Synthesis of an epimer of a 16-membered ring model for teicoplanin: Confirmation of epimerization during cycloamidation

Full text of DOI:10.1021/jo970508t

4536
J . Org. Chem. 1997, 62, 4536-4538
Sch em e 1
Syn th esis of a n Ep im er of a 16-Mem ber ed
Rin g Mod el for Teicop la n in : Con fir m a tion
of Ep im er iza tion d u r in g Cycloa m id a tion
Anthony J . Pearson,* Penglie Zhang, and Gilles Bignan
Department of Chemistry, Case Western Reserve University,
Cleveland, Ohio 44106
Received March 19, 1997X
Because of their unique structure and clinical impor-
tance, vancomycin and related antibiotics, including
teicoplanin and ristocetin, are currently the subject of
intensive synthetic studies.1 One of the difficulties with
their total synthesis is the atropisomerism (conforma-
tional isomerism) of the biaryl ether linkage and, in the
case of chloro-substituted derivatives, the peptido aryl
ether rings. In previous studies from our laboratory,
areneruthenium chemistry was utilized in the synthesis
of two 16-membered cyclic peptides 1 and 2, which
represent models for partial structures of teicoplanin and
ristocetin, designed to test the efficacy of our ruthenium-
promoted etherification methodology. Compound 1 was
obtained by a cycloamidation strategy using different
coupling methods, and two cyclized monomers, labeled
1-F 1 and 1-F 2, were obtained.2 We were uncertain
whether they were stable conformational isomers or
epimers at the phenylalanine residue. Ruthenium-
mediated cycloetherification recently reported from our
laboratory provided compound 2.3 Herein, we report the
unambiguous synthesis of compound 3, which is the
epimer of compound 2 in the phenylalanine unit. The
result shows that 1-F 1 and 1-F 2 obtained earlier are
epimers and not conformational isomers.
the preparation of compound 2.3 Thus, the one-step
reduction and in situ Boc protection of the azidation
product 44 gave compound 5, which was treated with
LiOOH to provide N-Boc-L-4-chlorophenylalanine (6).
Complexation of 6 using (CH3CN)3RuCpPF6 furnished
complex 7, which was coupled with dipeptide 8 to give
the tripeptide complex 9 in 69% yield. Cycloetherification
was effected by treatment with sodium 2,6-di-tert-bu-
tylphenoxide at -78 °C under high dilution conditions
(0.002 M), and photolytic demetalation of the product
(Rayonet, 350 nm, CH3CN) led to compound 3 in 31%
unoptimized overall yield. Higher field shift of H2 (6.12
ppm) in the NMR spectrum indicated the presence of the
cyclized product because H2 in ring B is shielded by ring
C. An important structural marker is the benzylic
methylene (Ha and Ha′). One of these protons appears
as a doublet of doublets at 3.23 ppm with J gem ) 12.4
Hz, J vic ) 5.1 Hz; the other is a triplet at 2.88 ppm,
resulting from large geminal and vicinal couplings (J gem
) J vic ) 12.4 Hz).
As reported previously, in the NMR spectrum of
compound 2,3 both benzylic H’s showed large geminal and
small vicinal coupling constants (J gem ) 13.8, J vic ) 4.8
Hz and J gem ) 13.8, J vic ) 3.0 Hz), which is in contrast
to that in compound 3. On the other hand, preparation
of compound 1 by cycloamidation using the pentafluo-
rophenyl (PFP) ester method gave a low yield (14%) of
two cyclized monomers labeled 1-F 1 and 1-F 2 in a ratio
Synthesis of 3 is outlined in Scheme 1, employing
cycloetherification procedures analogous to those used for
(3) Pearson, A. J .; Bignan, G.; Zhang, P.; Chelliah, M. J . Org. Chem.
1996, 61, 3940. In this work we obtained a 7:1 mixture of epimers in
favor of 2. More recently, we have observed that epimerization
apparently occurs during the aqueous quench after peptide coupling;
when this is performed carefully at 0 °C, a single isomer is obtained,
as described in the preparation of 9 and 3.
X Abstract published in Advance ACS Abstracts, J une 1, 1997.
(1) (a) Evans, D. A.; DeVries, K. M. in Glycopeptide Antibiotics;
Nagarajan, R., Ed.; Marcel Dekker: New York, 1994; p 63. (b) Rao, A.
V. R.; Gurjar, M. K.; Reddy, K. L.; Rao, A. S. Chem. Rev. 1995, 95,
2135.
