Synthesis and screening of peptoid arrays on cellulose membranes

Article abstract of DOI:10.1016/j.tet.2003.10.044

Rapid synthesis and screening of compound libraries enables the accelerated identification of novel protein ligands in order to support processes like analysis of protein interactions, drug target discovery or lead structure discovery. SPOT synthesis - a well established method for the rapid preparation of peptide arrays - has recently been extended to the field of nonpeptides. In this contribution we report on the systematic evaluation of the SPOT technique for the assembly of N-alkylglycine (peptoid) library arrays. In the course of this investigation bromoacetic acid 2,4-dinitrophenylester (1a) was identified to be the most suited agent for bromoacetylation in terms of yield and N-selectivity enabling straightforward submonomer synthesis on hydroxy-group rich cellulose membranes. The potential of this method for the rapid identification of novel nonpeptidic protein ligands was demonstrated by synthesis and screening of a library consisting of 8000 peptoids and peptomers (i.e. their hybrids with α-substituted amino acids) allowing the identification of micromolar ligands for the monoclonal antibody Tab-2.

Full text of DOI:10.1016/j.tet.2003.10.044

Tetrahedron 59 (2003) 9919–9930
Synthesis and screening of peptoid arrays on cellulose membranes
a,†
Niklas Heine, Thomas Ast, Jens Schneider-Mergener, Ulrich Reineke, Lothar Germeroth
b
a,b
b
b
b,
*
and Holger Wenschuh
a
Institute f. Med. Immunol., Charit e´ , Humboldt-University, Schumannstr. 20/21, 10117 Berlin, Germany
Jerini AG, Invalidenstr. 130, 10115 Berlin, Germany
b
Received 13 August 2003; revised 13 October 2003; accepted 13 October 2003
Abstract—Rapid synthesis and screening of compound libraries enables the accelerated identification of novel protein ligands in order to
support processes like analysis of protein interactions, drug target discovery or lead structure discovery. SPOT synthesis—a well established
method for the rapid preparation of peptide arrays—has recently been extended to the field of nonpeptides. In this contribution we report on
the systematic evaluation of the SPOT technique for the assembly of N-alkylglycine (peptoid) library arrays. In the course of this
investigation bromoacetic acid 2,4-dinitrophenylester (1a) was identified to be the most suited agent for bromoacetylation in terms of yield
and N-selectivity enabling straightforward submonomer synthesis on hydroxy-group rich cellulose membranes. The potential of this method
for the rapid identification of novel nonpeptidic protein ligands was demonstrated by synthesis and screening of a library consisting of 8000
peptoids and peptomers (i.e. their hybrids with a-substituted amino acids) allowing the identification of micromolar ligands for the
monoclonal antibody Tab-2.
q 2003 Elsevier Ltd. All rights reserved.
2
,3
1
. Introduction
based on solid phase methods. Two general techniques
have been established to generate complex libraries which
use either pooling strategies resulting in compound
The tremendous developments in the identification of novel
drug targets as well as the growing interest in understanding
protein–protein and protein–ligand interactions as central
biological processes has created an increased demand for
4
–7
mixtures
or highly paralleled synthesis procedures to
8
yield discrete compounds. A spatially addressed synthetic
methodology for single compounds offers the advantage
that the sequence of an oligomer can easily be identified by
its position and laborious tagging or deconvolution
procedures are not needed.
1
peptidic and peptidomimetic libraries for target validation
and drug discovery. Besides recent developments of
biological library techniques such as phage display,
ribosome display, and retroviral intracellular libraries,
rapid and effective parallel chemical synthesis of peptide
and furthermore peptidomimetic libraries is still a subject of
intense research and technology development. The majority
of approaches dealing with the chemical synthesis of large
numbers of peptides and related oligomeric compounds is
A technically simple and economic method was developed
9
according to the SPOT synthesis concept by Frank. This
method displays a high flexibility compared to other
multiple solid phase procedures especially with regard to
1
0 – 13
miniaturization and array geometries.
The basic
principle involved the positionally addressed delivery of
small volumes of reagent solutions directly onto a coherent
membrane sheet. The resulting droplets can be considered
as micro-reactors provided that a non-volatile solvent
system is used. The functional groups fixed on the
membrane surface react with the pipetted reagents and
conventional solid phase syntheses (SPS) occur. The
volumes dispensed as well as the physical properties of
the membrane surface and the solvent system define the size
of the resulting SPOTs and together with the minimum
distance between the SPOTs the number of oligomers which
can be synthesized per membrane area. The methodology
Keywords: combinatorial chemistry; epitope mimic; peptoid array; N-
selective acylation; solid-phase synthesis; SPOT synthesis.
Abbreviations: Ac, acetyl; ar, aromatic; Bu, butyl; Cy, cyclohexyl; DBU,
1
,8-diazabicyclo[5,4,0]undec-7-ene; DIC, diisopropylcarbo-diimide;
DIEA, diisopropylethylamine; DMF, N,N-dimethylformamide; Dnp, 2,4-
dinitrophenyl; EDTA, ethylendiaminetetraacetic acid; eq., equivalents; Et,
ethyl; Fmoc, 9-fluorenyl-methyloxycarbonyl; HBS-EP, HEPES-buffered
saline with EDTA- and polysorbate added; HEPES, 4-(2-hydroxyethyl)-
piperazine-1-ethanesulfonic acid; IgG, immune globuline G; Me, methyl;
NMP, N-methylpyrrolidone; Ph, phenyl; RU, resonance units; SPR, surface
plasmon resonance; TFA, trifluoroacetic acid; TIPS, triisopropylsilane;
TRIS, tris-(hydroxymethyl)-aminomethane.
14–20
has recently been reviewed.
*
Corresponding author. Tel.: þ49-30-97893-200; fax: þ49-30-97893-299;
e-mail: wenschuh@jerini.de
Present address: Boehringer Ingelheim Pharma GmbH & Co. KG, Dept.
of Chemical Research, D-88397 Biberach, Germany.
†
The general strategy for parallel assembly of oligomers on
cellulose membranes using the SPOT technique comprises
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2003.10.044

9920
N. Heine et al. / Tetrahedron 59 (2003) 9919–9930
Scheme 1. Key steps outlining the SPOT synthesis technique (PG¼side chain protecting group).
the following steps (Scheme 1): (i) cellulose derivatisation
enabling selective attachment of activated building blocks;
SPOT technique for the assembly of N-alkylglycine
2
9
(peptoid ) library arrays. Reaction conditions needed to
be adapted for SPOT synthesis protocols on cellulose
membranes to circumvent problems related to the different
kinetic situation on continuous surfaces and the large excess
of free hydroxy-functions on the cellulose. In the course of
this investigation bromoacetic acid 2,4-dinitrophenyl-ester
(1a) was found to be best suited for N-selective bromo-
acetylations being a central part in the submonomer
(
ii) attachment of linkers in cases where cleavage of
products is required for analysis or solution phase assays;
iii) SPOT synthesis of the oligomer; (iv) cleavage of side
(
chain protecting groups while retaining the covalent
membrane linkage; and (v) solid phase screening directly
on the membrane bound oligomers or (vi) cleavage from the
membrane for analysis.
3
0
synthesis concept. Most effective reaction conditions
were applied to the synthesis of an array consisting of
8000 hexapeptoids as well as -peptomers, their hybrids with
After numerous applications of the SPOT synthesis
1
4,15,17
concept for peptide synthesis,
recently extended to the synthesis of other chemical
the method was
3
1
a-substituted amino acids. This library was screened for
binding of the anti-transforming growth factor a (TGFa)
2
1,22
23
entities such as PNAs,
2
glycoconjugates, whole
protein domains, as well as small heterocyclic com-
4
32
monoclonal antibody (mab) Tab-2 in order to identify
2
5–27
pounds.
bioactive compounds de novo. Mab Tab-2 is a well
established model system used to evaluate a variety of
structurally different types of ligand arrays prepared by
2
7,28
In extension of preliminary communications the present
contribution reports on the systematic evaluation of the
1
1,12,25,33–36
SPOT synthesis.
29
30
Scheme 2. Monomer vs submonomer peptoid synthesis according to Simon et al. and Zuckermann et al., respectively.

