A progressive synthetic strategy for class B synergimycins

Article abstract of DOI:10.1016/j.tetlet.2004.01.048

Described are the syntheses of four macrocyclic peptides that are the core structure of class B synergimycins, and the synthesis of a final class B derivative. Our synthetic route to these synergimycin derivatives allows the incorporation of amino acid su

Full text of DOI:10.1016/j.tetlet.2004.01.048

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 2147–2150
A progressive synthetic strategy for class B synergimycins
Jennifer L. Robinson, Rachel E. Taylor, Lisa A. Liotta, Megan L. Bolla,
Enrique V. Azevedo, Irene Medina and Shelli R. McAlpine*
Department of Chemistry, 5500 Campanile Drive, San Diego State University, San Diego, CA 92182-1030, USA
Received 24 September 2003; revised 9 January 2004; accepted 12 January 2004
Abstract—Described are the syntheses of four macrocyclic peptides that are the core structure of class B synergimycins, and the
synthesis of a final class B derivative. Our synthetic route to these synergimycin derivatives allows the incorporation of amino acid
substitutions at all points in the macrocycle, leading to structurally diverse class B analogs.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Antibiotic resistance is an extreme public health con-
cern1 and previously treatable infectious diseases are
becoming life-threatening infections. In order to treat
these diseases a common strategy involves resurrecting
older classes of antibiotics via derivatization. A number
of potent antibiotics are large macrocyclic peptides.2
The synthesis of new antibiotics in these classes involves
constructing peptide derivatives of these macrocycles. In
an effort to design new antibiotics, we synthesized class
B synergimycin derivatives based on virginiamycin S1
(VS1) (Fig. 1).3 Our synthetic approach was intended to
simplify the exchange of amino acids in order to probe
the structure–activity relationship of each residue.
Herein we describe a synthesis strategy that provides
four core class B macrocycles and one initial class B
synergimycin derivative. Class A and class B syner-
gimycins are known to work synergistically to inhibit
protein synthesis in bacteria.4 The target site for the two
classes is the peptidyltransferase center of the 50S
ribosomal subunit.5 Renewed antibacterial activity is
known to occur upon structural changes to residues
within the macrocycle.6
The work described here utilizes a strategy where sub-
stitution of amino acids in any position is relatively
straightforward. In addition, the success of the recent
antibiotic synercid,6 which contains a class B derivative,
suggests the potential for making new antibiotics using
our strategy.
The specific binding site for class B on the ribosome is
well established.7 The first three residues (1, 2, and 3)
remain the same for this family of the synergimycins. It
is known that picolinic residue 1 is responsible for
binding to the ribosome and it is also thought that res-
idues 2 and 3 make critical contacts with the ribosome.8
Conformational studies of class B natural products
show these compounds all have a cis peptide configu-
ration between the 4 and 5 residue, a type VI b-turn
reinforced by an intramolecular hydrogen bond between
6 and 3, and a biased rotational state where the aromatic
ring of 5 is oriented toward residue 4.9 Due to the
structural importance of residues 1, 2, and 3 for binding
to the ribosome, only one small change was made to
residue 3.8 In order to simplify the synthesis of our
initial derivatives, the phenylglycine (3) in the natural
product was exchanged for a phenylalanine. It is not
clear what effect this will have on antibiotic activity.
N
OH
O
O
NH
H
N
H3C
O
N
O
O
NH
O
O
N
O
N
O
Figure 1. Virginiamycin S1 (VS1).
Keywords: Class B; Synergimycin; Antibiotic; Virginiamycin.
