Constrained peptidomimetics: Building bicyclic analogs of pyrazoline derivatives

Article abstract of DOI:10.1016/j.tet.2003.09.027

A synthetic route has been developed for incorporating pyrazoline derivatives as proline surrogates in constrained X-Pro peptidomimetics. The route allows for the synthesis of dipeptide building blocks having either a six or seven-membered-ring annulated

Full text of DOI:10.1016/j.tet.2003.09.027

Tetrahedron 59 (2003) 8515–8523
Constrained peptidomimetics: building bicyclic analogs of
pyrazoline derivatives
*
Bin Liu, John D. Brandt and Kevin D. Moeller
Department of Chemistry, Washington University, Campus Box 1134, St. Louis, MO 63130, USA
Received 10 July 2003; revised 10 September 2003; accepted 10 September 2003
Abstract—A synthetic route has been developed for incorporating pyrazoline derivatives as proline surrogates in constrained X-Pro
peptidomimetics. The route allows for the synthesis of dipeptide building blocks having either a six or seven-membered-ring annulated onto
the pyrazoline moiety, as well as for the asymmetric synthesis of analogs having substituents on N-terminal side of the building block.
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
meant synthesizing both analogs and then discarding the one
with the wrong bridgehead stereochemistry.
In 1997, Carreira and co-workers first reported a new class
of proline surrogates having the overall structure of 1
(Scheme 1).1 These analogs were analogous to the
previously reported six-membered ring pyridazine-3-
carboxylic acid derivatives 2,2 and in principle could be
used in a similar fashion to synthesize a variety of intriguing
new peptidomimetics.3 For example, efforts to utilize
bicyclic peptidomimetics to ‘map’ the three dimensional
requirements for the binding of thyrotropin releasing
hormone (TRH/3) to its endocrine receptor (TRH-R1)4,5
found that the success of the bicyclic analogs was strongly
dependent upon the stereochemistry at the bridgehead
position of the bicyclic peptide derivative (Scheme 1).
Analog 4 having an R-configuration at the bridgehead
carbon was over 400 times more active than the directly
analogous analog 5 with the opposite bridgehead stereo-
chemistry.5c Hence, it was clear that the construction of
future analogs would require the bridgehead configuration
found in 4 and not 5. This was a worrisome observation
because the bridgehead carbon in 4 had an R-configuration
and therefore possessed cis-stereochemistry about the
proline ring. In general, 5-substituted proline rings having
cis-stereochemistry can be difficult to synthesize and often
require the formation and subsequent separation of nearly
1:1 mixtures of both the cis- and trans-isomers.6 For the
initial synthesis of TRH analog 4, this was not a problem
because isomer 5 was also needed for probing questions
about bridgehead stereochemistry. But now that this
question was answered, synthesizing an analog like 4
It was tempting to suggest that these synthetic problems
could be avoided by incorporating a proline surrogate like 1
into the bicyclic peptidomimetic. Energy calculations7
suggest that the barrier for inversion of the bridgehead
nitrogen in analogs like 6 would range from 3 to 13 kcal/mol
Keywords: proline surrogates; pyrazoline; bicyclic analogs.
*
Corresponding author. Tel.: þ1-3149354270; fax: þ1-3149354481;
e-mail: moeller@wuchem.wustl.edu
Scheme 1.
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2003.09.027

8516
B. Liu et al. / Tetrahedron 59 (2003) 8515–8523
2. Initial studies
Six-membered ring aza pipecolic acid surrogates like 2 have
already been incorporated into bicyclic building blocks.10 In
this case, the amino acid derivative was reduced and N6
(pipecolic acid numbering) of the ring system protected as a
benzyl group. The amino acid derivative was then
transformed into the bicyclic compound using the chemistry
outlined in Scheme 2. While this strategy was effective and
suggested a route for constructing the analogous proline
based bicyclic lactam derivative, our desire to use the
analogs to probe the binding of TRH to the TRH-R
endocrine receptor required that the lactam ring bear a
functional group on the carbon beta to the amide carbonyl.
This requirement forced us to consider alternative starting
points for the N-terminal side of the building block. To this
end, the chemistry developed by Kazmaier and co-workers
appeared ideal.11 By utilizing chiral amines to influence the
stereochemical outcome of ester enolate Claisen rearrange-
ments, Kazmaier and co-workers developed a strategy for
rapidly constructing 3-substituted-2-amino-4-pentenoic
acid derivatives 12 having either absolute stereochemistry
(Scheme 3). If the stereochemistry of the olefin in 11 was
changed, then the relative stereochemistry of the substitu-
ents in the amino acid derivatives could also be varied. In
earlier efforts, these pentenoic acid derivatives proved to be
an excellent starting point for synthesizing the N-terminal
portion of bicyclic analogs such as 6.5,6 For this reason, we
also sought to take advantage of the Kazmaier chemistry in
the current synthetic efforts.
Scheme 2.
(3.2 kcal/mol for n¼1 and 12.8 kcal/mol for n¼2).8 Since a
barrier of approximately 20–25 kcal/mol is typically
needed for nitrogen isomers to be isolated at room
temperature,9 it would appear that the bridgehead amine
in analogs like 6 readily invert. Even at physiological pH
where the amine would be largely protonated, enough of the
free amine should be present to allow for this inversion.
Hence, the use of a bicyclic peptidomimetic like 6 would
allow for inversion of the bridgehead stereochemistry, and
thereby allow all of the molecules synthesized a chance to
populate the conformation with the correct bridgehead
stereochemistry for binding the receptor. But is such a
scenario really feasible, and if so how would the introduc-
tion of this bridgehead nitrogen alter the biological activity
of a constrained, bicyclic TRH analog? In order to begin
addressing these questions, we first needed a convenient
strategy for placing proline surrogates like 1 into bicyclic
peptide analogs like 6. We report herein a synthetic
approach for accomplishing this initial objective.
Two possible strategies were readily apparent (Scheme 4).
In the first, the bicyclic derivative was envisioned as arising
from a reductive amination reaction starting from 14. This
plan was initially favored because of the relative ease with
which pyrazoline derivative 1 underwent reductive amin-
ation and the fact that this route would avoid having to
Scheme 3.
Scheme 4.