N. Heine et al. / Tetrahedron 59 (2003) 9919–9930
9921
2
. Results and discussion
.1. Monomer and submonomer peptoid synthesis
Since peptoids were introduced to combinatorial chemistry,
large excess of cellulose OH-groups competing with the N-
terminal peptoid amino groups. In order to avoid undesired
acylation of the membrane support as well as unprotected
OH-groups of the growing oligomer chain an N-selective
acylation procedure would be favorable.
2
2
9
30
both monomer and submonomer approaches have been
successfully applied for their synthesis (Scheme 2).
Whereas the monomer strategy allows a direct application
of common Fmoc based peptide synthesis protocols due to
the use of preformed building blocks, the latter approach
makes use of a two step procedure comprising coupling of
bromoacetic acid and subsequent displacement of the
halogen by primary amines. Although the application of
Fmoc-N-alkylglycine building blocks appears to be useful
in some cases, e.g. to introduce functional groups
inaccessible by submonomer procedures, the monomer
strategy displays some major drawbacks such as limited
availability of building blocks and coupling difficulties
related to the high sterical hindrance at the alkylamino
function. Accordingly, adaptation of the submonomer
approach to the conditions of SPOT synthesis was attempted
to utilize the enormous number of readily available primary
amines as source of diversity and to avoid acylation
problems at the N-alkylglycine moiety. However, despite
of the advantages of the submonomer method for the
assembly of oligomers consisting exclusively of N-sub-
stituted glycines a combination with the monomer strategy
is useful in cases were peptoids and peptides are mixed in
one sequence (peptomers) or when Fmoc protected N-alkyl-
glycines are used to increase the diversity. In order to
perform both strategies in parallel within a single library a
Since active esters of carboxylic acids have been demon-
strated to show a significant N-selectivity in the presence of
4
1,42
free OH-groups,
several of such esters were prepared
for bromoacetic acid and tested for reactivity and selectivity
using SPOT synthesis conditions. Interestingly, it was found
that commonly used esters of lower reactivity such as
4
3,44
45,46
bromoacetic acid 4-nitrophenyl
or N-succinimidyl
ester gave significant amounts of side products resulting
from alkylation of the N-terminal secondary amino function
(Path B vs. A in Scheme 3). Accordingly, it was assumed
that reactivity of active esters needs to exceed a certain level
Scheme 3. Possible reaction pathways on the bromoacetylation of
gm r eo mu pb ) r. ane bound peptoid 2 with activated bromoacetic acid (X¼leaving
2
8
special Fmoc synthesis protocol is necessary as described.
2
.2. N-Selective bromoacetylation
The effective bromoacetylation of solid support bound
amines or N-alkylamines represents a critical reaction step
of the submonomer strategy. Several accounts reported on
the successful application of the carbodiimide activation of
3
0,31,37–40
bromoacetic acid.
While this procedure is the
Scheme 4. Synthesis of bromoacetic acid 2,4-dinitrophenylester (1a) and
for 1a;
method of choice in SPS on resin beads the situation on
cellulose membranes is different due to the presence of a
the corresponding pentafluorophenylester 1b. (a) pyridine/CH
DIEA/CH Cl for 1b.
Cl
2 2
2
2
Table 1. Tables 1 SPOT synthesis of tripeptoid 3 using different conditions for bromoacetylation of the hydroxy dipeptoid 2
a
b
c
Entry
–X
Yield 3 [%]
Purity 3 [%]
N-Selectivity (3: 4)
d
1
2
3
–ODnp (1a)
–OH/DIC
–OPfp (1b)
81
73
67
90
93:7
88:12
88:12
e
79
75
a
Membrane derivatisation with linker;100%, product was quantified by HPLC (peak area at 220 nm).
HPLC, 220 nm.
Ratio of yields of compounds 3 and 4.
b
c
d
e
See Figure 1 for corresponding HPLC-MS.
2
2
M Br-CH COOH, 1 M DIC.

9922
N. Heine et al. / Tetrahedron 59 (2003) 9919–9930
Scheme 5. SPOT synthesis of the hydroxytripeptoid 3 and the corresponding side chain O-acylated side product. (a) 1a (2£15 min); (b) n-butylamine
3£15 min); (c) ethanolamine (3£15 min); (d) Br-CH COX (for X see Table 1, 2£15 min); (e) piperidine (3£15 min); (f) TFA (95% in H O, 45 min) then
evaporation.
(
2
2
Figure 1. HPLC-chromatogram (220 nm) and MS-spectra corresponding to the annotated signals obtained from the SPOT synthesis of tripeptoid 3 using
bromoacetic acid 2,4-dinitrophenylester (1a) as acylating reagent (the signal at t¼2 min is the injection-peak).
to provide predominant formation of the acylation products.
Thus, the more reactive, stable 2,4-dinitro- and pentafluoro-
with piperidine (see path C in Scheme 3). The ratios of both
products were determined by HPLC and the N-selectivities
of the acylating species applied were subsequently deduced
‡
phenyl esters (1a and 1b) were synthesized in high yields
starting from bromoacetic acid bromide and the correspond-
k
(Table 1). It was found that all three reagents gave rise to
§
,{
ing electron deficient phenols (Scheme 4).
The
the desired N-acylated product 3 in purities between 75 and
90% (by HPLC, 220 nm) and yields ranging from 67 to 81%
(by quantitative evaluation of the UV-absorbtion). Bromo-
acetic acid 2,4-dinitrophenylester (1a) showed the best
results (see Fig. 1) and was superior to the other reagents
applied in terms of N-selectivity as well (93:7) therefore
being the reagent of choice for submonomer peptoid
synthesis on cellulose membranes.
reactivities and N-selectivities of these esters were
evaluated under conditions of the SPOT synthesis and
compared to carbodiimide activation using the synthesis of a
tripeptoid as a simple model system. The key step involved
acylation of the membrane bound dipeptoid 2 (Scheme 5)
with the corresponding active esters of bromoacetic acid.
Whereas the intended exclusive acylation of the N-alkyl-
amine followed by bromine substitution with piperidine
yields the tripeptoid 3; varying proportions of tripeptoid 4
result from partial O-acylation at the side chain and
subsequent nucleophilic substitutions of both bromines
2.3. Suitability of submonomer building blocks
In order to examine the applicability of different amines for
SPOT synthesis of peptoids, an array consisting of
5
0
‡
§
{
47
Application of 1a was previously published. However, no synthetic or
tripeptoids 5 was synthesized on a Rink-linker modified
cellulose membrane based on a variety of amines as
building blocks for the central position. The amines were
applied in large excess compared to the membrane loading:
analytical data were provided. The isolated ester 1b has—to our
knowledge—not been described yet, although it has been applied
48
generated in situ.
Ester 1a is crystalline while ester 1b was isolated as an oil. No
1
decomposition was observed ( H NMR) upon storage for at least 4
months at 208C under exclusion of light (1a) or at 48C in a sealed bottle
k
Compounds 3 and 4 were independently synthesized on Tentagel-S-
RAM-resin as HPLC-standards using the submonomer procedure.
37
˚
with some 4 A molecular sieve added (1b), respectively.
Attempts to synthesize bromoacetic acid esters with
hydroxybenzotriazole or 7-aza-hydroxy-benzotriazole using
carbodiimides gave insoluble products (NMP, DMSO, DMF). The
observed problems are in accordance with described unsatisfactory
results attempting to use bromoacetic acid and HOBt-based in situ
Determination of the molar absorptions of both trimers at 220 nm enabled
the calculation of absolute yields from HPLC analyses. Comparison of
both absorption coefficients revealed a neglectable contribution of the
side chain ester group of 4 to the total absorption, enabling the deduction
of the molar ratios of both compounds directly from their ratios in
absorption at 220 nm.
30,49
activation.

