Chemical tailoring of teicoplanin with site-selective reactions

Article abstract of DOI:10.1021/ja4038998

Semisynthesis of natural product derivatives combines the power of fermentation with orthogonal chemical reactions. Yet, chemical modification of complex structures represents an unmet challenge, as poor selectivity often undermines efficiency. The complex antibiotic teicoplanin eradicates bacterial infections. However, as resistance emerges, the demand for improved analogues grows. We have discovered chemical reactions that achieve site-selective alteration of teicoplanin. Utilizing peptide-based additives that alter reaction selectivities, certain bromo-teicoplanins are accessible. These new compounds are also scaffolds for selective cross-coupling reactions, enabling further molecular diversification. These studies enable two-step access to glycopeptide analogues not available through either biosynthesis or rapid total chemical synthesis alone. The new compounds exhibit a spectrum of activities, revealing that selective chemical alteration of teicoplanin may lead to analogues with attenuated or enhanced antibacterial properties, in particular against vancomycin- and teicoplanin-resistant strains.

Full text of DOI:10.1021/ja4038998

Article
pubs.acs.org/JACS
Chemical Tailoring of Teicoplanin with Site-Selective Reactions
Tejas P. Pathak and Scott J. Miller*
Department of Chemistry, Yale University, P.O. Box 208107, New Haven, Connecticut 06520-8107, United States
S
* Supporting Information
ABSTRACT: Semisynthesis of natural product derivatives
combines the power of fermentation with orthogonal chemical
reactions. Yet, chemical modification of complex structures
represents an unmet challenge, as poor selectivity often
undermines efficiency. The complex antibiotic teicoplanin
eradicates bacterial infections. However, as resistance emerges,
the demand for improved analogues grows. We have discovered
chemical reactions that achieve site-selective alteration of
teicoplanin. Utilizing peptide-based additives that alter reaction
selectivities, certain bromo-teicoplanins are accessible. These
new compounds are also scaffolds for selective cross-coupling
reactions, enabling further molecular diversification. These studies enable two-step access to glycopeptide analogues not available
through either biosynthesis or rapid total chemical synthesis alone. The new compounds exhibit a spectrum of activities, revealing
that selective chemical alteration of teicoplanin may lead to analogues with attenuated or enhanced antibacterial properties, in
particular against vancomycin- and teicoplanin-resistant strains.
Enterococcus (VRE) and vancomycin-resistant Staphylococcus
aureus (VRSA)), the pursuit of more effective antibiotics
assumes even greater significance.5 Teicoplanin (2), a
structurally more intricate glycopeptide, exhibits enhanced
potency and a different spectrum of activities relative to
vancomycin.6−8 In fact, 2 offers some promise in the context of
resistant strains with analogues advancing in clinical trials.9
Synthesis of analogues of 1 or 2 for biological evaluation
continues to fuel medicinal discoveries.10,11 Total syn-
thesis11−19 and semisynthesis3,20−24 (employing enzymatic
catalysts, synthetic catalysts, or chemical reagents) all provide
important avenues for research. In the case of 2, a number of
particular challenges exist. The structure is substantially more
complex than that of 1, rendering analogue access by total
synthesis more difficult. To date, while a total synthesis of the
teicoplanin aglycon has been reported,14,16,17 the complete total
chemical synthesis has not yet been reported. Moreover, the
enhanced complexity renders direct and selective functionaliza-
tion of 2 more challenging than the related goal of direct
functionalization of 1. Also, whereas 1 is readily available in
homogeneous form through a fermentation process,25 2 is
obtained as a complex mixture.26 Further purification is
therefore necessary to obtain a homogeneous starting material
for study of selective chemical reactions.
INTRODUCTION
■
Complex glycopeptides are important compounds in the
ongoing campaign against antibiotic resistance.1−4 Vancomycin
(1, Figure 1) in particular has been studied intensively due to
its efficacy against resistant strains of bacteria.1−4 Even so, as
vancomycin-resistance has emerged (e.g., vancomycin-resistant
Our laboratory has been pursuing a program that tests the
capacity of small-molecule catalysts to functionalize complex
molecules in a selective manner.27−29 This chemistry-based
approach offers the promise of enhancing access to analogues
of complex natural products and also presents challenges in the
Figure 1. Structures of glycopeptide antibiotics vancomycin (1),
teicoplanin (2), and teicoplanin A2-2 (3).
