Thermal atropisomerism of aglucovancomycin derivatives: Preparation of (M,M,M)- and (P,M,M)-aglucovancomycins

Article abstract of DOI:10.1021/ja981928i

The degradation of vancomycin to a series of aglucovancomycin derivatives containing modifications in key functional groups and a study of their thermal atropisomerism are detailed. In all of the cases, selective isomerism of the DE ring system atropisomers was observed under conditions where the CD and AB stereochemistries were unaffected. Competitive retro aldol ring cleavage of the CD and DE ring systems (CD > DE) was observed but could be minimized by the choice of solvent and thermal conditions (DE ring system) or precluded by alcohol protection (CD ring system). Similarly, competitive main chain succinimide formation through the loss of ammonia from the Asn residue could be minimized by the choice of thermal conditions or prevented by carboxamide protection. Resynthesis of natural aglucovancomycin, (M,M,M)-2, and its unnatural DE atropisomer (P,M,M)-2 from 6 are described. The comparative antimicrobial activity of the key derivatives and their unnatural DE ring system P-diastereomers are disclosed.

Full text of DOI:10.1021/ja981928i

8920
J. Am. Chem. Soc. 1998, 120, 8920-8926
Thermal Atropisomerism of Aglucovancomycin Derivatives:
Preparation of (M,M,M)- and (P,M,M)-Aglucovancomycins
Dale L. Boger,* Susumu Miyazaki, Olivier Loiseleur, Richard T. Beresis, Steven L. Castle,
Jason H. Wu, and Qing Jin
Contribution from the Department of Chemistry and The Skaggs Institute for Chemical Biology,
The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037
ReceiVed June 2, 1998
Abstract: The degradation of vancomycin to a series of aglucovancomycin derivatives containing modifications
in key functional groups and a study of their thermal atropisomerism are detailed. In all of the cases, selective
isomerism of the DE ring system atropisomers was observed under conditions where the CD and AB
stereochemistries were unaffected. Competitive retro aldol ring cleavage of the CD and DE ring systems (CD
> DE) was observed but could be minimized by the choice of solvent and thermal conditions (DE ring system)
or precluded by alcohol protection (CD ring system). Similarly, competitive main chain succinimide formation
through the loss of ammonia from the Asn residue could be minimized by the choice of thermal conditions or
prevented by carboxamide protection. Resynthesis of natural aglucovancomycin, (M,M,M)-2, and its unnatural
DE atropisomer (P,M,M)-2 from 6 are described. The comparative antimicrobial activity of the key derivatives
and their unnatural DE ring system P-diastereomers are disclosed.
Introduction
desired natural stereochemistry. Significantly, the DE atropi-
somer equilibration in the isolated ring system occurred much
more rapidly (10-20 min, 130 °C) than that in the stand alone
CD ring system (>10 h, 130 °C) and indicated that it may be
possible to preferentially equilibrate the vancomycin DE versus
CD atropisomers within the intact CDE ring system. We
recently implemented this preferential DE equilibration in a
synthesis of a vancomycin CDE ring system in a manner that
indirectly addresses the control of the atropisomer stereochem-
istry.12,13
Herein, we report studies of the thermal atropisomerism
of a series of vancomycin aglycons that define its scope,
limitations, and potential competitive reactions. These studies
define conditions that permit a selective DE versus CD
atropisomerism and have provided an appropriately protected
vancomycin aglycon which not only minimizes competitive
thermal reactions but also may serve as a relay intermediate in
the total synthesis of the natural product itself. Thus, thermal
atropisomerism of 6 provided a 1:1 mixture of (M,M,M)-6 and
Vancomycin (1, Figure 1) is the prototypical member of a
class of clinically important glycopeptide antibiotics1-4 used
for the treatment of methicillin-resistant Staphylococcus aureus
and against enterococci and bacterial infections in patients
allergic to â-lactam antibiotics.5 The structural complexity of
vancomycin, the interest in defining its structural features
responsible for inhibition of cell wall biosynthesis in sensitive
bacteria,6 and the emergence of clinical resistance7 have renewed
interest in 1 and related agents.
The inherent control of the vancomycin CD and DE ring
system atropisomer stereochemistries remains one of the most
significant challenges yet to be addressed in efforts directed at
its examination or total synthesis.8,9 In recent studies, we
defined conditions under which the isolated CD and DE ring
system atropisomers may be equilibrated, and this has allowed
the unnatural atropisomers to be thermally equilibrated,10,11
chromatographically reisolated, and recycled to provide the
(1) McCormick, M. H.; Stark, W. M.; Pittenger, G. E.; Pittenger, R. C.;
McGuire, G. M. Antibiot. Annu. 1955-1956, 606.
(9) For recent disclosures: (a) Evans, D. A.; Barrow, J. C.; Watson, P.
S.; Ratz, A. M.; Dinsmore, C. J.; Evrard, D. A.; DeVries, K. M.; Ellman,
J. A.; Rychnovsky, S. D.; Lacour, J. J. Am. Chem. Soc. 1997, 119, 3419.
(b) Konishi, H.; Okuno, T.; Nishiyama, S.; Yamamura, S.; Koyasu, K.;
Terada, Y. Tetrahedron Lett. 1996, 37, 8791. (c) Pearson, A. J.; Chelliah,
M. V. J. Org. Chem. 1998, 63, 3087. (d) Rao, A. V. R.; Gurjar, M. K.;
Lakshmipathi, P.; Reddy, M. M.; Nagarajan, M.; Pal, S.; Sarma, B. V. N.
B. S.; Tripathy, N. K. Tetrahedron Lett. 1997, 38, 7433. (e) Bois-Choussy,
M.; Vergne, C.; Neuville, L.; Beugelmans, R.; Zhu, J. Tetrahedron Lett.