N. Heine et al. / Tetrahedron 59 (2003) 9919–9930
9923
Figure 2. Building blocks applied for the synthesis of a library consisting of 8000 hexapeptoids and -peptomers. The purities annotated in (A) are HPLC-
purities (220 nm) of tripeptoids 5 derived from the corresponding amines after cleavage from the Rink-linker by TFA (therefore Boc and t-Bu-protecting
groups are cleaved in 5).
pp
when possible as 50% solutions in NMP or in cases of
lower solubility near saturation, at least 0.8 M (for details
see Section 4). The resulting products were analyzed by
HPLC-MS after cleavage. While most of the amines gave
rise to the desired tripeptoid 5 in purities above 75%
like glycine or b-alanine derivatives could be successfully
introduced into 5 using water instead of NMP as solvent
(0.05% Tween 20 was added in this case to decrease the
surface tension; entries 7 and 15–18). Selected heterocyclic
amines gave good results (entries 19–22). Some amines
such as those containing nucleophilic nitrogens in their side
chains, electron deficient anilines or a,a-dialkylhydrazines
turned out to be unsuited (data not shown).
(
220 nm), some building blocks were less suited (Fig. 2(A)).
We observed, that simple alkylamines generally gave good
results, as expected (entries 1–5). Anilines have to be more
electron rich than aniline itself to give reasonable results
(
entries 8–9). Aminoalkohols and -phenols as well as
2.4. Synthesis and screening of a peptoid library
hydroxylamine were successfully introduced without pro-
tection (entries 6–7 and 9–11). Boc or tert-butyl protection
was applied for amino or carboxylic acid side chains,
respectively (entries 12–16). Highly polar building blocks
The most widely used application of the SPOT synthesis
concept is parallel synthesis of peptide arrays for subsequent
evaluation of bioactivity especially affinity to a binding
1
7–20
partner
using solid phase screening directly with the
membrane linked compounds. Two successful examples
involved the de novo identification of the immunogenic
TGFa epitope recognized by the monoclonal anti-TGFa
pp
Most amine solutions were prepared v/v rather than in defined molarity
for ease of preparation. In a typical synthesis (membrane loading:
2
1
21
1
.0 mmol cm , reagent volume: 3.6 ml cm per cycle, three cycles)
using a representative amine (M¼100, d¼1.0) this corresponds to a
.0 M solution and results in 54 equiv. amine compared to the
membrane bound bromide.
32
antibody Tab-2
1
using synthetic cellulose-bound
36
2,33
hexamer or 15-mer peptide libraries. As an extension
of these results we synthesized a library consisting of 8000
5

9924
N. Heine et al. / Tetrahedron 59 (2003) 9919–9930
hexapeptoids and -peptomers aiming on the identification of
novel non-peptide epitope mimics by using a de novo
approach.
simultaneously. Whereas the hexapeptomers meant for
screening were directly attached to the cellulose without a
cleavable linker, eight hexapeptomers were synthesized in
parallel at the same membrane on a cleavable Rink-linker to
monitor the quality of synthesis. Cleavage yielded the
desired hexamers in an average HPLC-purity of 44%
(220 nm) with side products being truncated sequences as
expected. However the target product was the predominant
compound in six of the eight cases. Thus, the purity of the
library was considered to be sufficient for screening
purposes. This assumption was driven by the fact that
truncated sequences of low molecular weight are less
probable antibody ligands.
A set of 40 building blocks was selected for library synthesis
consisting of 35 amines (alkyl- and arylamines,
acylhydrazines and hydroxylamine) and five Fmoc-amino
acids chosen to introduce functional groups inaccessible by
submonomer procedures (Fig. 2). The selection of amines
was based on those which yielded the model trimer 5 in
purities above 60% (see above). In one case (2-amino-1,3-
propandiole, entry 11) we decided to tolerate a lower purity
in order to increase the diversity of the building block set. In
addition to these 22 amines 13 building blocks with similar
molecular structures and anticipated comparable synthesis
efficiencies were selected (Fig. 2(B)).
The cellulose-bound hexapeptoids and -peptomers were
incubated with a buffered aqueous solution of the
antibody Tab-2. Specifically bound antibody was
detected after washing by a second antibody labeled with
horseradish peroxidase and subsequent monitoring of
While it was not possible to synthesize all hexapeptoids and
-
(
peptomers based on the selected building block set
6
9
40 <4£10 ) we randomly chose 8000 members—a set
chemoluminiscence on treatment with a suitable
that can be synthesized efficiently on a single cellulose
membrane (19£28 cm). The unbiased, random sequences
and the corresponding sequence data file required to operate
the SPOT synthesizer were generated by LISA, a software
capable of handling monomer and submonomer procedures
substrate (Fig. 3). Several signals (i.e. black SPOTs in
Fig. 3) indicated binding of the antibody Tab-2 to
specific hexamers. Undesired binding with the second
antibody could be excluded in a control experiment without
Tab-2.
Figure 3. Binding of the antibody Tab-2 to a library consisting of 8000 cellulose bound hexapeptoids and -peptomers. Every SPOT determined by the grid
contains a compound with a specific sequence. Dark SPOTs indicate binding of the antibody—marked compounds were subjected to resynthesis and
quantification of binding (circles indicate resynthesized inactive compounds chosen as negative controls).