B. Liu et al. / Tetrahedron 59 (2003) 8515–8523
8517
protect and then deprotect the N5 nitrogen (proline
3. Building a functionalized TRH analog
numbering). Unfortunately, this initial strategy was
problematic. When the olefin in 12 was cleaved and the
resulting aldehyde protected as a dimethoxy acetal prior to
coupling with 1 (Scheme 3), the coupling reaction did not
proceed well. The dimethoxy acetal group was not stable to
any of the reaction conditions attempted. When a more
robust cyclic acetal was used, the coupling reaction
proceeded well. However, the subsequent reduction of the
imine in the pyrazoline ring was not successful.
With these initial studies completed, our interest in building
constrained TRH analogs focused attention on the synthesis
of analogs having either a phenyl or cyclohexyl substituent
beta to the carbonyl in the lactam ring. While an analog of
the natural TRH (p-GluHisPro-NH2) would require an
imidazole substituent on the lactam ring, the constrained
analogs studied to date have replaced this imidazole group
with either a phenyl or cyclohexyl ring.5c This was done in
order to simplify the synthesis of the analogs. The change
did significantly reduce the affinity of the analogs for the
TRH endocrine receptor, however the constrained analogs
synthesized did still bind the endocrine receptor and are
fully potent at high concentrations.5b In the time since these
initial analogs were studied, the methodology needed to
build bicyclic analogs having an imidazole ring substituent
has now been developed.12 Nevertheless, for the current
studies the phenyl and cyclohexyl substituents were still
targeted for study so that the newly synthesized TRH
analogs could be directly compared with the previously
synthesized conformational probes 4 and 5, as well as
previously synthesized seven-membered ring lactam based
probes.6
Because of these complications, the second approach
outlined in Scheme 4 quickly arose as the method of choice.
In analogy to the earlier syntheses of 10, this plan called for
first reducing the pyrazoline ring and then protecting N5
with a Cbz group. The reduced proline surrogate would then
be coupled to the pentenoic acid derivative to form 15 and
the aldehyde needed for the reductive amination reaction
released with the use of an ozonolysis reaction.
The reduced pyrazoline 17 (Scheme 5) needed for this
sequence was synthesized as outlined by Carriera.1a Using
this procedure, 16 was treated with sodium cyanoboro-
hydride, the N5 nitrogen protected, and the chiral auxiliary
used to synthesize the pyrazoline removed. With 17 in hand,
we initiated efforts to synthesize a bicyclic analog having a
1,6-diazabicyclo[4.3.0]nonane ring skeleton (Scheme 6).
This effort began with the coupling of the reduced
pyrazoline 17 to a phthalimide-protected vinylalanine
(18). The olefin in the resulting product 19 was cleaved
using an ozonolysis reaction and then the Cbz protecting
group removed and the intramolecular reductive amination
completed using hydrogen over palladium on carbon. The
ozonolysis-deprotection, reductive amination strategy
afforded the bicyclic product 20 in an 80% combined yield.
For the first analog, a phenyl substituent was selected
(Scheme 7). This effort began by synthesizing amino acid
derivative 21 from the known t-Boc protected 2-amino-3-
phenylpentenoic acid.11 The coupling of 21 to the reduced
pyrazoline 17 proceeded well. However, subsequent
cleavage of the olefin, removal of the Cbz group, and
reductive amination did not afford the desired product.
Instead, product 23 was obtained. Product 23 apparently
arose from an elimination of the pthalimide group followed
by reduction of the resulting olefin. An identical elimination
reaction was observed when a trifluoroacetate protecting
group was used in place of the pthalimide group in 21 and
22. A possible mechanism for this transformation is
suggested in Scheme 8. The key to this sequence is that
removal of the Cbz group leads to the formation of an
internal base. An intramolecular deprotonation reaction can
then trigger the b-elimination reaction that forms a styrene
derivative that is in turn reduced under the hydrogenation
conditions. Since the elimination reaction was not observed
Scheme 5.
Scheme 6.
Scheme 7.

8518
B. Liu et al. / Tetrahedron 59 (2003) 8515–8523
Scheme 8.
when the substrate without the phenyl ring (19/Scheme 6)
was utilized for the sequence, nor when the reduced
pyrazoline was replaced with proline, it would appear that
both the presence of the phenyl ring and the internal base
were required for the elimination reaction.
was isolated in a 90% yield as a single diastereomer.
Compound 29 was subjected to the ozonolysis–reduction
sequence described above in order to form an 80% isolated
yield of the constrained TRH building block. Clearly, the
use of the cyclohexyl substituent circumvented the earlier
elimination problems encountered with the phenyl-substi-
tuted substrate and allowed for construction of the desired
bicyclic building block.
At this point, it was not clear if the elimination would
remain a problem if a different substituent were employed
beta to the amide carbonyl. We suspected that the
elimination observed in Scheme 7 was aided by the
presence of the aromatic ring and the formation of the
styrene product. With this in mind, a TRH substrate having a
cyclohexyl substituent beta to the amide carbonyl was
targeted for synthesis. This effort began by constructing the
required 3-substituted cyclohexyl derivative 28 (Scheme 9)
by first coupling the known allylic alcohol 275c to a
trifluoroacetic acid protected glycine using DCC and DMAP
and then utilizing the chemistry developed by Kazmaier to
accomplish an asymmetric enolate Claisen reaction.11 The
absolute stereochemistry of the product was assumed to be
directly analogous to that obtained for the rearrangement
utilizing a phenyl substituent in place of the cyclohexyl. The
degree of enantiomeric excess was not determined at this
point, but instead the product was immediately coupled to
the reduced pyrazoline 17. The resulting amide product 29
4. Building seven-membered lactam ring analogs
Having established a rapid route for the synthesis of the
bicyclic building block, we sought to extend the work to
include the synthesis of building blocks having a 1,7-
diazabicyclo[5.3.0]decane ring skeleton (Scheme 10). The
synthesis started with the previously prepared intermediate
24. In this case, the olefin was converted into an aldehyde
using a hydroboration–oxidation sequence instead of the
previously utilized ozonolysis. The result was an aldehyde
one carbon longer that was perfectly set up for deprotection
of the Cbz group and then intramolecular reductive
amination to make the desired 32. Because of the extra
carbon in the chain, the elimination reaction that interfered
in the six-membered ring lactam case was not an issue, and
Scheme 9.