1997, 38, 5795. (f) Nicolaou, K. C.; Boddy, C. N. C.; Natarajan, S.; Yue,
T.-Y.; Li, H.; Braese, S.; Ramanjulu, J. M. J. Am. Chem. Soc. 1997, 119,
3421.
(10) Boger, D. L.; Borzilleri, R. M.; Nukui, S.; Beresis, R. T. J. Org.
Chem. 1997, 62, 4721. Boger, D. L.; Borzilleri, R. M.; Nukui, S. Bioorg.
Med. Chem. Lett. 1995, 5, 3091.
(11) Boger, D. L.; Castle, S. L.; Miyazaki, S.; Wu, J. H.; Beresis, R. T.;
Loiseleur, O. J. Org. Chem., in press.
(2) (a) Harris, C. M.; Kopecka, H.; Harris, T. M. J. Am. Chem. Soc.
1983, 105, 6915. (b) Williamson, M. P.; Williams, D. H. J. Am. Chem.
Soc. 1981, 103, 6580. (c) Sheldrick, G. M.; Jones, P. G.; Kennard, O.;
Williams, D. H.; Smith, G. A. Nature (London) 1978, 271, 223.
(3) Williams, D. H.; Kalman, J. R. J. Am. Chem. Soc. 1977, 99, 2768.
(4) Nagarajan, R. J. Antibiot. 1993, 46, 1181. Cooper, R. D. G.;
Thompson, R. C. Ann. Rep. Med. Chem. 1996, 31, 131. Malabarba, A.;
Nicas, T. I.; Thompson, R. C. Med. Res. ReV. 1997, 17, 69.
(5) Weidemann, B.; Grimm, H. In Antibiotics in Laboratory Medicine;
Lorian, V.; Ed.; Williams and Wilkins: Baltimore, 1996; pp 900-1168.
(6) Williams, D. H.; Searle, M. S.; Westwell, M. S.; Mackay, J. P.;
Groves, P.; Beauregard, D. A. Chemtracts: Org. Chem. 1994, 7, 133. Try,
A. C.; Sharman, G. J.; Dancer, R. J.; Bardsley, B.; Entress, R. M. H.;
Williams, D. H. J. Chem. Soc., Perkin Trans. 1 1997, 2911.
(7) Walsh, C. T.; Fisher, S. L.; Park, I.-S.; Prahalad, M.; Wu, Z. Chem.
Biol. 1996, 3, 21.
(8) Rao, A. V. R.; Gurjar, M. K.; Reddy, K. L.; Rao, A. S. Chem. ReV.
1995, 95, 2135. Evans, D. A.; DeVries, K. M. In Glycopeptide Antibiotics;
Nagarajan, R., Ed.; Marcel Dekker: New York; 1994; pp 63-104.
(12) Boger, D. L.; Loiseleur, O.; Castle, S. L.; Beresis, R. T.; Wu, J. H.
Bioorg. Med. Chem. Lett. 1997, 7, 3199.
(13) Boger, D. L.; Beresis, R. T.; Loiseleur, O.; Wu, J. H.; Castle, S. L.
Bioorg. Med. Chem. Lett. 1998, 8, 721.
S0002-7863(98)01928-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 08/22/1998

Atropisomerism of AglucoVancomycin DeriVatiVes
J. Am. Chem. Soc., Vol. 120, No. 35, 1998 8921
Scheme 1
Figure 1.
(P,M,M)-6 and their conversions to the natural vancomycin
aglycon 2 [(M,M,M)-2] and (P,M,M)-2, respectively, are de-
scribed. The former efforts constitute the final stages of an
anticipated total synthesis of the vancomycin aglycon, and the
latter have provided the corresponding key analogue possessing
the unnatural DE ring system (P)-stereochemistry.
Preparation of Substrates. Degradation of Vancomycin.
The degradation of vancomycin to a series of aglucovancomycin
derivatives is summarized in Scheme 1. Removal of the
disaccharide to provide aglucovancomycin 2 was accomplished
by TFA treatment14,15 (50 °C, 7.5 h), and shorter reaction periods
provided lower overall conversions to 3. Although aglucovan-
comycin (2) could be purified by semipreparative reverse phase
HPLC (C18, 2.5 × 10 cm, CH3CN-0.07% TFA/H2O, 17:83,
10 mL/min, Rt ) 18 min), it proved more convenient to carry
the crude material forward without purification. N-BOC forma-
tion (BOC2O, NaHCO3, dioxane-H2O, 25 °C, 2 h) followed
by methyl ester formation (CH3I, NaHCO3, DMF, 25 °C, 2 h)
and final exhaustive methyl ether protection of the phenols
(CH3I, K2CO3, DMF, 25 °C, 6 h) provided 3 in 37% overall
yield for the four steps. The intermediate N-BOC aglucovan-
comycin proved too polar to easily purify by normal phase
chromatography and selective diazomethane esterification, or
a single step exhaustive esterification and methyl ether formation
were not as effective as the two-step alkylative procedure, and
efforts to reverse the esterification (TMSCHN2) and N-BOC
protection proved unsuccessful. The most convenient procedure
entailed conducting the four steps on crude material with a final
the nitrile 6 (82%) was accomplished by treatment of 4 with
TFAA-pyridine in CH2Cl2 (0 °C, 10 min).16
Thermal Atropisomerism Studies. The now classic studies
leading to the structure elucidation of vancomycin which include
initial chemical degradation studies revealing the amino acid
composition and unusual amino sugar vancosamine,17 high-field
NMR studies that yielded further information on the unusual
amino acid structures, their connectivity linkages and stereo-
chemical relationships,3 and the X-ray structure of CDP-I,2c
a
degradation product produced by acidic thermal treatment (pH
4.2, 70-80 °C, 40 h), provided the basis for the first proposed
structure.2c This was shortly thereafter revised first to the correct
(M) versus (P) DE ring system atropisomer on the basis of
further NMR studies2b and finally to 1 with a revision for
backbone incorporation of an Asn versus iso-Asn residue.2a This
latter revision included chemical studies that documented the
intermediates in the conversion of vancomycin to its degradation
product CDP-I by way of an intermediate backbone succinimide
(Scheme 2). No DE ring-opened intermediates reflecting
cleavage of the backbone amide2a linking the central phenylg-
2
purification of the readily handled 3. O-Silylation of the C3
and C3 alcohols enlisting CF3CONMeTBDMS13 (50 equiv,
6
CH3CN, 50 °C, 9.5 h) provided 4 (66%) and a small amount of
the C36 mono-TBDMS ether 5 (10%). Shorter reaction periods
(4.5 h) at lower reaction temperatures (40 °C) provided a near
1:1 mixture of the readily separable 4 (54%) and 5 (44%). A
3
lycine and Asn residues or retro aldol2b at the C3 site were
2
third intermediate involving presumed silylation of the C4
carboxamide was also observed but was cleaved with an aqueous