N. Heine et al. / Tetrahedron 59 (2003) 9919–9930
9925
2
.5. Quantitative determination of binding
epitope (VVSHFND-NH ) in the surface plasmon reson-
2
ance experiments. Its dissociation constant was determined
4
to be 20 nM in accordance to previous ELISA and
isothermal titration calorimetry experiments. Thus we
could identifiy non-peptidic antibody ligands considerably
differing from the wild-type peptide epitope within a single
experiment starting from a na ¨ı ve library.
3
The sequences of the most active ligands were extracted
from the synthesis file and two of them (6 and 7) as well as
two inactive compounds (8 and 9) as negative controls were
resynthesized on resin beads applying published procedures
3
5
3
0,37
(
Fig. 4).
Binding with the antibody Tab-2 was
quantified in a series of surface plasmon resonance
experiments using a two channel setup (Biacore X).
Determination of paratope specific binding was ensured
5
1
by coupling of mab TE33, an antibody with the same
subtype but different specificity (anti-cholera toxin) in the
reference cell.
3. Conclusion
The SPS of peptoids—previously established using standard
polystyrene-based resins—could be successfully adapted to
cellulose membranes using the monomer as well as the
submonomer approach. The procedures had to be adjusted
to the special requirements related to the different kinetic
situation on continuous surfaces and the large number of
free hydroxy-functions on the cellulose. Within the
submonomer approach bromoacetic acid 2,4-dinitrophenyl-
ester (1a) was found to be ideally suited for efficient
acylation of the secondary amine of the growing peptoid
chain. In contrast to other applied reagents this ester neither
showed any N-alkylation nor significant OH-acylation.
Both selected ligands have dissociation constants (K ) in
D
the micromolar range (K ¼2.7 mM for 6 and 408 mM for 7)
D
whereas for the negative controls 8 and 9 no binding was
detected (Fig. 4). On comparison of both active compounds
6
shows better binding in the SPR experiment whereas 7
gives a stronger signal in the cellulose assay. A possible
explanation to this observation is the influence of the solid
phase on array bound compounds compared to the same
entities in solution as occurring in all heterogenous assay
systems. Depending on the degree and mode of confor-
mational restriction induced upon binding to the antibody
the fraction of the entropic term of the binding free energy
varies for the different assay systems. Another explanation
is based on the different C-termini of cellulose bound and
corresponding isolated compounds: if the C-terminal amide
binds to the antibody specifically the linked membrane
would disturb this interaction in the cellulose assay, whereas
if the linker is partly contributing to antibody binding the
affinity should be lower with the free compound in the SPR
assay. Nevertheless activity on cellulose led to compounds
binding the antibody in solution as well. In order to compare
the potency of the epitope mimics 6 and 7 to the native
ligand as well as to complement the quantification method,
we additionally included a hexapeptide covering the TGFa
A library consisting of 8000 hexapeptoids and -peptomers
was synthesized under optimized conditions. By screening
for binding with the monoclonal anti-TGFa-antibody Tab-2
ligands with micromolar affinity were identified. The
approach enabled the identification of non-peptide epitope
mimics de novo in a single step, i.e. without any input of
structural information derived from the epitope originally
recognized by the antibody. The dissociation constant of the
best compound from the library was only two orders of
magnitude worse compared to the wild-type peptide
epitope. These findings shows peptoid synthesis and
screening on cellulose membranes to be a valuable and
rapid tool for lead identification processes.
Figure 4. Hexapeptoids selected for quantification of binding to the antibody tab-2 (two ligands and two non-binders as indicated in Figure 3). The dissociation
constants shown were determined by SPR-experiments.

9926
N. Heine et al. / Tetrahedron 59 (2003) 9919–9930
4. Experimental
cence blotting substrate (POD) (Roche Diagnostics; Cat. #1
500 708) and 9 equiv. of TBS.
4
.1. General
4.3. Methods
NMR spectra were recorded on a Varian Unity-plus 300
13
1
spectrometer in CDCl at 258C ( H at 300 MHz, C at
3
4.3.1. General procedures. Washing. If not otherwise
noted, all membranes were washed in a stainless steel dish
between each reaction step with DMF, MeOH and Et O
2
1
9
7
shifts are given in parts per million (ppm) relative to internal
5.45 MHz and F at 284 MHz, respectively). Chemical
1
13
19
standards: Si(CH ) ( H and C) or CFCl ( F), respect-
3
three times each.
4
3
ively; coupling constants (J) are given in Hertz (Hz). FT-IR
spectra were performed on a Bruker IFS-66. HRMS-FAB
was performed on a Finnigan MAT 95ST (Trap) equipped
with a ‘Cs-gun’ using a glycerol/m-nitrobenzylalcohol
matrix and CsI/glycerol as standard. Elemental analyses
Amino derivatisation of cellulose. Cellulose membranes
2
2
with amino-loadings of 0.6–1.4 mmol cm were prepared
2
8
according to a previous report: a cellulose sheet
(19£28 cm) was immersed in a solution of epibromohydrine
and 15 M aq. perchloric acid in dioxane [10:1:90 (v/v/v)] for
3 h, washed with MeOH (2£) and dried in air. The
membrane was incubated with neat 4,7,10-trioxa-1,13-
tridecanediamine for 1 h at 808C in a stainless steel dish
on a heating plate. The membrane was washed with MeOH
(
(
EA) were carried out on a LECO CHNS-933 microanalyzer
for C,H,N) or by titration with 0.01N Hg(ClO ) /
4
2
diphenylcarbazone (Dirscherl/Erne) after Sch o¨ ninger-
decomposition (for Br), respectively. Analytical HPLC-
MS analysis was carried out on a Hewlett Packard 1100
HPLC system coupled to a Finnigan LCQ ion-trap ESI-mass
spectrometer. In a typical analysis-run approx. 50 nmol
substance was analyzed using a linear gradient (eluent A
(2£), 5 M NaOMe in MeOH, MeOH, H O (4£), MeOH, and
2
Et O and dried in air.
2
9
,15–17
0
.05% TFA in water, eluent B 0.05% TFA in acetonitrile;
Manual SPOT synthesis
30,37
of peptoids (submonomer
). A 2.0 ml amount of a solution of N-Fmoc-4-
2
1
gradient time: 25 min; flow rate: 0.3 ml min ) on a C18
RP-column (Vydac 218 TP5215, 2.1£150 mm). Preparative
separations were carried out on a Merck/Hitachi system
protocol
[amino-(2,4-dimethoxyphenyl)-methyl]-phenoxyacetic
acid, pentafluorophenol and diisopropylcarbodiimide in
NMP (0.2 M for all components) was pipetted to an
amino derivatized cellulose membrane on a 1£1 cm pencil
drawn grid after 30 min of preactivation. The procedure was
repeated after 15 min followed by washing and drying in air.
Capping of unreacted amino groups by acetylation was
(
(
6
L-6250, L-7400, D-7000) and a Merck column
LiChrospher 100, RP18, 10£250 mm, flow rate:
.0 ml min2 ) using an optimized gradient derived from
1
the analytical chromatogram. Melting points are
uncorrected. SPOT synthesis was done on the Auto Spot
Robot ASP 222-system (Abimed, Langenfeld, Germany);
the software LISA was used for generation of the sequence-
files (by W. -J. Wu, Inst. Med. Immunol., Charit e´ , Humboldt
University, Berlin, Germany; all rights at Jerini AG). SPR-
experiments were carried out on a Biacore X-system
achieved by immersing the membrane in a solution of Ac O
2
and DIEA in DMF [1:2:7 (v/v/v)] followed by washing and
drying. The degree of membrane derivatisation was
2
determined by punching a test-SPOT (0.23 cm ) and
quantifying the absorption of the dibenzofulvene-piperidine
adduct at 301 nm after release by piperidine (20% in
(
Biacore AB, Uppsala, Sweden).
5
2
DMF). The remaining membrane was deblocked by
treatment with piperidine (20% in DMF; 2£20 min)
followed by washing and drying in air. Submonomer
synthesis was performed by alternately pipetting either
2.0 ml of a solution of 1a (1.0 M in NMP) with one
repetition after 15 min or 2.0 ml of a solution of an amine (in
4
.2. Materials
Solvents and reagents were obtained from Sigma-Aldrich
Munich, Germany), VWR International (Darmstadt,
(
Germany) or Lancaster Synthesis (Frankfurt, Germany).
Amino acid derivatives were provided by Novabiochem
NMP or H O containing 0.05% Tween 20) with two
2
(
(
(
L a¨ ufelfingen, Swizerland); Tentagel-S-RAM-resin
0.23 mmol g2 ) was purchased from Rapp-Polymere
T u¨ bingen, Germany). SPOT synthesis was performed on
repetitions after 15 and 30 min, respectively. Excesses of
reagents were removed between each synthesis step by
washing. An extended washing procedure was applied after
treatment with amines: DMF (4£), MeOH (1£), 0.5 M aq.
NaOH, H O (5£), MeOH (2£), and Et O. After completion
1
Whatman 50 cellulose (Whatman, Maidstone, UK). The
antibodies Tab-2 and TE-33 were kindly provided by
W. H o¨ hne (Inst. for Biochemistry, Charit e´ , Humboldt-
Universit a¨ t, Berlin, Germany) and T. Scherf (Weizmann
Inst. of Science, Rehovot, Israel), respectively. The assays
were performed in 10 mM TRIS-buffered saline (TBS) at
pH¼8.0. The anti-mouse IgG antibody used for detection of
bound Tab-2 was provided by Sigma (Taufkirchen,
Germany). Antibodies bound to peptoids/peptomers on
cellulose membranes were detected using the recommended
luminescence solution (¼100 equiv. of luminescence sub-
strate solution A and 1 equiv. of the corresponding solution
B) with a Lumi-Imager system (Roche Diagnostics,
Mannheim, Germany). Blocking buffer consisted of
2
2
of the synthesis sequence, the SPOTs were punched out and
transferred into individual 2.0 ml Eppendorf tubes. The
peptoids were cleaved from the linker with TFA (95% in
H O; 70 ml; 45 min) and dissolved in acetonitrile (30% in
2
H O, 50 ml) immediate after evaporation in an vacuum
2
centrifuge (Eppendorf) at 458C. The solution was analyzed
by HPLC-MS.
4.3.2. 2,4-Dinitrophenyl-bromoacetate (1a). A solution of
anhydrous 2,4-dinitrophenol (54.3 mmol) and pyridine
(70.6 mmol) in CH Cl (80 ml) was cooled to 08C and
2
2
bromoacetylbromide (59.7 mmol) in CH Cl (25 ml) was
2
2
1 equiv. of the blocking reagent which was provided
together with the Boehringer Mannheim chemilumines-
added dropwise. After 1 h at 258C the organic layer was
extracted with H O and citric acid (10% in H O), dried over
2
2