Received: April 19, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja4038998 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
study of catalysis in environments of extreme molecular
complexity. One tenet of these studies involves documentation
of the intrinsic reactivity of the substrate toward a reagent of
interest in the absence of a catalyst or promoter. Then, the goal
becomes the identification of catalysts, conditions, or reagents
that alter the intrinsic reactivity to produce alternative
compounds.
structure of teicoplanin presents the additional challenge of
tuning the site of bromination to either the 5,7-biaryl region of
the structure (2, Figure 1, red) or the 1,3-biaryl-ether region (2,
Figure 1, blue). The intrinsic reactivities of these moieties,
relative to one another, were not completely clear at the outset.
Thus, one of the key goals of this study was to evaluate
catalysts/conditions that could provide site-selective bromina-
tion of either biaryl ring system. To initiate these studies, it was
necessary to undertake the purification of teicoplanin.8,22 We
targeted teicoplanin A2-2 (3, Figure 1) from the readily
available mixture of teicoplanins, which is a composite of
approximately six to nine molecular forms of teicoplanin
(Figure S1 in Supporting Information [SI]). The assignment of
the aromatic region of the NMR spectrum of 3 with modern
NMR techniques was also essential at the outset of these
studies22,33−35 as a critical step for analysis and determination
of the site of bromination within new products (Figures S1−S3
in SI).
Our recent study of the site-selective bromination of
vancomycin provides a case in point (Figure 2).29 We evaluated
RESULTS AND DISCUSSION
■
Studies of the site-selective bromination of 3 began with an
evaluation of its reaction in the presence of various quantities of
N-bromophthalimide (NBP). As shown in Figure 2A, when 3
was dissolved in MeOH/H2O (1:1), and 1.1 equiv of NBP was
employed, a major product was observed, with unreacted 3 also
prominent in the HPLC trace (Figure 3A(a)). When multiple
equivalents were employed, a highly complex mixture of
products was observed (Figure S4 in SI). The reaction with 1.1
equiv allowed for isolation and purification of the major
brominated species (7; Figure 3B). On preparative scale, 50.0
mg of 3 could be converted to 10.0 mg of analytically pure 7
with 6.0 mg of recovered 3 in a single operation. LC−MS
analysis revealed that the new compound was a monobromi-
nated form of teicoplanin A2-2. Figure 2C shows one of the
most revealing pieces of NMR data, an overlay of a diagnostic
region of the HSQC spectrum for both 3 and 7. Most notably,
the cross peak that we had assigned to the ring 7f (C−H)
correlation is absent in 7. At the same time, there is excellent
overlay of the overwhelming majority of the other peaks,
suggesting a minimum structural alteration. The cross peak for
the 7d and 5b (C−H) correlation has shifted slightly, as
highlighted in Figure 2C, suggesting substitution within the 5,7-
biaryl substructure. Additional data also support the assignment
of 7 as the ring 7f-Br variant of teicoplanin A2-2 (Figures S6−S8
in SI).
We then turned our attention to perturbation of the inherent
site-selectivity exhibited by 3 under our initial conditions. We
initially targeted catalysts like 4, designed in our earlier
studies29 to mimic the binding of 1 and 2 to their biological
target36,37 but outfitted with functional groups that might
accelerate brominaton.38 When 3 was exposed to bromination
conditions in the presence of peptide 4, we observed that
indeed bromination occurs, but the analytical HPLC trace
exhibits broad peaks, as shown in Figure 3A(b). The peak shape
renders analysis of reaction mixture and isolation of pure
materials difficult. We interpreted these features as the
formation of robust (i.e., too robust) 3−peptide complexes.
This assertion is consistent with the known, very high affinity of
3 for DAla-DAla-based peptides (Ka = 1.6 × 106 for Ac-Lys-
DAla-DAla at pH = 5.0).39 To remedy this issue in the context of
peptide-mediated brominations, we elected to replace one of
the DAla units with a DLeu. This strategy was projected to
attenuate the binding affinity of substrate and peptide, in accord
Figure 2. Summary of studies for the bromination of 1. HPLC traces
are shown for (a) the control reaction of 1 with no catalyst, (b) the
reaction conducted in the presence of peptide 4, and (c) the reaction
conducted in the presence of guanidine.
