Using hydroxymethylphenoxy derivates with the SPOT technology to generate peptides with authentic C-termini

Article abstract of DOI:10.1016/j.bmcl.2008.05.116

The SPOT technology can fulfill most requirements for highly parallel, multiple peptide synthesis of soluble peptides within the upper microgram range. Here, we report on an improved method using hydroxymethylphenoxyacetic acid (HMPA) for 19 amino acids and 4-(4-hydroxymethyl-3-methoxyphenoxy)-butyric acid (HMPB) for proline as acidic labile linkers in SPOT synthesis. Using this approach we could reduce side-chain reactions normally occurring during conventional alkaline peptide cleavage from cellulose membranes. All synthesis steps were adapted to fully-automated SPOT synthesis and therefore represent a time- and cost-saving procedure. Furthermore, the improved cleavage and washing steps resulted in peptides with authentic C-termini in a purity range of 60-95%. Our improved method is ideal for synthesizing many thousand different peptides subsequently used directly for different biological assays requiring authentic C-termini, such as CD8 T-cell epitope screening, vaccine immunization, or tumor imaging.

Full text of DOI:10.1016/j.bmcl.2008.05.116

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4038–4043
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Using hydroxymethylphenoxy derivates with the SPOT technology
to generate peptides with authentic C-termini
*
Bernhard Ay, Katja Landgraf, Mathias Streitz, Stephan Fuhrmann, Rudolf Volkmer, Prisca Boisguerin
Institut für Medizinische Immunologie, Charité—Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
The SPOT technology can fulfill most requirements for highly parallel, multiple peptide synthesis of
soluble peptides within the upper microgram range. Here, we report on an improved method using
hydroxymethylphenoxyacetic acid (HMPA) for 19 amino acids and 4-(4-hydroxymethyl-3-methoxy-
phenoxy)-butyric acid (HMPB) for proline as acidic labile linkers in SPOT synthesis. Using this approach
we could reduce side-chain reactions normally occurring during conventional alkaline peptide cleavage
from cellulose membranes. All synthesis steps were adapted to fully-automated SPOT synthesis and
therefore represent a time- and cost-saving procedure. Furthermore, the improved cleavage and washing
steps resulted in peptides with authentic C-termini in a purity range of 60–95%. Our improved method is
ideal for synthesizing many thousand different peptides subsequently used directly for different biolog-
ical assays requiring authentic C-termini, such as CD8 T-cell epitope screening, vaccine immunization, or
tumor imaging.
Received 4 April 2008
Revised 29 May 2008
Accepted 31 May 2008
Available online 5 June 2008
Keywords:
CDI
CDT
SPOT synthesis
T-cell epitope mapping
HMPA
HMPB
Ó 2008 Elsevier Ltd. All rights reserved.
Discovering new targets for vaccines and novel vaccination
methods are current research goals in immunology.1,2 Besides
the well-known combinatorial approaches,3 huge arrays of
sequence-based peptides (e.g., virus proteome-based) are powerful
tools for proteome-wide T-cell epitope mapping studies seeking to
deliver new targets for vaccine development. Recently, we explic-
itly demonstrated such a knowledge-based approach, revealing
novel peptide vaccines even in the case of CD8 T-cell responses.4
(up to 50%)4 and can be decomposed to the dehydroalanine moi-
ety.4,15 Another drawback is the necessity to neutralize the alkaline
peptide solutions, which result in high salt concentration impuri-
ties. Time-consuming desalting steps are essential not only for cor-
rectly determining the peptide concentration but also for salt-free
biological assays.4
To overcome these problems, we present here an improved
method to generate cleavable peptides with free C-termini follow-
ing the SPOT procedure, but using acid labile-linker molecules
(hydroxymethylphenoxy derivates). It was reported that these
linkers are used for SPPS on spherical cellulose resin,16 but to our
surprise we found no report on the use of hydroxymethylphenoxy-
acetic acid (HMPA) in SPOT technologies. We synthesized peptide
sequences containing 19 different C-terminal amino acids directly
on the HMPA-linker modified cellulose membrane with efficient
yields and purity using 1,10-carbonyl-di-imidazole (CDI) or 1,10-
carbonyl-di-(1,2,4-triazole) (CDT) as activators. For proline we
additionally tested 4-(4-hydroxymethyl-3-methoxyphenoxy)-bu-
tyric acid (HMPB) to avoid diketopiperazine formation. The synthe-
sis protocol was further adapted to the fully-automated SPOT