citric acid workup. The selective monosilylation was established
to occur with protection of the C36 alcohol by 1D and 2D 1H-
detected. Selective ring-expansion hydrolysis of the succinimide
with release of ring strain was suggested to account for the
exclusive formation of the CDP-I isomers, and the subsequent
atropisomerism between the two CDP-I isomers was shown to
2
1H NMR and the alternative C3 mono-TBDMS ether was not
detected. Clean dehydration of the Asn carboxamide to provide
(16) Both DCC-py and EDCI-py provided only the recovered starting
material under similar reaction conditions.
(17) Smith, G. A.; Smith, K. A.; Williams, D. H. J. Chem. Soc., Perkin
Trans. 1 1975, 2108 and references cited therein.
(14) Nagarajan, R.; Schabel, A. A. J. Chem. Soc., Chem. Commun. 1988,
1306.
(15) For the use of HCl, see: Marshall, F. J. J. Med. Chem. 1965, 8, 18.

8922 J. Am. Chem. Soc., Vol. 120, No. 35, 1998
Boger et al.
Scheme 2
occur readily (t1/2 ) 10 h, DMSO, 25 °C, 7:3 P/M),2a Scheme
2. Atropisomerism of the DE ring system was observed only
within the ring-expanded CDP-I isomers and not within the 16-
membered DE ring system of vancomycin. To date, the
assumption has been that the barrier to isomerism of either the
CD or DE ring system within vancomycin is too large to permit
their observation.2 Analogous to these efforts, our attempts to
thermally equilibrate vancomycin (1) or aglucovancomycin (2)
in aprotic polar (DMSO or DMF, 130 °C, 2.5 h) or nonpolar
solvents (o-Cl2C6H4, 130 °C, 2.5 h) led to extensive or partial
degradation, respectively, from which no information on the
atropisomerism could be established. Consequently, we exam-
ined a number of thermally less sensitive aglycon derivatives
which permitted their conventional handling (SiO2, organic
solvents) and allowed us to address the impact of the critical
substituents.
Figure 2.
observed in our studies of the isolated vancomycin CD, DE,
and CDE ring systems.10-13 Thus, C3 but not C3 alcohol
protection is required for clean thermal atropisomerism.
6
2
The most successful and the most extensively investigated
substrate was 6. For all of the substrates including 6, the thermal
atropisomerism appeared cleaner in a nonpolar (o-Cl2C6H4)
versus polar (DMF, DMSO) aprotic solvent although the effect
of other solvents was not surveyed. Selective and rapid DE
isomerism occurred at 110-140 °C (o-Cl2C6H4), and only traces
of competitive byproducts were observed (Figure 2). Given
the past assumption that interconversion of the vancomycin
atropisomers is not possible, the relative ease with which the
DE ring system isomerizes is especially notable (t1/2 ) 2.8 h,
130 °C). This atropisomerism is substantially faster than the
observed DE atropisomerism within the isolated CDE ring
system we have examined12,13 (t/12 ) 20.4 h, 130 °C) and
comparable to that observed within the isolated DE ring system
itself (t1/2 < 1 h, 130 °C).10,11 Even at 110 °C, a reasonable
rate of isomerization (t1/2 ) 8.2 h) is observed. Preparative
isomerism at 130 °C (5 h, 80% recovery) and 140 °C (1.5 h,
70% recovery) provided a separable 1:1 mixture of natural
(M,M,M)-6 and (P,M,M)-6 possessing the unnatural DE atro-
pisomer stereochemistry. The isomerism at 130 °C appeared
better than that at either the more vigorous conditions of
140 °C or the slower conditions at 110 °C and was enlisted for
our preparative studies. The assignment of the atropisomer
structure and stereochemistry followed from resubjection of
(P,M,M)-6 to the thermal conditions with reequilibration to the
In the conduct of the studies, two significant competitive
reactions were observed at the temperatures required for
atropisomerism, and only isomerism of the DE, not the CD,
ring system was observed.18 This latter observation is consistent
with expectations based on our prior studies of preferential DE
versus CD atropisomerism with the isolated ring systems.10-13
The two main competitive reactions were established to be retro
aldol ring cleavage preferentially of the CD versus DE ring
system which we will refer to as more polar products in the
following discussion and main chain succinimide formation (less
6
polar products). Protection of the C3 alcohol eliminated the
most competitive retro aldol ring cleavage reaction (more polar
products) isolating its most susceptible occurrence to the CD
ring system, and dehydration of the Asn carboxamide to a nitrile
prevented competitive succinimide formation (less polar prod-
ucts). That the most competitive retro aldol reaction is confined
to the CD ring system was established with the observation that
it is only detected with 3 and not 5 upon modest thermal
treatment in nonpolar solvents (o-Cl2C6H4, e140 °C). Notably,
2
5 still possesses a C3 free alcohol and the potential for retro
aldol DE ring cleavage, and this was not observed under the
thermal conditions most successful for atropisomerism (o-
Cl2C6H4, 130-140 °C) but was observed in more polar solvents
(DMF, 140 °C). An identical preferential sensitivity of the CD
versus DE ring system to retro aldol ring cleavage has been
1:1 mixture of atropisomers and 2D H-1H NMR (CD3OD,
313 K, 600 MHz). The latter revealed that the characteristic
1
(18) Although the AB ring system is capable of atropisomerism under
the thermal conditions examined, only the natural (M)-atropisomer was
detected, reflecting its thermodynamic preference, see: Evans, D. A.;
Dinsmore, C. J.; Ratz, A. M.; Evrard, D. A.; Barrow, J. C. J. Am. Chem.
