Building constrained peptidomimetics: A convenient approach to 3-phenyl-5-vinyl substituted proline derivatives

Article abstract of DOI:10.1016/S0040-4020(01)00532-4

An approach to the synthesis of 3-phenyl-5-vinyl substituted prolines starting from readily available 2,3-disubstituted pentenoic acid derivatives is reported. The route converts the pentenoic acid derived starting materials into cyclic N-acyliminium ion precursors with the use of a hydroboration-Swern oxidation sequence. This overall transformation worked best when the initial alkylborane product was treated sequentially with excess MCPBA and then acetone prior to the Swern oxidation. In this way, a 70% yield of the desired N-acyliminium ion precursor could be obtained. Once in hand, the cyclic N-acyliminium ion precursor was used to make the desired 3-phenyl-5-vinyl substituted prolines.

Full text of DOI:10.1016/S0040-4020(01)00532-4

TETRAHEDRON
Pergamon
Tetrahedron 57 52001) 6407±6415
Building constrained peptidomimetics: a convenient approach to
3-phenyl-5-vinyl substituted proline derivatives
Shengquan Duan and Kevin D. Moellerp
Department of Chemistry, Washington University, St Louis, MO 63130, USA
Received 11 January 2001; revised 1 March 2001; accepted 12 March 2001
AbstractÐAn approach to the synthesis of3-phenyl-5-vinyl substituted prolines starting from readily available 2,3-disubstituted pentenoic
acid derivatives is reported. The route converts the pentenoic acid derived starting materials into cyclic N-acyliminium ion precursors with
the use ofa hydroboration±Swern oxidation sequence. This overall transformation worked best when the initial alkylborane product was
treated sequentially with excess MCPBA and then acetone prior to the Swern oxidation. In this way, a 70% yield ofthe desired
N-acyliminium ion precursor could be obtained. Once in hand, the cyclic N-acyliminium ion precursor was used to make the desired
3-phenyl-5-vinyl substituted prolines. q 2001 Elsevier Science Ltd. All rights reserved.
As part ofan efort to provide general routes to constrained
peptidomimetics, we have reported the synthesis ofa variety
ofbicyclic peptide building blocks like I and II 5Scheme 1).1
In these examples, the N-terminal side ofthe molecule was
constructed by annulating a ring onto a proline derivative
52a, XˆH). This was accomplished by ®rst functionalizing
the proline derivative using an anodic methoxylation reac-
tion and then converting the resulting methoxylated product
52b, XˆOMe) into a 5-vinyl proline derivative 1. The vinyl
proline derivatives were used to build bicyclic structures
like I via an ole®n metathesis based strategy2 and bicyclic
structures like II via an intramolecular reductive amination
route.3
While both routes have proven to be very effective, little has
been done to apply the chemistry to the synthesis ofbuilding
blocks containing non-proline 5R±H) C-terminal groups.
Efforts in this area have been hindered by the lack of a
convenient method for synthesizing the required 3-substi-
tuted-5-vinyl proline starting materials 51). We found this
problem particularly perplexing in our efforts to build
analogs for the Phe7±Phe8 region of substance P.3 In this
work, a series ofbicyclic analogs oftype II was required
where both R and R1 were phenyl rings. In addition, it was
essential that the stereochemistry ofboth the proline
a-carbon and the phenyl substituent at C3 ofthe proline
ring be varied relative to the bridgehead carbon ofthe
bicyclic peptidomimetic. While several intriguing routes
to 3,5-disubstituted prolines had been reported 5Scheme
2),4 these approaches were not directly applicable to the
3-phenyl-5-vinyl substituted substrates 51, RˆPh) that
were needed. For example, the strategy outlined in Eq. 51)
5Scheme 2) provided a nice route to 3-carbomethoxy-5-
alkoxy prolines.4a However, the key reaction in this
sequence led to a 3:1 ratio ofstereoisomers and by necessity
afforded a product having a methyl ester at C3 ofthe proline
ring. Hence, the conversion ofa product obtained from this
route into a building block like 1 having a phenyl ring at C3
ofthe proline ring required a series ofadditional steps. The
need for these additional steps combined with the formation
ofstereoisomers in the key step rendered such an approach
cumbersome for making the quantities of material needed
for using the ®nal proline derivatives as starting materials
for the subsequent synthetic efforts. Similar problems
limited the utility ofthe approaches to 3,5-disubstituted
prolines outlined in Eqs. 52)4b and 53) 5Scheme 2).4c While
in both cases the chemistry developed leads to a 3,5-di-
substituted proline derivative, the length ofthe route
required to make substrates like 1 and the lack of
Scheme 1.
Keywords: peptidomimetics; vinyl substituted prolines; N-acyliminium
ions.
p
Corresponding author. Tel.: 11-314-935-4270; fax: 11-314-935-4481;
e-mail: moeller@wuchem.wustl.edu
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020501)00532-4

6408
S. Duan, K. D. Moeller / Tetrahedron 57 )2001) 6407±6415
Scheme 2.
stereochemical versatility in the syntheses again limited
the utility ofboth routes ofr building the starting
materials needed for constructing the desired bicyclic
peptidomimetics.
5Scheme 3).6 This route was enticing because both
disubstituted pentenoic acid derivatives like 3 and their
enantiomers can be readily synthesized in high optical
purity.7 In principle, conversion ofthese starting materials
into cyclic N-acyliminium ion precursors 54) would provide
rapid access to a number of3,5-disubstituted proline
derivatives. For example, compound 5 would be made
from 4 with the use ofa cuprate addition reaction,
compound 6 would be made from 5 by epimerization of
the carbon bearing the methyl ester, and compound 7
would be made from 4 by accomplishing the epimerization
reaction prior to the cuprate addition reaction. Ifthe
enantiomer of 3 were used as the starting material,7 then
the opposite enantiomer ofeach ofthe 3,5-disubstituted
proline derivatives would be available.
More general approaches to 3-substituted prolines like 2a
were available,5 and as in earlier studies the C5 position of
these products could be functionalized using an anodic
oxidation reaction.2,3 However, since the products from
these efforts were to be used as starting materials, we
hoped to develop a more ef®cient strategy for constructing
building blocks like 1 that would directly generate 3-substi-
tuted proline rings with functionality at C5 already in place.