N. Heine et al. / Tetrahedron 59 (2003) 9919–9930
9927
Na SO and evaporated under reduced pressure to give a
2
cleavage and purification by prep. HPLC (15 mg,
31,9 mmol, 73%).
4
yellow oil (16.2 g). The crude product was crystallized from
Et O (200 ml) at 2268C to yield 1a as clear yellow crystals
2
1
(
13.0 g, 42.6 mmol, 78%). Mp: 108–1098C; H NMR:
3
4.3.6. Synthesis of tripeptoid 4 on resin. Tripeptoid 4 was
synthesized on Tentagel-S-RAM-resin (190 mg) according
3
d¼4.19 (s), 7.56 (d, J¼8.9 Hz), 8.91 (dd, J¼8.9 Hz,
4
4
13
37
J¼2.7 Hz), 8.98 (d, J¼2.7 Hz); C NMR: d¼24.3, 121.8,
to a published general procedure. The following modi-
1
3
9
26.4, 129.3, 141.2, 145.5, 147.8, 164.2; IR (KBr) 3112,
071, 2957, 1783 (CvO), 1608, 1538, 1344, 1219, 1097,
19, 836, 733 cm2 ; EA (in %, calcd values in parenthesis):
fications were applied: bromoacetylations for both
C-terminal positions were performed with bromoacetic
acid (20 equiv.), DIC (10 equiv.), and 2,6-dimethylpyridine
(10 equiv.) in DMF using a single coupling step. The third
acylation yielding the N-terminus was performed with
bromoacetic acid (20 equiv.), DIC (10 equiv.), and 2,6-
dimethylpyridine (40 equiv.) in CH Cl in order to induce
1
C: 31.43 (31.50), H: 1.65 (1.65), N: 9.27 (9.18), Br: 26.32
26.19).
(
4.3.3. Pentafluorophenyl-bromoacetate (1b). Bromo-
acetylbromide (21.7 mmol) in CH Cl (30 ml) was cooled
2
2
side-chain esterification. The pure tripeptoid 4 was isolated
as a salt with TFA (2 equiv.) after cleavage and purification
by prep. HPLC (23 mg, 32.4 mmol, 74%).
2
2
to 08C. A solution of pentafluorophenol (21.7 mmol) and
DIEA (21.7 mmol) in CH Cl (30 ml) was added within
2
2
1
5 min. After 1 h at 258C the organic layer was extracted
with H O (2£), dried over Na SO and evaporated under
4.3.7. Suitability of submonomer building blocks. A
series of 22 different tripeptoids 5 was synthesized on 22
SPOTs on a cellulose membrane according to the general
procedure for manual SPOT synthesis. The first building
block was incorporated using n-butylamine (5 M in NMP).
The central position was applied by 22 different N-nucleo-
philes each of them introduced on a different SPOT (all
amines 50% in NMP, if not noted differently): 1:
n-butylamine; 2: rac-s-butylamine; 3: cyclohexylamine; 4:
3,3-diphenylpropylamine; 5: benzylamine; 6: ethanol-
2
2
4
reduced pressure to give 1b as a colourless oil (6.10 g,
2
1
13
0.0 mmol, 92%); H NMR: d¼4.15 (s); C NMR: d¼23.2,
1
9
1
24.2–125.0 (m), 135.8–143.2 (m), 163.7; F NMR:
3
3
d¼2162.16 (dd, J¼22, 18 Hz), 2157.25 (t, J¼22 Hz),
3
2
152.93 (d, J¼18 Hz); IR (film) 2969, 2672, 2464,
1
1
810/1790 (CvO), 1655, 1520, 1473, 1425, 1403, 1241,
222, 1145, 1095, 1000, 599 cm ; EA (in %, calcd
2
1
values in parenthesis): C: 31.36 (31.50), H: 0.77 (0.66), Br:
5
3
19
25.93 (26.20). Impurity pentafluorophenol: ,3% ( F
NMR).
amine; 7: hydroxylamine (50% in H O containing 0.05%
2
Tween 20); 8: 4-phenoxyaniline (2.0 M in NMP); 9:
4-aminophenol (saturated in NMP¼approx. 1 M); 10:
tyramine (1.0 M in NMP); 11: 2-amino-1,3-propandiole
(30% in NMP); 12: N-boc-1,2-diaminoethane; 13: N-boc-
1,4-diaminobutane; 14: tert-butyl carbazate (2 M in NMP);
4
.3.4. N-Selective bromoacetylation. The tripeptoid 3 was
synthesized on a cellulose membrane according to the
general procedure for manual SPOT synthesis of peptoids
using modified conditions during the second acylation step.
The amines n-butylamine, ethanolamine, and piperidine
were applied as building blocks (5 M in NMP each). For the
second bromoacetylation, conditions were varied according
to Table 1: either 1a, 1b (1 M in NMP each), or a solution of
bromoacetic acid (2 M in NMP) treated with 0.5 equiv. DIC
15: glycine tert-butyl ester acetate (2.5 M in H O containing
2
0.9 equiv. of NaOH and 0.05% Tween 20); 16: b-alanine
tert-butyl ester hydrochloride (3.0 M in H O containing
2
0.9 equiv. of NaOH and 0.05% Tween 20); 17: glycine
amide hydrochloride (5.0 M in H O containing 0.9 equiv. of
2
3
2
0 min prior to synthesis were applied (2.0 ml and another
.0 ml after 15 min). After completion of the synthesis
NaOH and 0.05% Tween 20); 18: b-alanine amide
hydrochloride (5.0 M in H O containing 0.9 equiv. of
2
cleavage yielded the crude product for analysis by HPLC-
MS. The contained amounts of tripeptoid 3 and its
derivative 4 were quantified by determination of the
corresponding peak areas in HPLC at 220 nm and
subsequent calculation of the absolute amounts using
calibration curves (authentic material for calibration was
synthesized on resin, see below). Chemical yields shown in
Table 1 result from the comparison of the amounts of
isolated products with the Rink-linker loading of the
membrane. All data shown result from at least five
independent experiments; the deviation of yields and
purities was #10% for 3 and #4% for 4.
NaOH and 0.05% Tween 20); 19: tryptamine (1.5 M in
NMP); 20: histamine (1.5 M in NMP); 21: 2-aminomethyl-
thiophene; 22: 2-aminothiazole (1.5 M in NMP). The trimer
was completed using piperidine (5 M in NMP). Each SPOT
was punched out and transferred into a 2 ml Eppendorf tube.
Cleavage with TFA yielded the products for analysis by
HPLC-MS. The purities (220 nm) of the products are listed
in Figure 2. The integrity of each compound was assured by
analysis of the ESI-mass spectra corresponding to the
þ
product HPLC-signal [shown data found for MþH (calcd
values in parenthesis)]: 1: 369.2 (369.29); 2: 369.3 (369.29);
3: 395.3 (395.30); 4: 507.3 (507.33); 5: 403.2 (403.27); 6:
357.1 (357.25); 7: 329.2 (329.22); 8: 481.2 (481.28); 9:
4
.3.5. Synthesis of tripeptoid 3 on resin. Tripeptoid 3 was
synthesized on Tentagel-S-RAM-resin (190 mg) according
405.1 (405.25); 10: 433.2 (433.28); 11: 387.2 (387.26); 12:
378.2 (356.27); 13: 384.2 (384.30); 14: 350.1 (328.23); 15:
371.2 (371.23); 16: 385.2 (385.25); 17: 392.1 (370.25); 18:
384.2 (384.26); 19: 456.2 (456.30); 20: 407.1 (407.28); 21:
409.2 (409.23); 22: 396.2 (396.21).
3
7
to a published general procedure. The following modi-
fications were applied: bromoacetylations for both
C-terminal positions were performed with bromoacetic
acid (20 equiv.), DIC (10 equiv.), and 2,6-dimethylpyridine
(
10 equiv.) in DMF using a single coupling step. The third
acylation yielding the N-terminus was performed using 1a
1.5 equiv.) in DMF in a single coupling step. The pure
tripeptoid 3 was isolated as a salt with TFA (1 equiv.) after
4.3.8. Automated SPOT synthesis of a library consisting
of 8000 hexapeptoids and -peptomers. A cellulose
2
2
(
membrane with an amino-loading of 0.5 mmol cm was
8
prepared based on our previous report: A cellulose sheet
2