Soc. 1997, 119, 3417.
nOe cross-peaks of the CD ring system (M)-stereochemistry
were unperturbed [C5a6/C2 (m) and C5a6/C3 (s)], while those
6
6
of the DE ring system were now diagnostic of the unnatural

Atropisomerism of AglucoVancomycin DeriVatiVes
J. Am. Chem. Soc., Vol. 120, No. 35, 1998 8923
Scheme 3
Scheme 4
Finally, initial studies conducted with 3 revealed that atro-
pisomerism in polar solvents including DMF (140 °C, 3 h, 40%
or 110 °C, 17 h, 41%) provided extensive conversion to the
polar products including the CD retro aldol ring cleavage
product20 (45-50%). At the lower temperature of 110 °C
(DMF), only one prominent retro aldol product arising from
cleavage of the CD ring system was observed. At 140 °C, two
retro aldol products were detected which most likely reflect both
CD and DE ring cleavage with the former appearing in less
than 30 min and the latter arising more slowly (2 h).20 The
competitive retro aldol reactions representing more polar
products were minimized in o-Cl2C6H4 (6-10%), and isomer-
ization at 110 °C (17 h, 60%) and 140 °C (3 h, 63%) under
conditions that also minimize succinimide formation provided
respectable recoveries of a 1:1 mixture of 3 atropisomers. Under
these conditions, 20% (110 °C) and 25% (140 °C) of the
succinimide atropisomers were also isolated. As in observations
made with 5 versus 4, the competitive generation of the
(P)-stereochemistry [C5b2/C32 (s) and C6b2/C4b4 (m)] rather than
the natural (M)-stereochemistry [C5a2/C32 (m), C5a2/C22 (w), and
4
C6b2/C4b (w)].
In contrast, 4 incorporating the free Asn carboxamide and
2
6
both the C3 and C3 alcohols protected as TBDMS ethers
exhibited a more complex atropisomerism complicated by
competitive succinimide formation. The competitive succin-
imide formation in o-Cl2C6H4 was minimal at the lower
temperature of 110 °C (10 h, 80% 4 and 5% 7) and it became
more prominent at 130 °C (1.8 h, 74% 4 and 14% 7). The two
atropisomers of the succinimide 7 were preparatively isolated
in superb yield by extending the thermal treatment (125 °C, 38
h, 82%) to provide a 2:1 mixture of (M)/(P) DE atropisomers
(Scheme 3). Although complicated by the competitive succin-
imide formation, the rate of atropisomerism by the carboxamide
4 appears to be slightly faster than that of the nitrile 6, and
analogous observations have been made with the isolated DE
ring system.11 The identity of the DE (P)-atropisomer was
3
succinimide appears more significant with 3, suggesting the C2
OTBDMS protection found in 4 may either slow its formation
or facilitate atropisomerism. Thus, useful interconversions are
possible even with 3, although those of 6 are much cleaner and
easier to predictably control.
Conversion of (M,M,M)-6 to Aglucovancomycin. The
preparation of natural aglucovancomycin from (M,M,M)-6 was
also examined in efforts to address two key issues associated
with a projected vancomycin aglycon total synthesis. The first
was whether conditions could be developed that would permit
hydration or hydrolysis of an Asn residue nitrile. The second
and far more significant issue was whether conditions could be
developed for aryl methyl ether cleavage. Thus, treatment of
(M,M,M)-6 with K2CO3-H2O2 (8 equiv/40 equiv) under modi-
fied Katritzky conditions21 (DMSO-H2O, 25 °C, 3.5 h) cleanly
provided the carboxamide 4 (73%) without cleavage of the
TBMDS ethers or C-terminal methyl ester (Scheme 4). If the
reaction times were extended, competitive TBDMS ether
cleavage was observed. Deprotection of the TBDMS ethers
(60 equiv of Bu4NF, 60 equiv of HOAc, THF, 25 °C, 29 h,
81%) followed by treatment of 3 with AlBr3 (40 equiv, EtSH,
25 °C, 5 h) provided (M,M,M)-2 (50%). Significantly, the final
conversion of 3 to 2 involves the cleavage of the four methyl
ethers and the methyl ester and N-BOC deprotection. The
sample of (M,M,M)-2 obtained by this sequence was identical
in all respects with authentic aglucovancomycin (2) derived from
the degradation of 1 (1H NMR, 1:1 comixture 1H NMR, HPLC
Rt and 1:1 comixture HPLC).