For this reason, we began to investigate what appeared to be
a very straightforward route to the required building blocks
With this in mind, the syntheses ofthe 3-phenyl-5-vinyl
substituted proline derivatives required for the substance P
efforts were undertaken.2c However, while the route
designed appeared straightforward and the starting methyl
ester 53) was readily assembled using the known
asymmetric Claisen rearrangement,7 the desired conversion
of 3 into the cyclic N-acyliminium ion precursor proved
problematic. Efforts to generate an alcohol intermediate
58) from the hydroboration step were complicated by a
competitive lactonization reaction 59) involving the methyl
ester 5Scheme 4). While this competitive cyclization could
be minimized during the reaction by oxidizing the inter-
mediate alkylborane with MCPBA, only very low yields
ofthe alcohol could be obtained ofllowing puri®cation.
Efforts to avoid puri®cation of the alcohol also met with
Scheme 3.

S. Duan, K. D. Moeller / Tetrahedron 57 )2001) 6407±6415
6409
Scheme 4.
MCPBA from the crude reaction product while avoiding the
need to purify the alcohol intermediate. To this end, we
found that adding acetone to the hydroboration±oxidation
sequence prior to workup was very helpful. The resulting
Baeyer±Villiger reaction generated methylacetate and
m-chlorobenzoic acid. The acid was separated from the
alcohol with the use ofa saturated sodium carbonate
wash. Following concentration, the crude alcohol left
behind could be used as the substrate for the Swern oxida-
tion without further puri®cation. In this fashion, starting
material 3 was converted into 4 in a 70% isolated yield
5Scheme 5).
Scheme 5.
limited success. Oxidation methods using PCC and TPAP to
directly convert alkylboranes into aldehydes led to the
generation of 4 and the formation of the imide derivative
510) following overoxidation. The overoxidation could be
avoided ifthe hydroboration using MCPBA in the workup
was employed and then the crude alcohol oxidized using
Swern conditions. Yet while successful to a certain degree,
this route required removing any ofexcess MCPBA rfom
the alcohol prior to the oxidation. The net result was a
delicate balance between having enough MCPBA present
to ef®ciently oxidize the alkylborane 5a reaction that
bene®ts from excess MCPBA) and limiting the excess
MCPBA in order to optimize the Swern oxidation. This
led to a dif®cult reaction that typically afforded only
30±40% yields ofthe desired 4.
Once compound 4 was in hand, attention was turned to
making building blocks 5±7. In an initial step, the alcohol
was exchanged for a methyl ether 512, Scheme 6). The
methyl ether was then treated with the cuprate reagent
derived from trans-1-lithio-1-propene and copper bromide
dimethylsul®de complex in the presence ofBF 3´Et2O. Like
all previous cuprate reactions ofthis type, 8 the vinyl group
was added to the face ofthe proline ring opposite that ofthe
methyl ester. None ofthe isomer having the vinyl group and
the methyl ester groups cis to each other was observed. This
was also true for the addition of 2-methylpropenyllithium
5leading to 5b) which could be used in place ofthe more
dif®cult to obtain trans-1-lithio-propene for building blocks
For these reasons, we sought a method for removing excess
Scheme 6.

6410
S. Duan, K. D. Moeller / Tetrahedron 57 )2001) 6407±6415
This work established both that the cis-stereochemistry
between the methyl ester and phenyl substituents was
preserved and that the stereochemistry at C5 was not altered
during the ozonolysis and reductive amination steps.
Assignment ofthe stereochemistry ofthe proline relative
to the known center in phenylalanine con®rmed that the
absolute stereochemistry of 5a and 5b had been assigned
correctly.
Figure 1.
Building block 5a was converted into stereoisomer 6 by
treating it with LDA at 2788C for 5 h 5Scheme 6). The
reaction was quenched with methanol to afford an 80%
yield ofthe epimerized product. The 5 h reaction time of r
the epimerization was important. Quenching the reaction
after 2 h led to a 1:1 mixture of 5a and 6, while longer
reaction times led to decreased yields ofthe product. The
progress ofthe epimerization could be monitored by the
proton chemical shift observed for the methyl ester. In 5a,
the methyl group was located in the shielding cone ofthe
aromatic ring and gave rise ofa signal with a chemical shift
ofca. 3.2 ppm in the proton NMR spectrum. The signal for
the methyl ester in 6 was found at a chemical shift of ca.
3.7 ppm. The relative stereochemistry ofthis isomer was
also con®rmed with the use ofa NOESY experiment. The
key interactions used to make the assignment are high-
lighted in Fig. 3.
where the steric bulk ofthe ole®n did not matter 5the less
substituted ole®ns are normally used in building blocks
made for ole®n metathesis based strategies). The relative
stereochemistry ofbuilding blocks was established with
the use ofa NOESY experiment. The key interactions
used to make this assignment are illustrated for 5a in Fig. 1.
Since the building blocks were to be used in conformational
probes, it was important to con®rm that the absolute stereo-
chemistry of 5a and 5b were assigned correctly and to
establish that the stereochemistry ofthe building block
would remain intact when it was incorporated into the
desired bicyclic peptidomimetic. To this end, 5b was
converted into the bicyclic lactam 14 5Scheme 7).
At no time during this synthesis was a second stereoisomer
observed, a fact that was constistent with the literature
observation7 that the preparation ofthe carboxylic acid
leading to 3 afforded an enantiomeric excess greater than
99%. As with 5a, the relative stereochemistry of 14 was
established with the use ofa NOESY experiment. The key
interactions used in this assignment are illustrated in Fig. 2.
A reversal ofthe cuprate addition±epimerization reaction
sequence led to stereoisomer 7 5Scheme 8). In this case, the
methoxylated compound 12 was treated with LDA and the
resulting enolate quenched with methanol after a period of
5 h. A 75% isolated yield ofthe epimerized product was
obtained. This material was then treated with the cuprate
reagent derived from trans-1-lithio-1-propene and copper
bromide dimethylsul®de complex in the presence of
BF3´Et2O. As in the earlier case, the vinyl group added to
the face of the incipient N-acyliminium ion that was trans to
the methyl ester. The relative stereochemistry ofthe product
was again established using a NOESY experiment 5Fig. 4).
The enantiomers ofbuilding blocks 5±7 5Fig. 5) were made
by starting with the enantiomer of 3.6 The chemistry used in
Figure 3.
Scheme 7.
Figure 2.
Scheme 8.