* Corresponding author. Tel.: +1-6195945580; fax: +1-6195944634;
e-mail: mcalpine@chemistry.sdsu.edu
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.01.048

2148
J. L. Robinson et al. / Tetrahedron Letters 45 (2004) 2147–2150
It is known that residue 4 is embedded into a lipophilic
region of the bacterial ribosome and that the alteration
of residue 4 to a more hydrophobic residue leads to
increased synergistic activity.10 In addition, studies
indicate the importance of p-stacking between the
ketone on residue 4 and the aromatic group of residue
5.9 Therefore, we chose four substitutions at position 4
(4a–d) in an effort to elucidate the influence of this res-
idue on p-stacking with residue 5. Finally, we chose two
substitutions in position 5. (L)-Tyrosine (5a) was chosen
because it resembled the amino acid in the natural
product (VS1), and (L)-tryptophan (5b) was chosen to
promote a p-stacking interaction between 4 and 5.11 In
this preliminary study, we chose to keep residue 6 the
base, commercially available L-threonine methyl ester
(2) and 3-hydroxy picolinic acid (1) were coupled in
methylene chloride (Fig. 3) to give fragment 1 (86%
yield). For fragment 2 synthesis we coupled L-phenyl-
alanine methyl ester (3) with the N-Boc protected resi-
due of 4 [(S)-(N)-Boc-1,2,3,4-tetrahydroisoquinoline-3-
carboxylic acid (4a), (N)-Boc-glycine (4b), (R)-(N)-Boc-
1,2,3,4-tetrahydroisoquinoline-3-carboxylic acid (4c),
and (S)-())-Boc-2-piperidine carboxylic acid (4d)]. This
gave four examples of dipeptide 3-4X-Boc (80–94%
yield).14 Deprotection of the amine on residue 4 using
20% TFA and 2 equiv of anisole in methylene chloride
gave the free amine 3-4X (ꢀquantitative yields). Cou-
pling of the dipeptides to N-(a)–Boc-
L
-tyrosine (5a) or
same as that in the natural product, (
L
)-proline, because
N-(a)–Boc-
L-tryptophan (5b) using HATU gave the
it forces a b-turn element into the macrocyclic archi-
tecture and it is responsible for hydrogen bonding to
residue 3. Residue 7 appears to play a small role in the
conformation and biological activity of the macrocycle.9
Based on these previous studies, we chose (L)-leucine as
residue 7.
eight desired tripeptides in high yields (65–94%).14
Deprotection of the acid using barium hydroxide in
methanol gave eight fragment 2 compounds in quanti-
tative yields. Fragment 3 was synthesized using HATU
as the activating agent and coupling the free amine of
L
- proline methyl ester (6) with N-(a)-Boc-leucine (7).
The three disconnections shown in our initial convergent
synthesis strategy (Fig. 2) allow us to readily exchange
amino acid residues. Macrocyclization was designed to
occur between the primary amine on residue 7 and the
free acid on residue 2. Previous successful syntheses of
the natural product have shown that ring closing can
occur between residues 3 and 4,12 and between the sec-
ondary amine 4 and 5.13
In our initial strategy, we coupled fragment 1 with
fragment 2 (Fig. 4), yielding five pentapeptides in yields
ranging from 45% to 90%.14 These pentapeptides were
then coupled to fragment 3 to give five linear hepta-
peptide precursors (yields 26–63%).14 Attempted cycli-
zation of the five linear heptapeptides failed using
conditions that had worked in a similar class of mac-
H3CO
O
Our efforts started with the construction of fragment 1.
Using O-(7-azabenzatrizol-1-yl)-N,N,N0,N0-tetramethyl-
uronium hexafluorophosphate (HATU) and HunigÕs
O
NH2
O
N
H3C
(a)
N
H
N
1
OCH3
+
OH
OH
OH
O
OH
OH
H3C
1-2
2
Fragment 1
86% yield
H3CO
O
O
O
BocHN
N
N
N
H
OH
OH
OCH3
NH2
OH
O
H
O
1. (a)
2. (b)
H
4a-d
O
H3C
HO
O
+
N
H3CO
Boc
4a-d
Fragment 1
Fragment 3
O
O
3-4
OH
N
3
4
1
HO
O
O
NHBoc
2
H3C
HN
NH
O
4 =
H
N
7
6
O
3-4-5
5a,
b
OH
O
N
NHBoc
4a-d
N
O
O
O
4
a
b
1. (a)
2. (c)
5a, b
5 =
O
Fragment 2
8 examples
65-94% yield
NH
O
a
3
NH
O
NH
OH
N
c
d
Boc
O
5
HN
OH
b
BocHN
HN
1. (a)
2. (c)
O
N
+
O
H
N
MeO
OH
N
O
NH
6
HO
6-7
7
O
O
O
Fragment 3
95% yield
NHBoc
4
Figure 3. Initial synthesis of fragments for class
B derivatives.