1
established by 2D H-1H NMR, and the succinimide atropi-
somers were easily distinguishable by additional changes in the
characteristic Asn CH2CONH2 resonances. Resubjecting 7 to
thermal atropisomerism revealed that the succinimide isomerizes
much more slowly than the parent compound 4 with little or
no isomerization observed at 130 °C and only very slow
isomerization at 140 °C (o-Cl2C6H4). At these temperatures
and reaction times, competitive decomposition was observed,
and only an approximate atropisomerism rate was established
(k ) 0.07 h-1, 140 °C). This slow rate of interconversion of
the succinimide atropisomers established that unnatural (P,M,M)-7
is most likely derived from a sequence involving first atropi-
somerism of 4 and subsequent succinimide formation and is
consistent with the lower temperature studies where atropisom-
erism is kinetically faster than competitive succinimide forma-
tion.
Similarly, 5 underwent an analogous atropisomerism of the
DE ring system with no evidence of CD isomerism. However,
the atropisomerism was complicated by competitive succinimide
formation in o-Cl2C6H4 analogous to observations made with
4, and under these conditions and up to temperatures of 140 °C
(2 h), no DE retro aldol ring cleavage was observed. Only at
110 °C where succinimide formation is slow could an ap-
proximate rate of DE atropisomerism be established. At the
higher temperatures, competitive succinimide formation pre-
cluded accurate assessments of the isomerization rates (Scheme
3). Interestingly and analogous to observations made with 3,
Conversion of (P,M,M)-6 to (P,M,M)-2. The Vancomycin
Aglycon with the Unnatural DE Atropisomer Stereochem-
istry. Analogous to the efforts described above, (P,M,M)-6
obtained by preparative thermal atropisomerism of (M,M,M)-6
1
(19) The product aldehyde exhibited a characteristic H NMR signal at
2
competitive succinimide formation appears to be slowed by C3
δ 9.85 (DMF-d7) and appeared at the following rates: 0.5 h (e5%), 1.5 h
(e10%), 4 h (ca. 50%). Extensive conversion to both polar and nonpolar
products provided a complex mixture.
OTBDMS protection with 4, and it was observed more
prominently with both 5 and 3. Conducting analogous studies
in DMF (140 °C) led to the observation of a slow competitive
retro aldol cleavage of the DE ring system.19
(20) The diagnostic -CHO signals in the 1H NMR (CD3OD, 313 K, 600
MHz) were observed at δ 9.87 (CD retro aldol) and δ 9.85 (DE retro aldol).
(21) Katritzky, A. R.; Pilarski, B.; Urogdi, L. Synthesis 1989, 949.

8924 J. Am. Chem. Soc., Vol. 120, No. 35, 1998
Boger et al.
Table 1. Antimicrobial Activity, MIC (µg/mL)
Staphylococcus aureus
Enterococcus faecium
Enterococcus faecium
(vancomycin-resistant)
compd
(ATCC 25923)
(ATCC 35667)
vancomycin (1)
(M,M,M)-2
(P,M,M)-2
(M,M,M)-3
(P,M,M)-3
(M,M,M)-4
(P,M,M)-4
(M,M,M)-6
(P,M,M)-6
(M,M,M)-9
1.25
0.7
14
2.5
2.8
28
250
180
>1800
>1700
>1400
>1900
>1800
>2100
>1400
45
13
110
>44
>120
>56
>130
>90
0.7
>44
>120
>56
>130
>90
1.4
Scheme 5
dechlorovancomycin22 which lacks the DE aryl chloride or the
related orienticins which exhibit essentially equipotent activity
unaffected by the lack of a DE aryl chloride. Thus, while the
presence of the chloride in the natural (M)-atropisomer does
not appear to significantly potentiate the antimicrobial proper-
ties,22 its presence in the unnatural (P)-atropisomer substantially
diminishes the antimicrobial potency.
In addition, (M,M,M)-3 exhibited surprisingly potent activity
despite N-acylation of the terminal N-methyl-Leu residue.
Although it was 10-20× less active than 2, the residual activity
is significant, given its critical role in binding N-acyl-D-Ala-D-
Ala, and similar observations have been implicated in prior
studies.4 Analogous to the observations made with 2, (M,M,M)-3
possessing the natural DE atropisomer stereochemistry was more
potent than (P,M,M)-3.
Significantly, permethylation of the four phenols with (M,M,M)-
9, which was obtained by N-BOC deprotection of (M,M,M)-3
(HCl-EtOAc, 86%, eq 1), provided an aglucovancomycin
was converted to (P,M,M)-2, aglucovancomycin possessing the
unnatural DE atropisomer stereochemistry. Thus, conversion
of the nitrile to the corresponding carboxamide under the
modified Katritzky conditions of K2CO3-H2O2 (85%) followed
by the cleavage of the TBDMS ethers (Bu4NF-HOAc, 71%)
and final exhaustive deprotection of the four methyl ethers,
methyl ester, and N-BOC group (AlBr3, EtSH, 51%) provided
(P,M,M)-2 (Scheme 5).
Antimicrobial Activity. The antimicrobial activities of
aglucovancomycin (2) and its unnatural (P,M,M)-atropisomer
were examined along with those of a number of the semisyn-
thetic derivatives and vancomycin itself against two vancomy-
cin-sensitive and one vancomycin-resistant bacterial strains
(Table 1). Aglucovancomycin (2) was found to be essentially
equipotent with vancomycin (1) against the three bacterial strains
examined, consistent with prior observations.4 Unnatural
(P,M,M)-2 was found to be 10× less potent than natural
(M,M,M)-2 and showed no enhanced selectivity against the
vancomycin-resistant Enterococcus faecium. This substantially
diminished activity of the unnatural DE atropisomer is an
interesting and important contrast to the properties of mono-
derivative that was at least as active if not slightly more potent
than 1 or 2 and slightly more effective against the vancomycin-
resistant bacteria. Such derivatives of the natural products have
not, to our knowledge, been disclosed or explored and may
prove to be an important class to pursue. Notably, the
C-terminal carboxylate of 9 is functionalized as a methyl ester,
and this derivatization site constitutes one of the more successful
used for modulating physiochemical and transport/absorption
properties without negatively impacting in vitro antimicrobial
activity.4 Thus, the assessment of the methyl ester with 9 (and
3, 4, and 6) versus the use of the free carboxylic acid would
not be expected to alter the in vitro antimicrobial properties
against the three bacterial strains examined.
(22) The removal of the vancomycin C6a2-Cl provides an antibiotic that
exhibits 70% of the antimicrobial activity of vancomycin and binds
representative di- and tripeptides slightly less effectively than vancomycin,
see: Harris, C. M.; Kannan, R.; Kopecka, H.; Harris, T. M. J. Am. Chem.
Soc. 1985, 107, 6652. In addition, the C6a2-Cl of the related antibiotic
eremomycin fits into a pocket at the antibiotic dimerization interface and
its removal reduces Kdim by a factor of 80, see: Mackay, J. P.; Gerhard,
U.; Beauregard, D. A.; Maplestone, R. A.; Williams, D. H. J. Am. Chem.
Soc. 1994, 116, 4573.

Atropisomerism of AglucoVancomycin DeriVatiVes
J. Am. Chem. Soc., Vol. 120, No. 35, 1998 8925
Table 2
compd
6
Experimental Section
conditions
t (h)
(M,M,M)/(P,M,M)
Degradation of Vancomycin. (M,M,M)-3. Vancomycin‚HCl (1.03
g, 0.71 mmol) was treated with CF3CO2H (25 mL), and the resulting
solution was stirred at 50 °C for 7.5 h. The mixture was concentrated
in vacuo, and the residue was triturated with EtOAc-hexane (1:1, 100
mL). The resulting precipitate was collected by filtration, washed with
EtOAc (50 mL), and dried under vacuum to afford aglucovancomycin
(2) as a crude residue. A sample of crude 2 (200 mg) prepared as
described above was purified by semipreparative reverse phase HPLC
(C18, 2.5 × 10 cm, CH3CN-0.07% TFA/H2O, 17:83, 10 mL/min, Rt
) 18 min) to afford pure aglucovancomycin (2, 88 mg) as a white
film.
A solution of the crude aglucovancomycin (2, from 1.03 g 1) in
dioxane-H2O (2:1, 15 mL) was treated sequentially with NaHCO3 (198
mg, 2.36 mmol) and BOC2O (388 mg, 1.78 mmol) at 0 °C. The
reaction mixture was slowly warmed to 25 °C and was stirred for 2 h.
The reaction mixture was cooled to 0 °C, quenched by the addition of
HOAc (0.135 mL, 2.36 mmol), concentrated in vacuo, and the residue
was triturated with EtOAc-hexane (1:1, 50 mL). The resulting
precipitate was collected by filtration, washed with EtOAc-hexane (1:
1, 50 mL), and dried under vacuum to afford N-BOC-aglucovancomycin
as a crude residue.
A solution of crude N-BOC-aglucovancomycin in DMF (10 mL)
was treated sequentially with NaHCO3 (198 mg, 2.36 mmol) and CH3I
(0.244 mL, 3.93 mmol) at 0 °C under Ar. The reaction mixture was
slowly warmed to 25 °C and stirred for 2 h. The reaction mixture was
cooled to 0 °C and quenched by the addition of H2O (200 mL) followed
by 10% aqueous HCl (5 mL). The resulting precipitate was collected
by filtration, washed with H2O (20 mL), and dried under vacuum to
afford N-BOC-aglucovancomycin methyl ester as a crude residue.
A solution of crude N-BOC-aglucovancomycin methyl ester in DMF
(10 mL) was treated sequentially with K2CO3 (342 mg, 2.47 mmol),
n-Bu4NI (114 mg, 0.31 mmol), and CH3I (0.386 mL, 6.2 mmol). The
reaction mixture was slowly warmed to 25 °C and stirred. After 3.5
h, additional K2CO3 (171 mg, 1.23 mmol) and CH3I (0.386 mL, 6.2
mmol) were added. The reaction mixture was stirred for 2.5 h,
quenched by the addition of 1% aqueous HCl (70 mL) at 0 °C, and
extracted with EtOAc (2 × 140 mL). The combined organic layers
were washed with H2O (30 mL) and saturated aqueous NaCl (30 mL),
dried (Na2SO4), and concentrated in vacuo. Chromatography (SiO2,
2.5 × 20 cm, 5-7% CH3OH-CHCl3 gradient elution) afforded
(M,M,M)-323 (340 mg, 929 mg theoretical, 37%) as a white solid.