(2) Pearson, A. J .; Lee, K. J . Org. Chem. 1995, 60, 7153
(4) Pearson, A. J .; Lee, K. J . Org. Chem. 1994, 59, 225.
S0022-3263(97)00508-2 CCC: $14.00 © 1997 American Chemical Society

Notes
J . Org. Chem., Vol. 62, No. 13, 1997 4537
of 7:3.2 The major component 1-F 1 (J gem ) 12.2, J vic
)
1H, J ) 7.0 Hz), 4.62-4.57 (m, 1H), 4.22-4.11 (m, 2H), 3.33
(dd, 1H, J ) 13.6, 1.3 Hz), 3.14 (dd, 1H, J ) 12.2, 2.0 Hz), 2.81-
2.74 (m, 2H), 1.38 (s, 9H); 13C NMR (CDCl3) δ 172.6, 155.1, 152.7,
135.0, 134.6, 132.9, 130.8, 129.4, 128.6, 127.4, 80.1, 66.5, 55.5,
54.1, 38.0, 37.5, 28.2; HRMS calcd for C20H18N2O4Cl (M+ - OBut)
385.0955, found 385.0953.
5.2 Hz and J gem ) J vic ) 12.2 Hz) has a splitting pattern
similar to that of 3, while the minor one 1-F 2 (J gem
14.0, J vic ) 5.2 Hz, J gem ) 14.0, J vic ) 2.9 Hz) was com-
parable to 2.
)
MM2 modeling has been reported for compound 1.2
Atropisomers along the biaryl ether bond and epimers
at each amino acid residue were considered. In the
atropisomer with all the stereochemical features corre-
sponding to that assigned to naturally occurring teico-
planin, the dihedral angles a/X1 and a′/X1 were 73° and
43°, respectively, which should give J vic’s consistent with
the values found in 1-F 2. The epimer of this atropisomer
at chiral center X1 showed dihedral angles of 61° and
179°. Alternatively, a conformational isomer of the
molecule could be produced, which showed dihedral
angles of 58° and 175°. It was proposed previously that
1-F 1 was the stable conformational isomer because of the
large energy difference between the epimeric compounds
(10.62 kcal/mol, 15.66 kcal/mol for the epimer). With
compounds 2 and 3 in hand, it now appears more likely
that 1-F 1 is in fact the X1 epimer.
(S)-N-[(1,1-Dim et h ylet h oxy)ca r b on yl]-4-ch lor op h en yl-
a la n in e (6). To a solution of compound 5 (110 mg, 0.25 mmol,
1 equiv) in 6 mL of THF/H2O (3:1) at 0 °C were added 30% H2O2
(1.48 mL, 6 equiv) and 1 N LiOH (0.5 mL, 2 equiv). The mixture
was stirred for 1 h and quenched with 1.5 M Na2SO3 (1.5 mL).
The solution was buffered to basic (pH ∼9) with saturated
NaHCO3. THF was evaporated, and the residue was extracted
with CH2Cl2 (5 mL × 3), acidified to pH ∼2 with 1 N HCl, and
extracted with EtOAc (5 mL × 4). The EtOAc layer was dried
with MgSO4 and evaporated give compound 6 (65.4 mg, 91%)
as a white solid: [R]22 ) +46.2 (c 0.90, CHCl3); 1H NMR
D
(DMSO-d6) δ 7.32 (d, 1H, J ) 8.5 Hz), 7.25 (d, 1H, J ) 8.5 Hz),
7.14 (d, 1H, J ) 8.4 Hz), 4.06 (m, 1H), 2.99 (dd, 1H, J ) 13.6,
4.4 Hz), 2.78 (dd, 1H, J ) 13.6, 10.5 Hz); 13C NMR (DMSO-d6)
δ 173.4, 155.4, 137.1, 131.0, 128.0, 78.1, 55.0, 35.7, 28.1; HRMS
calcd for C14H18NO4Cl 299.0924, found 299.0943.