selective halogenation as a way of mimicking the chemistry of
the tailoring halogenase enzymes with small-molecule cata-
lysts.30,31 The selective introduction of halide into the
glycopeptides offers a unique way to tune the properties of
the antibiotics. In addition, controlled introduction of halide
offers an opportunity to examine further chemical functional-
ization of these scaffolds, for example with metal-catalyzed
cross-coupling reactions.32 As shown in Figure 2A, when native
vancomycin (with no protecting groups) is exposed to N-
bromophthalimide (NBP), a very sluggish reaction takes place
in which one observes, after 2 h, mostly unreacted 1, but also
small, nearly equimolar amounts of several products. However,
in the presence of a designed peptide-based catalyst (4, Figure
2B), substantial rate acceleration is observed, the starting
material is completely consumed, and selective formation of
monobromo-vancomycin derivative 6 is observed. Alternatively,
as shown in Figure 2C, when peptide 4 is replaced with
guanidine as an additive, the selectivity is reversed, and
alternative monobromo-vancomycin derivative 5 is obtained.
Given the heightened biological activity and higher level of
molecular complexity of teicoplanin compared to that of
vancomycin, we sought to establish whether site-selective
halogenation of teicoplanin might be possible chemically. The
B
dx.doi.org/10.1021/ja4038998 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 3. (A) Effect of peptides on the site-selectivity of teicoplanin bromination compared to the control reaction (data collected on HPLC). (B)
Structure of brominated analogues 7 and 9. (C) Overlay of HSQC spectrum of 7 (gray) with 3 (blue).
with previous binding studies of DXaa-DXaa peptides with
glycopeptides like 1.37,40 Indeed, this strategy proved effective.
The corresponding bromination of 3 with Boc-Asn(Me)2-DLeu-
DAla-OH (8) as a promoter of the reaction produces a reaction
mixture that was readily analyzed by HPLC/LC−MS (Figure
3A(c)). Strikingly, peptide 8 diverts the reaction to give a new
monobrominated teicoplanin, which was not observed in
significant quantities in the absence of 8. As detailed below,
the structure of the new, 8-dependent monobrominated
teicoplanin may be assigned as the ring 3b-Br analogue 9.
The capacity of peptide 8 to redirect the site of bromination
away from the intrinsically more reactive 5,7-ring system to the
less reactive 1,3-ring system is a manifestation of nonenzymatic
control of site-selectivity in a highly complex molecular
environment. The reaction specificity is also highly dependent
on peptide structure and stereochemistry. Alteration of the
configuration of the D-Leu residue of the peptide to the L-
configuration, as in 11 (Boc-Asn(Me)2-Leu-DAla-OH), pro-
duced a very different result (Figure 3A(d)). In this case, the
major product was reverted to 7f-Br compound 7. Presumably,
these results are due to the reduced binding affinity of the
stereochemically mismatched peptide 11 to 3 during the
bromine transfer reaction.40 On preparative scale, in the
presence of peptide 12 (Figure 5), 160.0 mg of 3 provided
47.0 mg of analytically pure 9 (with 35.0 mg of recovered 3 and
16.0 mg of 10) in a single step. We note parenthetically that
while we employ a full equivalent of the peptide-based
promoter in these experiments, and therefore that the
“catalytic” species does not rigorously demonstrate turnover
under these conditions, the capacity of the peptide to alter
product selectivity is unambiguous. Moreover, when the
reaction is conducted with 50 mol % peptide 8, similar
selectivity is observed but the reaction proceeds to lower
conversion (∼90% conv. with 100 mol %, ∼75% conv. with 50
mol %; Figure 3A(e)).
The assignment of the structure for 9 is based on the
following NMR experiments. Shown in Figure 4A is the overlay
of the HSQC spectra for 3 (blue) and 9 (gray). Notably, the
correlation for the 3b (C−H) positions was not apparent in
compound 9. At the same time, we observed minor changes in
the chemical shift for the correlations corresponding to the 3f
and 3d (C−H) positions. A more dramatic chemical shift
perturbation was observed for the X3 (α-C−H) correlation
(the α-position the residue 3 aryl glycine; Figure S11 in SI).
These features could be induced by a change in the ring current
of aryl ring 3 upon bromination. To further illustrate the basis
of the assignment for structure 9, we show the key HMBC
(blue) and NOESY (green) correlations in Figure 4B. These
observations taken together culminate in our assignment of 9 as
the 3b-Br-teicoplanin. It is interesting to note that a more
detailed analysis of the HSQC spectrum of 9 reveals a doubling
of certain correlations. Figure 4C illustrates the presence of a
putative minor conformation, with the chemical shift
perturbations for the minor species most prominent for the
C
dx.doi.org/10.1021/ja4038998 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. (A) Overlay of HSQC spectrum of 9 (gray) with 3 (blue).