(M,M,M)-4. A solution of (M,M,M)-3 (189 mg, 144 µmol) in
anhydrous CH3CN (1.9 mL) was treated with CF3CONMeTBDMS (1.7
mL, 7.2 mmol) under Ar, and the resulting mixture was stirred at 50
°C for 9.5 h. The reaction mixture was poured into EtOAc-20%
aqueous citric acid (3:1, 40 mL) and stirred at 25 °C for 12 h. The
two layers were separated, and the aqueous phase was extracted with
EtOAc (2 × 40 mL). The combined organic layers were washed with
saturated aqueous NaHCO3 (15 mL) and saturated aqueous NaCl (15
mL), dried (Na2SO4), and concentrated in vacuo. Chromatography
(SiO2, 2.5 × 20 cm, 50-85% EtOAc-hexane and then 10% CH3OH-
CHCl3 gradient elution) afforded (M,M,M)-423 as a white film (147
mg, 222 mg theoretical, 66%) and (M,M,M)-523 as a white film (20.1
mg, 206 mg theoretical, 10%).
o-C6H4Cl2, 110 °C
0.62
1.62
3.77
7.77
13.77
0.50
1.00
1.50
2.50
3.50
4.50
5.50
0.28
0.53
0.87
1.20
1.53
1.00
2.00
5.00
9.00
0.67
1.33
2.50
2.98
96:4
91:9
86:14
75:25
69:31
97:3
6
6
o-C6H4Cl2, 130 °C
o-C6H4Cl2, 140 °C
92:8
88:12
79:21
69:31
62:38
58:42
90:10
83:17
72:28
65:35
50:50
94:6
4
4
o-C6H4Cl2, 110 °C
o-C6H4Cl2, 130 °C
93:7
87:13
82:15
88:12
78:22
64:36
60:40
solution of (M,M,M)-6 (2 mg, 1.3 µmol) in o-dichlorobenzene (0.1 mL)
saturated with Ar was warmed at 130 °C for 5.5 h. Periodically,
o-dichlorobenzene was evaporated under a flow of N2 and replaced
with CD3OD for the determination of the ratio of (M,M,M)-6 and
(P,M,M)-6 by 1H NMR (313 K, 600 MHz), Table 2. Removal of CD3-
OD and its replacement with o-dichlorobenzene allowed further heating
of (M,M,M)-6 and its emerging isomer (P,M,M)-6 with the unnatural
DE atropisomer stereochemistry. Upon completion of the kinetic
investigation, evaporation of o-dichlorobenzene and PTLC (SiO2, 3%
CH3OH-CHCl3) afforded the recovery of the starting material (M,M,M)-
623 (1.0 mg, 2 mg theoretical, 49%) and (P,M,M)-623 (0.7 mg, 2 mg
theoretical, 36%) as white films.
Analogous experiments enlisting (M,M,M)-6 were conducted at 110
°C and 140 °C in o-dichlorobenzene (Table 2).
Thermal Interconversion of the Atropisomers (M,M,M)-4 and
(P,M,M)-4. This was conducted following the general procedure. The
atropisomerism of 4 incorporating the free Asn carboxamide residue
exhibited a more complex atropisomerism, resulting from a competitive
succinimide formation. Investigated temperatures and heating time
frames were selected where the amount of succinimide formation
remains minimal (Table 2).
Formation of Succinimides (M,M,M)-7 and (P,M,M)-7. A solution
of (M,M,M)-4 (5 mg, 3.2 µmol) in o-dichlorobenzene (0.3 mL) saturated
with Ar was warmed at 125 °C for 38 h. Evaporation of the solvent
and PTLC (SiO2, 3% CH3OH-CHCl3) afforded (M,M,M)-723 (2.7 mg,
4.9 mg theoretical, 55%) and (P,M,M)-723 (1.3 mg, 4.9 mg theoretical,
27%) as white solids.
Thermal Interconversion of the Atropisomers (M,M,M)-5 and
(P,M,M)-5. This was conducted following the general procedure. The
atropisomerism of 5 in o-dichlorobenzene was complicated by competi-
tive succinimide formation analogous to the observation made with 4.
Only at 110 °C where succinimide formation is slow could an
approximate rate of DE atropisomerism be established.
Formation of Succinimides (M,M,M)-8 and (P,M,M)-8. A solution
of (M,M,M)-5 (2.5 mg, 1.8 µmol) in o-dichlorobenzene (0.2 mL)
saturated with Ar was warmed at 140 °C for 2 h. Evaporation of the
solvent and PTLC (SiO2, 2% CH3OH-CHCl3) afforded (M,M,M)-823
(0.8 mg, 2.5 mg theoretical, 33%) and (P,M,M)-823 (0.4 mg, 2.59 mg
theoretical, 17%) as white solids.
(M,M,M)-6. A solution of (M,M,M)-4 (137 mg, 88.6 µmol) in
anhydrous CH2Cl2 (3 mL) at 0 °C was treated sequentially with pyridine
(42.9 µL, 530 µmol) and TFAA (50.0 µL, 354 µmol). The resulting
mixture was stirred at 0 °C for 10 min, diluted with EtOAc (20 mL),
and quenched by the addition of 1% aqueous HCl (2 mL). The two
layers were separated, and the aqueous phase was extracted with EtOAc
(2 × 5 mL). The combined organic layers were washed with H2O (5
mL) and saturated aqueous NaCl (10 mL), dried (Na2SO4), and
concentrated in vacuo. Chromatography (SiO2, 1.5 × 15 cm, 50-
60% EtOAc-hexane gradient elution) afforded (M,M,M)-623 (111 mg,
135 mg theoretical, 82%) as a white film.
Thermal Interconversion of the Atropisomers (M,M,M)-3 and
(P,M,M)-3. The atropisomerism of 3 in o-dichlorobenzene was
complicated by competitive formation of succinimide and other
unidentified compounds which precluded assessment of the isomer-
ization rates.
Thermal Atropisomerism Studies. General Procedure. Thermal
Interconversion of the Atropisomers (M,M,M)-6 and (P,M,M)-6. A
(23) Full characterization data and diagnostic and strong 2D 1H-1H nOe’s
are provided in the Supporting Information.

8926 J. Am. Chem. Soc., Vol. 120, No. 35, 1998
Boger et al.
In DMF, upon heating at 110-140 °C, rapid formation of a
compound more polar than (M,M,M)-3 was observed and shown to
arise from retro aldol CD ring cleavage.
