A novel strategy for oligopeptide synthesis using a polymer-supported ammonium fluoride

Article abstract of DOI:10.1016/j.tetlet.2006.05.129

A novel method for the preparation of oligopeptides with a PS-ammonium fluoride in the solution phase is reported. The synthesis of lipid II pentapeptide is efficiently synthesized via a PS-ammonium fluoride without chromatographic purifications. The method reported here is very convenient to synthesize a relatively large amount of oligopeptides with abundantly available Fmoc-protected amino acids in a time efficient manner.

Full text of DOI:10.1016/j.tetlet.2006.05.129

Tetrahedron Letters 47 (2006) 5325–5328
A novel strategy for oligopeptide synthesis using a
polymer-supported ammonium fluoride
Michio Kurosu* and Dean C. Crick
Department of Microbiology, Immunology, and Pathology, Colorado State University, 1682 Campus Delivery,
Fort Collins, CO 80523-1682, USA
Received 1 March 2006; revised 17 May 2006; accepted 19 May 2006
Available online 12 June 2006
Abstract—A novel method for the preparation of oligopeptides with a PS-ammonium fluoride in the solution phase is reported. The
synthesis of lipid II pentapeptide is efficiently synthesized via a PS-ammonium fluoride without chromatographic purifications. The
method reported here is very convenient to synthesize a relatively large amount of oligopeptides with abundantly available Fmoc-
protected amino acids in a time efficient manner.
Ó 2006 Elsevier Ltd. All rights reserved.
In connection with our ongoing efforts to identify and
validate new drug targets in Mycobacterium tuberculosis
(Mtb), we initiated studies on peptidoglycan (PG) bio-
synthesis in which we demonstrated that mycobacterial
lipid II is composed of a complex mixture of modified
muramyl and peptide moieties linked to decaprenyl
phosphate and that most of these modifications take
place on lipid II rather than a precursor such as UDP-
MurNAc-pentapeptide (Park’s nucleotide).1 Although
the machinery for PG biosynthesis has been considered
a crucial target, which step of PG biosynthesis repre-
sents the most promising target for binding small mole-
cules that can eventually be developed into a new drug
target has not been defined. In order to develop drug
target in Mtb, we have been interested in membrane-
associated proteins, phospho-N-acetylmuramoyl-penta-
peptide-transferase (MraY), N-acetylglucosamine trans-
ferase (MurG), and specialized proteins termed
‘flippases’ that translocate lipid II from the cytoplasmic
side to the external surface of the cell membrane. The
development of economical high-throughput screens
for Mtb MraY and MurG, and identification of lipid
II flippases require significant amounts of Mycobacteria
compatible Park’s nucleotide, lipid I, and lipid II. The
enzymes necessary for enzymatic synthesis of lipid II
(i.e. MurA-F, MraY and MurG) are available in our
laboratories, however, the specificity of MraY limits
the practicality of a completely enzymatic synthesis of
appropriate amounts of lipid II analogues.2 Although
elegant chemical and chemoenzymatic synthesis of lipid
II have been reported,3 their syntheses involve a signifi-
cant number of intermediate purification steps and re-
sult in low overall yields. In order to supply gram
quantities of (1) Park’s nucleotide lipid I for the devel-
opment of high-throughput screening and (2) lipid II
which will be utilized in screening and studying lipid II
flippases, it is indispensable to eliminate laborious chro-
matographic purification steps after coupling of building
blocks in rapid and practical synthesis.
In an efficient synthesis of lipid II analogues, we
attempted to eliminate laborious purification steps nec-
essary after coupling of each building block. Our syn-
thetic strategy includes (1) a practical synthesis of the
pentapeptide (building block 1) in solution phase (step i),
(2) transfer the pentapeptide to the polymer-support
(step ii), (3) conjugation of appropriately protected
MurNAc (building block 2) to the polymer-supported
pentapeptide (step iii), (4) introduction of lipid ana-
logues (building block 3) (step iv), and (5) simultaneous
deprotection to form lipid I analogues followed by
MurG catalyzed GlcNAc addition to lipid I analogues
leading to lipid II analogues3a (step v) as illustrated in
Scheme 1. We have established MurG promoted Glc-
NAc addition to lipid I analogues with UDP-GlcNAc.
In order to synthesize lipid I analogues in a time efficient
manner, the laborious synthesis of pentapeptide con-
taining hydrophilic residues needed to be simplified. A
*
Corresponding author. Tel.: +1 970 491 7628; fax: +1 970 491
1815; e-mail: michio.kurosu@colostate.edu
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.05.129

5326
M. Kurosu, D. C. Crick / Tetrahedron Letters 47 (2006) 5325–5328
Building block 1 (protected Mtb pentapeptide)
NH2-L-Ala-γ-D-Glu(OMe)-meso-DAP(COCF3)(OMe)-D-Ala
HO
D-Ala-OP
P: Protecting group
or polymer-support
O
R1O
O
MurNAc
AcHN
Me
Building block 2 (MurNAc derivative)
O
O
OR2
step i to iv
P
P
O
p-PMBO
+
O
O- O-
O
HN
L-Ala
O
p-PMBO
Me
R2=
O
O
AcHN
O
Me
O
O
OAllyl
OAllyl
NH
P
7
HO2C
HO
D-Glu
p-PMB= p-methoxybenzoyl
NH2
NH
Building block 3 (lipid monophosphate)
O
O
CO2H
2-
+ R2OPO2
meso-DAP
NH
UDP-GlcNAc
H
R1= H: Lipid I
R1=
HO
D-Ala
Me
N
CO2H
D-Ala
step v
MurG
NHAc
UDP
O
Me
HO
O
: Lipid II
OH
Scheme 1.
large number of natural and unnatural N-9-fluorenyl-
methoxycarbonyl (Fmoc)-protected amino acids are
now commercially available, which have been used in
solid phase syntheses. However, depending on physical
properties of the amino acid residues, low overall yields
caused by incomplete deprotection of Fmoc and by-
product formation during the Fmoc deprotection are
often observed in the solid phase synthesis of oligo- to
long-peptides.4 Deprotection of Fmoc-protected oligo-
peptides with a large excess of secondary amines in solu-
tion causes a significant problem with isolation of the
deprotected free amines. In order to be able to utilize
the abundantly available Fmoc-protected amino acids
as building blocks for solution phase syntheses of oligo-
to long peptides, reliable and convenient Fmoc-depro-
tection conditions needed to be developed. In this letter,
we report a novel method for the synthesis of
oligopeptides in solution using a polymer-supported
ammonium fluoride.
formation of an ammonium-amide complex (a in
Scheme 2). Such a complex would easily be isolated
from the reaction mixtures. Tetrabutylammonium fluo-
ride (TBAF) is known to deprotect Fmoc group.6 How-
ever, the hygroscopic ammonium fluoride is very
difficult to apply to the isolation of an amide anion as
an ammonium complex. Taking advantage of the hydro-
phobic nature of polystyrene polymer and by varying
the structure of quaternary amine, we identified a non-
hygroscopic polymer-supported ammonium fluoride
possessing excellent F ion donor activity and better
cationic properties than the tetrabutylammonium ion.
We synthesized several polymer-supported ammonium
fluorides in over 80% yield, which were determined by
elemental analyses of the remaining Cl atoms.7 As
shown in Scheme 3, (chloromethyl)polystyrene (PS)
was quaternized with tertiary amines. Commercially
unavailable tertiary amines were synthesized in 70–
95% yields from the corresponding secondary amines
As illustrated in Scheme 2, we envisioned capturing the
deprotected amides onto the polymer-support and wash-
ing out all by-products derived from Fmoc group. The
unfavored equilibrium observed (b in Scheme 2),5 to
which inadequate amounts of secondary amines were
applied, can be shifted in the desired direction by the
via
a
condition of diisopropyl azodicarboxylate
(DIAD)/PS-triphenylphosphine/THF. Under these con-
ditions, no overalkylations to form quaternary amines
were observed.8 The counter-ions of the quaternized
PSs were then exchanged to fluoride ions by the treat-
ment with saturated KF in MeOH.9
R3
R2
R1
N
O
O
b
a
H2N
N
FmocHN
wash
O
N
H
HN
H
R1
R1
N
+
Sc
slow
DBF
H
R1
b: Deblocking
reagents
with or without
a: Polymer-supported
ammonium ion
DBF
DBF scavenger (Sc)
Scheme 2.

M. Kurosu, D. C. Crick / Tetrahedron Letters 47 (2006) 5325–5328
5327
1a : R1=Me, R2=
1b : R1=Me, R2=
1c : R1=Me, R2=butyl
1d : R1=Me, R2=benzyl
n
n
-decyl
-octyl
1)
Cl
R1
Me
R2
N
N
H
N
DIAD, PS-Ph3P
R2I / THF
R1
R2
F-
toluene
2) KF/MeOH
80~90%
Me
R1
1e : R1=ethyl, R2=
n
-butyl
Me
1f : R1= -butyl, R2=
n
n
-butyl
90~95%
Scheme 3.
The efficiency of the PS-ammonium fluorides (1a–f) in
deprotection of Fmoc-D-Ala-D-Ala-OMe and hydrolytic
stability of their ammonium-amide anion complexes was
investigated. The rate of the deprotection reaction
increased in the order: 1c < 1b ꢀ 1a < 1d ꢁ 1e 6 1f.
The Fmoc-deprotection with 1f (1.5 equiv at 0.05 M
concentrations) was completed within 1 min in DMF.
However, its PS-N,N,N-dibutylmethyl-amide complex
showed hydrolytic instability; the free amine was lea-
ched from the polymer during washing with DMF. On
the other hand, Fmoc deprotection of Fmoc-D-Ala-D-
Ala-OMe with 1a required 2–3 h to complete (1.5 equiv
at 0.05 M concentrations). The rate of Fmoc-deprotec-
tion with 1b was noticeably slower than that of 1a, but
the deprotection was completed within 4 h. It was con-
firmed that the D-Ala-D-Ala-OMe amide strongly bound
to PS-ammonium ion derived from 1a and 1b and was
not dissociated in DMF, THF, or the other aprotic
solvents. Thus, DBF and its derivatives could easily be
removed from the reaction mixtures. The PS-ammo-
nium amide complexes could be efficiently uncomplexed
with a stoichiometric amount of pyridineÆHCl complex
in DMF to provide HClÆH-D-Ala-D-Ala-OMe in 95%
yield10 and PS-ammonium chloride could then be recon-
verted to the corresponding PS-ammonium fluoride by
the treatment with KF in MeOH. Thus, the polymer
bound ammonium fluoride can easily be recycled.9
We then applied PS-ammonium fluoride 1b11 to the
deprotection of Fmoc-protected amines, mono- and
oligopeptides. Representative results are summarized
in Table 1. Regardless of the nature of amino acid
residue, deprotection of the Fmoc group was completed
within 4 h with 1.5 equiv of 1b. Because PS-ammonium
fluoride reacts with Fmoc-protected amino acids selec-
tively to capture the amides, unreacted free amines (cou-
pling counter parts) and excess reagents are removed by
a simple filtration. Therefore, it is possible to synthesize
oligopeptides without chromatographic purification
after each peptide coupling by using the PS-ammonium
fluoride. We demonstrate the Fmoc/PS-ammonium
fluoride strategy of synthesizing bacterial lipid II penta-
peptide as a model study in which L-Lys(COCF3)
was utilized instead of meso-DAP(COCF3)(OMe) in
Scheme 1. HClÆH-D-Ala-D-Ala-OAllyl was coupled with
0.9 equiv of Fmoc-L-Lys(COCF3)-OH via PyBOP12 as a
coupling reagent to provide Fmoc-L-Lys(COCF3)-D-
Ala-D-Ala-OAllyl, which was then deprotected with
1b. The generated complex was then decomplexed with
pyridineÆHCl, after removing DBF derivatives, affording
HClÆH-Lys(COCF3)-D-Ala-D-Ala-OAllyl in 95% yield.
This was subjected to the coupling reaction with the
N-hydroxysuccinimide (NHS) ester of Fmoc-c-D-
Glu(OMe)-OH followed by (1) scavenging the liberated
NHS13 with Si-carbonate,14 (2) Fmoc-deprotection with
Table 1.
KF/MeOH
Me
+
Me
N
Me
Me
N
C8H17
Cl-
N
HCl
1b / DMF
C8H17
FmocNH R
HN
R
HCl-NH2
R
DBF
amide anion catpture
with PS-ammonium ion
Entry
Fmoc-protected amine or peptides (FmocNH-R)
PS-ammonium F
Reaction time (h)
Yield (%)a,b
1
2
3
4
5
6
7
8
Fmoc-NHC6H4Cl-p
Fmoc-L-Phe
1b
1b
1b
1b
1b
1b
1b
1b
0.5
2
4
4
4
4
4
4
98
99
95
96
95
94
93
92
Fmoc-L-Lys(COCF3)-OMe
Fmoc-D-Ala-D-Ala-OAllyl
Fmoc-D-Ala-D-Ala-OMe
Fmoc-L-Ala-c-D-Glu(OMe)2
Fmoc-L-Lys(COCF3)-D-Ala-D-Ala-OAllyl
Fmoc-L-Ala-L-Phe-L-Gly-L-Gly-OMe
All reactions were carried out with 1.5 equiv of PS-ammonium F (1.5–1.7 mmol/g).
All products were isolated as their HCl salts.
a Yield was established by 1H NMR analysis.
b No Fmoc-derived by-products were observed in 1H NMR.

5328
M. Kurosu, D. C. Crick / Tetrahedron Letters 47 (2006) 5325–5328
5. (a) Carpino, L. A.; Mansour, E. M. E.; Knapczyk, J. J.
Org. Chem. 1983, 48, 666; (b) Sheppeck, J. E., II; Kar, H.;
Hong, H. Tetrahedron Lett. 2000, 41, 5324.
6. Ueki, M.; Tsurusaki, T.; Okumura, J. Pept. Chem. 1994,
213.
7. A general procedure for the synthesis of the polymer-
supported ammonium fluoride. (Chloromethyl)polystyrene
(ꢂ1.7 mmol/g, crosslinked with 1% DVB, 10 g) and the
tertiary amine (3–5 equiv) in toluene (50 mL) were heated
at 90 °C for 3 h. The resins were washed with MeOH,
THF-water(3/1), THF, and EtOAc. The quaternized PS-
ammonium chlorides were soaked in sat. KF in MeOH
(40 mL) for 20 min and the supernatant was removed.
This process was repeated thrice. The resins were washed
with MeOH-water (3/1), MeOH, and THF and dried
under high vacuum.
1b, (3) washing the complex, and (4) decomplexation
with pyridineÆHCl to provide the desired HClÆH-c-D-
Glu(OMe)-L-Lys(COCF3)-D-Ala-D-Ala-OAllyl in 90%
yield. In a similar manner, the protected lipid II penta-
peptide, HClÆH-L-Ala-c-D-Glu(OMe)-L-Lys(COCF3)-D-
Ala-D-Ala-OAllyl, was synthesized by coupling with
Fmoc-L-Ala-OH. Overall yield of the synthesis of the
pentapeptide from H-D-Ala-OAllyl was 65–70%.15
Thus, the lipid II pentapeptide was readily synthesized
without chromatographic purification.
In conclusion, we have developed a novel Fmoc-depro-
tection method with a polymer-supported ammonium
fluoride for synthesizing oligopeptides in a time efficient
manner. This method is very practical for the synthesis
of relatively large amounts of the target oligopeptides
without chromatographic purification. Moreover, the
polymer-supported ammonium fluoride utilized in the
reactions can be recovered as its ammonium chloride
and regenerated to the ammonium fluoride without loss
of reactivity.
8. Kurosu, M.; Dey, S. S.; Crick, D. C. Tetrahedron Lett.
2006, 47, 4871–4875.
9. Microscopic analyses of surface of these polymer-sup-
ported ammonium fluorides (1a–f) revealed that neither
section nor fragmentation was observed on the spheres of
polymers.
10. No by-products were detected in 1H NMR of HClÆNH2-D-
Ala-D-Ala-OMe.
11. Because N,N-dimethyloctylamine is an inexpensive build-
ing block, 1b was utilized for further studies.
12. (a) Chen, S.; Xu, J. Tetrahedron Lett. 1992, 33, 647; (b)
Carpino, L. A.; Imazumi, H.; El-Faham, A.; Ferrer, F.
J.; Zhang, C.; Lee, Y.; Foxman, B. M.; Henklein, P.;
Acknowledgements
We thank the National Institutes of Health (NIAID
grant AI049151) and Colorado State University for gen-
erous financial support.
Hanay, C.; Mugge, C.; Wenschuh, H.; Klose, J.; Beyer-
¨
mann, M.; Bienert, M. Angew. Chem. Int. Ed. 2002, 41,
442.
13. A large excess of NHS cause decomplexation to leach the
free amine.
References and notes
14. Sauer, D. R.; Kalvin, D.; Phelan, K. M. Organic Lett. 5,
2721.
1. (a) Mahapatra, S.; Yagi, T.; Belisle, J. T.; Espinosa, B. J.;
Hill, P. J.; McNeil, M. R.; Brennan, P. J.; Crick, D. C. J.
Bact. 2005, 187, 2747; (b) Mahapatra, S.; Scherman, H.;
Brennan, P. J.; Crick, D. C. J. Bact. 2005, 187, 2341.
2. Experiments with Mtb MraY indicated that this enzyme
has a high degree of specificity for endogenous prenyl-
phosphate and doses not accept exogenously supplied
prenylphosphate.
3. (a) Ha, Sha.; Chang, E.; Lo, M.-C.; Men, H.; Park, P.; Ge,
M.; Walker, S. J. Am. Chem. Soc. 1999, 121, 8416; (b)
VanNieuwenhze, M. S.; Mauldin, S. C.; Zia-Ebrahimi,
M.; Winger, B. E.; Hornback, W. J.; Saha, S. L.; Aikins, J.
A.; Blaszczak, L. C. J. Am. Chem. Soc. 2002, 124, 3656.
4. (a) We encountered difficulty in purifying the desired
protected pentapeptide (L-Ala-c-D-Glu(OMe)-L-Lys-
(COCF3)-D-Ala-D-Ala-OMe) from by-products generated
on the polymer-support by using a standard Fmoc
strategy (piperidine as a deblocking reagent); (b) Zhang,
A. J.; Russell, D. H.; Zhu, J.; Burgess, K. Tetrahedron
Lett. 1998, 39, 7439.
15. A general procedure for the synthesis of oligopeptide with
the polymer-supported ammonium fluoride 1b. Fmoc-pro-
tected oligopeptides were synthesized using PyBOP
(1.1 equiv) and N-methylmorpholine (2.0 equiv) in DMF.
After standard water work-up, the crude product was
dried under high vacuum. Into the solution of crude
Fmoc-protected oligopeptide in DMF (0.1–0.05 M) was
added 1b (1.5 equiv), and the reaction mixture was gently
stirred for 1 h. The reaction mixture was filtered through a
glass-filter and the remaining PS-ammonium amide anions
were washed with DMF (thrice). The PS-ammonium
amide anions were dried under high vacuum and swelled
in DMF (0.1 M). Into the reaction mixture pyridineÆHCl
(1.0 equiv) was added. After 10 min, the reaction mixture
was filtered through a glass-filter and the solvent was
removed under high vacuum to give the oligopeptide HCl
salt. The recovered PS-ammonium chloride was subjected
to the ion exchange reaction with KF in MeOH to
afford 1b.