(B) Key NOESY (green) and HMBC (blue) correlation supportive of
the structural assignment of 9. (C) HSQC spectrum of 9 showcasing
the presence of the minor conformer.
Figure 5. Effect of varying the position of the catalytic residue on
product distribution.
4b and X3 (α-C−H) correlations. These observations suggest
possible conformational heterogeneity in the peptide backbone
region between residue 5 and residue 3. This observation has
also been made previously for teicoplanin A2-2 itself, albeit with
quite low intensity for the putative minor conformer.33,41 The
amplification of conformational heterogeneities as a function of
site-selective modification of complex natural products is a
phenomenon that appears to be quite common, based on these
and other recent studies.28
To further assess the capacity of peptide-based catalysts/
promoters to influence the site of bromination, we examined
different sequences wherein the position of the Asn(Me)2
moiety was varied within the tripeptide. Our initial hypothesis
was that localization of a putative bromine-directing side chain
might alter the site of bromination, affording access to
alternative teicoplanin A2-2-derived bromides. However, as
shown in Figure 4, our observations did not support this initial
assertion. Instead, both peptides 12 and 13, with the Asn(Me)2
as the central residue (12, Figure 5B) or as the residue in the C-
terminal position (13, Figure 5C), deliver highly selective
brominations of teicoplanin to deliver the 3b-Br teicoplanin 9,
as is observed in the presence of peptide 8 (Figure 5A).
Notably, all three peptides in the series offer the same striking
reversal of selectivity relative to the control reaction in the
absence of peptide.
Our interpretation of these observations is that the binding
of DXaa-DYaa peptide fragments to teicoplanin may occur in
accord with well-established models in the literature (Figure 6,
Adapted from PDB 3VFJ). If so, this event may result in a
conformational change of the glycopeptide that renders the 3b-
position of the substrate more reactive toward uncatalyzed
bromination, relative to the 7f-position, which reacts first in the
absence of peptide. In this situation, the possible bromination-
accelerating effects of the Asn(Me)2 side chain may be muted,
Figure 6. X-ray crystal structure of 6C-dechloro-3 bound to Lys-DAla-
DAla-OH. Adapted from PDB 3VFJ.
relative to the intrinsically high reactivity of the resorcinol
moieties in the glycopeptide. It is also possible that other amide
bond carbonyl groups of the ligand or the urethane carbonyl
group may also assist in directing the delivery of Br. It is also
difficult to exclude the possible direction of bromination from
D
dx.doi.org/10.1021/ja4038998 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
carbonyl groups in teicoplanin A2-2 itself. In each scenario, the
unique selectivity exhibited by the putative teicoplanin A2-2/
peptide adduct does support a delivery event that derives from
supramolecular complex formation.
Pronounced peptide-dependent effects were also observed as
one examines further bromination of teicoplanin A2-2. As
shown in Figure 7A, when 3 was exposed to an excess of NBP
Our analysis of the NMR spectra revealed the new
compound to be the illustrated 7d,3b,3d-Br3 teicoplanin
analogue (14, Figure 7B). A first-order analysis of the overlaid
HSQC spectra of 3 (blue) and 14 (gray) reveals a larger
number of chemical shift perturbations for many of the C−H
correlations than we had observed for compounds 7 and 9,
possibly reflecting greater structural reorganization upon
tribromination. Among the most diagnostic observations were
the absence of the C−H correlations for the 7d, 3b, and 3d
positions. In addition, the perturbation of chemical shifts for
nearest neighbors was also apparent for all three of these
positions. For example, X3 (α-C−H), 7f (C−H), and 3f (C−H)
correlations perturbations may all be observed. Notably,
significant changes in the chemical shifts of the unique
mannose ring (M) were apparent, consistent with bromination
of 7d-position (Figure S20 in SI). These features further
suggested a significant conformational reorganization and
prompted additional NMR studies to unambiguously assign
tribromide analogue. NOESY and HMBC experiments were
performed, and the data support the assignment of 14 (Figure
S21 in SI). On a preparative scale, 15 mg of 14 could be
obtained from 80 mg of 3 in a single operation.