By using the same procedure, we obtained (P,M,M)-323 (3.4 mg,
4.8 mg theoretical, 71%) as a white film from (P,M,M)-4 (5.6 mg, 3.6
µmol).
Retro Aldol CD Ring-Cleaved 3. A solution of (M,M,M)-3 (5.0
mg, 3.8 µmol) in DMF (0.2 mL) saturated with Ar was warmed at 140
°C for 0.5 h. Evaporation of the solvent and PTLC (SiO2, 5% CH3-
OH-CHCl3) afforded the retro aldol CD ring-cleaved product23 (0.8
mg, 2.5 mg theoretical, 33%) as a white solid.
Conversion of (M,M,M)-6 to Aglucovancomycin [(M,M,M)-2] and
(P,M,M)-6 to (P,M,M)-2. (P,M,M)-6. A solution of (M,M,M)-6 (39.9
mg, 2.62 µmol) in o-dichlorobenzene (3 mL) was stirred at 120 °C for
16.5 h. The reaction mixture was concentrated in vacuo. PTLC (SiO2,
40% EtOAc-CHCl3) afforded the recovery of starting material
(M,M,M)-6 (17.5 mg, 39.9 mg theoretical, 44%) and (P,M,M)-623 (15.4
mg, 39.9 mg theoretical, 39%) as a white film.
Preparation of Aglucovancomycin 2 from 3. A vial charged with
(M,M,M)-3 (5.0 mg, 3.8 µmol) was treated with AlBr3 (38.1 mg, 0.143
mmol) in EtSH (0.2 mL) under Ar. The resulting mixture was stirred
at 25 °C for 5 h, diluted with CHCl3 (0.5 mL), cooled to 0 °C, and
quenched by the addition of CH3OH (0.1 mL). The solvent was
removed by a stream of N2. The crude mixture was purified by PTLC
(SiO2, 55% CH3OH-EtOAc) and semipreparative reverse phase HPLC
(CH3CN-0.07% TFA/H2O, 17:83, 10 mL/min, Rt ) 18 min) to afford
(M,M,M)-223 (2.4 mg, 4.8 mg theoretical, 50%) as a white film.
By using the same procedure, we obtained (P,M,M)-223 (2.0 mg,
3.9 mg theoretical, 51%) as a white film from (P,M,M)-3 (4.1 mg,
3.1 µmol).
Preparation of 4 from 6. A solution of (M,M,M)-6 (3.5 mg, 2.3
µmol) in DMSO (0.6 mL) at 25 °C was treated sequentially with H2O2
(10 µL, 98 µmol) and 10% aqueous K2CO3 (25.0 µL, 20 µmol). The
resulting mixture was stirred at 25 °C for 3.5 h, diluted with EtOAc
(15 mL), and quenched by the addition of 0.1% aqueous HCl (2 mL).
The two layers were separated, and the aqueous phase was extracted
with EtOAc (2 × 10 mL). The combined organic layers were washed
with H2O (10 mL) and saturated aqueous NaCl (10 mL), dried (Na2-
SO4), and concentrated in vacuo. PTLC (SiO2, 6% CH3OH-CHCl3)
afforded (M,M,M)-423 (2.6 mg, 3.5 mg theoretical, 73%) as a white
film.
(M,M,M)-9. A solution of (M,M,M)-3 (30.3 mg, 23.1 µmol) in
anhydrous dioxane (1 mL) was treated with 4 N HCl-EtOAc (2 mL)
under Ar, and the resulting mixture was stirred at 25 °C for 0.5 h. The
reaction mixture was concentrated in vacuo, and the residue was
triturated with EtOAc (5 mL) to give (M,M,M)-923 (24.9 mg, 28.8 mg
theoretical, 86%) as a white solid.
Acknowledgment. We gratefully acknowledge the financial
support of the National Institutes of Health (CA41101), the
Skaggs Institute for Chemical Biology, the award of a sabbatical
leave for S.M. (Japan Tobacco), the award of a Swiss National
Foundation postdoctoral fellowship and a Ciba-Geigy Jubila¨ums
Stiftung (O.L.), the award of a National Institutes of Health
postdoctoral fellowship (R.T.B., CA71102), and the award of
a NSF predoctoral fellowship (S.L.C., GER-9253922).
By using the same procedure, we obtained (P,M,M)-423 (2.5 mg,
2.9 mg theoretical, 85%) as a white film from (P,M,M)-6 (2.9 mg, 1.9
µmol).
Preparation of 3 from 4. A solution of (M,M,M)-4 (1.6 mg, 1.04
µmol) in THF (50 µL) was treated with n-Bu4NF-HOAc (1:1, 1 M
solution in THF, 31 µL, 31 µmol) at 25 °C, and the resulting mixture
was stirred at 25 °C. After 11 h, additional n-Bu4NF-HOAc (1:1, 1
M solution in THF, 31 µL, 31 µmol) was added. The reaction mixture
was stirred for 18 h, quenched by the addition of H2O (2 mL) at 0 °C,
and extracted with EtOAc (2 × 15 mL). The combined organic layers
were washed with H2O (2 × 15 mL) and saturated aqueous NaCl (20
mL), dried (Na2SO4), and concentrated in vacuo. PTLC (SiO2, 9%
CH3OH-CHCl3) afforded (M,M,M)-323 (1.1 mg, 1.4 mg theoretical,
81%) as a white film.
Supporting Information Available: Full characterization
of (M,M,M)- and (P,M,M)-2-8, (M,M,M)-9, and diagnostic or
strong 2D 1H-1H nOe’s are provided (7 pages, print/PDF). See
any current masthead page for ordering information and Web
access instructions.
JA981928I