B. Liu et al. / Tetrahedron 59 (2003) 8515–8523
8519
Scheme 10.
the synthesis was perfectly compatible with the presence of
the phenyl substituent beta to the amide carbonyl.
the reaction mixture quenched by the addition of water. The
aqueous layer was extracted with dichloromethane and the
combined organic layers were washed with brine and dried
over Na2SO4. The solvent was removed in vacuo and the
residue submitted to purification by flash chromatography
through silica gel (3:1 hexane/EtOAc) to afford the
carbamate product (350 mg, 29% over two steps) as an oil.
5. Conclusions
A synthetic route has been developed for rapidly incorpor-
ating pyrazoline derivatives into substituted bicyclic lactam
peptidomimetics. The route, which is analogous to earlier
efforts to synthesize related pipecolic acid based substrates,
capitalizes on the availability of chiral 3-substituted-2-
amino-4-pentenoic acid derivatives for constructing the
N-terminal portion of the peptidomimetics. Bicyclic build-
ing blocks having either a six- or seven-membered ring
lactam constraining the N-terminal side of the building
block can by synthesized. In addition, the approach taken
allows for the asymmetric synthesis of analogs that are
substituted beta to the amide carbonyl. For the six-
membered ring lactam cases, a competitive elimination
reaction interfered with the final reductive amination
reaction in the synthesis of analogs having a phenyl
substituent beta to the lactam carbonyl. The elimination
was not a problem for either substrates having a cyclohexyl
substituent beta to the carbonyl or substrates leading to
seven-membered ring lactams.
1H NMR (300 MHz, CDCl3): d 7.42–7.29m, 5H), 5.26 (a of
ab, J¼12.6 Hz, 1H), 5.15 (b of ab, J¼12.6 Hz, 1H), 4.39 (t,
J¼7.5 Hz, 1H), 3.89 (dd, J¼7.5, 5.4 Hz, 1H), 3.74–3.58 (m,
2H), 3.54 (a of ab, J¼14.1 Hz, 1H), 3.45 (b of ab,
J¼14.1 Hz, 1H), 2.61– 2.57 (m, 1H), 2.19–2.06 (m, 3H),
1.94–1.88 (m, 3H), 1.43–1.22 (m, 2H), 1.45 (s, 3H), 0.98
(s, 3H); 13C NMR (75 MHz, CDCl3): d 170.0, 156.0, 136.5,
128.4, 128.3, 128.0, 127.9, 127.8, 67.4, 65.4, 59.9, 53.0,
48.9, 47.8, 46.8, 44.5, 38.1, 32.7, 32.5, 26.4, 20.7, 19.9; IR
(neat/NaCl): 2960, 2885, 1697, 1455, 1411, 1335,
1136 cm21; LRMS (FAB) m/e (rel. intensity 454.1
(MþLiþ, 8), 313.1 (20), 160.1 (100); HRMS (FAB)
calculated for C22H29O5N3SLi (MþLiþ) 454.1988, found
454.1980.
6.1.2. 1-Benzyl 3-methyl (3S)-pyrazolidine-1,3-di-
carboxylate (17). To a solution of the carbamates
synthesized above (100 mg, 0.22 mmol) in CH2Cl2/MeOH
(1:1, 3 mL) was added a 10% solution of Mg(OCH3)2 in
MeOH (390 mL) at 08C. The reaction mixture was stirred at
same temperature for 1 h and monitored by TLC until the
starting material disappeared. At the end of the reaction,
CH2Cl2 (5 mL) was added followed by a saturated NH4Cl
solution (5 mL). The aqueous phase was extracted with
CH2Cl2 several times and the combined organic phases
were dried over Na2SO4, filtered, and concentrated in vacuo.
The residue was purified using flash chromatography
through silica gel (hexane/EtOAc, 1:1) to afford compound
17 (44 mg, 74%) as a foamy solid.
6. Experimental
6.1. Data for compounds
6.1.1. 4-{[(3S)-1-Benzylpyrazolidin-3-yl] carbonyl}-
10,10-dimethyl-4-azatricyclo [5.2.1.01,5]-decane. To a
solution of 2D-pyrazoline 16 (830 mg, 2.67 mmol) was
added NaCNBH3 (419 mg, 6.67 mmol). The reaction
mixture was stirred at room temperature for 1 h at which
time it was diluted with EtOAc and quenched with the
addition of a saturated aqueous solution of potassium
carbonate. The organic layer was washed with brine and
dried over Na2SO4. The solvent was removed using a rotary
evaporator and the resulting product dissolved in CH2Cl2
(10 mL). To this solution was added CbzCl (460 mL,
3.2 mmol) and triethylamine (408 mL, 2.9 mmol). The
reaction was stirred at room temperature for 2 h and then
1H NMR (300 MHz, CDCl3): d 7.42–7.26 (m, 5H), 5.26 (a
of ab, J¼12.0 Hz, 1H), 5.15 (b of ab, J¼12.0 Hz, 1H), 3.90
(t, J¼7.5, 1H), 3.77–3.71 (m, 1H), 3.75 (s, 3H), 3.59–3.50
(m, 1H), 2.46–2.39 (m, 1H), 2.24–2.13 (m, 1H); 13C NMR
(75 MHz, CDCl3): d 171.7, 155.9, 136.4, 128.3, 128.0,
127.9, 67.3, 59.6, 52.4, 45.8, 31.8; IR (Neat/ NaCl): 3256,

8520
B. Liu et al. / Tetrahedron 59 (2003) 8515–8523
2959, 1737, 1696, 1437, 1400, 1348, 1214, 1118 cm21
;
7.66 (m, 2H), 4.94 (dd, J¼10.5, 7.8 Hz, 1H), 4.63 (dd,
J¼9.3, 6 Hz, 1H), 3.77 (s, 3H), 3.50–3.45 (m, 1H), 3.33–
3.28 (m, 1H), 2.94–2.80 (m, 2H), 2.78–2.69 (m, 1H), 2.55–
2.48 (m, 1H), 2.32–2.20 (m, 2H); 13C NMR (75 MHz,
CDCl3): d 170.8, 167.7, 167.5, 161.6, 134.6, 134.2, 132.3,
131.9, 123.9, 123.8, 57.3, 55.4, 52.3, 47.1, 28.6, 26.6; IR
(neat/NaCl): 3472, 2953, 2848, 1716, 1643, 1393,
1175 cm21; GC–MS (EI) m/e (rel. intensity) 343 (Mþ,
20), 229 (5), 186 (5), 143 (10), 83 (100).