S. Duan, K. D. Moeller / Tetrahedron 57 )2001) 6407±6415
6411
1. Experimental9
1.1. Data for compounds
1.1.1. Methyl +2S,3S)-2-[N-+tert-butyloxycarbonyl)
amino]-3-phenyl-4-pentenate +3). 52S,3S)-2-[N-5tert-
butyloxycarbonyl) amino]-3-phenyl-4-pentenoic acid7
52.72 g, 9.3 mmol) was dissolved in 40 mL ofCH OH and
Figure 4.
3
4 mL ofwater was added. The solution was titrated to
pHˆ7.0 5pH paper) with a 20% aqueous solution of
Cs2CO3 5,8 mL). The mixture was evaporated to dryness
and the residue was reevaporated twice from 15 mL of tolu-
ene. The yellowish solid cesium salt obtained was stirred
with 1.16 mL 518.6 mmol) ofiodomethane in 15 mL of
DMF for 1 day. The mixture was partitioned between
EtOAc 5150 mL) and water 5150 mL). The organic phase
was separated and washed with saturated NaCl solution
twice, dried over MgSO4, and evaporated to dryness. The
crude product was chromatographed through 200 g ofsilica
gel that was slurry packed with a 4:1 hexane in ether solu-
tion. Elution with the same solvent afforded 2.7 g 592%) of
the desired product 3 as a white crystal: TLC 5Et2O/hexane
1
51:4)) Rfˆ0.2; H NMR 5CDCl3/300 MHz) d 1.37 5s, 9H),
Figure 5.
3.66 5s, 3H), 3.76 5t, 1H, Jˆ7.8 Hz), 4.67 5t, 1H, Jˆ7.8 Hz),
4.83 5d, 1H, Jˆ8.7 Hz), 5.17 5d, 1H, Jˆ11.4 Hz), 5.18 5d,
1H, Jˆ15.6 Hz), 6.03±6.13 5m, 1H), 7.17±7.35 5m, 5H);
13C NMR 5CDCl3/75 MHz) d 171.8, 155.0, 138.7, 136.4,
128.5, 127.9, 127.1, 117.3, 79.9, 57.3, 54.2, 53.4, 52.2,
51.8, 51.4, 27.9; IR 5neat/NaCl) 3028, 2976, 1744, 1715,
1496, 1367, 1165, 700 cm21; LRMS 5FAB) 306 5MH1, 15),
250 545), 206 550), 185 537), 137 537), 117 522), 93 5100);
HRMS C17H24NO4 5MH1) calcd 306.1705, found 306.1707.
this effort was identical to that described above, and the
absolute stereochemistry for the series con®rmed by
converting the enantiomer of 5b 55b-ent) into a bicyclic
lactam derivative 15 5Scheme 9). As with the earlier dia-
stereomer 14, the stereochemistry of 15 was assigned using
a 2D-NOESY experiment 5Fig. 6).
In conclusion, we have found that a series of 5-vinyl substi-
tuted proline derivatives needed for generating bicyclic
lactam peptidomimetics can be rapidly synthesized from
known chiral pentenoic acid derivatives. The key to this
transformation was a hydroboration±oxidation sequence
that bene®ted from the use of MCPBA followed by acetone
in the workup ofthe hydroboration step.
1.1.2. +3S)-N-tert-Butyloxycarbonyl-5-hydroxy-3-phenyl-
l-proline methyl ester +4). To a solution of 3 5156 mg,
0.5 mmol) in 2 mL ofTHF at 0 8C was added 1 mL of
1.0 M BH3´THF solution 51.0 mmol). The mixture was stir-
red and allowed to slowly warm to room temperature. After
1 h, the reaction was then cooled to 08C and 1.3 g
54.0 mmol) ofMCPBA in 8 mL ofTHF added. The mixture
was allowed to warm slowly to room temperature and stir-
red overnight. Then 1 mL ofacetone was added and the
reaction was stirred for 10 min. The reaction was then
diluted with ether, transferred to a separatory funnel,
washed three times with saturated Na2CO3 solution, and
then washed with saturated NaCl solution. The organic
layer was dried over MgSO4 and concentrated in vacuo.
This residue was redissolved in 2 mL ofCH 2Cl2 and cooled
to 2788C. To this solution was added 80 mL ofDMSO
51.1 mmol), then followed with 60 mL ofrfeshly distilled
5COCl)2 50.55 mmol). After 15 min, 0.35 mL of triethyl-
amine 52.50 mmol) was added and the reaction was allowed
to warm to room temperature. Water 55 mL) was added and
the reaction was then transferred to a separatory funnel. The
organic phase was separated and the inorganic phase was
extracted with additional CH2Cl2 525 mL£3). The combined
organic phase was washed successively with brine, 1% HCl,
H2O and 5% Na2CO3, dried over MgSO4 and concentrated
in vacuo. The crude product was chromatographed through
20 g ofsilica gel that was slurry packed with a 1:1 hexane in
ether solution. Elution with the same solvent afforded
110 mg 570%) ofa 7:4 mixture ofisomers 5 4) as a yellowish
oil: TLC 5Et2O/hexane 51:1)) Rfˆ0.21, 0.14; 1H NMR
Scheme 9.
Figure 6.