9928
N. Heine et al. / Tetrahedron 59 (2003) 9919–9930
(
and 15 M aq. perchloric acid in dioxane [10:1:90 (v/v/v)] for
19£28 cm) was immersed in a solution of epibromohydrine
37: Fmoc-sarcosine; 38: Fmoc-L-alanine; 39: Fmoc-L-
proline; 40: rac-Fmoc-pipecolic acid. Bromoacetylations
were carried out with 2,4-dinitrophenyl-bromoacetate (1a,
1.0 M in NMP). All acylations were repeated once, whereas
bromine displacements by amines were repeated twice.
Excesses of reagents were removed between each synthesis
step by washing as described above (‘Manual SPOT
synthesis of peptoids’). After completion of the hexamer
synthesis, the ‘screening array’ was separated from the
‘analysis array’ and treated twice with TFA, and TIPS in
1
h, washed with MeOH (2£) and dried in air. The
membrane was incubated with 4,7,10-trioxa-1,13-tri-
decanediamine (20% in DMF) for 2 h at 258C in a stainless
steel dish. The membrane was washed with MeOH (2£),
5
M NaOMe in MeOH, MeOH, H O (4£, MeOH, and Et O
2
2
and dried in air. The synthesis of the hexapeptomers was
performed semi-automatically using the ABIMED autospot
system configured to address an array consisting of 100£80
SPOTs evenly spaced on the membrane (“screening array”,
diameter of the SPOTs: 1.2 mm, distance between the
SPOTs: 2.2 mm). In addition eight hexapeptoids were
synthesized at the edge of the array (“analysis array”,
SPOT-distance: 1.0 cm) applying the 45-fold reagent
volume thus yielding SPOTs with a diameter of 9 mm
enabling HPLC-MS-analysis of products after cleavage
from the linker to monitor the quality of synthesis. In a first
step 40 nl of Fmoc-b-alanine pentafluorophenylester (0.3 M
in DMSO) for SPOTs of the screening array and 1.8 ml of a
solution of N-Fmoc-4-[amino-(2,4-dimethoxyphenyl)-
methyl]-phenoxyacetic acid, pentafluorophenol and diiso-
propylcarbodiimide in NMP (0.2 M for all components) for
SPOTs of the analysis array were pipetted to the membrane
using the Auto Spot Robot. The procedure was repeated
after 15 min followed by detachment of the membrane from
the synthesizer, washing in a stainless steel dish, and drying
in air. Capping of unreacted amino groups by acetylation
was achieved by immersing the membrane in a solution of
H O [94:2:3 (v/v/v) with 1 g phenol/100 ml added] for
2
15 min and once with TFA, TIPS, and H O in CH Cl
2
2
2
[50:2:3:45 (v/v/v/v) with 1 g phenol/100 ml added] for 2 h
in order to remove side chain protecting groups. The
membrane was washed with CH Cl (2£), DMF, Et N (2%
2
2
3
in DMF), DMF (2£), MeOH (2£), and Et O and dried in air.
2
The SPOTs on the ‘analysis array’ were punched out and
transferred into individual 2.0 ml Eppendorf tubes. The
peptoids and peptomers were cleaved from the linker with
TFA (95% in H O; 70 ml; 20 min) and dissolved in
2
acetonitrile (30% in H O; 50 ml) immediate after evapor-
2
ation of TFA in a vacuum centrifuge (Eppendorf) at 458C.
The solutions were analysed by HPLC-MS. The sequences
(from N- to C-terminus, numbers see Fig. 2), purities
(220 nm, HPLC, mp¼target is the Main Product), and
þ
p
ESI-MS-data (MþH , calcd values in parenthesis) are: 41 -
p
05-01-18-05-01, 63% (mp), 791.4 (791.48); 41 -05-01-18-
05-25, 76% (mp), 819.4 (819.48); 39-31-33-23-16-10, 42%
(mp), 764.3 (764.42); 39-02-12-17-18-31, 44% (mp), 685.2
(685.40); 31-08-06-37-29-38, 55% (mp), 715.2 (715.38);
14-35-35-37-39-24, 47% (combined purity of 2 peaks
caused by 4 diastereomers, mp), 765.2 (765.43); 20-04-
33-22-19-08, 11%, 1100.3 (1100.48); 11-32-15-24-30-23,
Ac O and DIEA in DMF [1:2:7 (v/v/v)] followed by
2
washing. Amino groups of the membrane bound amino
acids were deblocked by treatment with piperidine (20% in
DMF; 2£20 min, followed by washing) and visualized by
staining with bromophenolblue (0.01% in MeOH). SPOTs
at the corners of the array were subsequently marked with a
pencil in order to enable a proper reposition of the
membrane on the synthesizer prior to each synthesis step.
p
10%, 822.3 (822.44) [ 41: piperidine (50% in NMP)].
4.3.9. Screening of 8000 cellulose bound hexapeptoids
and -peptomers for binding with the antibody Tab-2. The
cellulose membrane with peptomers covalently bound was
washed with MeOH and TBS (3£) followed by incubation
with blocking buffer (1 h). The solvents were decanted and
the wet membrane was incubated with mab Tab-2
(1 mg ml2 in 100 ml of blocking solution; 2 h). The
membrane was washed with TBS containing 0.05%
Tween 20 (T-TBS; 3£). Bound antibody was detected
after incubation with a second peroxidase (POD)-labeled
anti-mouse IgG antibody (1 mg ml2 in 100 ml of blocking
solution; 2 h). After washing with T-TBS (6£) POD activity
was measured and quantified on a LumiImager after
incubation with the chemoluminesence substrate (40 ml;
1 min). The result is shown in Figure 3 (10 min exposure
time).
The hexamers were synthesized according to a sequence file
generated by the program LISA prior to synthesis contain-
ing 8000 sequences randomly and unbiased chosen based on
a set of 40 building blocks (see Fig. 2). Since monomer and
submonomer syntheses were applied on the same membrane
in parallel a previously published procedure comprising the
1
2
8
following steps was used: all acylations [with Fmoc-
amino acids (monomers) and bromoacetic acid (sub-
monomers)] were performed in in a first step whereas
Fmoc removal (by 4% DBU in NMP, monomers) and
bromosubstitution (by primary amines, submonomers) were
carried out in a second step after washing. In both steps, the
synthesizer was controlled by the sequence file. See Section
1
4
.3.7 for building blocks no. 1 to 22. Additionally the
following reagents were used: (all amines 50% in NMP, if
not noted differently): 23: iso-butylamine; 24: cyclohexyl-
methylamine; 25: rac-tetrahydrofurfuryl-amine; 26:
o-chlorbenzylamine; 27: p-methoxybenzylamine; 28: rac-
4.3.10. Synthesis of hexapeptomers 6, 7, 8, and 9 on resin.
The hexamers 6, 7, 8, and 9 were synthesized on Tentagel-
S-RAM-resin (400 mg) using modifications of known
3
7
protocols for the introduction of submonomer
and
3
1
1
-phenylethylamine; 29: N-boc-1,3-diaminopropane; 30:
monomer building blocks, respectively. After initial
removal of the Fmoc-group [(i) piperidine (20% in DMF),
4.0 ml, 1£40 min; (ii) DMF, 5.0 ml, 5£1 min] each of the 6
building blocks was introduced according to the desired
sequence (Table 2) using either the submonomer or the
monomer protocol. Submonomer procedure: The resin
bound amine was bromo-acetylated [(i) 1a (0.3 M in
nicotinic hydrazide (0.8 M in NMP); 31: 2-methoxy-
ethylamine; 32: 3-amino-1-propanol; 33: rac-2-amino-1-
propanol; 34: rac-3-amino-1,2-propandiol (30% in NMP);
3
5: rac-2-amino-1-phenylethanol. All Fmoc-amino acids
were applied as 0.6 M solutions in NMP, activated with
.55 equiv. DIC 30 min prior to synthesis: 36: Fmoc-glycin;
0