synthesis technology, allowing fast and efficient synthesis of solu-
ble peptides with authentic C-termini. Finally, the novel synthesis
approach was successfully validated in a biological assay that
determined the stimulatory activity of the CD8 T-cell epitope
VTEHDTLLY (VPAP protein, SWISS-PROT Accession No. P16790)
from human cytomegalovirus (CMV).4
A
modified SPOT technology synthesis protocol enabled the
synthesis of thousands of human cytomegalovirus virus-derived
peptides with free and authentic C-termini, a prerequisite for a
CD8 T-cell response.5,6
In addition to its original use—the synthesis and screening of
cellulose membrane-bound peptide arrays (for a review see7–9)—
SPOT technology10 has emerged as an efficient micro-scale synthe-
sis method. Thousands of soluble peptides can be synthesized in a
short time with microgram yields and purities sufficient for biolog-
ical screening assays11–14—one advantage compared with com-
monly used large-scale SPPS. Usually, peptides are cleaved from
the cellulose support by alkaline hydrolysis or amminolysis11,13
:
the former providing peptides with a carboxylated C-termini, the
latter producing the corresponding carboxyl-amide. However,
alkaline conditions lead to several unwanted side reactions, pri-
marily oxidation, to different extents, of cysteine and methionine
sulfur atoms. Cysteine is especially affected by strong oxidation
The well-known linker molecule HMPA is an ideal anchor
for the SPOT synthesis technology due to its stability against
* Corresponding author.
E-mail address: prisca.boisguerin@charite.de (P. Boisguerin).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.05.116

B. Ay et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4038–4043
4039
piperidine and sensitivity to acidic conditions, for example, >50%
trifluoroacetic acid (TFA).18 Scheme 1 describes the modification
steps applied to a cellulose membrane (detailed protocol is given
in Supplementary material). First, the membrane was treated with
Fmoc-b-alanine as described17 resulting in the amino-functional-
ized support, called matrix 1 (800–1000 nmol/cm2). Second, the
bifunctional HMPA molecule (0.6 M in DMF) was attached spot-
wise to the surface of 1. Several activation reagents were tested;
finally 1.1 equiv of 2-ethoxy-1-ethoxycarbonyl-1,2-dihydroquino-
line (EEDQ) was chosen as the most favored candidate due to its
long-lived active ester (Scheme 1, step i). Figure 1A shows that a
3- or 4-times coupling approach is sufficient for HMPA attachment.
Third, in analogy to our recent publication14 the ester bound be-
brane-bound HMPA (2) was formed with the aid of 3 equiv of
CDI or 3 equiv of CDT (Scheme 1, step ii). CDT was used for the ami-
no acids glutamic acid, proline, and tyrosine. The coupling effi-
ciency was determined by measuring the Fmoc-piperidine adduct
concentration as described previously.17 Figure 1B summarizes
the coupling efficiency of all 20 Fmoc-L-amino acids. We obtained
coupling yields between 140 and 450 nmol/cm2 generally using 4-
times coupling, but specifically 8-times coupling for glutamic acid,
proline, and tyrosine using freshly prepared reagents. Fourthly,
after an additional acetylating step (Scheme 1, step iii) peptides
were synthesized according to the standard SPOT synthesis proto-
col.17 Finally, the spots were punched out and transferred into
eppendorf tubes or deep-well plates. The peptides were released
from the support and side chains were deprotected by the follow-
tween the first L-amino acid (0.4 M, in DMF) and cellulose mem-
Scheme 1. Reaction scheme for the synthesis of cleavable peptides with authentic C-termini using acid-label HMPA on cellulose membrane. Reaction conditions: (i) 0.6 M
HMPA in DMF activated with 1.1 equiv of EEDQ was directly coupled (3–4 times) onto the b-alanine-modified cellulose membrane (1) resulting in 2. (ii) After acetylation [2%
Ac2O in DMF] of the remaining amino-group of the b-alanine, 0.4 M Fmoc-L-amino acid-OH in DMF activated with 3 equiv of CDI/CDT was coupled (4–8 times) to generate the
cleavable site (3). (iii) The remaining OH-groups of HMPA were acetylated [2% Ac2O, 1% DIPEA in DMA]. Thereafter, peptide elongation was carried out by standard SPOT
synthesis17 (4). (iv) Using our improved cleavage/washing step procedure [2.5 h 60% TFA, 3% triisobutylsilane (TIBS), 2% H2O in DCM; vacuum dry; 1.0 h 90% TFA, 3% TIBS, 2%
H2O in DCM; vacuum dry; 1ꢀ H2O wash, 3ꢀ tert-butylmethylether wash, vacuum dry and dissolve in degassed acetonirile/H2O (1:1)] we obtained peptides with authentic
C-termini.
Figure 1. Determining the coupling efficiencies of the first two steps during SPOT synthesis with HMPA. (A) 0.6 M HMPA in DMF activated with 1.1 equiv of EEDQ was
coupled 1–4 times on a b-alanine-modified cellulose membrane. Thereafter, 0.4 M Fmoc-
L
-glycine-OH in DMF activated with 3 equiv of CDI was coupled 4-times.
-amino acid-OH in DMF activated with 3 equiv
Quantification was achieved by measuring the Fmoc-piperidine complex17 cleaved from one spot (0.25 cm2). (B) 0.4 M Fmoc-
L
of CDI, coupled 4 times. Amino acids denoted with * were coupled 8 times and were activated with 3 equiv of CDT. Quantification was achieved by measuring the Fmoc-
piperidine complex17 cleaved from one spot (0.25 cm2). The horizontal line determines the threshold of 100 nmol/cm2.

4040
B. Ay et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4038–4043
ing protocol (Scheme 1, step iv): (a) Peptides were cleaved from
the support by treatment with a mixture of 60% TFA for 2.5 h. (b)
The cellulose spots were then removed and the solution was sub-
sequently dried in a vacuum. (c) The remaining residuum was once
again treated with a mixture of 90% TFA for 1 h and again dried in a
vacuum. (d) Remaining TFA–salt complexes were removed with
H2O by azeotropic distillation. (e) The residue was then washed 3
times with tert-butylmethylether and dried. It should be men-
tioned that following exactly the order of the TFA-cleavage steps
is important to obtain optimal yields. During the 60% TFA solution
procedure the ester linkage between HMPA and the peptide is
cleaved, but without releasing the HMPA moiety from the cellulose
membrane. After separation from the cellulose spot we further en-
sure complete peptide side-chain deprotection with 90% TFA. Sep-
aration of the cellulose spot from the first cleavage mixture (60% in
DCM) significantly reduces the amount of HMPA complex impuri-
ties18, while the washing steps are necessary to remove byproducts
and TFA–salt complexes. This is confirmed in Figure 2A, which
shows the HPLC elution profile and ESI mass spectra of the synthe-
sized peptide WKL-G. The HPLC chromatogram depicts the pep-
tide’s excellent purity, while its integrity is demonstrated by
mass spectrometry. Greater amounts of impurities are not detected
in either HPLC or mass spectrometry.
using the standard Fmoc-protocol.17 To enable comparison with
our previously reported method14, we sequentially choose as X,
the amino acids alanine, aspartic acid, methionine, lysine, and
tryptophan. We analyzed the HPLC retention time of each pure
di-peptide and each mixture (L/L and L/D). Then we synthesized
the same L–X di-peptide on a HMPA/b-alanine-modified cellu-
lose membrane using our improved conditions. These di-pep-
tides were subsequently cleaved using standard TFA cleavage
conditions17 and also analyzed by HPLC. We determined the
percentage of racemization by peak area integration, and only
detected a small amount, between 1% and 8%, of the diastereo-
meric
L/D-peptide (Supplementary material Table 1). This is
clearly below the racemization found with the coupling re-
agents MSNT and HOBt-ester or the symmetric anhydride19
and within the range of the directed cellulose coupling
method.4,14
The improved protocol was used to synthesize the peptides
WKL-X, where X represents all 20 gene-encoded amino acids. Mea-
sured and expected masses of these peptides, putative byproducts,
as well as peptide purities are presented in Table 1. Our analyses
revealed that purities ranged between 60% and >95%. As expected,
our results demonstrate that the formation of dehydroalanine in
the cysteine-containing peptide4,15 was totally suppressed by the
novel protocol, and that oxidation was reduced to 10%. Methionine
oxidation, which occurs during standard solid-phase synthesis,20
was reduced to less than 50% (Table 1). Furthermore, we could ex-
clude any C-terminal oligomerization of the synthesized peptides.
One known problem of amino acid carboxy-group activation
is the resulting racemization. To determine racemization levels
during coupling of the first amino acid to the HMPA molecule,
we synthesized di-peptides
(L)L–(L)X and (L)L–(D)X on resin
Figure 2. Model peptides synthesized using the improved method. (A) HPLC chromatogram and ESI mass spectra of the model peptide WKL-G revealed no C-terminal
elongation during CDI-coupling. For details of peptide synthesis see Scheme 1. (B) HPLC chromatograms of the peptides APINSNPPNTFV, WVYFPPQYAYLT and YSQLAKLGINPY
showing purities >60%.

B. Ay et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4038–4043
4041
Table 1
were measured as Böhringer Light Units (BLU in Table 2). In all
cases, we could measure different signal intensities that approxi-
mately correspond to peptide yields. Based on these results, for
automated SPOT synthesis we recommend using a HMPB linker if
proline is the C-terminal amino acid (HMPB is 2-fold more expen-
sive than HMPA).
To assess the quality of SPOT synthesized peptides using the
HMPA-linker approach, we used the peptide VTEHDTLLY, known
as a CMV-derived CD8 T-cell epitope (VPAP protein), as a model.
T-cell stimulation by this peptide was determined by flow
cytometry (detailed protocol is given in Supplementary mate-
rial). First, the peptide sequence VTEHDTLLY was synthesized
by conventional Fmoc-chemistry on resin and purified by HPLC.
Then the same peptide sequence was synthesized by our novel
HMPA-linker-aided SPOT technology. Subsequently, both pep-
tides were tested in parallel for T-cell stimulation measured by
flow cytometry.
Characterization of model peptides by ESI mass spectrometry and analytical HPLC
Sequences
Expected [M] Measured [M+H+]+ Puritya (%) Byproductsb
WKL-A
WKL-C
WKL-D
WKL-E
WKL-F
WKL-G
WKL-H
WKL-I
516.30
548.28
560.29
574.30
592.33
502.28
582.32
558.34
573.35
558.34
576.30
559.30
573.32
601.36
532.29
546.31
544.33
631.34
608.32
1269.62
517.28
549.22
561.24
575.22
593.27
503.26
583.28
559.28
574.28
559.27
577.28
560.29
574.29
602.35
533.22
547.29
545.28
632.32
609.30
1270.57
1547.83
1366.79
>95
80
<1
15
<1
n.d.
<1
<1
n.d.
<1
n.d.
<1
10
n.d.
n.d.
n.d.
n.d.
<1
<1
n.d.
<1
>95
>95
>95
>95
>95
>95
>95
>95
65
WKL-K
WKL-L
WKL-M
WKL-N
WKL-Q
WKL-R
WKL-S
WKL-T
WKL-V
WKL-W
WKL-Y
APINSNPPNTFV
>95
80
>95
>95
>95
>95
80
Peripheral blood mononuclear cells (PBMC) were obtained
from CMV (AD169) sero-positive healthy donors (HLA 0101*).
The PBMC assay (5 million cells, stimulated overnight) was per-
formed as described previously4 using a peptide concentration of
90
70
n.d.
n.d.
n.d.
WVYFPPQYAYLT 1546.74
YSQLAKLGINPY
79
1365.71
68
1
lg/ml. A positive signal corresponds to CD8 and interferon
c
a
Purities were determined by HPLC peak area integration.
Byproducts represented C-terminal elongations and/or oxidation of cysteine/
(INF ) signals over a threshold of 0.03% of the total CD8 T-cells.
c
b
Figure 4 presents the stimulation effect of both synthesized pep-
tides (Fig. 4C and D) compared to the controls, DMSO (as a sol-
vent, Fig. 4A), and HMPA (as a potential impurity, Fig. 4B).
Neither control had any impact on T-cells (both signals below
<0.03%). Both the peptide prepared by standard resin-based
Fmoc-chemistry (Fig. 4C) and the peptide synthesized according
to the HMPA-linker-aided SPOT technology (Fig. 4D) resulted in
high T-cell stimulation. Furthermore, the stimulation values of
0.270% and 0.210% for these peptides are within the same range,
thus directly validating our synthesis strategy.
Taken together, using both the hydroxymethylphenoxy derived
linkers and the automated synthesizer, it is now possible to gener-
ate huge sets of soluble peptides without restriction, in the upper
microgram range (approximately 5000 peptides) within one week,
for example, for screening CD8 T-cell epitopes. This represents a
time- and cost-saving advantage to the conventional large-scale
SPPS.
methionine; n.d., not detectable.
Such a side reaction was previously described as a problem during
CDI/CDT activation.14
In addition, these results could be reproduced with three differ-
ent, randomly chosen 12-mer peptide sequences, whose purities
are also >65% (Fig. 2B and Table 1) and many other sequences (data
not shown).
Synthesis of the WKL-P sequence turned out to be more compli-
cated due to the well-known formation of diketopiperazine. The
extent of this side reaction is generally limited, but increases mark-
edly when the cis-conformation of the peptide bond is favored, for
example, in peptides with a C-terminal proline.21,22
To prevent diketopiperazine formation we analyzed another
acid-cleavable linker: 4-(4-hydroxymethyl-3-methoxyphenoxy)-
butyric acid (HMPB). First, we determined the best activation re-
agent, which proved to be 1.1 equiv of EEDQ for HMPB, just as
for HMPA.
Thereafter, we synthesized the sequences WKL-G (as a control),
KDT-P (the protected T should avoid cleavage), and WKL-P. By
measuring the Fmoc-piperidine adducts of the first C-terminal
amino acid, we could follow the coupling efficiencies of the first
amino acid on the linker (data not shown). MS measurements re-
vealed the expected mass for WKL-G and KDT-P (although peak
intensities were very low) using both linkers. The mass peak of
WKL-P could only be detected using HMPB.
Nevertheless, there should still be enough peptide for a biolog-
ical screen. To prove this hypothesis, we synthesized a His-tag on
the three sequences using SPOT synthesis. The cleaved peptides
were then printed on a solid support (e.g., PVDF-membrane) and
detected by antibody incubation (Fig. 3). The signal intensities
Here, we demonstrate in detail that introducing the HMPA lin-
ker during the SPOT synthesis procedure, combined with an opti-
mized cleavage/washing procedure, results in peptides with
authentic C-termini without racemization or oligomerization of
the C-terminal amino acid, and with excellent purity. For peptides
with proline as C-terminal amino acid, we advise using HMPB to
Table 2
Characterization of proline sequences in comparison to glycine sequences
Linker
Seq.
Tag
Massexp.
Massfound
BLU
HMPA
HMPA
HMPB
HMPB
HMPA
HMPA
HMPB
HMPB
HMPA
HMPA
HMPB
HMPB
WKL-P
WKL-P
WKL-P
WKL-P
KDT-P
KDT-P
KDT-P
KDT-P
WKL-G
WKL-G
WKL-G
WKL-G
—
542.3
1227.6
542.3
1227.6
459.2
1144.5
459.2
1144.5
502.3
—
—
HHHHH
—
HHHHH
—
HHHHH
—
HHHHH
—
16,111
543.31
410.22**
460.22
382.53**
460.23
573.30*
503.21
594.8*
503.28
594.8*
73,588
29,574
171,811
617,774
1,455,500
HHHHH
—
HHHHH
1187.6
502.3
1187.6
Massexp., expected mass; Massfound, measured mass using ESI mass spectrometry
[M+H+]+; BLU, Boehringer Light Units detected by blot analysis and measured using
LuniImiger.
Figure 3. Test for biological activity. His-tagged peptides as listed in Table 2 were
spotted on PFDV-membranes and detected using anti-polyHis/anti-mouse-HRP
antibodies. (For details see Supplementary material.)
*
[M+2H+]++
.
[M+3H+]+++
.
**

4042
B. Ay et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4038–4043
Figure 4. Validating the biological activity of the peptide VTEHDTLLY. The plots of the FACS measurements show the interferon
depicted on the top left panel, we could divide the graph into four parts: (1) lymphocytes without INF or CD8 signals; (2) T-cells without INF
lymphocytes with INF but without CD8 signals and (4) lymphocytes with both INF and CD8 signals. A signal is significant if more than 0.03% of the total measured T-cells
gave a positive signal for INF and CD8. (A) T-cell stimulation with DMSO as a negative control (0.011%). (B) T-cell stimulation with HMPA (0.006%) as a second negative
c
(IFNc
) signal versus the CD8 signal. As
c
c
but with CD8 signals; (3)
c
c
c
control. (C) T-cell stimulation with the resin synthesized peptide as a positive control (0.270%) is in same range as the cellulose synthesized peptide generated using our
improved conditions (0.210%) (D).
decrease diketopiperazine formation and increase peptide yields.
Additionally, the acidic cleavage conditions decrease cysteine and
methionine oxidation and also represent a time-saving step due
to simultaneous cleavage and side-chain deprotection of the pep-
tides. These advantages extend the SPOT technique to an ideal pep-
tide synthesis tool for any biological assay.
References and notes
1. Paston, S. J.; Dodi, I. A.; Madrigal, J. A. Hum. Immunol. 2004, 65, 544.
2. De Groot, A. S. Drug Discovery Today 2006, 11, 203.
3. Beck-Sickinger, A. G.; Jung, G. From multiple peptide synthesis to peptide
libraries. In Combinatorial Peptide and Nonpeptide librariesLibraries; Jung, G., Ed.;
VCH Verlagsgesellschaft: Weinheim, 1996; p 79.
4. Ay, B.; Streitz, M.; Boisguerin, P.; Schlosser, A.; Mahrenholz, C. C.; Schuck, S. D.;
Kern, F.; Volkmer, R. Biopolymers 2007, 88, 64.
5. Kiecker, F.; Streitz, M.; Ay, B.; Cherepnev, G.; Volk, H. D.; Volkmer-Engert, R.;
Kern, F. Hum. Immunol. 2004, 65, 523.
Acknowledgments
6. Kern, F.; Surel, I. P.; Brock, C.; Freistedt, B.; Radtke, H.; Scheffold, A.; Blasczyk,
R.; Reinke, P.; Schneider-Mergener, J.; Radbruch, A.; Walden, P.; Volk, H. D. Nat.
Med. 1998, 4, 975.
7. Frank, R. J. Immunol. Methods 2002, 267, 13.
8. Wenschuh, H.; Volkmer-Engert, R.; Schmidt, M.; Schulz, M.; Schneider-
Mergener, J.; Reineke, U. Biopolymers 2000, 55, 188.
Financial support from Deutsche Forschungs-gemeinschaft (VO
885/3-1) is gratefully acknowledged. We thank Ines Kretzschmar
for the continuous synthesis of high quality peptides.
9. Reineke, U.; Volkmer-Engert, R.; Schneider-Mergener, J. Curr. Opin. Biotechnol.
2001, 12, 59.
10. Frank, R. Tetrahedron 1992, 48, 9217.
Supplementary data
11. Kamradt, T.; Volkmer-Engert, R. Mol. Divers. 2004, 8, 271.
12. Hilpert, K.; Volkmer-Engert, R.; Walter, T.; Hancock, R. E. Nat. Biotechnol. 2005,
23, 1008.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.05.116.

B. Ay et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4038–4043
4043
13. Bhargava, S.; Licha, K.; Knaute, T.; Ebert, B.; Becker, A.; Grotzinger, C.;
Hessenius, C.; Wiedenmann, B.; Schneider-Mergener, J.; Volkmer-Engert, R.
J. Mol. Recognit. 2002, 15, 145.
14. Ay, B.; Volkmer, R.; Boisguerin, P. Tetrahedron Lett. 2007, 48, 361.
15. Nakanishi, T.; Sato, T.; Sakoda, S.; Yoshioka, M.; Shimizu, A. Biochim. Biophys.
Acta 2004, 1698, 45.
membranes. In Combinatorial Chemistry; Fenniri, H., Ed.; Oxford–NY, 2000;
pp 95–116.
18. Bui, C. T.; Ercole, F.; Pham, Y.; Campbell, R.; Rasoul, F. A.; Maeji, N. J.; Ede, N. J.
J. Pept. Sci. 2000, 6, 534.
19. Blankemeyer-Menge, B.; Nimtz, M.; Frank, R. Tetrahedon Lett. 1990, 31,
1701.
16. Englebretsen, D. R.; Harding, D. R. Int. J. Pept. Protein Res. 1994, 43, 546.
17. Wenschuh, H.; Gausepohl, H.; Germeroth, L.; Ulbricht, M.; Matuschewski, H.;
Kramer, A.; Volkmer-Engert, R.; Heine, N.; Ast, T.; Scharn, D.; Schneider-
Mergener, J. Positionally addressable parallel synthesis on continuous
20. Grant, G. A. Synthetic Peptides: A User’s Guide; W.H. Freemann and Company:
New York, 1992.
21. Khosla, M. C.; Smeby, R. R.; Bumpus, F. M. J. Am. Chem. Soc. 1972, 94, 4721.
22. Gisin, B. F.; Merrifield, R. B. J. Am. Chem. Soc. 1972, 94, 3102.