6412
S. Duan, K. D. Moeller / Tetrahedron 57 )2001) 6407±6415
5CDCl3/300 MHz) d 1.38 5s, 3.6H), 1.39 5s, 3.6H), 1.47 5s,
0.9H), 1.58 5s, 0.9H), 2.10±2.20 5m, 0.5H), 2.40±2.74 5m,
1.5H), 3.20 5s, 0.3H), 3.22 5s, 1.2H), 3.25 5s, 0.3H), 3.27 5s,
1.2H), 3.54±3.80 5m, 0.7H), 4.01±4.19 5m, 0.3H), 4.48±
4.68 5m, 1H), 5.57±5.83 5m, 1H), 7.18±7.34 5m, 5H); 13C
NMR 5CDCl3/75 MHz) d 171.1, 154.0, 136.2, 136.1, 129.1,
129.0, 128.9, 128.4, 128.3, 128.1, 127.9, 127.8, 127.7,
127.6, 127.5, 82.4, 82.2, 81.7, 81.5, 81.2, 81.1, 64.6, 64.4,
51.5, 51.3, 45.3, 44.9, 44.4, 44.0, 38.2, 36.3, 35.9, 35.2,
28.4, 28.3, 28.2, 28.1, 28.0; IR5neat/NaCl) 3300, 2950,
with a 4:1 hexane in ether solution. Elution with the same
solvent afforded 120 mg 585.7%) of the desired product as a
colorless oil: TLC 5Et2O/hexane 51:1)) Rfˆ0.40;
[a]D ˆ215.98 5c 0.252, ether); 1H NMR 5CDCl3/
20
600 MHz) d 1.38 5s, 5.4H), 1.44 5s, 3.6H), 1.73 5d, 3H,
Jˆ6.6 Hz), 1.85±1.88 5m, 1H), 2.73±2.83 5m, 1H), 3.23
5s, 1.2H), 3.26 5s, 1.8H), 3.68±3.82 5m, 1H), 4.48 5d,
0.6H, Jˆ9.0 Hz), 4.55 5d, 0.4H, Jˆ9.0 Hz), 4.59 5t, 0.4H,
Jˆ6.6 Hz), 4.72 5t, 0.6H, Jˆ6.6 Hz), 5.48±5.68 5m, 2H),
7.20±7.31 5m, 5H); 13C NMR 5CDCl3/150 MHz) d 171.6,
153.4, 136.5, 136.4, 131.3, 131.2, 128.3, 128.3, 127.9,
127.8, 127.5, 127.4, 125.6, 125.5, 80.1, 80.0, 64.9, 64.3,
58.8, 58.4, 51.3, 51.2, 45.2, 44.3, 34.2, 33.4, 28.4, 28.2,
17.6, 17.5; IR 5neat/NaCl) 3028, 2976, 2950, 1746, 1705,
1692, 1452, 1392, 1367, 1209, 1176, 1132, 964, 783,
698 cm21; LRMS 5FAB) 346 5MH1, 6), 344 58), 290 558),
246 5100), 230 557), 204 526), 186 556), 93 534); HRMS
C20H28NO4 5MH1) calcd 346.2018, found 346.2022.
1720, 1659, 1320, 1300, 1050, 725 cm21
.
The product was carried to the next step without further
characterization.
1.1.3. +3S)-N-tert-Butyloxycarbonyl-5-methoxy-3-phenyl-
l-proline methyl ester +12). To the solution of 4 5104 mg,
0.33 mmol) in 1 mL ofmethanol, 15 mg PPTS 50.06 mmol)
was added and the reaction was stirred under room tempera-
ture overnight. Then the reaction mixture was diluted with
10 mL ofethyl acetate, successively washed with 1% HCl,
saturated NaHCO3 and brine, dried over MgSO4, concen-
trated in vacuo. The crude product was chromatographed
through 20 g ofsilica gel that was slurry packed with a
4:1 hexane in ether solution. Elution with the same solvent
afforded 90 mg 590, 63% over the three steps from 3) ofthe
desired product 12 as a colorless oil: TLC 5Et2O/hexane
1.1.5.
+3S,5S)-N-tert-Butyloxycarbonyl-3-phenyl-5-+2-
methyl-1-propenyl)-l-proline methyl ester +5b). Vinyl
substituted proline derivative 5b was made in a fashion
identical to that described for the synthesis of 5a. TLC
1
5Et2O/hexane 51:2)) Rfˆ0.3; H NMR 5CDCl3/300 MHz)
d 1.38 5s, 5.4H), 1.44 5s, 3.6H), 1.73 5s, 2.1H), 1.74 5s,
2.1H), 1.79 5s, 0.9H), 1.80 5s, 0.9H), 1.76±1.85 5m, 1H),
2.76±2.92 5m, 1H), 3.24 5s, 0.34H), 3.28 5s, 0.66H), 3.77±
3.90 5m, 1H), 4.52 5d, 0.6H, Jˆ9.0 Hz), 4.60 5d, 0.4H,
Jˆ9.0 Hz), 4.86 5t, 0.4H, Jˆ9.0 Hz), 4.96 5t, 0.6H,
Jˆ9.0 Hz), 5.31 5t, 1H, Jˆ9.0 Hz), 7.20±7.34 5m, 5H);
13C NMR5CDCl3/75 MHz) d 171.8, 154.4, 153.4, 136.7,
136.6, 133.5, 132.1, 128.3, 127.9, 127.8, 127.4, 126.5,
79.9, 79.8, 64.9, 64.3, 55.4, 55.3, 51.3, 51.2, 45.8, 44.8,
35.7, 35.0, 28.4, 28.3, 25.8, 25.6, 18.1, 18.0; IR 5neat/
NaCl) 3054, 3028, 2971, 2945, 1744, 1705, 1692, 1452,
1
51:1)) Rfˆ0.5; H NMR 5CDCl3/300 MHz) d 1.38 5s, 6H),
1.50 5s, 3H), 2.05 5dd, 1H, Jˆ5.4, 12.3 Hz), 2.58 5ddd, 1H,
Jˆ5.4, 13.8, 26.4 Hz), 3.19 5s, 1.1H), 3.22 5s, 1.9H), 3.42 5s,
1.07H), 3.46 5s, 1.93H), 4.01±4.10 5m, 1H), 4.48 5d, 0.66H,
Jˆ8.7 Hz), 4.52 5d, 0.34H, Jˆ8.7 Hz), 5.25 5d, 0.34H,
Jˆ5.1 Hz), 5.45 5d, 0.66H, Jˆ4.8 Hz), 7.18±7.39 5m, 5H);
13C NMR5CDCl3/75 MHz) d 171.2, 154.0, 136.2, 136.1,
129.1, 128.3, 128.3, 127.7, 127.8, 127.5, 127.5, 89.0, 88.8,
81.1, 80.8, 64.8, 64.2, 56.4, 56.2, 51.4, 51.2, 44.9, 44.0,
35.1, 34.1, 28.3, 28.1; IR 5neat/NaCl) 3028, 2971, 2950,
1744, 1708, 1387, 1367, 1318, 1204, 1173, 1085,
700 cm21; LRMS 5FAB) 304 5MH12CH3OH, 45), 204
5100), 144 585); HRMS C17H22NO4 5MH1CH3OH) calcd
304.1549, found 304.1548.
1390, 1364, 1253, 1207, 1170, 1126, 899, 791, 698 cm21
;
LRMS 5FAB) 366 5M1Li, 22), 358550), 304560), 258578),
2045100), 144534), 91532); HRMS C21H29NO4 5M1Li)
calcd 366.2256, found 366.2252.