5
Reagents and conditions: (a) HATU (1.2 equiv), HunigÕs base
(3 equiv), CH3CN; (b) TFA (20%), CH2Cl2, anisole (2 equiv); (c)
Ba(OH)2 (4 equiv), MeOH.
Fragment 2
Figure 2. Amino acids used in class B derivative synthesis.

J. L. Robinson et al. / Tetrahedron Letters 45 (2004) 2147–2150
2149
HO
O
HO
O
H3CO
NH
NH
H3CO
O
O
O
O
O
1. (a)
2. (b)
O
NHBoc
+
5 a, b
O
NHBoc
+
Cbz
N
N
H
1. (a)
2. (b)
4a-d
4a-d
3-4-5
N
H
OH
OH
3-4-5
H3C
5a, b
OH
H3C
Fragment 2
2
1-2
Fragment 1
Fragment 2
Cbz
HN
2
N
1
Boc
H
3C
OCH3
NH
HO
O
O
O
O
HN
2
N
O
Boc
H3C
NH
HO
OCH3
O
NH
HO
3
O
O
O
O
N
O
Fragment 3
NHBoc
5a, b
4a-d
O
(a)
O
Tetrapeptide
8 examples
31-67% yield
NH
3
Fragment 3
O
O
NHBoc
(a)
4a-d
Cbz
HN
Fragment 1and 2
5 examples
45-90% yield
5a, b
Cbz
OCH3
H3C
HN
2
H3C
BHocN
7
O
O
N
HN
O
1. (c)
2. (b)
3. (a)
O
6
O
O
N
O
O
N
N
1
NH
HO
O
HO
O
O
O
O
3
NH
HN
NH
1. (c)
2. (b)
3. (a)
O
HN
2
H3C
4a-d
OCH3
O
NH
O
H3C
7
O
HN
BocNH
5a, b
6
4a-d
O
O
N
O
O
O
N
O
Linear hexapeptide Precursor
8 examples
43%-75% yields
5a, b
O
NH
Macrocyclic ClassB Derivatives
4 examples
16%-35% yields
3
NH
O
O
O
O
O
NH
O
NH
4a-d
4a-d
Figure 5. Modified synthesis of macrocyclic derivatives. Reagents and
conditions: (a) HATU (1.2 equiv), TBTU, PyBop, and/or DEPBT used
as coupling reagents (1.2 equiv), HunigÕs base (3 equiv), CH3CN; (b)
TFA (20%), CH3CN, anisole (2 equiv); (c) Ba(OH)2 (4 equiv), MeOH.
5a, b
Linear Precursor
5 examples
26-63%
5a, b
Anticipated Macrocyclic Derivatives
Figure 4. Initial synthesis of macrocyclic derivatives. Reagents and
conditions: (a) HATU (1.2 equiv), TBTU, PyBop, and/or DEPBT used
as coupling reagents (1.2 equiv), HunigÕs base (3 equiv), CH3CN; (b)
TFA (20%), CH3CN, anisole (2 equiv); (c) Ba(OH)2 (4 equiv), MeOH.
31–67% yields.14 Then, the amine on residue 5 was
deprotected and was coupled to fragment 3 (43–75%
yields for eight examples of hexapeptides). These hexa-
peptides were then subjected to the macrocyclization
conditions used in the previous strategy and four mac-
rocycles were obtained in reasonable yields (16–
35%).14;17
rocyclic peptides.15 These conditions involved depro-
tecting the acid using 4 equiv of barium hydroxide, and
quenching with TFA in acetonitrile with 2 equiv of
anisole. Upon deprotection of the amine we subjected
the crude, dry product to HATU, TBTU, and DEPBT
coupling reagents (1.2 equiv each), with HunigÕs base
(3 equiv) in acetonitrile.16 The final macrocyclizations
ran over approximately 4 days at low concentration
(0.005–0.01 M) in order to maximize the yields. How-
ever, upon LCMS analysis at each stage of this in situ
deprotection and macrocyclization, it appeared that
residue 1 was base sensitive and was labile under the
barium hydroxide deprotection conditions. Although
other conditions were explored for this deprotection,
only basic conditions successfully removed the methyl
ester, which also removed residue 1. In addition, a
protection strategy was investigated for the hydroxyl on
residue 1. However, these conditions were not straight-
forward13 and another modified synthetic route was
developed.
A class B derivative was achieved by removing the Cbz
protecting group using hydrogenation and Pd/C and
then coupling residue 1 to the core macrocycle (Fig. 6),18
thus successfully validating our revised synthetic strat-
egy.
In summary, we have developed a useful synthetic route
to class B synergimycin derivatives that will allow the
incorporation of amino acid substitutions at all points in
the macrocycle.19 This will lead to structurally diverse
class B synergimycins that have the potential to be
antibiotics against resistant strains of bacteria. The
successful synthesis of an initial class B derivative using
this route establishes the viability of our approach. We
are currently carrying out the synthesis of other deriv-
atives and upon completion, antibacterial assays will be
run on these compounds in order to establish a struc-
ture–activity relationship.
We modified our strategy and used a Cbz-protected
residue 2 in place of fragment 1 (Fig. 5). Coupling of
residue 2 to fragment 2 makes the eight tetrapeptides in

2150
J. L. Robinson et al. / Tetrahedron Letters 45 (2004) 2147–2150
battista, M.; Ize, G.; Engelborghs, Y.; Cocito, C. J. Biol.
Chem. 1984, 259, 6334–6339.
Cbz
HN
N
O
O
O
H
N
O
N
9. (a) Anteunis, M. J. O.; Sharma, N. K. Bull. Soc. Chim.
Belg. 1988, 97, 281–292; (b) Anteunis, M. J. O.; Auwera,
C. V. d.; Vanfleteren, L.; Borremans, F. Bull. Soc. Chim.
Belg. 1988, 97, 135–148.
10. (a) DiGiambattista, M.; Sharma, N. K.; Anteunis, M. J.
O. Bull. Soc. Chim. Belg. 1990, 99, 195–211; (b) Anteunis,
M. J. O.; Callens, R. E. A.; Tavernier, D. K. Eur. J.
Biochem. 1975, 58, 259–268.
OH
HN
H3C
H
N
1. (a)
2. (b)
N
O
O
O
O
H3C
O
O
NH
O
O
O
O
N
H
N
O
NH
N
N
H
OH
OH
Macrocyclic precursor
2-3-4c-5a-6-7
11. Zhang, W.; Anteunis, M. J. O.; Borremans, F. Bull. Soc.
Chim. Belg. 1988, 97, 419–429.
Macrocycle Final Class B Derivative
1 example
60% yield (2steps)
12. Anteunis, M. J. O.; Auwera, C. V. d.; Vanfleteren, L.;
Borremans, F. Bull. Soc. Chim. Belg. 1988, 97, 135–148.
13. (a) Anteunis, M. J. O.; Auwera, C. V. d.; Vanfleteren, L.;
Borremans, F. Bull. Soc. Chim. Belg. 1988, 97, 135–148;
(b) Sharma, N. K.; Anteunis, M. J. O. Bull. Soc. Chim.
Belg. 1989, 98, 355–356.
Figure 6. Modified synthesis of macrocyclic derivatives. Reagents and
conditions: (a) H2, Pd/C, CH3CN; (b) HATU (1.2 equiv), HunigÕs base
(3 equiv), CH3CN.
14. Yield ranges reflect those for each individual amino acid
monomer that is used in that particular step of the
synthesis, that is, yields vary depending on the unique
amino acid used in the reaction at that stage of the
synthesis.
15. Bolla, M. L.; Azevedo, E. V.; Smith, J. M.; Taylor, R. E.;
Ranjit, D. K.; Segall, A. M.; McAlpine, S. R. Org. Lett.
2003, 5, 109–112.
16. Bolla, M. L.; Azevedo, E. V.; Smith, J. M.; Taylor, R. E.;
Ranjit, D. K.; Segall, A. M.; McAlpine, S. R. Org. Lett.
2003, 5, 109–112, Ring-closing reactions are slow and
typically low yielding. Unpublished results from the Guy
Laboratory at UCSF, and our laboratory have found that
the use of several coupling reagents facilitates ring-closing
reactions by providing a choice of reagents for the specific
substrate. This is in lieu of optimizing each individual
reaction for each individual coupling agent.
Acknowledgements
We thank Pfizer, La Jolla, for equipment and financial
donations as well as their fellowships to M.L.B. (sum-
mer 2002) and E.V.A. (2001–2003) and I.M. (2003–
2004). We thank ARCS foundation for a fellowship to
J.L.R. (2003–2004). We thank the McNair program for
their generous support for M.L.B. (summer 2002). We
thank the MIRT program for their support of M.L.B.
E.V.A., and I.M. (2003–2004). We thank San Diego
State University and Boehringer-Ingelheim Pharma-
ceuticals for financial support. We thank Dr. Indrawan
McAlpine (Pfizer, La Jolla), Professor Rodney Guy
(UCSF), and Dr. Greg Roth (Abbott) for helpful dis-
cussions of this work.
17. The four macrocyclic peptides in Figure 5 are listed as
numerical structures, MS data, and yields as follows (note:
MS data is given as major peaks with +45[2 · Naþ ) 1],
+23 [Naþ], and +1 being those peaks):
References and notes
(a)
(MW ¼ 914)MS: 960.40, 939.45, 936.50, 915.5.
Yield: 16%
1. (a) Neu, H. C. Science 1992, 257, 1064; (b) Kaufman, M.;
Friday, M. Washington Post 2000, 45, 319–323, March
17th, p A01.
2. Kerns, R.; Dong, S. D.; Fukuzawa, S.; Carbeck, J.;
Kohler, J.; Silver, L.; Kahne, D. J. Am. Chem. Soc. 2000,
122, 12608–12609.
3. Cocito, C. Microbiol. Rev. 1979, 43, 145.
4. Ennis, H. L. J. Bacteriol. 1965, 90, 1102–1109.
5. Depardieu, F.; Courvalin, P. Antimicrob. Agents Chemo-
ther. 2001, 45, 319–323.
6. Aventis Pharmacueticals website: www.Aventispharma-
US.com.
7. (a) Champney, W. S.; Tober, C. L. Curr. Microbiol. 2000,
41, 126–135; (b) Porse, B. T.; Garrett, R. A. J. Mol. Biol.
1999, 286, 375–387.
8. (a) DiGiambattista, M.; Sharma, N. K.; Anteunis, M. J.
O. Bull. Soc. Chim. Belg. 1990, 99, 195–211; (b) DiGiam-
Purity: ꢀ80%
(b)
(c)
(d)
(MW ¼ 812)MS: 857.7, 835.8, 813.9.
Yield: 35%
Purity: ꢀ85%
(MW ¼ 914)MS: 956.4, 933.2, 915.1.
Yield: 20%
Purity: ꢀ75%
(MW ¼ 890)MS: 937.7, 913.6, 891.5.
Yield: 17%
Purity: ꢀ75%.
18. The final macrocyclic class B derivative has the following
data: (MW ¼ 901) MS: 946.5. 924.5, 924.4. Yield: 60% for
two steps Purity: ꢀ80%.
19. All compounds were characterized using NMR and
LCMS.