At this stage, we wished to establish whether the new
brominated variants of 3 were substrates for metal-catalyzed
Suzuki cross-coupling reactions.32,42 At the beginning of the
study, we encountered poor results due to the apparent binding
of Pd reagents to teicoplanin that resulted in inhibition of
catalysis. This hypothesis derived from the observation of
molecular ions corresponding to [substrate+Pd] in the LC−
MS of the reaction mixture. However, we found that use of
higher loadings of Pd (50 mol %) along with the water-soluble
phosphine ligand 15 (100 mol %) allowed cross-coupling to
occur with useful efficiencies. For example, when bromo-
teicoplanin 9 was subjected to these conditions, furan-
containing teicoplanin analogue 16 was obtained in 28% yield
(73% conversion, Figure S39 in SI), in a single step, with high
purity after reverse phase HPLC purification (Figure 8A).
Intriguingly, analysis of the reaction mixture prior to
purification by LC−MS revealed the presence of a minor,
doubly functionalized product with two furyl groups, and
possessing only a chlorine atom, suggesting functionalization of
one of the indigenous chlorines of 3. This compound was
assigned as 17, and notably it can be isolated in 35% yield
under conditions optimized for its formation. In this vein, we
note that bromo-teicoplanin 7, under analogous conditions,
may be converted to compound 18 or 19, where substitution of
the typically less reactive (toward metal-catalyzed cross-
coupling) C−Cl bond has occurred, rather than at the generally
more reactive C−Br bond (Figure 8B). The unexpected high
reactivity of the ring 2c position of 3 stimulated examination of
the Pd-catalyzed cross-coupling of native 3 under related
conditions. Indeed, we found that compounds 20, 21, and 22
could be obtained in 43% (55% conv., Figure S24 in SI), 30%
(79% conv., Figure S28 in SI), and 31% (58% conv., Figure S31
in SI) isolated yield, respectively (Figure 8C). We conclude at
this stage that metal-catalyzed cross-coupling reactions of
bromo-teicoplanins are possible in a site-selective manner, but
we also recognize that normative hierarchies of haloarene
reactivity are subject to case-specific study as molecular
complexity alters reactivity.
Figure 7. (A) Effect of peptides on the site-selectivity of teicoplanin
tribromination compared to the control reaction (data collected on
HPLC). (B) Assigned structure of tribrominated teicoplanin (14). (C)
Overlay of HSQC spectrum of 14 (gray) with 3 (blue).
(3.3 equiv), either in the absence or in the presence of different
peptides (8, 12, and 13, with variable loci of the Asn(Me)2 side
chain) different product distributions were obtained. In the
absence of a peptide-based promoter, a highly complex mixture
of products was obtained (Figure 7A(a)). One implication is
that there appears to be a comparable level of reactivity for
many sites within teicoplanin A2-2 as further functionalization
occurs. Even so, the reaction mixtures are substantially less
complex when peptide-based promoters were evaluated (Figure
7A(b−d)), exhibiting their capacity to perturb site-selectivity
among the less reactive sites. When a peptide bearing the N-
terminal Asn(Me)2 side chain was employed (8), the reaction
mixture contains a number of products, including a prominent
HPLC peak that contains an inseparable mixture of di- and
tribromides (Figure 7A(b)). When the Asn(Me)2 is central to
the tripeptide (12), two peaks were observed in the HPLC
trace that once again contain mixtures (Figure 7A(c)). Yet,
when the Asn(Me)2 is in the C-terminal position (13), a
striking, peptide-dependent outcome is observed, with a new,
homogeneous peak apparent in the HPLC trace (Figure
7A(d)). LC−MS analysis reveals the compound to be a new
tribrominated species.
The chemistry described above enabled direct synthesis of
eleven previously unknown analogues of 3. Given that several
exhibit conformational perturbations, and all possess altered
E
dx.doi.org/10.1021/ja4038998 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 8. (A) Cross-coupling of compounds 9. (a) Conditions: 50 mol % Pd(OAc)2, 100 mol % water-soluble-SPHOS, K2CO3 (10 equiv), boronic
acid (10 equiv), H2O/MeCN (2:1), 35 °C for 16 and 50 °C for 17. (B) Cross-coupling of compounds 7. (b) Conditions: 50 mol % Pd(OAc)2, 100
mol % water-soluble-SPHOS, K2CO3 (10 equiv), boronic acid (10 equiv), H2O/MeCN (2:1), 35 °C for 18 and 50 °C for 19. (C) Cross-coupling of
compounds 3. (c) Conditions: 100 mol % Pd(OAc)2, 100 mol % water-soluble-SPHOS, K2CO3 (10 equiv), boronic acid (10 equiv), H2O/MeCN
(2:1), 50 °C. Conversions are not corrected for response factor.