1.1.6. +3S,5S)-N-+Cbz±Phe)-3-Phenyl-5-+2-methyl-1-pro-
penyl)-l-proline methyl ester +13). The vinyl substituted
proline 5b 5490 mg, 1.3 mmol) was dissolved in 5 mL of3N
HCl in ethyl acetate 5VHCl/VEtOAcˆ1:3) and the reaction was
stirred at room temperature for 30 min. The solvent was
removed and the residue was dried in vacuo. The residue
was then dissolved in 5 mL ofCH 2Cl2. To this solution was
added 0.33 mL of N-ethylmorpholine 52.6 mmol) and 0.9 g
ofCbz±Phe±F 52.6 mmol) in 5 mL ofCH 2Cl2. The stirring
was continued for one day. The reaction was then diluted
1.1.4. +3S,5S)-N-tert-Butyloxycarbonyl-3-phenyl-5-+trans-
1-propenyl)-l-proline methyl ester +5a). In a ¯ame-dried
25 mL round bottom ¯ask under argon, 23 mg oflithium
wire 5containing 0.5% Na, 3.27 mmol) was cut and washed
three times with hexane. Anhydrous ether 53.5 mL) was
added and the mixture was cooled to 2208C. To this solu-
tion was added 0.14 mL of trans-1-bromo-1-propene
51.64 mmol). After 2 h at 2208C, this solution was cannu-
lated into a 50 mL ¯ask containing 0.34 g CuBr´Me2S
51.64 mmol) in 1 mL ofanhydrous ether at 2408C. The
resulting dark brown solution was stirred at 2408C f or
1 h, cooled to 2788C, and then treated with 0.20 mL of
BF3´Et2O 51.64 mmol). After 5 min, 137 mg of 12
50.41 mmol) was added to the solution and the cooling
bath was removed. After 15 min, the reaction mixture was
quenched with a 1:1 NH4OH/saturated NH4Cl solution and
diluted with CH2Cl2. The layers were separated and the
inorganic phase was extracted three times with CH2Cl2.
The combined organic layers were dried over MgSO4 and
concentrated in vacuo. The crude product was chromato-
graphed through 20 g ofsilica gel that was slurry packed
with 15 mL ofCH Cl2 and washed successively with 5%
2
citric acid 5two times), 5% NaHCO3 5two times), and brine.
The resulting organic layer was dried over MgSO4 concen-
trated in vacuo. The crude product was chromatographed
through 50 g ofsilica gel that was slurry packed with 1:1
ether in hexane. Elution with the same solvent afforded
450 mg 560%) ofthe coupling product 13 as a pale yellow
solid: TLC 5Et2O/hexane 51:1)) Rfˆ0.14; 1H NMR 5CDCl3/
300 MHz) d 1.67±1.70 5m, 6H), 2.01±2.15 5m, 2H),
2.38±2.48 5m, 1H), 2.82±3.24 5m, 2H), 3.25±3.26 5m,
3H), 3.65±3.79 5m, 2H), 4.09±4.14 5m, 1H), 4.21±4.25
5m, 1H), 4.62±4.88 5m, 1H), 5.03±5.21 5m, 1H), 5.24±5.42
5m, 1H), 7.14±7.31 5m, 15H); 13C NMR 5CDCl3/75 MHz) d

S. Duan, K. D. Moeller / Tetrahedron 57 )2001) 6407±6415
6413
173.4, 171.3, 170.3, 155.6, 140.3, 136.6, 136.3, 135.9, 134.3,
129.4, 128.4, 128.3, 128.2, 127.9, 127.7, 127.5, 126.8, 126.5,
126.2, 70.1, 66.7, 65.4, 64.3, 55.4, 53.0, 51.3, 47.6, 44.0, 39.5,
39.1, 36.3, 33.4, 32.9, 25.8, 18.1; IR 5neat/NaCl) 3374, 3286,
3023, 2945, 1741, 1715, 1640, 1491, 1439, 1232, 1044, 732,
695 cm21; LRMS 5FAB) 547 5M1Li, 100), 487 56), 296 510),
266 518), 210 516), 160 520), 91562); HRMS C33H36N2O5Li
5M1Li) calcd 547.2784, found 547.2783.
10 mL ofpear ¯ask ®tted with N 2 inlet and magnetic stir bar,
0.14 mL ofdiisopropylamine 51.0 mmol) was dissolved in
1 mL ofrfeshly distilled THF, cooled to
2788C, then
0.44 mL n-BuLi 52.5 M in hexane, 1.1 mmol) was added.
After 30 min, this solution was cannulated into a 25 mL of
round bottom ¯ask containing the starting material 5a
560 mg, 0.2 mmol) in 1 mL ofTHF at 2788C. The reaction
was allowed to slowly warm to room temperature. After 5 h,
5 mL ofanhydrous methanol was added and the reaction
was stirred for 8 h. The reaction mixture was then
neutralized with dilute HCl to pHˆ7 and 5 mL ofwater
was added. The organic layer was separated and the
inorganic layer was extracted three times with CH2Cl2.
The combined organic layer was washed with saturated
NaCl solution, dried over Na2SO4, and concentrated in
vacuo. The crude product was chromatographed through
10 g ofsilica gel that was slurry packed with a 4:1 hexane
in ether solution. Elution with the same solvent afforded
48 mg 580%) ofthe desired product 6 as a colorless oil:
1.1.7. +3S,6S,8S,9S)-1,4-Diaza-3-benzyl-9-carbomethoxy-
8-phenyl-2-oxo bicyclo-[4, 3, 0]nonane +14). In a 50 mL of
round bottom ¯ask ®tted with stir bar, 440 mg ofthe
coupling product 13 50.83 mmol) was dissolved in 5 mL
ofCH Cl2. The solution was cooled to 2788C and then
2
ozone bubbled through until a blue color persisted. After
the solution turned blue, the ozone was bubbled for addi-
tional 10 min. The ¯ow ofozone was then terminated and
oxygen bubbled through the solution till the blue color
disappeared. The reaction was then treated with 0.3 mL of
Me2S 54.1 mmol) at 2788C, allowed to warm to room
temperature, and stirred overnight. The solvent was
removed in vacuo and the crude product was chromato-
graphed through 20 g ofsilica gel using 4:1 ether in hexane
as elutant to afford 370 mg 590%) of product 5two isomers,
,5:3) as a white solid: TLC 5Et2O/hexane 51:1)) Rfˆ0.30,
20
TLC 5Et2O/hexane 51:1)) Rfˆ0.40; [a]D ˆ131.58 5c
1
0.127, ether); H NMR 5CDCl3/300 MHz) d 1.42 5s, 9H),
1.75 5d, 3H, Jˆ6.3 Hz), 2.09 5m, 1H), 2.22±2.32 5m, 1H),
3.44±3.54 5m, 1H), 3.68 5s, 3H), 4.21±4.56 5m, 2H), 5.30±
5.87 5m, 2H), 7.22±7.35 5m, 5H); 13C NMR 5CDCl3/
75 MHz) d 173.2, 153.3, 139.2, 131.2, 130.8, 128.7,
127.1, 126.5, 80.2, 66.6, 65.8, 59.7, 59.2, 51.8, 47.9, 47.0,
40.2, 29.6, 28.3, 17.8; IR 5neat/NaCl) 3033, 2971, 2935,
1752, 1692, 1452, 1392, 1364, 1253, 1196, 1165, 1121,
964, 762, 698 cm21; LRMS 5FAB) 346 5MH1, 10), 246
544), 230 518), 204 515), 185 550), 154 542), 137 532), 93
5100); HRMS C20H28NO4 5MH1) calcd 346.2018, found
346.2021.