LRMS (FAB) m/e (rel. intensity) 271 (MþLiþ, 100), 160
(90); HRMS (FAB) calculated for C13H16O4N2Li 271.1270,
found 272.1263.
6.1.3. Methyl 2-[2-(1,3-dihydro-2H-isoindol-2-yl) pent-4-
enoyl]-1-benzylpyrazolidine-3-carboxylate (19). To a
solution of 18 (110 mg, 0.45 mmol) in anhydrous ether
(1 mL) was added PCl5 (93.6 mg, 0.45 mmol) at 08C. The
reaction mixture was stirred at 08C for 1 h until all of the
solid dissolved. The solvent was then removed using a
rotary evaporator and the residue placed on a vacuum line
for 1 h in order to afford the acid chloride. The acid chloride
was taken up in toluene (1.5 mL) and then added to a
mixture of 17 (80 mg, 0.30 mmol) in toluene (1 mL) and
10% aqueous solution of NaHCO3 (1 mL). The resulting
reaction mixture was stirred overnight at room temperature.
When the reaction was complete, EtOAc (20 mL) was added
to the mixture and the organic phase washed with saturated
NaHCO3 and brine. The combined organic layers were then
driedoverNa2SO4, filtered, and concentratedinvacuoinorder
to provide a crude product that was purified by chromato-
graphy through silica gel (hexane/EtOAc, 1:1) to afford the
pure coupled product 19 (130 mg, 87%) as an oil.
6.1.5. 2-(1,3-Dioxo-1,3-dihydro-2H-isoindol-2-yl)-3-
phenylpent-4-enoic acid (21). The titled compound was
made from the corresponding t-Boc protected amino acid
derivative. The t-Boc protected amino acid11 (1.27 g,
4.36 mmol) was dissolved in 3 M HCl–EtOAc (10 mL)
and stirred at room temperature for 2 h. The solvents were
removed under reduced pressure and the residue was again
dissolved in a 5% Na2CO3 solution (20 mL). N-Ethoxy-
carbonylphthalimide (1.0 g, 4.6 mmol) was then added and
the resulting solution was stirred at room temperature for 2 h.
At the end of the reaction, the reaction mixture was washed
with ether. The pH of the solutionwas then adjusted to 1 bythe
addition of 6N HCl and the aqueous phase was extracted three
times with CH2Cl2. The combined organic phases were dried
over Na2SO4 and concentrated in vacuo to provide phthal-
imide 21 (1.0 g, 76%) as an oil. The purity of the product thus
obtained was confirmed by NMR analysis.
1H NMR (300 MHz, CDCl3): d 7.87–7.70 (m, 4H), 7.45–
7.26 (m, 5H), 5.70–5.59 (m, 1H), 5.28–5.18 (m, 3H), 5.02–
4.91 (m, 3H), 4.02–3.94 (m, 1H), 3.62 (s, 3H), 2.84–2.80
(m, 2H), 2.65 (q, J¼8.1 Hz), 2.31– 2.11 (m, 2H); 13C NMR
(75 MHz, CDCl3): d?170.6, 169.1, 167.2, 134.5, 134.2,
133.4, 131.3, 128.6, 128.5, 128.4, 123.7, 123.5, 118.6, 69.2,
64.9, 57.8, 52.5, 50.7, 47.1, 33.5, 29.4, 25.5; IR (Neat/
NaCl): 2953, 1772, 1716, 1387, 1200 cm21; LRMS (FAB)
m/e (rel. intensity) 498 (MþLiþ, 100), 313 (10), 160 (65);
HRMS (FAB) calculated for C26H25O7N3Li 498.1853,
found 498.1834.
1H NMR (CDCl3, 300 MHz): d 7.71–7.60 (m, 4H), 7.14–
7.12 (m, 4H), 7.10–7.01 (m, 1H), 6.20 (ddd, J¼17.4, 9.9,
8.1 Hz, 1H), 5.27–5.18 (m, 2H), 4.59 (dd, J¼10.8, 7.5 Hz);
13C (CDCl3, 75 MHz): d 173.8, 167.1, 139.1, 137.9, 134.0,
131.0, 128.4, 127.9, 127.0, 123.4, 117.3, 54.9, 48.6; IR
(neat/NaCl): 3084, 3063, 1832, 1775, 1720, 1386, 1068,
719 cm21; LRMS (FAB) m/e (rel. intensity) 322.1 (MþHþ,
20), 154 (100), 136 (90), 107 (45), 89 (90); HRMS (FAB)
calculated for C19H16O4N (MþHþ) 322.1079, found
322.1078.
6.1.4. Methyl (1S,7S)-7-(1,3-dioxo-1,3-dihydro-2H-iso-
indol-2-yl)-8-oxohexahydro-1H-pyrazolo[1,2-a]pyrid-
azine-1-carboxylate (20). Ozone was bubbled through to a
solution of 19 (120 mg, 0.25 mmol) in CH2Cl2 (5 mL) at
2788C until a purple blue color persisted. Oxygen was then
passed through the solution to remove the excess ozone.
After the color of the solution disappeared, dimethyl sulfide
(180 mL, 2.5 mmol) was added at 2788C, the reaction
mixture allowed to warm to room temperature, and the
mixture stirred overnight. The solvent was then removed
using a rotary evaporator and the residue passed through a
short plug of silica gel in order to afford the aldehyde. The
aldehyde was carried on without further purification.
6.1.6. Methyl 1-benzyl-2-[2-(1,3-di-oxo-1,3-dihydro-2H-
isoindol-2-yl)-3-phenylpent-4-enoyl] pyrazolidine-3-
carboxylate (22). To a solution of 21 (282 mg, 0.88 mmol)
in anhydrous ether (3 mL) at 08C was added PCl5 (183 mg,
0.88 mmol). The reaction mixture was stirred at 08C for 1 h
until the entire solid dissolved. The solvent was then
removed using a rotary evaporator and the residue put on a
vacuum line for another 1 h to afford the acid chloride. The
acid chloride was taken up in toluene (3 mL) and then added
to a mixture of 17 (120 mg, 0.45 mmol) in toluene (1 mL)
and 10% aqueous NaHCO3 (2 mL). The resulting reaction
mixture was stirred overnight at room temperature. At the
end of the reaction, EtOAc (30 mL) was added to reaction
mixture and the organic phase washed with saturated
NaHCO3 and brine. The organic layer was then dried over
Na2SO4, filtered, and concentrated in vacuo in order to
provide the crude product. The crude product was purified
by chromatography through silica gel (hexane/EtOAc, 1:1)
to afford the coupled product 22 (230 mg, 90%) as an oil.