[η6-[4-Ch lor o-1-[(2S )-2-[N -[(1,1-d im e t h yle t h oxy)ca r -
bon yl]a m in o]-3-oxo-3-h yd r oxyp r op yl]ben zen e]](η5-cyclo-
p en ta d ien yl)r u th en iu m Hexa flu or op h osp h a te (7). A solu-
tion of compound
6 (100 mg, 0.33 mmol, 1 equiv) and
Because of the favored formation of 1-F 1 over 1-F 2,
we deduce that, during the cycloamidation reaction using
PFP active ester together with Et3N, the X1 epimer is
kinetically preferred in the cyclization process, consider-
ing the dominance of the acyclic epimer is very unlikely
under such mild reaction conditions.5 Similar phenom-
ena have recently been observed in the synthesis of the
cycloisodityrosine subunit of deoxybouvardin and RA-VII
by Inoue et al.6 and Boger et al.7
In conclusion, by synthesizing the X1 epimer of com-
pound 2, we have demonstrated the formation of the
epimer during the cycloamidation reaction. Even though
racemization at the phenylalanine chiral center is usually
not observed, the favorable orientation of the reactive
groups in the acyclic epimer in the transition state for
the cycloamidation step is a plausible reason for the
dominant formation of the cyclic epimer. Moreover, the
present work demonstrates the capability of NMR to
easily distinguish these two epimers and helps to confirm
the stereochemical assignments that have been made for
teicoplanin.8
(CH3CN)3RuCpPF6 (189 mg, 1.3 equiv) in 6 mL of 1,2-dichloro-
ethane was purged at rt with Ar for 25 min and then refluxed
for 5 h. The cooled solution was filtered through a thin layer of
Celite to remove Ru residues and evaporated to give compound
7 (222 mg, 100%) as a brown solid. The complex was used in
the next step without further purification: 1H NMR (CDCl3) δ
6.70-6.20 (m, 4H), 5.83 (br, 1H), 5.54 (s, 5H), 4.47 (m, 1H), 3.15
(dd, 1H, J ) 13.9, 4.4 Hz), 2.86 (dd, 1H, J ) 13.9, 9.6 Hz), 1.48
(s, 9H).
[η6-[4-Ch lor o-1-[(2S )-2-[N -[(1,1-d im e t h yle t h oxy)ca r -
b on yl]a m in o]-3-oxo-3-[[(1S)-1-p h en yl-2-oxo-2-[[(1R)-1-(3-
h yd r oxy-4-m et h oxyp h en yl)-2-et h oxy-2-oxoet h yl]a m in o]-
eth yl]a m in o]p r op yl]ben zen e]](η5-cyclop en ta d ien yl)r u th e-
n iu m Hexa flu or op h osp h a te (9). To a stirred solution of the
areneruthenium complex 7 (30.5 mg, 0.05 mmol, 1 equiv) and
HOBt (10.2 mg, 1.5 equiv) in freshly distilled DMF (1.0 mL) at
0 °C was added EDCI (11.0 mg, 1.1 equiv). In a separate
reaction flask, the dipeptide 8 (16.9 mg, 1.0 equiv) and diiso-
propylethylamine (9.6 µL, 1.1 equiv) in 1.2 mL of DMF were
stirred at 0 °C for 25 min. The resulting solution was added
dropwise to the solution of 7, and the mixture was stirred for 4
h at 0 °C and 20 h at rt. The solution was then diluted with 3
mL of H2O at 0 °C and extracted with CH3CN/CH2Cl2, (1/3, 6
mL × 5). The organic layers were combined and washed with
saturated NaHCO3 (4 mL × 2), 1 N KHSO4 (4 mL), and brine (4
mL). The organic layer was then dried over Na2SO4, concen-
trated in vacuo, and treated with diethyl ether to precipitate
the product. The mixture was then filtered, and the residue was
washed with cold diethyl ether and then dried under vacuum
to provide the Ru complex 6 (33 mg, 69%) as a pale brown solid
foam. Proton NMR showed it to be sufficiently pure for next
step: 1H NMR (CD3CN) δ 7.76 (bd, 1H, J ) 7.4 Hz), 7.57 (bd,
1H, J ) 7.5 Hz), 7.34 (s, 5H), 6.90-6.65 (m, 5H), 6.45-6.35 (m,
2H), 6.09 (m, 2H), 5.34 (s, 5H), 5.45-5.30 (m, 1H), 4.35-4.20
(m, 1H), 4.10 (m, 2H), 3.80 (s, 3H), 2.95-2.60 (m, 2H), 1.35 (s,
9H), 1.15 (t, 3H, J ) 7.4 Hz).
Cycloeth er ifica tion Rea ction . Syn th esis of 3. Sodium
hydride (25.0 mg, 60% in oil) was stirred with 128 mg of 2,6-
di-tert-butylphenol in 10 mL of dry THF for 20 min to give a
yellow solution. An aliquot (624 µL, 1.1 equiv) of the resulting
solution was diluted with 12 mL of precooled (-78 °C) THF, and
a solution of 9 (33 mg, 0.035 mmol) in 5.5 mL freshly distilled
THF was added at -78 °C via a syringe pump over 4.5 h and
the mixture then stirred for an additional 20 h at rt. The
solution was concentrated in vacuo and was redissolved in 15
mL of freshly distilled CH3CN, degassed by bubbling with N2
for 25 min, and then irradiated (Rayonet, 350 nm) for 24 h under
N2. The mixture was cooled to rt, concentrated in vacuo to ca.
0.5 mL, and treated with excess Et2O. After filtration, the
ethereal filtrate was concentrated under reduced pressure to give
a pale yellow residue that was purified first by thin-layer
chromatography (SiO2, EtOAc/hexanes, 1:1) to give compound
Exp er im en ta l Section
Gen er a l m eth od s are as described elsewhere.4
(4S)-3-[(2S)-2-[N-[(1,1-Dim eth yleth oxy)ca r bon yl]a m in o]-
3-(4-ch lor op h en yl)-1-oxop r op ion yl]-4-ben zyl-2-oxa zolid i-
n on e (5). A suspension of 10% Pd/C (∼12 mg) in 4 mL of dry
EtOAc was stirred vigorously under H2 atmosphere for 20 min.
To the mixture were added compound 4 (100 mg, 0.28 mmol, 1
equiv) and (Boc)2O (73 mg, 1.2 equiv). After being stirred at rt
under H2 for 3 h, the mixture was filtered through Celite, and
the filtrate was evaporated to give 110 mg (90%) of compound 5
as a white solid: mp 150-151 °C; [R]22D ) +80.1 (c 0.49, CHCl3);
Rf 0.25 (hexane/EtOAc, 7/3); IR (CHCl3) 1784, 1703 cm-1 1H
;
NMR (CDCl3) δ 7.37-7.18 (m, 9H), 5.72-5.65 (m, 1H), 5.15 (bd,
(5) It was shown interconversion between 1-F 1 and 1-F 2 was not
successful, which rules out the possibility that 1-F 1 was generated
from 1-F 2 after the cyclization; see ref 2.
(6) Inoue, T.; Sasaki, T.; Takayanagi, H.; Harigaya, Y.; Hoshino, O.;
Hara, H.; Inaba, T. J . Org. Chem. 1996, 61, 3936.
(7) (a) Boger, D. L.; Zhou, J . J . Org. Chem. 1996, 61, 3938. (b) Boger,
D. L.; Zhou, J .; Borzilleri, R. M.; Nukui, S. Bioorg. Med. Chem. Lett.
1996, 6, 1089. (c) Boger, D. L.; Zhou, J .; Borzilleri, R. M.; Nukui, S.;
Castle, S. L. J . Org. Chem. 1997, 62, 2054.
(8) (a) Hunt, A. H.; Molloy, R. M.; Occolowitz, J . L.; Marconi, G. G.;
Debono, M. J . Am. Chem. Soc. 1984, 106, 4891. (b) Barna, J . C. J .;
Williams, D. H.; Stone, D. J . M.; Leung, T.-W. C.; Doddrell, D. M J .
Am. Chem. Soc. 1984, 106, 4895.

4538 J . Org. Chem., Vol. 62, No. 13, 1997
Notes
3 (6.5 mg, 31%). Further purification was achieved by HPLC
(column: silica gel Whatman Magnum 9; eluant: hexanes/
EtOAc, 1:1; flow rate: 3 mL/min; tR ) 18.9 min): [R]22D ) +43.1
(c 0.3, CHCl3); Rf 0.51 (hexane/EtOAc, 1/1); 1H NMR (CD3CN) δ
7.50-6.80 (m, 12H), 6.12 (s, 1H), 5.60 (br, 1H), 5.39 (d, 1H, J )
12.9 Hz), 5.31 (d, 1H, J ) 10.8 Hz), 4.35 (br, 1H), 4.20 (m, 2H),
3.90 (s, 3H), 3.23 (dd, 1H, J ) 12.4, 5.1 Hz), 2.88 (dd, 1H, J )
12.4, 12.4 Hz), 1.40 (s, 9H), 1.25 (t, 3H, J ) 7.3 Hz); HRMS calcd
for C33H37N3O8 603.2581, found 603.2603.
acknowledged. We are grateful to Professor Dale Boger
for useful suggestions and discussions.
Su p p or tin g In for m a tion Ava ila ble: 1H, and 13C NMR
and MS spectra for the compounds 5, 6, 7, 9, and 3 (9 pages).
This material is contained in libraries on microfiche, im-
mediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
Ack n ow led gm en t. Financial support from the Na-
tional Institutes of Health (GM 36925) is gratefully
J O970508T