. Reduction of
the ester 2 with a two molar excess of diisobutyl aluminum
hydride (DIBAL-H) at -70 °C gave the allylic alcohol 319
in 81% yield after chromatography.20 Alkylation of ethyl (S)-
2-ethoxy-3-(4-hydroxyphenyl)propanoate 4,12,21 which was
on hand from another development project, with 3 under
Mitsunobu22 conditions with tri-n-butylphosphine/1,1′-(azodi-
(14) Etgen, G. J.; Oldham, B. A.; Johnson, W. T.; Broderick, C. L.; Montrose,
C. R.; Brozinick, J. T.; Misener, E. A.; Bean, J. S.; Bensch, W. R.; Broks,
D. A.; Shuker, A. J.; Rito, C. A.; McCarthy, J. R.; Ardecky, R. A.; Tyhonas,
J. S.; Dana, S. L.; Bilakovics, J. M.; Paterniti, Jr., J. R.; Ogilvie, K. M.;
Liu, S.; Kauffman, R. F. A. Diabetes 2002, 51, 1083-1087.
(15) Sauerberg, P.; Bury, P. S.; Mogensen, J. P.; Deussen, H.-J.; Pettersson, I.;
Fleckner, J.; Nehlin, J.; Frederiksen, K. S.; Albrektsen, T.; Din, N.; Svensson,
L. A.; Ynddal, L.; Wulff, E. M.; Jeppesen, L. J. Med. Chem. 2003, 46,
4883-4894.
(22) Hughes, D. L. Org. Prepr. Proced. Int. 1996, 28, 127-164.
(23) The medicinal chemistry yields are not corrected for purity.
(24) The aim is to produce this batch speedily to obtain an evaluation of the
project value as quickly as possible, rather than to develop a perfect
production route.
(16) A GMP-batch synthesized immediately after drug candidate selection will
be used for preclinical pharmacology and toxicicology studies and thereby
at the same time qualify for first human dose.
(25) Silver, L. Chem. Eng. Progr. 1967, 63, 43-49.
(26) Whiting, M. C. Chem. Eng. News 1972, 50, 86-87.
(27) As impurities, even traces, often effect the safety windows for acetylenic
compounds, a comprehensive safety evaluation should be conducted before
scaling to pilot plant.
(17) Krause, N. Chem. Ber. 1990, 123, 2173-2180.
(18) An isolated yield of surprising 91% (analytically pure) after chromatography
followed by Kugelrohr distillation was claimed in ref 17 on a 60 mM scale.
(19) Takeuichi, R.; Tanabe, K.; Tanaka, S. J. Org. Chem. 2000, 65, 1558-1561.
(20) Yoon, N. M.; Gyoung, Y. S. Org. Chem. 1985, 50, 2443-2450.
(21) Ebdrup, S.; Deussen, H.-J.; Zundel, PCT-Appl. WO 01/11072.
(28) Howk, B. W.; Sauer, J. C. Organic Syntheses; John Wiley & Sons: New
York, 1963; Collect. Vol. IV, pp 801-803.
(29) Allen, C. F. H.; Edens, C. O., Jr. Organic Syntheses; John Wiley & Sons:
New York, 1955; Collect. Vol. III, pp 731-733.
364
•
Vol. 8, No. 3, 2004 / Organic Process Research & Development

Scheme 2. Scaled GMP-kilo-laboratory synthesis of phenylpropargyl aldehyde 1
Scheme 3. Scaled GMP-kilo-laboratory synthesis of the PPAR-agonist NNC 61-4655
For the synthesis of 2 (Scheme 3) we replaced NaH by
NaOEt as this reagent was safer to handle on large scale
and foaming was avoided during deprotonation. Other bases,
such as KOtBu, NaOH, KOH, K2CO3, and Et3N were tested
in THF, MTBE, and toluene with the aim to improve yield
and purity of the product. In-process HPLC analyses
indicated an average of 18% side products with KOtBu and
28% with NaOEt in toluene. Although KOtBu was most
promising with respect to the purity of the product, this base
was discarded due to gel formation of the reaction mixture
in toluene.31 The total cis-isomer of 2 formed, as detected
solvent ratio during workup was optimized to achieve
satisfactory phase separation. We knew from small-scale
experiments starting with pure 2 that the DIBAL-H reduction
was a very clean reaction, and we never detected any cis-
trans isomerization. Neither an excess of DIBAL-H nor
prolonged reaction time led to over-reduction or more
byproduct formation. The yield over both steps {1 f2 f3}
was 41%, mainly due to losses during the distillation of 3.33
The most straightforward solution to avoid the Mitsunobu-
coupling was to replace the hydroxyl group in 3 by a leaving
group to allow C-O bond formation by nucleophilic attack
of the phenolate anion of 4. The conversion of allylic alcohols
to allylic chlorides without allylic rearrangement, by using
a mixture of methanesulfonyl chloride, lithium chloride, and
collidine in DMF, had been described in the literature.34 Our
process research revealed that lithium chloride and collidine
were not necessary for the reaction. We assume that the
transformation proceeded via the iminium salt generated in
situ from methanesulfonyl chloride and DMF in contrast to
the mechanism described by Collington and Meyers in ref
34.35
The reaction during the addition of methanesulfonyl
chloride to a solution of 3 in DMF was exothermic. The inner
temperature of the reaction mixture immediately rose gradu-
ally upon addition of methanesulfonyl chloride but could be
easily kept at around 70 °C by controlling the addition rate.