functional group display on the teicoplanin scaffold, we elected
to determine minimum inhibitory concentrations (MICs) for
the new analogues as antibacterial agents. We tested the
compounds against five bacterial strains, including methicillin-
resistant S. aureus (MRSA) and vancomycin-resistant Enter-
ococcus (VRE; VanB exhibits vancomycin resistance, but it is
teicoplanin susceptible; VanA is both vancomycin and
teicoplanin resistant). As shown in Table 1, in comparison to
control compounds (entry 1−3), the newly synthesized
analogues 7, 9, and 10 (entries 4−6) are quite similar in
potency to 3 against all five bacterial strains. In contrast,
tribrominated teicoplanin A2-2 (14, entry 7) exhibits a decrease
in activity with four of the five strains. It is perhaps related that
we observed the most dramatic conformational perturbations
with 14 relative to native teicoplanin A2-2, as exhibited by
NMR spectroscopy.
Table 1. MIC Data for New Analogues
a
b
MIC values reported in μg/mL. MSSA = methicillin-susceptible S.
c
On the other hand, the analogues obtained through cross-
coupling (entries 8−14) demonstrated comparable or increased
potency against several of the bacterial strains, in comparison to
vancomycin and teicoplanin. Compound 16, for example, with
aureus, ATCC 29213. MRSA = methicillin-resistant S. aureus, ATCC
d
43300. VSE = vancomycin-susceptible enterococci, ATCC 29212.
e
f
VRE = vancomycin-resistant enterococci, ATCC 51299. MMX 486.
See Supporting Information for additional details.
F
dx.doi.org/10.1021/ja4038998 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Notes
furyl substitution at the 3b-position, exhibited higher potency
against the MRSA strain (entries 1−3 vs 8). Relocation of furyl
substituent from the 3b-position to the 2c-position (compound
20, entry 9) resulted in enhancement of activity against VRE
strains (entries 1−3 vs 9). Compound 17 (entry 10), with both
2c- and 3b-positions substituted with a furyl group, exhibits a
similar activity profile in comparison to 16 and 20 (entry 8 and
9). Compound 18 (entry 11) possessing the 7f-bromine
substituent and the 2c-furyl group also maintains an analogous
profile. A striking and different profile was observed with
compounds 19, 21, and 22 (entries 12−14). Substitution of the
2c-position of 3 with biphenyl functionality (compound 21)
results in substantial activity against VRE (VanA) strain.
Simultaneously, however, compound 19 exhibits a loss of
potency when evaluated against the MSSA and MRSA strains.
Compound 19 (entry 12), with a 7f-Br and a 2c-biphenyl
functionality, also exhibits this trend. Compound 22, with 2c-
octenyl substitution, exhibits the trend as well, while showing
significant potency against the vancomcin- and teicoplanin-
resistant strain (Van A, entries 1−3 vs 14). These data are
compared to antibacterial behaviors of the antibiotic linezolid in
entry 15. The unique behaviors of biphenyl-containing
compounds 19, 21, and 22 may suggest a change in the
mechanism of action.1,43−45 The data presented in Table 1
demonstrate that altering the structure of teicoplanin with
either bromination or cross-coupling reactions of either
brominated teicoplanins (7 and 9) or teicoplanin A2-2, itself
(3), can lead to compounds with significant antibacterial
activity against strains that exhibit vancomycin and teicoplanin
resistance.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful to the National Institutes of General Medical
Sciences of the National Institute of Health (GM-068649) for
support.
REFERENCES
■
(1) Kahne, D.; Leimkuhler, C.; Lu, W.; Walsh, C. Chem. Rev. 2005,
105, 425.
(2) Jeya, M.; Moon, H.-J.; Lee, K.-M.; Kim, I.-W.; Lee, J.-K. Curr.
Pharm. Biotechnol. 2011, 12, 1194.
(3) Ashford, P.-A.; Bew, S. P. Chem. Soc. Rev. 2012, 41, 957.
(4) Li, T.-L.; Liu, Y.-C.; Lyu, S.-Y. Curr. Opin. Chem. Biol. 2012, 16,
170.