1
0.14; H NMR 5CDCl3/300 MHz) d 2.15±2.28 5m, 1H),
2.48±2.52 5m, 1H), 3.11±3.30 5m, 2H), 3.31 5s, 3H), 3.44
5s, 1H), 3.74±3.87 5m, 1H), 4.60±4.62 5m, 1H), 4.96 5d, 1H,
Jˆ8.7 Hz), 5.17 5s, 2H), 5.46 5s, 1H), 7.04±7.46 5m, 15H);
13C NMR 5CDCl3/75 MHz) d 169.9, 167.5, 155.8, 155.6,
138.2, 135.5, 135.0, 129.8, 129.3, 128.9, 128.6, 128.5,
128.4, 128.2, 128.1, 128.0, 127.8, 127.7, 127.5, 126.9,
126.7, 81.0, 68.9, 67.8, 66.8, 63.2, 60.3, 57.3, 51.7, 44.8,
38.2, 37.0, 33.1, 31.5, 22.6, 14.0; IR 5neat/NaCl) 3400,
3028, 2950, 1739, 1705, 1679, 1449, 1405, 1307, 1209,
749, 698 cm21; LRMS 5FAB) 521 5M1Li, 46), 312 516),
210 520), 160 518), 121 512), 91 5100); HRMS
C30H30N2O6Li 5M1Li) calcd 521.2264, found 521.2259.
1.1.9. +3S)-N-tert-Butyloxycarbonyl-5-methoxy-3-phenyl-
d-proline methyl ester. A procedure identical to that used
to prepare 6 from 5a was used. To this end, 130 mg of 12
afforded 97 mg 575%) of the epimerized product as a color-
less oil. TLC 5Et2O/hexane 51:4)) Rfˆ0.12; 1H NMR
5CDCl3/300 MHz) d 1.43 5s, 6.3H), 1.52 5s, 2.7H), 2.11±
2.16 5m, 1H), 2.24±2.30 5m, 1H), 3.48 5s, 0.9H), 3.51 5s,
2.1H), 3.69 5s, 3H), 3.71±3.77 5m, 1H), 4.27 5d, 0.7H,
Jˆ9.6 Hz), 4.42 5d, 0.3H, Jˆ9.3 Hz), 5.30 5d, 0.3H,
Jˆ4.5 Hz), 5.40 5d, 0.7H, Jˆ4.5 Hz), 7.27±7.36 5m, 5H);
13C NMR 5CDCl3/75 MHz) d 172.6, 154.0, 139.2, 128.7,
127.3, 88.2, 87.9, 81.1, 80.8, 66.3, 65.5, 55.6, 55.2, 52.1,
52.00, 47.4, 46.6, 41.8, 40.6, 28.3, 28.1, 25.00; IR 5neat/
NaCl) 2971, 2945, 1754, 1708, 1452, 1385, 1364, 1315,
1199, 1170, 1085, 765, 698 cm21; LRMS 5FAB) 304
5MH12CH3OH, 16), 204 5100), 144 528), 93 526); HRMS
C17H22NO4 5MH12CH3OH) calcd 304.1549, found
304.1544.
To ®nish the synthesis of 14, a solution of220 mg ofthe
ozonolysis product 50.44 mmol) in 3 mL ofanhydrous
methanol was treated with 80 mg of5% palladium on
BaSO4. The reaction was then placed under hydrogen
balloon for one day. Following this period, the reaction
mixture was ®ltered through celite and the ®ltrate concen-
trated in vacuo. The crude product was chromatographed
through 20 g ofsilica gel using ether/methanol 59:1) as
the elutant to afford 121 mg 580%) of pure product 14 as a
yellowish oil: TLC 5CH3OH/Et2O 51:50)) Rfˆ0.16; 1H
NMR 5CDCl3/300 MHz) d 1.84 5br. s, 1H), 2.11±2.19 5m,
1H), 2.36 5dd, 1H, Jˆ11.4, 24.3 Hz), 2.98 5dd, 1H, Jˆ11.4,
13.8 Hz), 3.03 5dd, 1H, Jˆ11.4, 21.9 Hz), 3.17 5dd, 1H,
Jˆ3.6, 12 Hz), 3.22 5s, 3H), 3.26 5dd, 1H, Jˆ3.6,
14.1 Hz), 3.65±3.76 5m, 2H), 3.88±3.98 5m, 1H), 4.72 5d,
1H, Jˆ9.0 Hz), 7.22±7.36 5m, 10H); 13C NMR 5CDCl3/
75 MHz) d 170.3, 169.6, 138.6, 135.8, 129.4, 128.6,
128.3, 127.9, 127.6, 126.5, 63.7, 59.9, 58.1, 51.5, 45.4,
43.8, 37.4, 32.6; IR5neat/NaCl) 3462, 3328, 3059, 3023,
2945, 1741, 1643, 1496, 1452, 1431, 1369, 1323, 1207,
1.1.10. +3S,5R)-N-tert-Butyloxycarbonyl-3-phenyl-5-+trans-
1-propenyl)-d-proline methyl ester +7). In a procedure
similar to that used to prepare 5a from 12, 64 mg ofthe
methoxy compound prepared in the proceeding experiment
was converted into 48 mg 575%) of 7 as a colorless oil. TLC
20
5Et2O/hexane 51:1)) Rfˆ0.36; [a]D ˆ221.18 5c 0.10,
1
ether); H NMR 5CDCl3/300 MHz) d 1.40 5s, 4.5H), 1.41
1176, 783, 729, 695 cm21
;
5s, 4.5H), 1.65 5d, 3H, Jˆ5.1 Hz), 1.87±1.97 5m, 1H), 2.44±
2.50 5m, 1H), 3.33±3.41 5m, 1H), 3.68 5s, 3H), 4.36±4.48
5m, 2H), 5.25±5.66 5m, 2H), 7.21±7.39 5m, 5H); 13C NMR
5CDCl3/75 MHz) d 173.2, 172.3, 154.5, 153.2, 140.2, 139.9,
1.1.8. +3S,5S)-N-tert-Butyloxycarbonyl-3-phenyl-5-+trans-
1-propenyl)-d-proline methyl ester +6). In a ¯ame-dried

6414
S. Duan, K. D. Moeller / Tetrahedron 57 )2001) 6407±6415
132.5, 131.8, 128.7, 128.6, 128.0, 127.1, 126.1, 80.1, 79.9,
66.8, 66.1, 60.8, 52.0, 51.9, 47.9, 47.2, 41.0, 40.5, 28.2,
17.6, 17.4; IR 5neat/NaCl) 3033, 2971, 2935, 1749, 1710,
1690, 1452, 1392, 1364, 1253, 1199, 1173, 959, 757,
698 cm21; LRMS 5FAB) 346 5MH1, 6), 246 520), 204
517), 185 554), 137 5500), 93 5100); HRMS C20H28NO4
5MH1) calcd 346.2018, found 346.2013.