The mixture was stirred at 70 °C for 1-5 h followed by the
addition of water and extraction of the product into toluene
to give 8 in 96% yield and 91% purity. The purity obtained
for 8 solely depended on the purity of the starting material
1
by H NMR, was always less than 3%, independent of the
reaction conditions. Attempts to purify crude 2 by distillation
resulted in unacceptable losses, presumably due to decom-
position. To improve the overall yield, 2 was consequently
used without purification in the next step.
The procedure for the DIBAL-H reduction of 2 was
changed significantly in comparison to the medicinal chem-
istry procedure, leading to a much safer process.32 The excess
of DIBAL-H was reduced from one molar to 10%. THF was
replaced by toluene, and the volume was reduced 100 times.
The reduction was carried out between 15 and 20 °C
(allowing a better control of the exothermic reaction as the
added DIBAL-H reacted instantaneously). Quenching of the
reaction mixture was done by reversed sequence of addition
(again to gain improved control). However, unpurified 2
contained phosphoric organic impurities, which were reduced
to phosphine. Therefore, the reaction should only be carried
out in a well-ventilated fume hood. The aqueous/organic
(30) To obtain higher yields, the distillation steps might be omitted during future
campaigns without significantly compromising the purities.
(31) Due to the time pressure of the project, we were not able to carry out more
experiments with KOtBu to avoid gel formation, e.g. to test that base with
other solvents on a multigram scale.
(33) A maximum of 63% yield had been obtained earlier on a 250-g scale.
(34) Collington, E. W.; Meyers, A. I. J. Org. Chem. 1971, 36, 3044-3045.
(35) Taylor, E. C., Ed. AdVances in Organic Chemistry: Methods and Results;
Iminium Salts in Organic Chemistry. Bo¨hme, H., Viehe, H. G., Eds.; Vol.
9, Parts 1 and 2; John Wiley & Sons: New York, 1979.
(32) Experiments to replace DIBALH with LiALH4 and Red-Al only resulted
in degradation of 2.
Vol. 8, No. 3, 2004 / Organic Process Research & Development
•
365

Scheme 4. Pilot-plant synthesis of (S)-2-ethoxy-3-(4-hydroxyphenyl)propanoic acid isopropyl ester 921,36,37
3 (additional cis-trans isomerization was usually e1%) and
did not change with minor variations of reaction parameters.
As short-path distillation of 8 did not improve its purity, the
compound was used without purification in the next step.
The key intermediate phenol 9 was readily available in
kilogram quantities from a pilot-plant campaign of our
ragaglitazar development project.12,21 Therefore, and not due
to chemical reasons, the isopropyl ester 9 was used for
scaling instead of the ethyl ester 4 used in the original
medicinal chemistry synthesis. The pilot-plant process on a
40-50-kg scale towards rac-4, starting from 4-benzyloxy-
benzaldehyde, and the enzymatic hydrolysis to 11 has been
published before (Scheme 4).21,36,37 The key intermediate 9
was prepared by acid-catalyzed esterification of 11 in excess
2-propanol.12 This procedure had been further optimized for
pilot-plant scale and was run on a 42-kg scale to yield 9 in
99% chemical and 99.4% optical purity.
Alkylation of the phenol 912,21 with the allylic chloride 8
in acetone, with a catalytic amount of potassium iodide, using
potassium carbonate as a base, readily gave the ester 10 in
96% yield and 92% purity as an oil, which was used without
purification in the next step. NNC 61-4655 was then obtained
from 10 by simple saponification with aqueous sodium
hydroxide in 98% yield and 94% purity. The amount of
solvents used during saponification was significantly (ca. 6×)
reduced in comparison to the original medicinal chemistry
procedure. To achieve practical solubility and phase separa-
tion at these high concentrations, the hydrolysis had to be
carried out at 70 °C along with the workup at 55-60 °C.
NNC 61-4655 was a low-melting wax (mp g 71 °C), which
could not be crystallized and was therefore used directly in
the following salt formation. No cis-trans isomerization and
less than 1% racemization was observed under the reaction
conditions during the transformation {8 + 9 f10 f NNC
61-4655}.
second a salt would be preferred for pharmaceutical develop-
ment.38,39
To identify a suitable salt of NNC 61-4655 for pharma-
ceutical development, various bases, which are commonly
used in pharmaceutical products, were screened for salt
formation with the carboxylic acid NNC 61-4655 (pKa )
3.6) in a range of solvents. Precipitates of NNC 61-4655 on
milligram scale were obtained with potassium hydroxide,
sodium hydroxide, L-arginine, L-lysine, 2-aminoethanol,
ethylenediamine, benethamine, and benzathine. The precipi-
tates were tested by differential scanning calorimetry (DSC),
thermogravimetric analysis (TGA), powder X-ray diffraction
(XRD), and dynamic vapor sorption (DVS) to elucidate their
melting points, thermal properties, crystallinities, and hygro-
scopicities, respectively. Additionally the following properties
were calculated by semi-emperical calculations: intrinsic
solubility, pH solubility profile, and ionisable properties (pKa
values). Four crystalline salts (L-arginine, 2-aminoethanol,
benethamine, and benzathine) were then selected as first-
round salt candidates on the basis of crystal and thermo-
dynamic properties, hygroscopicity, and aqueous solubilities
(see Table 1).
During the progression of the project, more data relevant
for salt selection were generated in parallel, and the final
salt candidate was elected: The results of a four-week
stability study40 revealed a possible instability of the beneth-
amine salt, which was excluded as a candidate. The otherwise
overall favorable L-arginine was not chosen due to its
degradation susceptibility towards light exposure. The benz-
athine salt was discarded due to its low melting point and
its unfavorable acid ratio. The 2-aminoethanol salt of NNC
61-4655 (Scheme 5) was chosen as the preferred candidate
for further development. Its high melting point together with
a demonstrated integrity of the crystal form, a good chemical
stability, and its non-hygroscopicity made it appropriate for
tablet formulation.41 Furthermore, this salt had the highest
acid ratio and sufficient solubility of 3.0 mg/mL (pH 7) to
Salt Assessment and Preparation. A suitable salt of
NNC 61-4655 was highly desirable at this stage of the
project. First, a salt of NNC 61-4655 might give the
possibility of purifying the final product by crystallization,
(38) Stahl, H. W.; Wermuth, C. G. Handbook of Pharmaceutical Salts, Properites,
Selection, and Use; Wiley-VCH: Weinheim, 2002.
(39) Bastin, R. J.; Bowker, M. J.; Slater, B. J. Org. Process Res. DeV. 2000, 5,
(36) Deussen, H.-J.; Zundel, M.; Valdois, M.; Lehman S. V.; Weil, V.; Hjort,
C., M.; Østergaard, P. R.; Marcussen, E.; Ebdrup, S. Org. Process Res.
DeV. 2003, 7, 82-88.
(37) Rahbek Østergaard, P.; Mailand Hjort, C.; Deussen, H.-J.; Zundel, M.,
Ebdrup, S.; Christensen, S.; Patkar, S. PCT-Appl. WO 02/12472.
427-435.
(40) Performed at three different conditions, 40 °C/75% relative humidity, 60
°C/ambient humidity, ambient humidity and temperature/ light exposure,
and analyzed with the following methods: XRD, Karl Fischer, HPLC purity
and assay, DSC, and TGA.
366
•
Vol. 8, No. 3, 2004 / Organic Process Research & Development

Table 1. Comparison of first-round NNC 61-4655 salt candidates
NNC 61-4655
L-arginine
NNC 61-4655
aminoethanol
NNC 61-4655
benethamine
NNC 61-4655
benzathine
test
appearance of powder
melting point DSC onset (°C)
white
208.5
yellow
147.9
pale yellow
120.2
white
109.2
preliminary polymorphism study no other form detected no other form detected no other form detected no other form detected
other thermal behavior
aqueous solubility (mg/mL)
hygroscopicity
nothing detected
2.0
non-hygroscopic
67%
light-sensitive
13% decrease
nothing detected
3.0
non-hygroscopic
85%
nothing detected
0.06
non-hygroscopic
62%
4-6% decrease at
40 and 60 °C
nothing detected
0.1
non-hygroscopic
59%
acid ratio
four-week stability study40
stable
stable
Scheme 5. Scaled GMP-kilo-laboratory synthesis of the
salt NNC 61-4655, 2-aminoethanol
enhanced in comparison to that from the original medicinal
chemistry route from 24% yield23 to 37%. The quantity and
quality of the batch was adequate to allow complete
preclinical and phase I clinical studies with the same batch.
A pharmaceutically acceptable salt with excellent properties
was identified early in the development process, and this
should facilitate the whole future development program.
However, the current synthesis has some drawbacks with
regard to a future production on a commercial scale. The
synthesis of the allylic chloride 8 requires five steps and only
gives 8 in 18% overall yield. Although this yield might be
improvable by more optimization work (e.g. by omitting
distillation steps), we decided to search for a new synthetic
route towards 8 due to the following reasons: (a) The current
synthesis for the intermediate 3 requires DIBAL-H, which
is pyrophoric and rather expensive. A literature search and
our initial investigations indicated that DIBAL-H might not
be exchangeable with another reducing reagent.32 (b) The
existing procedure for the ester 2 would have to be
significantly improved with regard to the current process and
workup. The formation of phosphine during the reduction
of 2 would be a safety issue on a commercial production
scale. Conversely, it was obvious to retain the sequence {8
+ 9 f 10} due to the availability of the key intermediate
9.36 Therefore, we continued with our development program
just after the preparation of the first GMP batch to find a
shorter route towards the second key intermediate 8, which
will be described in the following section.
Development Work after the GMP-Kilo-Laboratory
Synthesis and Final Conclusions. The synthesis of 5-
phenylpent-1-en-4-yn-3-ol 12 (Scheme 6) from phenyl-
acetylene 1 and acrolein has been described in the literature
although without giving experimental details.44 Chou et al.45
have recently illustrated that 12 can be transformed into 8
by different halogenating agents in varying yields and E/Z-
ratios. By improving the described methods and adapting
the procedures to become applicable for large-scale produc-
tion we developed a straightforward two-step synthesis for
the key intermediate 8 on a laboratory scale (44 g). The
allylic alcohol 12 was obtained by addition of deprotonated
phenylacetylene to acrolein in 86-93% yield as a raw
product of 65-70% purity. The major obstacle of this
reaction was undesired polymerization of acrolein, which
satisfy standard dissolution requirements. Another advantage
was that it was easiest to crystallize and the obtained crystals
had the best filtration properties on large scale.
Ten solvents (EtOAc, EtOH, acetone, CH3CN, iPrOAc,
THF, iPrOH, tBuOAc, 1-PrOH, MTBE) were tested to find
the optimum solvent for the preparation and crystallization
of NNC 61-4655, 2-aminoethanol, first on a milligram- and
then on a multigram scale. The same crystal form was always
found, independent of the crystallization conditions and the
solvent. We finally choose ethanol as the solvent of choice,
which allowed a good control of the crystallization process,
yielded filterable crystals, enhanced the relative purity of the
salt by 4-5% in comparison to the crude acid, and gave a
reasonable isolated yield (g86%) of NNC 61-4655, 2-amino-
ethanol on scaling of the crystallization procedure.42 Thus,
the crude NNC 61-4655 from the GMP-kilo-synthesis was
transformed into its 2-aminoethanol salt NNC 61-4655,
2-aminoethanol in 86% yield and 98.7% HPLC purity.43
This allowed us to explore the final salt form of the future
drug as early as in preclinical development and for the first
human dose.
Conclusions from the Scaled GMP-Kilo-Laboratory
Synthesis. A scalable synthetic route was developed for the
synthesis of the title compound on the basis of a modified
medicinal chemistry route. More than 900 g of NNC 61-
4655, 2-aminoethanol was finally synthesized in 32%
overall yield starting from phenylpropargyl aldehyde 1 under
GMP conditions only three months after this compound had
been selected as a drug candidate for pharmaceutical
development. The overall yield towards NNC 61-4655 was
(41) A mechanical grinding stability experiment and an eight-week stability study
under accelerated storage conditions (60 °C/ambient humidity) performed
after the preparation of the GMP-batch confirmed the selection of the NNC
61-4655, 2-aminoethanol salt as development candidate.
(42) No polymorphism was observed for any of the batches produced.
(43) HPLC: 0.31% cis-NNC 61-4655; two major impurities of unknown structure
(0.10% + 0.77%); CE: e1.7% R-NNC 61-4655; KF water: 0.07%; 0.14%
solvent residue (EtOH).
(44) Mahrwald, R.; Quint, S. Tetrahedron Lett. 2000, 56, 7463-7468.
(45) Chou, S.-Y.; Tseng, C.-L.; Chen, S.-F. Tetrahedron Lett. 2000, 41, 3895-
3898.
Vol. 8, No. 3, 2004 / Organic Process Research & Development
•
367

Scheme 6. Improved two-step synthesis of the key-intermediate 8
could be suppressed by optimizing the reaction conditions,
mainly by keeping the reaction temperature below 0 °C. The
crude compound 12 was then chlorinated with thionyl
chloride to give a 5/1 E/Z-mixture of 8. By two subsequent
distillation steps the trans-isomer 8 could be obtained in 38%
isolated yield and 97% purity containing less than 3% of
the cis-isomer. It is expected that this yield will be improved
by up to 15% on further scaling due to residual trans-isomer
8 in the distillation remainder, which could not be transferred
in our equipment on our employed scale. Thus, a scalable
two-step synthesis for 8 was developed (g33-35% overall
yield) on a laboratory scale, which could substitute the five-
step (18% overall yield) synthesis, used for the first scaling,
during the following campaigns and in future production
(Important note: A safety investigation should be carried
out for this reaction sequence before further scale up!).
With respect to the coupling step of 8 and 9 and the
following steps, further optimization of the process might
be achievable. If coupling would be possible with the acid
11 instead of the ester 9, the esterification step from 11 f
9 could be omitted. The coupling reaction might be carried
out in the same solvent in which 8 or 9/11 are obtained to
avoid an additional solvent-stripping step. More generally,
the overall procedure from the coupling of 8 and 9 towards
NNC 61-4655, 2-aminoethanol might be further improved.
Despite the promising results, the synthetic route described
in this article was no longer an option for future campaigns
due to patent application by a competitor claiming compound
9.46 Future laboratory activities were initiated to find a novel
proprietary scalable route which does not employ 9 for the
synthesis of NNC 61-4655, 2-aminoethanol to be used in
future campaigns (including production of phase II material).
on a Fisons instrument EA 1108 elemental analyzer. The
capillary electrophoresis (CE) analysis was performed on a
HP3DCE instrument (Agilent, Waldborn, Germany) equipped
with an auto sampler, a capillary cartridge, a high-voltage
power supply, a diode array detector, electrodes, and a
hydrostatic injection system. The electrophoretic data system
was the HP Chemstation software, and the data were
collected with a frequency of 10 Hz. The CE separations
were carried out with untreated fused-silica capillaries from
Agilent with the following dimensions: 80.5 cm total length
with 72.0 cm effective length, 50 µm inner diameter, and
extended light path with an inner diameter of 150 µm at the
detector window. The electrolyte was prepared by dissolving
3.0% (w/v) sulfobutyl ether-â-cyclodextrin (Advasep 4,
Cydex, Inc., Overland Park, KS) and 0.50% (w/v) dimethyl-
â-cyclodextrin (Agilent, Waldborn, Germany) both in 50 mM
borate buffer pH 9.3 (Agilent) followed by filtering through
a 0.45 µm polypropylene filter. To this solution was added
5% (v/v) MeCN to give the final electrolyte. The electro-
phoresis was carried out in normal polarity mode. The
electrophoretic conditions were as follows: voltage, 21 kV;
current, 50 µA; capillary temperature controlled at 25 °C;
injection was 50 mbar for 4.0 s; detection, UV at 205 nm
with reference of 380 nm. The sample concentration was
0.05 mg/mL in 1/5 acetonitrile/5 mM borate buffer pH 9.3.
The capillary was conditioned with 0.1 N NaOH for 20 min
daily and flushed with 0.1 N NaOH (3 min), water (2 min),
and electrolyte (3 min) between each run. HPLC: (a)
detection: 273 nm; flow: 1.0 mL/ min; column: Water
Symmetri C18, 5 µm, 3 mm × 150 mm; 35 °C, gradient
(15-65%): MeCN, 0.25 M phosphate buffer pH 7; (b)
detection: 250 nm; flow: 1.0 mL/ min; column: Merck
LiChrospher RP-18, 5 µm, 4 mm × 250 mm; 35 °C; gradient
(20-80%): MeCN/water (20:80, v:v) + 0.1% H3PO4, MeCN
+ 0.1% H3PO4. (c) detection: 220 nm; flow: 1.0 mL/ min;
column: Merck LiChrospher RP-18, 5 µm, 4 mm × 250
mm; 35 °C, isocratic MeCN/water (26:74, v:v) + 0.1%
H3PO4. Powder X-ray diffraction patterns were recorded on
a Bruker D8 Advance diffractometer equipped with a
multilayer mirror that selects the Cu KR radiation (λ )
1.5418 Å). Samples were mounted in flat plate reflection
geometry, and the diffracted intensities were detected with
a scintillation counter. Scan ranges were 2-30° in diffraction
angle 2θ with steps of 0.03° and a speed of 6 s/step. All the
aforementioned salts were crystalline. Differential scanning
analysis (DSC) for melting-point determination was executed
on a MDSC 2920, TA Instruments. Samples (approximately
2-5 mg) were heated in pinhole-crimped aluminium pans
from 25 to 280 °C at a rate of 5 °C/min. The DSC measuring
chamber was continuously purged with dry nitrogen during
the runs, and the instrument was routinely calibrated with
Experimental Section
General Procedures. Experimental details are described
on the largest scale carried out. To produce sufficient
amounts of the title compound, some reactions were run
multiple times. Batches were then pooled before proceeding
with the next reaction step, whenever appropriate. All
1
chemicals, unless otherwise stated, were from Aldrich. H
NMR spectra were recorded at either 200 MHz on a Bruker
Avance DPX 200 or at 400 MHz on a Bruker Avance DRX
400 instrument. Mass spectra were recorded on a Finnigan
LCQ quadrupole ion-trap mass spectrometer with an elec-
trospray ion source. The sample in the form of an MeCN:
water (1:1) solution (121 mg/L) was introduced using a
syringe pump inlet (flow: 3 µL/min). Source voltage: 3.3
kV. Capillary voltage: 8.56 V. Capillary temp: 200 °C.
Collision energy: 36%. Elemental analyses were performed
(46) Andersson, K. PCT-Appl. WO 99/62872.
368
•
Vol. 8, No. 3, 2004 / Organic Process Research & Development

indium and tin. DSC for safety evaluation was executed from
25 to 500 °C at a rate of 10 °C/min without nitrogen purging
under air. Thermogravimetric analysis (TGA) experiments
were executed on a Thermo Gravimetric Analyser, model
2959, TA Instruments. Samples (approximately 5 mg) were
heated in an open platinum pan from 25 to 250 °C at a rate
of 10 °C/min. The TGA measuring chamber was continu-
ously purged with dry nitrogen during the runs, and the
instrument was routinely calibrated with indium and alu-
minium. The hygroscopicity properties were investigated by
using a dynamic vapor sorption instrument as approximately
15 mg of sample was placed on the sample pan of the
instrument. A full cycle of isotherms (sorption and desorption
at 25 °C) with equilibrating stations at 10, 20, 30, 40, 50,
60, 70, 80, 90, and 98% relative humidity (RH) was
performed. Prior to analysis, the apparatus was calibrated
using a built-in calibration program and a certified 100-mg
weight. The pH solubility profile of the free acid of NNC
61-4655 and of the various counterions and the corresponding
pKa values were calculated by use of the ACD Solubility
and pKa program modules, both version 4.0.47 Gas chro-
matograms of compound 12 and 8 were recorded on a
Hewlett-Packard HP6890 equipped with an FID detector. A
temperature ramp (60 to 290 °C at 15 °C/min) was applied.
Gas flow: 1.4 mL (He)/min. Column: Optima 5 (Macherey
& Nagel), 25 m, 320 µm, film thickness 0.25 µm. Samples
were diluted with dichloromethane (to approximately 1%),
and 1 µL portions thereof were analyzed using a 20:1 split
mode.
(99%). Sulfuric acid (4 N, 4.8 L) was added, and the reaction
mixture was slowly heated with stirring. The ethanol formed
was distilled off at such a reflux rate that the still-head
temperature was kept between 79 and 85 °C (inner temper-
ature to 86-96 °C). A total of 2750 mL of distillate was
collected over a period of about 5 h. The reaction mixture
was cooled to room temperature and extracted twice with
3.4 L of MTBE. The combined MTBE phases were washed
with 2.7 L of water, and the MTBE was removed in vacuo.
The obtained oil was then distilled (short path) at 84-106
°C inner temperature (bp: 78-80 °C/4.0 mbar) to give 2.117
kg (67% yield based on 93% HPLC-purity) of 1 (exothermic
DSC-onset: 175 °C) as a colorless oil containing 4% (HPLC)
1
of 7. H NMR (CDCl3, 200 MHz) δ 1.28 (t, 6H), 3.74 (m,
4H), 5.49 (s, 1H), 7.31 (m, 3H), 7.48 (m, 2H).
E-5-Phenyl-pent-2-en-4-ynoic Acid Ethyl Ester (2).
NaOEt (96%) (699 g, 9.86 mol) was suspended in 7.25 L of
dry toluene under a nitrogen atmosphere. Triethyl phos-
phonoacetate (Fluka 99%) (2.112 kg, 9.42 mol) dissolved
in 1.2 L of dry toluene was added dropwise to the stirred
suspension, keeping the inner temperature between 21 and
24 °C. The mixture was stirred for an additional 30 min at
that temperature after complete addition. Phenylpropargyl-
aldehyde 1 (93%) (1.115 kg, 8.22 mol) dissolved in 2.0 L
of dry toluene was added dropwise over 1.5 h to the stirred
mixture, keeping the inner temperature between at 20-25
°C. The mixture was stirred for an additional 60 min at that
temperature after complete addition. Sulfuric acid (4 N, 6.6
L) was then added cautiously, keeping the inner temperature
between 0 and 25 °C. The organic phase was separated and
washed twice with 4.4 and 2.2 L of aqueous NaHCO3 (10%).
The toluene was then removed in vacuo to give 2.013 kg
(100% ) 1.645 kg) of crude 2 as brown oil,48 which was
used in the next step without purification. 1H NMR (CDCl3,
200 MHz) δ 1.31 (t, 3H), 4.24 (q, 2H), 6.29 (d, 1H, J ) 15
Hz), 6.98 (d, 1H, J ) 15 Hz), 7.3 (m, 3H), 7.48 (m, 2H).
E-4-Phenylbut-1-en-3-yn-1-ol (3). Caution: This step
should only be carried out in a well-ventilated fume hood
due to the formation of phosphine originating from phosphor-
containing impurities! Raw E-5-phenyl-pent-2-en-4-ynoic
acid ethyl ester 2 (1.266 kg, ∼5.17 mol) dissolved in 540
mL of dry toluene was added dropwise over 1.5 h to 6.48
kg (11.39 mol) of diisobutylaluminium hydride (DIBAL-H
25 wt % solution in toluene), keeping the inner temperature
between 15 and 20 °C under a nitrogen atmosphere (gas
eVolution!). The reaction mixture was stirred at 15-20 °C
inner temperature for another 30 min after complete addition.
The reaction mixture was then added cautiously over 2.5 h
to 9.8 L of hydrochloric acid (4 N) with stirring, keeping
the inner temperature between 15 and 25 °C (exothermic!).
The organic phase was separated and washed with 4.0 L of
water. The toluene was removed in vacuo to give 903 g of
an oil, which was distilled (short path) at 134-170 °C oil
bath temperature (bp: 114-118 °C/1.5-3.0 mbar) to give
356 g (41% yield based on 95% HPLC-purity over two
steps: {1f2f3} of 3 (exothermic DSC-onset: 253 °C) as
Phenylpropargylaldehyde Diethyl Acetal (7). Into a
reaction flask equipped with a thermometer and a 25-cm
fractionating column (Raschig ring filled) were charged 2.90
kg (28.4 mol) of phenylacetylene 6 (exothermic DSC-onset:
189 °C) (98%), 4.21 kg (28.4 mol) of triethyl orthoformate
(99%), and 170.5 g (0.53 mol) of zinc iodide (>98%).
Ethanol was slowly distilled from the reaction mixture, which
was heated to about 117 °C inner temperature before
refluxing in the still-head was begun. A total of 1.9 L of
distillate (bp 76-80 °C) was collected over a period of 5 h
as the inner temperature of the reaction mixture gradually
rose to 196 °C (due to external heating). The reaction mixture
was cooled to room temperature and filtered through a pad
of Celite. The filtrate was then distilled under reduced
pressure at 3.5 mbar. A total of 700 g of fore-run was
discarded until a boiling point of 111 °C was reached. The
remaining main fraction was then distilled at 112-130 °C
inner temperature (bp: 102-104 °C/2.0 mbar) to give 4.00
kg (69% yield, HPLC-purity: 99%) of 6 (exothermic DSC-
1
onset: 293 °C) as a colorless oil. H NMR (CDCl3, 200
MHz) δ 7.40 (m, 2H), 7.48 (m, 1H), 7.59 (m, 2H), 9.39 (s,
1H).
Phenylpropargylaldehyde (1). Into a reaction flask
equipped with a nitrogen inlet, a thermometer, and a variable
takeoff splitter distillation head fitted with a 27-cm frac-
tionating column (Raschig ring filled) was charged 4.676
kg (22.9 mol) of phenylpropargylaldehyde diethyl acetal 7
1
a colorless oil. H NMR (CDCl3, 200 MHz) δ 1.88 (br s,
(47) Advance Chemistry Development Inc., 90 Adelaide Street West, Toronto,
Ontario M5H 3V9, Canada.
(48) Exothermic DSC-onset of pure 2 (after distillation): 248 °C.
Vol. 8, No. 3, 2004 / Organic Process Research & Development
•
369

1H), 4.25 (dd, 2H), 5.96 (dd, 1H), 6.34 (dd, 1H), 7.27 (m,
3H), 7.42 (m, 2H).
MTBE. The MTBE of the combined organic phases was
removed in vacuo to give 1.257 kg of NNC 61-4655 (98%
yield based on 94% HPLC-purity) as an oil, which solidified
on standing to give a brown wax. NNC 61-4655 was used
in the following crystallization without purification. 1H NMR
400 MHz (acetone-d6) δ 1.10 (3H, t, J ) 7 Hz), 2.90 (1H,
dd, J ) 14 Hz, 8 Hz), 3.01 (1H, dd, J ) 14 Hz, 4 Hz), 3.38
(1H, dq, J ) 7 Hz, 9 Hz), 3.63 (1H, dq, J ) 7 Hz, 9 Hz),
4.04 (1H, dd, J ) 8 Hz, 4 Hz), 4.69 (2H, dd, J ) 5.3 Hz,
1.8 Hz), 6.14 (1H, dt, J ) 16 Hz, 1.8 Hz), 6.43 (1H, dt, J )
16 Hz, 5.3 Hz), 6.90 (2H, d, J ) 8.5 Hz), 7.21 (2H, d, J )
8.5 Hz), 7.36-7.40 (3H, m), 7.43-7.47 (2H, m). 13C NMR
100 MHz (acetone-d6) δ 15.41, 38.77, 66.37, 68.09, 80.53,
88.03, 90.89, 112.38, 115.19, 124.01, 129.30, 129.39, 130.96,
131.32, 132.20, 139.66, 158.04, 173.38.
(S)-E-2-Ethoxy-3-[4-(5-phenylpent-2-en-4-ynyloxy)phen-
yl]propanoic Acid, 2-Aminoethanol (NNC 61-4655, 2-
Aminoethanol). (S)-E-2-Ethoxy-3-[4-(5-phenylpent-2-en-4-
ynyloxy)phenyl]propanoic acid NNC 61-4655 (94%) (977.0
g, 2.62 mol) was dissolved in 4.6 L of ethanol (96%). The
solution was heated to reflux, and 178.7 g (2.90 mol) of
2-aminoethanol (99%) dissolved in 280 mL of ethanol (96%)
was added dropwise to the boiling mixture. The mixture was
slowly allowed to cool to ambient temperature overnight with
stirring (crystallization starts at 70 °C). The formed crystals
were filtered off and washed 3× with 500 mL of ethanol
(96%). The crystals were dried at 30 °C in vacuo overnight
and sieved (1.0 mm) to give 935.8 g of NNC 61-4655,
2-aminoethanol (86% yield based on 98.7% HPLC-purity)
as light-yellow crystals (DSC onset: 152.5 °C, DSC melt-
ing: 155.1 °C) (containing 0.31% cis-isomer and e1.7%
R-isomer). Anal. Calcd for C, H, N [corrected for 0.07%
water (Karl Fischer)]: C, 70.00; H, 7.11; N, 3.40. Found:
C, 70.04; H, 7.34; N, 3.36. ESI-MS: 351 (MH+). 1H NMR
400 MHz (DMSO-d6) δ 1.00 (3H, t), 2.64-2.70 (1H, m),
2.79 (2H, t), 2.84-2.88 (1H, m), 3.13-3.21 (1H, m), 3.52-
3.58 (3H, m), 3.63-3.67 (1H, m), 4.65 (2H, d), 6.14 (1H,
d), 6.41 (1H, dt), 6.84 (2H, d), 7.14 (2H, d), 7.37-7.41 (3H,
m), 7.43-7.48 (2H, m).
5-Phenylpent-1-en-4-yn-3-ol (12). Phenylaetylene 6 (98%)
(128.7 g, 1.23 mol) was added dropwise over 1 h to 350.0
g (1.2 mol) of n-hexyllithium in n-hexane (2.4 N) under an
inert atmosphere, keeping the inner temperature between 5
and 10 °C. MTBE (148 g) was added to the suspension,
which was stirred for 2 h at room temperature. The mixture
was cooled to -5-0 °C, and a solution of 56.1 g (0.95 mol)
of acrolein (95%) in 222 g of MTBE was added over 1 h at
that temperature. The suspension slowly became a homo-
geneous mixture after it slowly (30 min) had been heated to
room temperature. This mixture was slowly added to a slurry
of 79 g of acetic acid (98%) and 290 g of ice, keeping the
inner temperature below 30 °C. The organic phase was
separated, and the water phase was extracted with 55 g of
MTBE. The combined organic phases were washed with 100
mL water, and the solvent was removed in vacuo to give
198.9 g of 12 (86-93% yield based on 65-70% GC purity)
as orange oil, which was used in the next step without
purification (Important note: A thorough safety investigation
E-5-(Chloropent-3-en-1-ynyl)benzene (8) from Phenyl-
but-1-en-3-yn-1-ol 3. E-4-Phenylbut-1-en-3-yn-1-ol 3 (95%)
(678.3 g, 4.07 mol) was dissolved in 1.9 L of DMF.
Methanesulphonyl chloride (540.0 g, 4.71 mol) was added
dropwise with stirring, and the inner temperature rose to 70
°C (ice-bath cooling is necessary). The reaction mixture was
further stirred at that temperature for 1.5 h. The reaction
mixture was cooled to room temperature, and 1.9 L of water
was added within 30 min (temperature rose to 33 °C). The
mixture was extracted twice with 1.3 L of toluene. The
combined organic phases were washed twice with 1.3 L of
water. The toluene was removed in vacuo to give 755.0 g
of raw 9 (exothermic DSC-onset: 219 °C) (96% yield based
on 91% HPLC-purity) as an oil (containing <4% cis-isomer
and e3% 3), which was used in the next step without
1
purification. H NMR (CDCl3, 200 MHz) δ 4.12 (dd, 2H),
5.98 (dd, 1H), 6.28 (dd, 1H), 7.30 (m, 3H), 7.42 (m, 2H).
(S)-E-2-Ethoxy-3-[4-(5-phenylpent-2-en-4-ynyloxy)phen-
yl]propanoic Acid Isopropyl Ester (10). (S)-2-Ethoxy-3-
(4-hydroxyphenyl)propanoic acid isopropyl ester 9 (99%,
containing 0.3% R-isomer) (1.042 kg, 4.09 mol) and 700.0
g (3.60 mol) of E-5-(chloropent-3-en-1-ynyl)benzene 8 (91%)
were dissolved in 2.4 L of acetone followed by the addition
of 821.0 g (5.94 mol) of potassium carbonate and 66.0 g
(0.396 mol) of potassium iodide. The mixture was refluxed
for 5 h and then allowed to cool to room temperature. MTBE
(1.35 L), and 4.6 L of water were added to the mixture. The
organic phase was separated and washed twice with 1.2 L
of aqueous NaOH (1 N) and with 1.2 L of saturated NaCl
solution. The solvents were removed in vacuo to give 1.468
kg of 10 (96% yield based on 92% HPLC purity) as an oil,
1
which was used in the next step without purification. H
NMR 400 MHz (CDCl3) δ 1.15 (3H, d, J ) 7 Hz), 1.16
(3H, t, J ) 7 Hz), 1.22 (3H, d, J ) 7 Hz), 2.94 (2H, d, J )
6.6 Hz), 3.35 (1H, m), 3.59 (1H, m), 3.94 (1H, t, J ) 6.6
Hz), 4.59 (2H, dd, J ) 5.3 Hz, 1.8 Hz), 4.98-5.07 (1H, m),
6.05 (1H, dt, J ) 15.8 Hz, 1.8 Hz), 6.36 (1H, dt, J ) 15.8
Hz, 5.3 Hz), 6.82 (2H, d, J ) 9 Hz), 7.16 (2H, d, J ) 9 Hz),
7.28-7.30 (3H, m), 7.41-7.43 (2H, m). 13C NMR 100 MHz
(CDCl3) δ 15.51, 22.23, 38.83, 66.48, 68.05, 68.75, 80.81,
87.55, 91.01, 112.66, 114.92, 123.54, 128.73, 130.12, 130.92,
131.93, 138.17, 157.50, 172.52.
(S)-E-2-Ethoxy-3-[4-(5-phenylpent-2-en-4-ynyloxy)phen-
yl]propanoic Acid (NNC 61-4655). A mixture of 1.466 kg
(3.44 mol) of (S)-E-2-ethoxy-3-[4-(5-phenylpent-2-en-4-
ynyloxy)phenyl]propanoic acid isopropyl ester 10 (92%), 4.9
L of ethanol (96%), and 5.6 L of aqueous NaOH (1 N)
solution was heated for 3 h at 70 °C inner temperature with
stirring. The ethanol was distilled off under reduced pressure,
and 20.6 L of saturated NaCl solution was added to the
reaction mixture, which was kept at 60 °C to obtain a clear
solution. The basic water phase was then washed 3× with
2.0 L of MTBE at 55 °C. MTBE (2.0 L) was added to the
mixture, and the water phase was acidified to pH 2 with
concentrated hydrochloric acid. The organic phase was
separated, and the water phase was extracted with 2.0 L of
370
•
Vol. 8, No. 3, 2004 / Organic Process Research & Development

should be carried out for this reaction before further scale
up). 1H NMR 200 MHz (CDCl3) δ 3.1 (1H, s br), 5.14 (1H,
d, J ) 5 Hz), 5.26 (1H, d, J ) 10 Hz), 5.52 (1H, d, J ) 16
Hz), 6.09 (1H, ddd, J ) 5/10/16 Hz), 7.23-7.36 (3H, m),
7.40-7.52 (2H, m).
(bp: 96-103 °C/0.6-0.8 mbar) to give 92 g of crude 8.
The crude 8 was fractionally redistilled with a Sulzer column
(20 theoretical plates) at e130 °C inner temperature (bp:
82-85 °C/0.7-1.0 mbar) to give 44.0 g of 8 (38% yield
based on 97% HPLC-purity) as an oil (containing e3% cis-
isomer).
E-5-(Chloropent-3-en-1-ynyl)benzene (8) from 5-Phen-
ylpent-1-en-4-yn-3-ol (12). Thionyl chloride (112.5 g, 0.96
mol) was added dropwise over 1 h to a solution of 145.3 g
(0.63 mol) of crude 5-phenylpent-1-en-4-yn-3-ol (12) (69%)
dissolved in 155 g of toluene, keeping the inner temperature
between 40 and 45 °C (gas eVolution: SO2, HCl!). The
mixture was then stirred for a further 12 h at that temperature
and then slowly added (remainders were transferred with an
additional 50 g of toluene) to a slurry of 139.8 g of potassium
carbonate in 50 g of water and 250 g of ice with vigorous
stirring (inner temperature e 40 °C). If necessary, the pH
was adjusted to 7, and the organic phase was separated. The
organic phase was then washed with 100 mL of water at
35-40 °C, and the solvent was removed in vacuo to give
151.7 g of an oil (containing 59% 8 and 12% cis-isomer),
which was distilled (short path) at e120 °C inner temperature
Acknowledgment
We acknowledge scientific advice and fruitful discussions,
in particular concerning salt preparation and selection with
Anette Frost Jensen. We thank Per Steen Klifforth, Susanne
Bjerggaard Laustsen, Pernille Rasmussen Flemming Gun-
dertofte, Vibeke Bødstrup, Anne Ryager, and Thomas Seigert
for excellent technical assistance. We thank Lars Iversen,
Paul Stanley Bury, Helle Brødholt, Ole Ekelund, and Akio
Bergholdt for their scientific advice and for fruitful discus-
sions.
Received for review April 7, 2003.
OP034048J
Vol. 8, No. 3, 2004 / Organic Process Research & Development
•
371