To a solution of the aldehyde (70 mg, 0.14 mmol) in EtOAc
(50 mL) was added 5% Pd/C (70 mg). The reaction mixture
was stirred under H2 atmosphere until no starting material
was observed by TLC. The catalyst was removed by
filtration through a short plug of silica gel and then the silica
gel was washed with EtOAc. Removal of the solvents
yielded the crude product. The crude product was
chromatographed through silica gel (hexane/EtOAc/MeOH,
3:2:1) to afford product 20 (80% over two steps) as a foamy
solid.
1H NMR (300 MHz, CDCl3): d 7.66–7.02 (m, 9H), 5.92–
5.89 (m, 1H), 5.46 (dd, J¼10.5, 10.5 Hz, 2H), 5.20 (d,
J¼5.4 Hz, 1H), 5.10–4.95 (m, 3H), 4.60 (dd, J¼11.1,
6.6 Hz, 1H), 4.00–3.91 (m, 1H), 3.65 (s, 3H), 2.47–2.41
1H NMR (300 MHz, CDCl3): d 7.84–7.80 (m, 2H), 7.71–

B. Liu et al. / Tetrahedron 59 (2003) 8515–8523
8521
(m, 1H), 2.20–2.10 (m, 2H); 13C NMR (75 MHz, CDCl3): d
170.6, 167.9, 166.6, 157.2, 138.8, 137.9, 137.4, 135.2,
134.1, 130.8, 128.8, 128.6, 128.5, 128.4, 128.1, 126.8,
123.2, 122.8, 116.5, 69.2, 57.5, 53.0, 52.5, 47.2, 46.4, 29.8;
IR (neat/NaCl): 2953, 1719, 1384, 1199 cm21; LRMS
(FAB) m/e (rel. intensity) 574 (MþLiþ, 100), 313 (10), 160
(60); HRMS (FAB) calculated for C32H29O7N3Li 574.2166,
found 574.2142.
1182 cm21; LRMS (FAB) m/e (rel. intensity) 593.2
(2MþLiþ, 5), 300.1 (MþLiþ, 30), 231.2 (15), 178.0
(100); HRMS (FAB) calculated for C13H18F3O3NLi
(MþLiþ) 300.1399, found 300.1405.
6.1.9. (2S,3S)-3-Cyclohexyl-2-[(trifluoroacetyl)amino]-
pent-4-enoic acid (28). To a slurry solution of the ester
synthesized in the preceding experiment (2.93 g, 10 mmol),
quinidine (6.48 g, 20 mmol), and Al(OiPr)3 (2.25 g,
11 mmol) in THF (20 mL) at 2408C was added slowly
LiHMDS (55 mmol, prepared by adding 22 mL of 2.5 M
nBuLi in hexane to 13.7 mL of HMDS at 2208C in 20 mL
of THF). The resulting solution was stirred at 2408C for an
additional 1 h before it was allowed to warm to room
temperature and stirred overnight. At the end of the reaction,
1N KHSO4 was added to quench the reaction and the
aqueous phase extracted with ether four times. The com-
bined organic phase was dried over Na2SO4 and concen-
trated in vacuo in order to provide a crude product that was
purified by recrystallization with R-methylbenzylamine to
afford amino acid 28 (2.0 g, 68%) as a foamy solid.
6.1.7. Methyl 8-oxo-6-phenylhexahydro-1H-pyra-
zolo[1,2-a]pyridazine-1-carboxylate (23). Ozone was
bubbled through a solution of 22 (200 mg, 0.35 mmol) in
CH2Cl2 (15 mL) at 2788C until a purple blue color
persisted. Oxygen was then passed through the solution to
remove the excess ozone. After the color of the solution
disappeared, dimethyl sulfide (256 mL, 3.5 mmol) was
added to the 2788C reaction mixture. The reaction was
then allowed to warm to room temperature and stirred
overnight. The solvent was then removed using a rotary
evaporator and the resulting residue passed through a
short silica gel pad and washed with Et2O to afford the
aldehyde. The aldehyde was carried on without further
purification.
1H (CDCl3, 300 MHz): d 6.81 (br d, J¼7.8 Hz, 1H), 5.52
(td, J¼16.8, 10.2 Hz, 1H), 5.26 (dd, J¼10.2, 2.1 Hz, 1H),
5.11 (dd, J¼17.1, 1.8 Hz, 1H), 4.88 (dd, J¼8.7, 5.1 Hz, 1H),
2.16 (ddd, J¼10.2, 7.8, 2.1 Hz, 1H), 2.03 (br d, J¼8.6 Hz,
1H), 1.8–0.8 (m, 10H); 13C (CDCl3, 75 MHz): d 173.9,
134.8, 120.5, 53.7, 51.9, 37.2, 31.1, 30.8, 26.2, 25.9; IR
To a solution of aldehyde synthesized above in EtOAc
(60 mL) was added 5% Pd/C (100 mg). The reaction
mixture was stirred under an atmosphere of hydrogen until
TLC analysis no longer indicated the presence of starting
material. The catalyst was removed by filtration through a
short silica gel pad and was washed with EtOAc. Removal
of the solvent and purification by flash chromatography
through silica gel (Hexane/EtOAc, 1:3) yielded product 23
(53 mg, 55%) as a foamy solid.
(neat/NaCl): 3283, 2928, 2854, 1713, 1548, 1170 cm21
;
LRMS (FAB) m/e (rel. intensity) 294 (MþHþ, 50), 248
(20), 198 (100), 166 (50), 154 (30), 123 (30); HRMS (FAB)
calculated for C13H18O3F3N 294.1318, found 294.1314.