(5) Payne, D. J.; Gwynn, M. N.; Holmes, D. J.; Pompliano, D. L. Nat.
Rev. Drug Discovery 2007, 6, 29.
(6) Bardone, M. R.; Paternoster, M.; Coronelli, C. J. Antibiot. 1978,
31, 170.
(7) Parenti, F.; Beretta, G.; Berti, M.; Arioli, V. J. Antibiot. 1978, 31,
276.
(8) Borghi, A.; Coronelli, C.; Faniuolo, L.; Allievi, G.; Pallanza, R.;
Gallo, G. G. J. Antibiot. 1984, 37, 615.
(9) Malabarba, A.; Goldstein, B. P. J. Antimicrob. Chemother. 2005,
55, ii15.
(10) Guskey, M. T.; Tsuji, B. T. Pharmacotherapy 2010, 30, 80.
(11) James, R. C.; Pierce, J. G.; Okano, A.; Xie, J.; Boger, D. L. ACS
Chem. Biol. 2012, 7, 797.
(12) Evans, D. A.; Wood, M. R.; Trotter, B. W.; Richardson, T. I.;
Barrow, J. C.; Katz, J. L. Angew. Chem., Int. Ed. 1998, 37, 2700.
(13) Nicolaou, K. C.; Mitchell, H. J.; Jain, N. F.; Bando, T.; Hughes,
R.; Winssinger, N.; Natarajan, S.; Koumbis, A. E. Chem−Eur. J. 1999,
5, 2648.
(14) Boger, D. L.; Kim, S. H.; Miyazaki, S.; Strittmatter, H.; Weng, J.-
H.; Mori, Y.; Rogel, O.; Castle, S. L.; McAtee, J. J. J. Am. Chem. Soc.
2000, 122, 7416.
(15) Boger, D. L. Med. Res. Rev. 2001, 21, 356.
(16) Boger, D. L.; Kim, S. H.; Mori, Y.; Weng, J.-H.; Rogel, O.;
Castle, S. L.; McAtee, J. J. J. Am. Chem. Soc. 2001, 123, 1862.
(17) Evans, D. A.; Katz, J. L.; Peterson, G. S.; Hintermann, T. J. Am.
Chem. Soc. 2001, 123, 12411.
(18) Xie, J.; Okano, A.; Pierce, J. G.; James, R. C.; Stamm, S.; Crane,
C. M.; Boger, D. L. J. Am. Chem. Soc. 2011, 134, 1284.
(19) Okano, A.; James, R. C.; Pierce, J. G.; Xie, J.; Boger, D. L. J. Am.
Chem. Soc. 2012, 134, 8790.
(20) Losey, H. C.; Jiang, J.; Biggins, J. B.; Oberthur, M.; Ye, X.-Y.;
Dong, S. D.; Kahne, D.; Thorson, J. S.; Walsh, C. T. Chem. Biol. 2002,
9, 1305.
(21) Pinter, G.; Batta, G.; Keki, S.; Mandi, A.; Komaromi, I.; Takacs-
Novak, K.; Sztaricskai, F.; Roth, E.; Ostorhazi, E.; Rozgonyi, F.;
Naesens, L.; Herczegh, P. J. Med. Chem. 2009, 52, 6053.
(22) Liu, Y.-C.; Li, Y.-S.; Lyu, S.-Y.; Hsu, L.-J.; Chen, Y.-H.; Huang,
Y.-T.; Chan, H.-C.; Huang, C.-J.; Chen, G.-H.; Chou, C.-C.; Tsai, M.-
D.; Li, T.-L. Nat. Chem. Biol. 2011, 7, 304.
CONCLUSIONS
■
In summary, we have identified small-molecule promoters that
enable access to unique brominated forms of teicoplanin A2-2.
The approach allows control over whether the 5,7-biaryl ring
system, or the 1,3-biaryl ether sector of the natural product
undergoes bromination. NMR studies facilitated assignments of
the site of bromination and analysis of conformational
consequences. Metal-catalyzed cross-coupling reactions of
brominated teicoplanin analogues and teicoplanin A2-2 itself
have enabled access to further diversified compounds. These
studies also unveil unexpected hierarchies of site-selectivity for
cross-coupling reactions in complex molecular environments
(C−Cl site over C−Br site). Evaluation of the antibacterial
properties of the new compounds reveals both positive and
negative modulation of activity against various strains of
resistant bacteria. The activity of compound 22 may be
particularly notable for its activity against the VanA-type strain
of VRE. Chemistry-dependent, site-selective alteration of
teicoplanin has thus led to unique compounds of altered and
notable biological activity. Generally, this approach may allow
for diversification and analysis of SAR for quite complex
molecules.