552); HRMS C20H28NO4 5MH1) calcd 346.2018, found
346.2017.
1.1.14. +3S,6R,8R,9R)-1,4-Diaza-3-benzyl-9-carbomethoxy-
8-phenyl-2-oxo bicyclo-[4, 3, 0] nonane +15). Starting from
5b-ent, the bicyclic analog 15 was prepared in a fashion
identical to that described for the preparation of 14 from
1
5b. TLC 5Et2O/CH3OH 59:1)) Rfˆ0.30; H NMR 5CDCl3/
600 MHz) d 1.78 5bs, 1H), 2.02 5dt, 1H, Jˆ10.8, 15.6 Hz),
2.29±2.34 5m, 1H), 2.56 5dd, 1H, Jˆ12.6, 14.4 Hz), 3.04
5dd, 1H, Jˆ10.2, 16.8 Hz), 3.24 5s, 3H), 3.28 5dd, 1H,
Jˆ10.8, 15.6 Hz), 3.39 5dd, 1H, Jˆ4.8, 16.8 Hz), 3.68
5dd, 1H, Jˆ11.4, 3.6 Hz), 3.77 5dt, 1H, Jˆ4.8, 10.8 Hz),
4.19±4.25 5m, 1H), 5.16 5d, 1H, Jˆ10.8 Hz) 7.17±7.36
5m, 10H); 13C NMR 5CDCl3/150 MHz) d 169.6, 168.9,
139.2, 137.8, 129.3, 128.5, 128.2, 128.0, 127.2, 126.6,
63.8, 59.6, 59.2, 51.4, 48.8, 44.2, 38.5, 36.2; IR 5neat/
NaCl) 3486, 3222, 3027, 2945, 1743, 1647, 1496, 1452,
1431, 1370, 1202, 1178, 754, 699 cm21; LRMS 5FAB)
371 5M1Li, 100), 337 510), 280 530), 218 512), 134 510),
91 516); HRMS C22H24N2O3Li 5M1Li) calcd 371.1947,
found 371.1937.
1.1.11. +3R,5R)-N-tert-Butyloxycarbonyl-3-phenyl-5-+trans-
1-propenyl)-d-proline methyl ester +5a-ent). Starting
from
52R,3R)-2-N-5tert-butyloxycarbonyl)
amino]-3-
phenyl-4-pentenoic acidaa the enantiomer of 5a 55a-ent)
was synthesized in a fashion identical to that described
above for 5. TLC 5Et2O/hexane 51:1)) Rfˆ0.40;
20
[a]D ˆ114.68 5c 0.205, ether); 1H NMR 5CDCl3/
300 MHz) d 1.38 5s, 5.4H), 1.44 5s, 3.6H), 1.71±1.74 5m,
3H), 1.83±1.90 5m, 1H), 2.70±2.86 5m, 1H), 3.22 5s, 1.2H),
3.26 5s, 1.8H), 3.71±3.84 5m, 1H), 4.49±4.74 5m, 2H),
5.50±5.70 5m, 2H), 7.12±7.33 5m, 5H); 13C NMR 5CDCl3/
75 MHz) d 171.6, 171.5, 153.3, 136.5, 136.4, 131.1, 128.3,
127.9, 127.8, 127.4, 125.5, 80.0, 79.9, 64.9, 64.2, 58.3, 51.1,
45.2, 44.3, 34.2, 33.3, 28.3, 28.2, 17.7, 17.5; IR 5neat/NaCl)
3033, 2981, 2950, 1749, 1708, 1695, 1390, 1367, 1176,
1132, 786, 698 cm21; LRMS 5FAB) 346 5MH1, 6), 344
58), 290 558), 246 5100), 230 557), 204 526), 186 556), 93
534); HRMS C20H28NO4 5MH1) calcd 346.2018, found
346.2023.
Acknowledgements
We thank the National Institutes ofHealth 5GM5324001)
for their generous ®nancial support. In addition, we thank
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 NIH RR00954, for their assistance.
1.1.12. +3R,5R)-N-tert-Butyloxycarbonyl-3-phenyl-5-+trans-
1-propenyl)-l-proline methyl ester +6-ent). The enantio-
mer of 6 56-ent) was synthesized in a fashion identical to
that used to prepare 6. TLC 5Et2O/hexane 51:1)) Rfˆ0.45;
[a]D ˆ225.08 5c 0.16, ether); 1H NMR 5CDCl3/600 MHz)
20
d 1.42 5s, 5.4H), 1.44 5s, 3.6H), 1.75 5d, 3H, Jˆ6.6 Hz), 2.08
5m, 1H), 2.28 5m, 1H), 3.48±3.52 5m, 1H), 3.67 5s, 1.8H),
3.69 5s, 1.2H), 4.22±4.57 5m, 2H), 5.62±5.90 5m, 2H),
7.23±7.35 5m, 5H); 13C NMR 5CDCl3/150 MHz) d 173.2,
154.2, 153.3, 140.1, 139.3, 131.3, 130.8, 128.7, 128.7,
128.5, 128.3, 128.2, 127.8, 127.3, 127.1, 126.4, 80.2, 80.0,
66.6, 65.8, 59.7, 59.2, 52.1, 51.8, 47.9, 47.0, 40.2, 39.6,
29.6, 28.2, 17.8, 17.6; IR 5neat/NaCl) 3033, 2971, 2935,
1752, 1702, 1695, 1454, 1392, 1364, 1250, 1196, 1168,
1121, 964, 760, 698 cm21; LRMS 5FAB) 346 5MH1, 10),
290 535), 246 5100), 230 544), 204 532), 186 540), 157 520),
117 520), 91 530); HRMS C20H28NO4 5MH1) calcd
346.2018, found 346.2021.
Appendix A. Supplementary material available
The proton and carbon spectra are included for all new
compounds along with 2D-NOESY data for compounds
5a, 6, 7, 14, and 15 542 pages).
References
1. For a recent review concerning the use oflactam based pepti-
domimetics see: 5a) Hanessian, S.; McNaughton-Smith, G.;
Lombart, H.-G.; Lubell, W. D. Tetrahedron 1997, 53, 12789.
For more recent references see: 5b) Polyak, F.; Lubell, W. D.
J. Org. Chem. 1998, 63, 5937. 5c) Curran, T. P.; Marcaurell,
L. A.; O'Sullivan, K. M. Org. Lett. 1999, 1, 1998. 5d) Gosselin,
F.; Lubell, W. D. J. Org. Chem. 2000, 65, 2163 as well as
references therein.
1.1.13. +3R,5S)-N-tert-Butyloxycarbonyl-3-phenyl-5-+trans-
1-propenyl)-l-proline methyl ester +7-ent). The enantio-
mer of 7 57-ent) was prepared in a fashion identical to that
described for the preparation of 7. TLC 5Et2O/hexane 51:1))
Rfˆ0.36; [a]D ˆ115.98 5c 0.315, ether); 1H NMR 5CDCl3/
20
2. 5a) Beal, L. M.; Liu, B.; Chu, W.; Moeller, K. D. Tetrahedron
2000, 56, 10113. 5b) Beal, L. M.; Moeller, K. D. Tetrahedron
Lett. 1998, 39, 4639. For a complementary N-acyliminium ion
cyclization route to analogs like I see: 5c) Li, W.; Hanau, C. E.;
d'Avignon, A.; Moeller, K. D. J. Org. Chem. 1995, 60, 8155.
5d) Li, W.; Moeller, K. D. J. Am. Chem. Soc. 1996, 118, 10106.
5e) Chu, W.; Moeller, K. D. Tetrahedron Lett. 1999, 40, 7939.
3. 5a) Tong, Y.; Fobian, Y. M.; Wu, M.; Boyd, N. D.; Moeller,
K. D. J. J. Org. Chem. 2000, 65, 2484. 5b) Fobian, Y. M.;
Moeller, K. D. In Methods in Molecular Medicine: Peptidomi-
metic Protocols, Kazmierski, W. M., Ed.; Humana: Totawa,
1999; pp 259.
600 MHz) d 1.40 5s, 4.5H), 1.42 5s, 4.5H), 1.65±1.67
5m, 3H), 1.90±1.95 5m, 1H), 2.44±2.51 5m, 1H), 3.35±
3.39 5m, 1H), 3.69 5s, 3H), 4.35±4.50 5m, 2H), 5.27±5.65
5m, 2H), 7.25±7.33 5m, 5H); 13C NMR 5CDCl3/600 MHz)
d 173.3, 172.8, 154.5, 153.3, 140.3, 140.0, 132.6, 131.9,
128.6, 127.1, 126.2, 126.1, 80.1, 79.9, 66.9, 66.2, 60.8,
52.1, 51.9, 48.0, 47.2, 41.1, 40.5, 28.2, 17.7, 17.4; IR
5neat/NaCl) 3033, 2971, 2935, 1749, 1710, 1692, 1452,
1395, 1367, 1253, 1199, 1168, 959, 757, 698 cm21
;
LRMS 5FAB) 346 5MH1, 6), 344 58), 290 548), 246
5100), 204 520), 186 547), 1579 516), 117 518), 93

S. Duan, K. D. Moeller / Tetrahedron 57 )2001) 6407±6415
6415
4. 5a) Cotton, R.; Johnstone, A. N. C.; North, M. Tetrahedron
1995, 51, 8525. 5b) Mulzer, J.; Meier, A.; Buschmann, J.;
Luger, P. Synthesis 1996, 123. 5c) Lorthiois, E.; Marek, I.;
Normant, J. F. J. Org. Chem. 1998, 63, 2442.
7. 5a) Kazmaier, U.; Krebs, A. Angew. Chem., Int. Ed. Eng. 1995,
34, 2012. 5b) Kazmaeir, U.; Maier, S. Tetrahedron 1996, 52,
941.
8. 5a) Wistrand, L.-G.; Skrinjar, M. Tetrahedron 1991, 47, 573.
5b) Collado, I.; Ezquerra, J.; Pedregal, C. J. Org. Chem. 1995,
60, 5011 See also Ref. 3a.
5. For the general approaches to 3-substituted prolines see:
5a) Damour, D.; Pulicani, J.-P.; Vuilhorgne, M.; Mignani, S.
Synlett 1999, 786. 5b) Carpes, M. J. S.; Miranda, P. C. M. L.;
Correia, C. R. D. Tetrahedron Lett. 1997, 38, 1869.
5c) Mosberg, H. I.; Kroona, H. B. J. Med. Chem. 1992, 35,
4498. 5d) Chung, J. Y. L.; Wasicak, J. T.; Arnold, W. A.;
May, C. S.; Nadzan, A. M.; Holladay, M. W. J. Org. Chem.
1990, 55, 270. 5e) Sarges, R.; Tretter, J. J. Org. Chem. 1974, 39,
1710.
9. For a description ofgeneral experimental details see: Wong,
P. L.; Moeller, K. D. J. Am. Chem. Soc. 1993, 115, 11434. The
NOESY spectrum resulted from a 2£220£2048 data matrix
with 16 scans per t1 value. Spectra were recorded using mixing
times of220 and 450 ms. Optimal results were obtained with
the mixing time of450 ms. The time domain data were zero
®lled to yield a 2K£2K data matrix and were Fourier trans-
formed using sine bell and Gausssian weighing function in
both the t1 and t2 dimensions.
6. For a related route to 3-substituted prolines see: Evans, M. C.;
Johnson, R. L. Tetrahedron Symposium in Print )Peptidomi-
metics) 2000, 56, 9801.