6.1.10. Methyl (3S)-2-{(2S,3S)-3-cyclohexyl-2-[(trifluoro-
acetyl) amino] pent-4-enoyl}-1-benzylpyrazolidine-3-
carboxylate (29). To a solution of 28 (282 mg, 0.88 mmol)
in anhydrous ether (3 mL) at 08C was added PCl5 (183 mg,
0.88 mmol). The reaction mixture was stirred at 08C until
the all of the solid dissolved. The solvent was then removed
using a rotary evaporator and the resulting residue placed on
a vacuum line for another 1 h to afford the acid chloride. The
acid chloride was taken up in toluene (3 mL) and then added
to a mixture of 17 (120 mg, 0.45 mmol) in toluene (1 mL)
and 10% NaHCO3 in H2O (2 mL). The resulting reaction
mixture was stirred overnight at room temperature. At the
end of the reaction, EtOAc was added to the mixture and the
organic phase washed with saturated NaHCO3 and brine.
The organic layer was then dried over Na2SO4, filtered, and
concentrated in vacuo in order to provide the crude product.
The crude product was purified by chromatography through
silica gel (hexane/EtOAc, 1:1) to afford the product 29
(230 mg, 90%) as an oil.
1H (CDCl3, 300 MHz): d 7.40–7.25 (m, 5H), 4.68 (dd,
J¼11.1, 2.2 Hz, 1H), 4.8 (s, 3H), 3.42–3.38 (m, 2H), 3.25–
3.18 (m, 1H), 3.05–2.96 (m, 2H), 2.78 (a of abx, J¼15.0,
8.1 Hz, 1H), 2.62 (b of abx, J¼15.0, 8.1 Hz, 1H), 2.54–2.42
(m, 1H), 2.18–2.12 (m, 1H); 13C (CDCl3, 75 MHz): d
171.2, 164.0, 141.7, 128.8, 127.2, 127.1, 58.6, 56.4, 54.1,
52.6, 38.9, 36.2, 28.0; IR (neat/NaCl): 2952, 1746, 1647,
1197 cm21; GC–MS (EI): m/e (rel. intensity) 274 (Mþ, 80),
215 (20), 187 (25), 143 (100), 83 (70).
6.1.8. (2E)-3-Cyclohexylprop-2-enyl [(trifluoroacetyl)
amino] acetate. To a solution of allyl alcohol 27 (4.0 g,
28.6 mmol) and TFA protected glycine (4.88 g, 28.6 mmol)
in CH2Cl2 (120 mL) was added DCC (6.48 g, 31.5 mmol)
and DMAP (1.74 g, 14.3 mmol). The reaction mixture
was stirred at room temperature for 20 h. A white solid was
formed that was then removed by filtration. The filtrate was
washed with 1 M KHSO4, 5% NaHCO3, and brine. The
organic phase was then dried over Na2SO4, and concen-
trated. The residue was purified by chromatography through
silica gel (hexane/EtOAc, 3:1) to afford the ester product
(8.0 g, 95%) as a pale yellow oil.
1H (CDCl3, 300 MHz): d 6.91 (br d, J¼9.6 Hz, 1H), 5.62
(td, J¼17.1, 6.9 Hz, 1H), 5.42 (d, J¼12.0 Hz, 1H), 5.28 (t,
J¼10.2, 1H), 5.08–4.94 (m, 3H), 4.64 (dd, J¼6.3, 6.0 Hz,
1H), 4.36–4.28 (m, 1H), 3.47 (s, 3H), 3.20 (q, J¼10.2 Hz,
1H), 2.38–1.98 (m, 3H), 1.82–1.58 (m, 5H), 1.42–0.92 (m,
6H); 13C (CDCl3, 75 MHz): d 171.6, 169.9, 157.9, 134.8,
134.0, 128.3, 128.1, 128.0, 119.4, 69.4, 57.3, 54.9, 52.5,
50.7, 47.4, 37.4, 32.0, 29.9, 27.6, 26.6, 26.3; IR (neat/NaCl):
2924, 2851, 1734, 1661, 1208, 1174 cm21; LRMS (FAB)
m/e (rel. intensity) 546.2 (MþLiþ, 65), 412 (10), 313 (30),
1H (CDCl3, 300 MHz): d 7.00 (br s, 1H), 5.75 (dd, J¼15.7,
6.3 Hz, 1H), 5.24 (td, J¼15.7, 5.4 Hz, 1H), 4.62 (d,
J¼6.6 Hz, 1H), 4.12 (d, J¼5.4 Hz, 1H), 2.02–1.92 (m,
1H), 1.88–1.62 (m, 5H), 1.35–1.01 (m, 5H); 13C (CDCl3,
75 MHz) 168.0, 143.7, 120.0, 67.6, 41.4, 40.3, 42.4, 26.0,
25.8; IR (neat/NaCl): 3337, 2927, 2853, 1716, 1555, 1450,

8522
B. Liu et al. / Tetrahedron 59 (2003) 8515–8523
160 (100); HRMS (FAB) calculated for C26H32O6F3N3Li
(MþLiþ) 546.2403, found 546.2376.
J¼5.7 Hz, 2H), 2.44–2.38 (m, 1H), 2.22–2.08 (m, 2H). The
compound was carried onto the next step without further
characterization.
6.1.11. Methyl (1S,6R,7S)-6-cyclohexyl-8-oxo-7-[(tri-
fluoroacetyl) amino] hexahydro-1H-pyrazolo [1,2-a]
pyridazine-1-carboxylate (30). Ozone was bubbled
through a solution of 29 (200 mg, 0.37 mmol) in CH2Cl2
(15 mL) at 2788C until a purple blue color persisted.
Oxygen was then passed through the solution to remove the
excess ozone. After the color of the solution disappeared,
dimethyl sulfide (270 mL, 3.7 mmol) was added to the
reaction mixture. The resulting solution was then allowed to
warm up to room temperature and stirred overnight. The
solvent was removed using a rotary evaporator and the
resulting residue passed through a short plug of silica gel to
afford the aldehyde. The aldehyde was carried on to the next
step without further purification.
To the above alcohol in CH2Cl2 (3 mL) was added Dess–
Martin reagent (CH2Cl2 solution, 15% by weight) and the
resulting solution stirred at room temperature for 4 h. 10%
NaHCO3 and saturated Na2S2O3 were then added until both
the aqueous and organic phases became clear. The organic
phase was dried over Na2SO4 and concentrated in vacuo.
The crude product was purified by flash chromatography
through silica gel (Hexane/Et2O, 3:1) to afford the aldehyde
31 (115 mg, 51 % over two steps) as an oil.
1H NMR (CDCl3, 300 MHz): d 9.47 (s, 1H), 5.47 (d,
t¼12.0 Hz, 1H), 5.28–5.5.14 (m, 2H), 4.92 (dd, J¼8.4,
5.7 Hz, 1H), 4.42 (dt, J¼10.5, 3.3 Hz, 1H), 4.00–3.92 (m,
1H), 3.70 (s, 3H), 2.87–2.80 (m, 1H), 2.42–2.10 (m, 4H);
13C (CDCl3, 75 MHz): d 200.0, 170.5, 167.5, 166.5, 156.4,
138.2, 135.2, 134.1, 128.9, 128.8, 128.7, 128.3, 128.2,
127.2, 123.3, 69.4, 57.6, 53.7, 52.6, 47.2, 46.6, 38.6, 29.7;
To a solution of the aldehyde synthesized above in EtOAc
(70 mL) was added 5% Pd/C (80 mg). The reaction mixture
was stirred under an atmosphere of hydrogen until no
starting material remained evident by TLC. The catalyst was
removed by filtration through a short plug of silica gel and
the silica gel then washed with EtOAc. The resulting
solution was concentrated in vacuo and the resulting residue
purified by chromatography through silica gel (hexane/
EtOAc, 2:3) to afford 30 (80%) as an oil.
IR (neat/NaCl): 2947, 2913, 2846, 1720, 1381, 1200 cm21
;
LRMS (FAB) m/e (rel. intensity) 590 (MþLiþ, 20), 269
(20), 160 (100); HRMS (FAB) calculated for C32H29O8N3Li
(MþLiþ) 590.2115, found 590.2124.
6.1.13. Methyl (1S,8S)-8-(1,3-dioxo-1,3-dihydro-2H-iso-
indol-2-yl)-9-oxo-7-phenylhexahydro-1H,5H-pyrazolo
[1,2-a][1,2]diazepine-1-carboxylate (32). To a solution of
aldehyde 31 (60 mg, 0.10 mmol) in EtOAc (70 mL) was
added 5% Pd/C (60 mg). The reaction mixture was stirred
under an atmosphere of hydrogen until TLC analysis
indicated that no starting material remained. The cata-
lyst was removed by filtration through a short plug of
silica gel and the silica gel washed with EtOAc. Removal
of the solvents and purification by chromatography
through silica gel (hexane/EtOAc, 2:3) afforded 32
(35 mg, 80%).
1H (CDCl3, 300 MHz):d 7.45 (br s, 1H), 4.66 (dd, J¼9,
1.8 Hz, 1H), 4.56 (dd, J¼8.7, 5.7 Hz, 1H), 3.78 (s, 3H), 3.42
(dt, J¼6.9, 2.4 Hz, 1H), 3.25 (t, J¼10.2 Hz, 1H), 3.11 (dq,
J¼9.9, 3.3 Hz, 1H), 2.94–2.83 (m, 2H), 2.56–2.42 (m, 1H),
2.37–2.29 (m, 1H), 1.73–0.92 (m, 11H); 13C (CDCl3,
75 MHz): d 170.4, 164.2, 56.5, 54.8, 53.5, 52.8, 51.3, 40.6,
37.1, 30.7, 29.8, 26.6, 26.2, 26.0; IR (neat/NaCl): 3248,
2924, 2851, 1745, 1720, 1653, 1448, 1211, 1180,
1155 cm21; GC–MS (EI) m/e (rel. intensity): 391 (Mþ,
30), 332 (Mþ2COOCH3, 10), 308 (Mþ2C6H11, 20), 248
(25), 143 (25), 83 (100); HRMS (FAB) calculated for
C17H24O4N3F3Li (MþLiþ) 398.1879, found 398.1874.
1H (CDCl3, 300 MHz): d 7.67–7.63 (m, 2H), 7.58–7.52 (m,
2H), 7.27–7.01 (m, 5H), 5.86 (d, J¼7.5 Hz, 1H), 4.64 (t,
J¼7.8 Hz, 1H), 3.99 (dt, J¼9.9 Hz, 1H), 3.80 (s, 3H), 3.50–
3.41 (m, 1H), 3.37–3.33 (m, 1H), 3.18–3.11 (m, 1H), 2.94–
2.87 (m, 1H), 2.50–2.15 (m, 4H); 13C (CDCl3, 75 MHz): d
171.8, 167.8, 166.2, 142.8, 133.5, 128.3, 127.6, 126.5,
123.0, 58.5, 56.5, 55.3, 54.9, 52.7, 42.8, 38.0, 29.8; IR
(neat/NaCl): 3030, 2952, 1745, 1720, 1667, 1437, 1382,
1200 cm21; GC–MS (EI) m/e (rel. intensity): 433 (Mþ, 25),
405 (Mþ2CO, 20), 372 (20), 346 (50), 317 (30), 284
(Mþ2PhthvN, 100), 207 (45), 185 (50), 143 (90), 104
(80), 83 (100); HRMS (FAB) calculated for C24H23O5N3Li
440.1798, found 440.1789.
6.1.12. Methyl (3S)-2-[(2S)-2-(1,3-dioxo-1,3-dihydro-2H-
isoindol-2-yl)-3-phenyl-5-oxopentanoyl]-1-benzylpyra-
zolidine-3-carboxylate (31). To a solution of BH3.MeS
(100 mL, 10 M, 1 mmol) in THF (1 mL) at 08C was slowly
added 2-methyl-2-butene (212 mL, 2 mmol). The resulting
solution was stirred at 08C for 1 h after which the
vinylphenylalanine derivative 24 (227 mg, 0.4 mmol) in
THF (1 mL) was added via a cannula. The resulting solution
was allowed to warm to room temperature and stirred for
4 h. The reaction was diluted with THF (5 mL) and satur-
ated NaOAc (2 mL) and then H2O2 (2 mL) was added. The
resulting mixture was stirred at room temperature for 8 h.
Ether was then added to dilute the reaction, the layers
separated, and the aqueous phase exacted three times with
ether. The combined organic phases were washed with brine,
dried over Na2SO4, and concentrated in vacuo. The resulting
residue was purified using flash chromatography through
silica gel (hexane/ether, 1:2) in order to afford the alcohol.
Acknowledgements
We thank the National Science Foundation (CHE-9023698)
for their generous support of our programs. We also
gratefully acknowledge the Washington University High
Resolution NMR facility, partially supported by NIH grants
RR02004, RR05018, and RR07155, and the Washington
University Mass Spectrometry Resource Center, partially
supported by NIHRR00954, for their assistance.
1H NMR (CDCl3, 300 MHz): d 7.63–7.01 (m, 9 aromatic
H0s), 5.48 (d, J¼11.7 Hz, 1H), 5.30–5.22 (m, 2H), 4.94 (dd,
J¼8.5, 5.7 Hz, 1H), 4.02–3.95 (m, 2H), 3.71 (s, 3H), 3.37 (t,

B. Liu et al. / Tetrahedron 59 (2003) 8515–8523
8523
Moeller, K. D. J. Med. Chem. 1996, 39, 1571–1574. (b)
Laakkonen, L.; Li, W.; Perlman, J. H.; Guarnieri, F.; Osman,
R.; Moeller, K. D.; Gershengorn, M. C. Mol. Pharmacol.
1996, 49, 1092–1096. (c) Li, W.; Moeller, K. D. J. Am. Chem.
Soc. 1996, 118, 10106. For fully constrained analogs see: (d)
‘Thyrotropin Releasing Hormone Analogs: A Building Block
Approach to the Construction of Tetracyclic Peptidomimetics’
Chu, W.; Perlman, J. H.; Gershengorn, M. C.; Moeller, K. D.
Bioorg. Med. Chem. Lett. 1998, 8, 3093.
References
1. (a) Mish, M.; Guerra-Martinez, F.; Carreira, E. M. J. Am.
Chem. Soc. 1997, 119, 8379. (b) Guerra, F. M.; Mish, M. R.;
Carreira, E. M. Org. Lett. 2000, 2, 4265.
2. For the initial isolation of the acid see: (a) Morimoto, K.;
Shimada, N.; Naganawa, H.; Takita, T.; Umezawa, H.
J. Antibiot. 1981, 34, 1615. For the synthesis of 2 and
representative pyridazine acids see: (b) Attwood, M. R.;
6. Beal, L. M.; Liu, B.; Chu, W.; Moeller, K. D. Tetrahedron
2000, 56, 10113.
¨
Hassall, C. H.; Krohn, A.; Lawton, G.; Redshaw, S. J. Chem.
Soc., Perkin Trans. 1 1986, 1011. (c) Hughes, P.; Clardy, J.
J. Org. Chem. 1989, 54, 3260. (d) Schmidt, U.; Riedl, B.
Synthesis 1993, 8, 809. (e) Nakamura, Y.; Ito, A.; Shin, C.
Bull. Chem. Soc. Jpn 1994, 67, 2151. (f) Ciufolini, M. A.;
Shimizu, T.; Swaminathan, S.; Xi, N. Tetrahedron Lett. 1997,
38, 4947. (g) Xi, N.; Alemany, L. B.; Ciufolini, M. A. J. Am.
Chem. Soc. 1998, 120, 80. (h) Ciufolini, M. A.; Xi, N. Chem.
Soc. Rev. 1998, 27, 437. (i) Aspinall, I. H.; Cowley, P. M.;
Mitchell, G.; Stoodley, R. J. J. Chem. Soc., Chem. Commun.
1993, 15, 1179. (j) Aspinall, I. H.; Cowley, P. M.; Mitchell, G.;
Raynor, C. M.; Stooley, R. J. J. Chem. Soc., Perkin Trans. 1
1999, 2591.
7. To find the lowest energy ground state conformation of the
molecule, a conformation distribution search was run at the
molecular mechanics level. The most favorable structures
were then submitted to an equilibrium geometry calculation at
the AM1 semi-empirical level. To find the energy level for the
transition state, a transition geometry calculation was run at
the AM1 level. This resulted in a structure with one imaginary
vibration above 100 Hz. This vibration was animated in order
to show the nitrogen vibrating up and down in a manner
consistent with the motion leading to inversion. At this point, a
single point energy calculation at the 3-21G* ab initio level
was run for both the lowest energy conformation and the
transition state structure. The energy difference for the two
conformations was taken as the barrier for the inversion.
8. For TRH analogs constrained with a seven-membered ring
lactam see Ref. 6.
3. For a review concerning the use of lactam based peptidomi-
metics see: (a) Hanessian, S.; McNaughton-Smith, G.;
Lombart, H.-G.; Lubell, W. D. Tetrahedron 1997, 53,
12789. For more recent references see: (b) Polyak, F.; Lubell,
W. D. J. Org. Chem. 1998, 63, 5937. (c) Curran, T. P.;
Marcaurell, L. A.; O’Sullivan, K. M. Org. Lett. 1999, 1, 1225.
(d) Gosselin, F.; Lubell, W. D. J. Org. Chem. 2000, 65, 2163.
(e) Polyak, F.; Lubell, W. D. J. Org. Chem. 2001, 66, 1171.
(f) Feng, Z.; Lubell, W. D. J. Org. Chem. 2001, 66, 1181.
4. For modeling studies leading to predicted ‘active confor-
mation’ for TRH binding to its endocrine receptor TRH-R see:
(a) Marshall, G. R.; Gorin, F. A.; Moore, M. L. Annual Reports
in Medicinal Chemistry; Clarke, F. H., Ed.; Academic: New
York, 1978; Vol. 13, p 227 and references therein. (b) Font, J.,
‘Computer-Assisted Drug Design.’, Ph.D. Thesis, Washington
University in St. Louis, 1986.
9. Hehre, W. J.; Shusterman, A. J.; Huang, W. W. A Laboratory
Book of Computational Organic Chemistry; Wavefunction,
1985; pp 85.
10. For the synthesis of analogs having six and seven-membered
ring lactams fused to the pyridazine acid ring see Ref. 2b. For a
review concerning the pharmacology of analogs having a
seven-membered ring lactam fused to the pyridazine acid
(such a the angiotensin converting enzyme inhibitor cilazapril)
see: Belz, G. G.; Breithaupt, K.; Erb, K. J. Cardiovasc.
Pharmacol. 1994, 24(suppl. 2), 14.
11. Kazmaier, U.; Krebs, A. Angew. Chem. Int. Ed. Eng. 1995, 34,
2012.
12. Chu, W.; Moeller, K. D. Tetrahedron Lett. 1999, 40, 7939.
5. For partially constrained analogs of TRH see: (a) Rutledge,
L. D.; Perlman, J. H.; Gershengorn, M. C.; Marshall, G. R.;