̈
(23) Jeya, M.; Moon, H.-J.; Lee, K.-M.; Kim, I.-W.; Lee, J.-K. Curr.
Pharm. Biotechnol. 2011, 12, 1194.
(24) Losey, H. C.; Peczuh, M. W.; Chen, Z.; Eggert, U. S.; Dong, S.
D.; Pelczer, I.; Kahne, D.; Walsh, C. T. Biochemistry 2001, 40, 4745.
(25) Jung, H.-M.; Kim, S.-Y.; Moon, H.-J.; Oh, D.-K.; Lee, J.-K. Appl.
Microbiol. Biotechnol. 2007, 77, 789.
(26) Jung, H.-M.; Jeya, M.; Kim, S.-Y.; Moon, H.-J.; Kumar, S. R.;
Zhang, Y.-W.; Lee, J.-K. Appl. Microbiol. Biotechnol. 2009, 84, 417.
(27) Lewis, C. A.; Miller, S. J. Angew. Chem., Int. Ed. 2006, 118, 5744.
(28) Fowler, B. S.; Laemmerhold, K. M.; Miller, S. J. J. Am. Chem. Soc.
2012, 134, 9755.
ASSOCIATED CONTENT
■
S
* Supporting Information
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
scott.miller@yale.edu
(29) Pathak, T. P.; Miller, S. J. J. Am. Chem. Soc. 2012, 134, 6120.
G
dx.doi.org/10.1021/ja4038998 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(30) Neumann, C. S.; Fujimori, D. G.; Walsh, C. T. Chem. Biol. 2008,
15, 99.
(31) Dong, C.; Flecks, S.; Unversucht, S.; Haupt, C.; van Pee
́
, K.-H.;
Naismith, J. H. Science 2005, 309, 2216.
(32) Nakama, Y.; Yoshida, O.; Yoda, M.; Araki, K.; Sawada, Y.;
Nakamura, J.; Xu, S.; Miura, K.; Maki, H.; Arimoto, H. J. Med. Chem.
2010, 53, 2528.
(33) Heald, S. L.; Mueller, L.; Jeffs, P. W. J. Magn. Reson. 1987, 72,
120.
(34) Hunt, A. H.; Molloy, R. M.; Occolowitz, J. L.; Marconi, G. G.;
Debono, M. J. Am. Chem. Soc. 1984, 106, 4891.
(35) 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.
(36) Nitanai, Y.; Kikuchi, T.; Kakoi, K.; Hanamaki, S.; Fujisawa, I.;
Aoki, K. J. Mol. Biol. 2009, 385, 1422.
(37) Economou, N. J.; Nahoum, V.; Weeks, S. D.; Grasty, K. C.;
Zentner, I. J.; Townsend, T. M.; Bhuiya, M. W.; Cocklin, S.; Loll, P. J.
J. Am. Chem. Soc. 2012, 134, 4637.
(38) Denmark, S. E.; Burk, M. T. Proc. Natl. Acad. Sci. U.S.A. 2010,
107, 20655.
(39) Malabarba, A.; Trani, A.; Tarzia, G.; Ferrari, P.; Pallanza, R.;
Berti, M. J. Med. Chem. 1989, 32, 783.
(40) Nieto, M.; Perkins, H. R. Biochem. J. 1971, 124, 845.
(41) Westwell, M. S.; Gerhard, U.; Williams, D. H. J. Antibiot. 1995,
48, 1292.
(42) Roy, A. D.; Gruschow, S.; Cairns, N.; Goss, R. J. M. J. Am. Chem.
̈
Soc. 2010, 132, 12243.
(43) Ge, M.; Chen, Z.; Russell, H.; Onishi; Kohler, J.; Silver, L. L.;
Kerns, R.; Fukuzawa, S.; Thompson, C.; Kahne, D. Science 1999, 284,
507.
(44) Goldman, R. C.; Baizman, E. R.; Longley, C. B.; Branstrom, A.
A. FEMS Microbiol. Lett. 2000, 183, 209.
(45) Chen, L.; Walker, D.; Sun, B.; Hu, Y.; Walker, S.; Kahne, D.
Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 5658.
H
dx.doi.org/10.1021/ja4038998 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX