Practical and scalable synthesis of a selective CCK1 receptor antagonist

Article abstract of DOI:10.1021/jo1017684

We describe a practical and scalable route to compound (Z)-1, a selective CCK₁ receptor antagonist. Notable features of this concise route are (1) a regioselective construction of the pyrazole core through the reaction of an aryl hydrazine and

Full text of DOI:10.1021/jo1017684

pubs.acs.org/joc
be regulated through agonism of the CCK1 receptor. There-
Practical and Scalable Synthesis of a Selective CCK1
Receptor Antagonist
fore, a number of CCK1 antagonists have been evaluated in
the clinic for potential application in irritable bowel syn-
drome (IBS), nonulcer dyspepsia, biliary colic, chronic con-
stipation, and pancreatic cancer.3,2a Our laboratories have
also disclosed a series of selective, nonpeptide CCK1 antago-
nists.4 Recent efforts in this area identified compound (Z)-1
(Scheme 1) as a molecule of interest for the continued study
of CCK1 antagonism and its applications within the mam-
malian system.
Christopher M. Mapes, Neelakandha S. Mani,
Xiaohu Deng,* Chennagiri R. Pandit, Kelly J. McClure,
Marna C. W. Pippel, Clark A. Sehon, Laurent Gomez,
Shirin Shinde, J. Guy Breitenbucher, and Todd K. Jones
Johnson & Johnson Pharmaceutical Research and
Development, LLC, 3210 Merryfield Row, San Diego,
California 92121, United States
SCHEME 1. The Initial Synthesis of Compound (Z)-1
xdeng@its.jnj.com
Received September 8, 2010
The original synthesis of compound (Z)-1 involves a six-
step sequence.5 The most challenging step of the synthesis
was the construction of the trisubstituted Z olefin. Perkin
condensation6 of aldehyde 2 with 3-chlorophenyl acetic acid
afforded exclusively the thermodynamically more stable
(E)-1. Photoisomerization followed by HPLC separation
was required to produce the desired (Z)-1 in low yield. An
alternative stereoselective route was recently reported fea-
turing a sequential functionalization of a geminal dibromo
vinyl precursor.7 However, the extreme cryogenic conditions
(-116 °C) and the low yield precluded its use for the further
up scaling. Herein, we report efforts that culminated in a
practical and scalable synthesis of compound (Z)-1.
We recognized that the key issue to address for the
synthesis of (Z)-1 was installation of the Z configured trisub-
stituted olefin. Stereoselective synthesis of thermodynami-
cally disfavored Z-R,β-unsaturated acid derivatives is a
classic problem in organic synthesis. Careful examina-
tion of the available olefination methods in the literature
indicated that the Horner-Wadsworth-Emmons (HWE)
We describe a practical and scalable route to compound
(Z)-1, a selective CCK1 receptor antagonist. Notable
features of this concise route are (1) a regioselective
construction of the pyrazole core through the reaction
of an aryl hydrazine and an elaborated acetylenic ketone,
(2) a Tf2O/pyridine mediated Z-selective dehydration of
an R-hydroxyester, and (3) a stereoselective hydrolysis.
The sequence is high-yielding and amenable for large-
scale synthesis.
Cholecystokinin (CCK) was first identified from extracts
of porcine intestinal mucosa in 1966. A 33 amino acid
peptide hormone, CCK regulates a variety of physiological
processes in response to food intake.1 These processes in-
clude gall bladder contraction, pancreatic enzyme secretion,
gastric acid secretion, gastric emptying, and duodenal/colonic
motility. In addition, CCK is abundantly found in the CNS and
is thought to be involved in aspects of nociception, satiety,
anxiogenesis, memory, and learning.2
The biological actions of CCK are mediated through two
G-protein coupled receptors, CCK1 (formerly CCK-A) and
CCK2 (formerly gastrin/CCK-B). The actions of CCK on
gall bladder contraction, pancreatic enzyme secretion, duo-
denal motility, and gastric emptying rate are believed to
(3) (a) Varga, G. Curr. Opin. Invest. Drugs 2002, 3, 621. (b) Varga, G.;
Balint, A.; Burghardt, B.; D’Amato, M. Br. J. Pharmacol. 2004, 141, 1275. (c)
Peter, S.; D’Amato, M.; Beglinger, C. Dig. Dis. 2006, 24, 70.
(4) (a) McClure, K.; Hack, M.; Huang, L.; Sehon, C.; Morton, M.; Li, L.;
Barrett, T.; Shankley, N.; Breitenbucher, J. Bioorg. Med. Chem. Lett. 2006,
16, 72. (b) Sehon, C.; McClure, K.; Hack, M.; Morton, M.; Gomez, L.; Li, L.;
Barrett, T.; Shankley, N.; Breitenbucher, J. Bioorg. Med. Chem. Lett. 2006,
16, 77. (c) Gomez, L.; Hack, M.; McClure, K.; Sehon, C.; Huang, L.;
Morton, M.; Li, L.; Barrett, T.; Shankley, N.; Breitenbucher, J. Bioorg.
Med. Chem. Lett. 2007, 23, 6493.
(5) (a) Murray, W.; Wachter, M. J. Heterocycl. Chem. 1989, 26, 1389. (b)
Murray, W. V.; Hadden, S. K.; Wachter, M. P. J. Heterocycl. Chem. 1990, 27,
1933. (c) Barrett, T.; Breitenbucher, J.; Gomez, L.; Hack, M.; Huang, L.;
McClure, K.; Morton, M.; Sehon, C.; Shankley, N. WO 2004007463.
(6) (a) Perkin, W. H. J. Chem. Soc. 1868, 53, 181. (b) Perkin, W. H.
J. Chem. Soc. 1877, 31, 388.
(1) Jorpes, E.; Mutt, V. Acta Physiol. Scand. 1966, 66, 196.
(2) (a) Herranz, R. Med. Res. Rev. 2003, 23, 559. (b) Tullio, P.; Delarge, J.;
Pirotte, B. Expert Opin. Invest. Drugs 2000, 9, 129.
(7) Gomez, L.; Wu, J.; Mani, N. S.; Basu, S.; Moravek, J.; Breitenbucher,
J. Tetrahedron Lett. 2010, 51, 1110.
7950 J. Org. Chem. 2010, 75, 7950–7953
Published on Web 10/27/2010
DOI: 10.1021/jo1017684
r
2010 American Chemical Society

Mapes et al.
JOCNote
TABLE 1. The HWE Reaction with Different Phosphonatesa
SCHEME 2. The Stereoselective Hydrolysis
entry
R1
R
Z/E ratio
yield, %
1
2
3
4
Et
Et
Me
Me
Et
3:7
4:1
3:1
4:1
90
84
95
50
CF3CH2
CF3CH2
Ph
TABLE 2. Hydrolysis of Pure (Z)-3
aConditions: 1.2 equiv of KOBut, 3 equiv of 18-crown-6, THF, 0
°C-rt.
reaction appeared most promising for stereoselective Z-
olefination. With easily accessible aldehyde 2 in hand, we
decided to study the HWE reaction in detail. The classic
HWE conditions with diethyl phosphonate favored the
formation of the E isomer (Table 1, entry 1). Still and
Gennari have previously demonstrated that electronically
modified HWE reagents can have profound effects on the
stereochemical course of these reactions.8 Specifically,
employment of electron-withdrawing bistrifluoroethyl phos-
phonate reagents was shown to promote Z-olefin forma-
tion. Indeed, in our case, the Still-Gennari variant of the
HWE reagent gave rise to a 4:1 Z:E mixture (entry 2). We
also investigated the effect of changes to R1; substitution of
ethyl with methyl afforded a lower Z to E ratio but a higher
yield (entry 3). The Ando modification of the HWE reaction9
resulted in comparable Z selectivity, but in a lower yield
(entry 4). Our attempts to further improve the Z selectivity of
the Still-Gennari protocol by the use of additives, changes
in addition order or time, and solvent switches were largely
unsuccessful.
With the 3:1 Z:E mixture of 3 in hand, we reasoned that
a significant rate difference in hydrolysis of the Z and E olefinic
acid esters10 might provide a facile isolation of the desired
Z isomer. Indeed, we found that under carefully controlled
conditions (room temperature, 2.0 equiv of LiOH, 3:1:1 THF:
EtOH:H2O), the undesired ester (E)-3 was preferentially hydro-
lyzed to acid (E)-1 whereas the desired ester (Z)-3 remained
intact. The polarity difference of acid (E)-1and ester (Z)-3made
isolation rather easy (Scheme 2).
With pure ester (Z)-3 in hand, the final hydrolysis to
compound (Z)-1 was surprisingly difficult because of the
thermodynamic instability of the Z olefin.11 Aqueous alka-
line hydrolysis under reflux conditions resulted in the com-
plete isomerization of the double bond to (E)-1 (Table 2,
entry 1). Different hydroxide counterions (Liþ, Naþ, and
Kþ) did not significantly impact the outcome (entries 2-4).
Changing the solvent systems to MeOH/H2O (B) or THF/
H2O (C) greatly slowed the hydrolysis reaction probably due
to the poor solubility of (Z)-3 (entries 5-7). Increasing the
quantity of base to 10 equiv sped up the hydrolysis reaction
but also the isomerization process (entry 7). Switching to
an alternate solvent system, 1,4-dioxane/H2O (D) greatly
entry
base
equiv
solventsa
T, °C
(Z)-3:(E)-1:(Z)-1b
1
2
3
4
5
6
7
8
9
10
LiOH
KOH
NaOH
LiOH
LiOH
LiOH
LiOH
LiOH
LiOH
LiOH
5
5
5
5
5
5
10
5
5
3
A
A
A
A
B
C
C
D
D
D
reflux
50
50
50
50
50
55
55
75
0:100:0
5:20:75
35:12:53
15:21:64
90:1:9
90:2:8
0:95:5
10:0:90
0:3:97
0:3:97
75
aSolvent A: 2-Me-THF/MeOH/H2O. Solvent B: MeOH/H2O. Sol-
vent C: THF/H2O. Solvent D: 1,4-dioxane/H2O. bOn the basis of
integration of HPLC peaks.
facilitated the hydrolysis process (entry 8). Increasing the
temperature to 75 °C drove the reaction to completion but
also resulted in ∼3% isomerization (entry 9). The amount of
base could be reduced to 3.0 equiv to provide an almost
identical result (entry 10). Fortunately, the small amount of
undesired (E)-1 could be easily excluded by trituration from
a mixed solvent of DCM/EtOAc/hexanes. Therefore, the
first chromatography-free route for the synthesis of com-
pound (Z)-1 on the multigram scale was developed.
Although the above route provided compound (Z)-1 on
a multigram scale, the moderate 3:1 Z:E selectivity in the key
olefination step and the reliance on the noncommercially
available Still-Gennari phosphonate negatively affected the
overall throughput of this process. We embarked on pursu-
ing a new synthetic strategy to construct the Z olefin. Almost
all the classic olefination reactions such as the HWE reaction
rely on selective dehydration of a β-hydroxyester to set the
olefin geometry (Figure 1). This often favors the thermo-
dynamically more stable E isomer, as we experienced above
and witnessed in many examples in the literature. Recently,
we have demonstrated that Tf2O/pyridine-mediated dehy-
dration of R-hydroxyesters provides a powerful method for
stereoselective synthesis of Z-R-arylacrylates on simple
substrates.12 Whether this new method could apply to fully
elaborated substrate 4 remained to be investigated.
To answer this question, an easy access to precursor 4 was
required. Exploiting our previous experience on the synthesis of
analogous pyrazole compounds,13 we devised a concise route
for key precursor 4 (Scheme 3). Starting from inexpensive
3-chloromandelic acid, protection of the R-hydroxyacid as
(8) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405.
(9) Ando, K. J. Org. Chem. 1998, 63, 8411.
(10) (a) Schmid, R.; Partali, V.; Anthonsen, T.; Anthonsen, H. W.;
Kvittingen, L. Tetrahedron Lett. 2001, 42, 8543. (b) Kuroda, C.; Sunakawa,
T.; Muguruma, Y. Helv. Chim. Acta 2008, 91, 888.
(12) Mani, N.; Mapes, C.; Wu, J.; Deng, X.; Jones, T. J. Org. Chem. 2006,
13, 5039.
(13) Liang, J.; Mani, N.; Jones, T. J. Org. Chem. 2007, 72, 8243.
(11) The Z isomer was found to be photosensitive. Light exclusion was
necessary during the hydrolysis process.
J. Org. Chem. Vol. 75, No. 22, 2010 7951

Mapes et al.
JOCNote
The whole sequence is high-yielding and amenable for large-
scale synthesis.
Experimental Section
HWE Synthesis of Compound 3 (Table 1, 3:1 Z:E Mixture). In
a 250-mL, two-necked, round-bottomed flask, [bis-(2,2,2-tri-
fluoroethoxy)phosphoryl]-(3-chlorophenyl)acetic acid methyl
ester (6.5 g, 15.0 mmol, 1.1 equiv) and 18-crown-6 (10.9 g,
41 mmol, 3.0 equiv) were dissolved in anhydrous THF (83 mL).
At 0 °C, KOBut (1.0 mol/L in THF, 16.5 mL, 1.2 equiv) was
added dropwise over 15 min. After the mixture was stirred at
0 °C for 0.5 h, 5-benzo[1,3]dioxol-5-yl-1-(2,5-dichloro-phenyl)-
1H-pyrazole-3-carbaldehyde 2 (5.0 g, 13.8 mmol, 1.0 equiv) was
added portion wise as a solid. Upon completion of the addition,
the reaction solution was stirred at room temperature overnight.
Saturated NH4Cl aqueous solution (50 mL) was added to
quench the reaction and the resulting suspension was filtered
through a pad of Celite and washed with EtOAc. The layers were
separated and the aqueous layer was extracted with EtOAc (2 ꢀ
50 mL). The organic layers were combined, washed with brine,
dried over MgSO4, filtered, and concentrated to yield 3 as a red-
brown oil (7.3 g, 13.8 mmol, 100%) as a mixture of 3:1 Z:E
isomers (the isomeric ratio was determined by the 1H NMR
peak integration of the crude product).
FIGURE 1. The alternative dehydration strategy.
SCHEME 3. Second-Generation Synthesis of Compound (Z)-1
Stereoselective Hydrolysis to Remove (E)-1 (Scheme 2). In a
250-mL, round-bottomed flask, the 3:1 Z:E mixture of com-
pound 3 (7.3 g, 13.8 mmol, 1.0 equiv) was dissolved in THF/
MeOH (3:1, 120 mL) and the LiOH (0.66 g, 0.027 mol, 2.0 equiv)
solution in water (30 mL) was added. After the mixture was
stirred at room temperature for 3 h, 1 mol/L of HCl was added
to adjust the pH to 4. The organic solvents (MeOH and THF)
were removed under vacuum and the remaining aqueous solu-
tion was extracted with EtOAc (3 ꢀ 50 mL). The organic layers
were combined, washed with brine, dried over MgSO4, filtered,
and concentrated. The crude oily product was directly loaded
onto a short pad of silica gel and washed with 10-20% EtOAc/
hexanes to isolate pure ester (Z)-3 (4.9 g, 67% yield) as a pale
yellow oil.
Second-Generation Synthesis of Compound (Z)-1 (Scheme 3):
Compound 8 In a 500-mL, one-necked, round-bottomed flask
fitted with a reflux condenser and magnetic stir bar was charged
compound 7 (crude, 22.6 g, 56 mmol, 1.0 equiv) prepared in
the previous step (see the Supporting Information), then EtOH
(300 mL) was added. To this suspension was added 2,5-dichloro-
phenylhydrazine (10 g, 56.5 mmol, 1.0 equiv) as a solid. The
resulting suspension was heated to form a homogeneous solution
and stirred at reflux temperature under air for 5 h (end point
based on periodic HPLC analysis of the reaction). The solution
was cooled to room temperature and the solvent was evaporated
under reduced pressure. The crude product was passed through a
short pad of silica gel with 10% EtOAc/hexanes as eluents and
then recrystallized from hot MeOH to afford pure 8 as a white solid
(24.2 g, 42.3 mmol, 77%, for 2 steps). Mp 152-153 °C. 1H NMR
(500 MHz, CDCl3) δ 7.72 (s, 1 H), 7.68-7.62 (m, 1 H), 7.38-7.30
(m, 5 H), 6.70 (d, J = 8.2 Hz, 1 H), 6.66-6.60 (m, 2 H), 6.36 (s,
1 H), 5.95 (s, 2 H), 3.44 (d, J = 14.6 Hz, 1 H), 3.26 (d, J = 14.6 Hz,
1 H), 1.42 (s, 6 H). 13C NMR (151 MHz, CDCl3) δ 172.1, 147.9,
147.8, 147.75, 145.7, 141.7, 138.8, 134.5, 133.1, 131.1, 130.4, 130.3,
130.0, 129.8, 128.4, 125.4, 123.5, 123.2, 121.9, 110.9, 108.5, 108.2,
107.0, 101.3, 82.9, 40.7, 27.9, 27.5. HRMS-ESI (m/z) [M þ H]þ
calcd for C28H22N2Cl3O5 571.0589, found 571.0570.
a 1,3-dioxolanone followed by a base-mediated alkylation
with propargyl bromide afforded 6 in excellent yield. Sono-
gashira cross-coupling reaction of 6 with 1,3-benzodioxole-
5-carbonyl chloride followed by pyrazole formation reaction
with 2,5-dichlorohydrazine provided compound 8 in 77%
yield over 2 steps. It is worth noting that the pyrazole
formation reaction was highly regiospecific with only the
desired regioisomer isolated. Deprotection of the 1,3-dioxo-
lanone ring provided the requisite R-hydroxyester precursor
4 in good yield. Gratifyingly, we found that the standard
stereoselective dehydration conditions (Tf2O, pyridine,
DCM) developed previously worked well with compound 4
to provide an excellent 30:1 Z to E selectivity and 85%
isolated yield of 3 on multigram scales. At last, the same
stereoselective hydrolysis protocol afforded (Z)-1 in excel-
lent yield and selectivity. After trituration from DCM/EtOAc/
hexanes, pure compound (Z)-1 was obtained in 7 linear steps
and in a 41% overall yield.
In conclusion, two synthetic routes were developed dur-
ing large-scale synthetic campaigns of compound (Z)-1,
a selective CCK1 receptor antagonist. The classic Horner-
Wadsworth-Emmons reaction and its variants provided
modest Z-E stereoselectivity (3:1) for the construction of
the trisubstituted olefin. A stereoselective hydrolysis was
then developed for nonchromatographic isolation of the Z
isomer. The second route capitalized on the novel stereo-
selective R-hydroxyester dehydration reaction and the regio-
selective pyrazole synthesis previously developed in this
lab to formulate a new synthetic strategy. In the end, com-
pound (Z)-1 was produced from commercially available
materials in 7 linear steps with an overall yield of 41%.
Compound 4. In a 1-L, round-bottomed flask, compound 8
(20.2 g, 0.035 mol, 1.0 equiv) was dissolved in anhydrous
methanol (500 mL) and NaOMe (2.10 g, 0.039 mol, 1.1 equiv)
was added in one portion. After the mixture was stirred at room
temperature overnight, water (500 mL) was added. The reaction
mixture was extracted with EtOAc (3 ꢀ 200 mL). The organic
7952 J. Org. Chem. Vol. 75, No. 22, 2010

Mapes et al.
JOCNote
layers were combined, washed with brine, dried over MgSO4,
filtered, and concentrated. The crude product was passed
through a short pad of silica gel with 25% EtOAc in hexanes
as the eluents to yield pure 4 as a light yellow oil (16.4 g, 0.030
123.1, 122.2, 108.5, 108.3, 105.8, 101.4, 52.6. HRMS-ESI (m/z)
[MþH]þ calcd for C26H18Cl3N2O4 527.0327, found 527.0349.
Compound (Z)-1. To a 500-mL, round-bottomed flask were
added compound 3 (14.0 g, 0.0265 mol, 1.0 equiv), 1,4-dioxane
(140 mL), water (140 mL), and LiOH (3.17 g, 0.132 mol,
5.0 equiv) sequentially. The flask was then submerged into a
preheated oil bath at 75-77 °C for 12 h.14 After the reaction was
complete, the reaction solution was cooled to 0 °C and the pH
was adjusted to ∼4 with 6 N HCl. The acidified solution was
extracted with EtOAc (3 ꢀ 100 mL). The organic layers were
combined, washed with brine, dried over MgSO4, filtered, and
concentrated.15 The crude product was dissolved in CH2Cl2
(100 mL) and 10% EtOAc in hexanes (100 mL) was slowly
added with stirring. Precipitation began immediately. The stir
bar was removed and the flask was left undisturbed at room
temperature for several hours. The solids were collected and
dried in a vacuum oven at 60 °C to yield the title compound
(11.1 g, 82% yield) as an off white crystalline powder. Mp
174-175 °C. 1H NMR (400 MHz, CDCl3) δ 7.51-7.47
(m, 1H), 7.46-7.35 (m, 4H), 7.35-7.28 (m, 2H), 7.05 (s, 1H),
6.76-6.71 (m, 1H), 6.70 (s, 1H), 6.69-6.64 (m, 2H), 5.98
(s, 2H). 13C NMR (101 MHz, CDCl3) δ 167.5, 148.6, 148.0,
147.5, 146.6, 140.9, 137.4, 134.6, 134.1, 133.4, 131.5, 131.2,
130.4, 129.6, 129.4, 128.3, 128.2, 126.7, 126.5, 122.4, 121.9,
108.7, 108.5, 108.3, 101.6. HRMS-ESI (m/z) [M þ H]þ calcd
for C25H16Cl3N2O4 513.0170, found 513.0195.
1
mol, 85% yield). H NMR (400 MHz, CDCl3) δ 7.73 (s, 1H),
7.59 (dt, J = 6.9, 2.0 Hz, 1H), 7.43-7.27 (m, 5H), 6.75-6.55 (m,
3H), 6.32 (s, 1H), 5.96 (s, 2H), 4.73 (s, 1H), 3.74 (s, 3H), 3.70
(d, J = 14.7 Hz, 1H), 3.26 (d, J = 14.7 Hz, 1H). 13C NMR (101
MHz, CDCl3) δ 174.4, 149.8, 148.4, 148.2, 145.9, 143.5, 139.1,
134.8, 133.4, 131.6, 131.3, 130.8, 130.3, 130.0, 128.5, 126.5,
124.3, 123.6, 122.5, 108.9, 108.7, 107.4, 101.8, 78.4, 53.7, 39.4.
HRMS-ESI (m/z) [MþH]þ calcd for C26H19Cl3N2O5 545.0432,
found 545.0441.
Compound 3 (30:1 Z:E). In a 500-mL, round-bottomed flask,
compound 4 (16.38 g, 0.030 mol, 1.0 equiv) was dissolved in
anhydrous CH2Cl2 (160 mL). At 0 °C under N2, triflic anhydride
(16.9 g, 0.060 mol, 2.0 equiv) was added dropwise. After the
mixture was stirred for 10 min, anhydrous pyridine (11.8 g, 0.15
mol, 5.0 equiv) was added dropwise. The reaction solution was
stirred at 0 °C for 1 h and then warmed to room temperature for
another 48 h. Water (250 mL) was added to quench the reaction.
The aqueous layer was extracted with CH2Cl2 (2 ꢀ 100 mL). The
organic layers were combined, washed with brine, dried over
MgSO4, filtered, and concentrated. The crude product was
purified by column chromatography (eluting with a gradient
of 10-25% EtOAc/hexanes) to yield 3 with 30:1 Z to E
selectivity (determined by 1H NMR) as a light orange foam
(14.0 g, 88% yield). 1H NMR (600 MHz, CDCl3) δ 7.50-7.44
(m, 2H), 7.39-7.34 (m, 3H), 7.33-7.28 (m, 2H), 7.05 (s, 1H),
6.72 (d, J=8.0 Hz, 1H), 6.69-6.64 (m, 2H), 6.60 (s, 1H), 5.96
(s, 2H), 3.91 (s, 3H). 13C NMR (151 MHz, CDCl3) δ 169.0,
148.3, 148.1, 147.8, 146.0, 138.6, 138.2, 134.7, 133.7, 133.1,
131.3, 130.55, 130.46, 130.0, 129.8, 128.4, 126.4, 124.5, 123.6,
Acknowledgment. We wish to thank Prof. Scott E. Denmark
for helpful discussions; Dr. Jiejun Wu and Ms. Heather
McAllister for analytical support; and Dr. Daniel J. Pippel
for proofreading the manuscript.
Supporting Information Available: General methods, syn-
thetic procedures for the Still-Gennari phosphonate and com-
pounds 5-7, and NMR spectra of compounds 1 and 3-8. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(14) Preheating the oil bath is important to avoid significant isomeriza-
tion. At lower temperatures, the hydrolysis process is very slow so that the
isomerization is favored.
(15) A small amount (∼3%) of the undesired E isomer was always
observed in the crude final material.
J. Org. Chem. Vol. 75, No. 22, 2010 7953