N. Heine et al. / Tetrahedron 59 (2003) 9919–9930
9929
Table 2. Analytical data for hexapeptomers 6–9
a
Sequence (N- to C-terminus)
b
Purity (%)
þ
Yield
ESI-MS Calcd
(MþH ) Found
c
d
6
7
8
9
27-29-24-38-27-24
24-24-36-15-36-24
39-38-29-02-13-27
20-40-31-15-01-27
54 mg (54%)
7.4 mg (10%)
98
98
98
97
863.54
706.45
718.46
800.43
863.4
e
706.3
718.3
800.2
f
31 mg (32%)
c
33 mg (35%)
a
b
c
d
e
f
Numbers correspond to building blocks listed in Figure 2.
HPLC, 220 nm.
Salt with 2 equiv. TFA.
þ
High resolution FAB-MS (MþH ): 863.5402 (calcd: 863.53949).
Salt with 1 equiv. TFA.
Salt with 3 equiv. TFA.
DMF), 1.5 ml, 1£30 min; (ii) DMF, 5.0 ml, 5£1 min] and
treated with a solution of the appropriate primary amine in
DMF [(i) 1.5 ml amine solution, 1£45 min; (ii) DMF,
state kinetics plot and K -values were determined with the
D
‘BIAevaluation’ Software (version 3.0.1, Biacore AB,
Uppsala, Sweden). Binding of the peptide VVSHFND-
NH2 was calculated from the corresponding kinetic
association- and dissociation-constants, respectively
(‘BIAevaluation’ Software, 1:1 Langmuir-binding, concen-
5
.0 ml, 5£1 min]. The following amines were used
(
concentrations in parentheses): 1: n-butylamine (5 M); 2:
rac-s-butylamine (5 M); 13: N-boc-1,4-diaminobutane
3 M); 15: glycine-tert-butylester acetate (3 M in H O
5
21 21
(
tration range: 16 nM–4.0 mM, k ¼1.10£10 M
s
;
2
a
k ¼2.18£10 s21). Determined values for x were
2
3
2
containing 0.9 equiv. of NaOH and 0.05% Tween 20; the
resin was additionally washed with H O prior to and after
d
below 1.2. The integrity of the chip was ensured by
control-measurements in the course of approx. 130
measurement cycles yielding stable response.
2
amine treatment); 20: histamine (1.5 M); 24: cyclohexyl-
methylamine (5 M); 27: 4-methoxybenzylamine (5 M); 29:
N-boc-1,3-diamino-propane (3 M); 31: 2-methoxyethyl-
amine (5 M). Monomer procedure: The resin bound amine
was acylated with the appropriate Fmoc-amino acid after
preactivation for 30 min with 0.55 equiv. of DIC [(i) Fmoc-
amino acid (0.6 M in DMF), 1.5 ml, 1£30 min; (ii) DMF,
Acknowledgements
The authors gratefully acknowledge technical assistance
from A. Rosenberg, U. Beutling, and M. Schlief. G. Hoehne,
Technical University, Berlin is thanked for performing the
high resolution MS spectra. Financial support was provided
by Fonds der Chemischen Industrie.
5
.0 ml, 5£1 min] followed by Fmoc-removal [(i) piperidine
(
5
20% in DMF), 2.0 ml, 1£20 min; (ii) DMF, 5.0 ml,
£1 min]. The following amino-acids were used: 36:
N-Fmoc-glycine; 38: N-Fmoc-L-alanine; 39: N-Fmoc-L-
proline; 40: N-Fmoc-D/L-pipecolic acid.
After completion of the sequence the resin was washed
extensively [piperidine (20% in DMF); DMF; H O; DMF;
References
2
MeOH; CH Cl ; Et O, each 5.0 ml, 3£1 min] and dried.
2
2
2
Cleavage from the resin as well as deprotection was
achieved by treatment with TFA (95% in H O; 6.0 ml;
1. Latham, P. W. Nat. Biotechnol. 1999, 17, 755–757.
2. Jung, G.; Beck-Sickinger, A. G. Angew. Chem., Int. Ed. Engl.
1992, 31, 367–383.
2
3
0 min). The solution was filtered, the resin washed with
acetonitrile and the combined filtrates were evaporated to
dryness. The crude product was purified by prep. HPLC and
analyzed by HPLC-MS (results see Table 2).
3. Jung, G. Combinatorial Peptide and Nonpeptide Libraries;
VCH: Weinheim, Germany, 1996.
4. Geysen, H. M.; Meloen, R. H.; Barteling, S. J. Proc. Natl.
Acad. Sci. USA 1984, 81, 3998–4002.
4.3.11. Analysis of mab Tab-2 binding by surface
plasmon resonance (SPR) spectroscopy. The K values
were determined using an SPR protocol for a double-
channeled Biacore X instrument. In our binding experi-
ments the immobilized phase was generated by coupling
5. Furka, A.; Sebestyen, F.; Asgedom, M.; Dibo, G. Int. J. Pept.
Protein Res. 1991, 37, 487–493.
D
5
4
6. Houghten, R. A. Proc. Acad. Sci. USA 1985, 82, 5131–5135.
7. Geysen, H. M.; Rodda, S. J.; Mason, T. J. Mol. Immunol. 1986,
23, 709–715.
monoclonal mouse-IgG-antibodies (measuring cell: Tab-2,
1
8. Pirrung, M. C. Chem. Rev. 1997, 97, 473–488.
9. Frank, R. Tetrahedron 1992, 48, 9217–9232.
10. Frank, R. Bioorg. Med. Chem. Lett. 1993, 3, 425–430.
11. Kramer, A.; Volkmer-Engert, R.; Malin, R.; Reineke, U.;
Schneider-Mergener, J. Pept. Res. 1993, 6, 314–319.
12. Kramer, A.; Schuster, A.; Reineke, U.; Malin, R.; Volkmer-
Engert, R.; Landgraf, C.; Schneider-Mergener, J. Methods
(Comp. Meth. Enzymol.) 1994, 6, 388–395.
5
TE-33, 75 mg ml
0 mg ml2 in NaOAc-buffer, pH¼5.0; reference cell:
2
1
in NaOAc-buffer, pH¼4.0) to a
N-hydroxysuccinimide activated dextrane-coated sensor
chip (CM5) using the Biacore amine immobilisation kit
(resulting in an antibody-loading between 5000 and 6000
RU). Binding was analysed at concentrations between
2
1
00 nM and 2.0 mM in HBS-EP buffer (10 mM HEPES,
50 mM NaCl, 3 mM EDTA, 0.005% polysorbate 20,
13. Frank, R. J. Biotechnol. 1995, 41, 259–272.
14. Frank, R.; Overwin, H. Methods Mol. Biol. 1996, 66,
149–169.
pH¼7.4). The chip was regenerated with glycine/HCl-
buffer (10 mM, pH¼2.0) prior to each measurement. The
ammount of peptomer-binding at equilibrium was deter-
mined. Equilibrium values were analyzed using the steady-
15. Kramer, A.; Schneider-Mergener, J. Methods Mol. Biol. 1998,
87, 25–39.

9930
N. Heine et al. / Tetrahedron 59 (2003) 9919–9930
1
6. Kramer, A.; Reineke, U.; Dong, L.; Hoffmann, B.; Hoffm u¨ ller,
U.; Winkler, D.; Volkmer-Engert, R.; Schneider-Mergener, J.
J. Pept. Res. 1999, 54, 319–327.
cation Potential; Cortese, R., Ed.; Walter de Gruyter: Berlin,
Germany, 1996; pp 53–68.
35. Hahn, M.; Winkler, D.; Welfle, K.; Misselwitz, R.; Welfle, H.;
Wessner, H.; Zahn, G.; Scholz, C.; Seifert, M.; Harkins, R.;
Schneider-Mergener, J.; Hoehne, W. J. Mol. Biol. 2001, 314,
293–309.
1
7. Wenschuh, H.; Gausepohl, H.; Germeroth, L.; Ulbricht, M.;
Matuschewski, H.; Kramer, A.; Volkmer-Engert, R.; Heine,
N.; Ast, T.; Scharn, D. In Combinatorial Chemistry: A
Practical Approach; Fenniri, H., Ed.; Oxford University:
Oxford, UK, 2000; pp 95–116.
36. Reineke, U.; Ivascu, C.; Schlief, M.; Landgraf, C.; Gericke, S.;
Zahn, G.; Herzel, H.; Volkmer-Engert, R.; Schneider-
Mergener, J. J. Immunol. Methods 2002, 267, 37–51.
37. Figliozzi, G. M.; Goldsmith, R.; No, S. C.; Banville, S. C.;
Zuckermann, R. N. Methods Enzymol. 1996, 267, 437–447.
38. Nguyen, J. T.; Turck, C. W.; Cohen, F. E.; Zuckermann, R. N.;
Lim, W. A. Science 1998, 282, 2088–2092.
1
1
2
8. Reineke, U.; Volkmer-Engert, R.; Schneider-Mergener, J.
Curr. Opin. Biotechnol. 2001, 12, 59–64.
9. Reimer, U.; Reineke, U.; Schneider-Mergener, J. Curr. Opin.
Biotechnol. 2002, 13, 315–320.
0. Frank, R.; Schneider-Mergener, J. In Springer Laboratory
Manual; Koch, J., Mahler, M., Eds.; Springer: Heidelberg,
Germany, 2002; pp 1–22.
39. Heizmann, G.; Hildebrand, P.; Tanner, H.; Ketterer, S.;
Pansky, A.; Froidevaux, S.; Beglinger, C.; Eberle, A. N.
J. Recept. Signal Transduct. Res. 1999, 19, 449–466.
40. Olivos, H. J.; Alluri, P. G.; Reddy, M. M.; Salony, D.;
Kodadek, T. Org. Lett. 2002, 4, 4057–4059.
2
2
2
2
1. Weiler, J.; Gausepohl, H.; Hauser, N.; Jensen, O. N.; Hoheisel,
J. D. Nucleic Acids Res. 1997, 25, 2792–2799.
2. Matysiak, S.; Hauser, N. C.; W u¨ rtz, S.; Hoheisel, J. D.
Nucleosides Nucleotides 1999, 18, 1289–1291.
3. Jobron, L.; Hummel, G. Angew. Chem., Int. Ed. Engl. 2000,
41. Girin, S. K.; Shvachkin, Y. P. J. Gen. Chem. USSR (Engl.
Transl.) 1979, 49, 395–399.
39, 1621–1624.
42. Kisfaludy, L.; Mohacsi, T.; Low, M.; Drexler, F. J. Org.
Chem. 1979, 44, 654–656.
4. Toepert, F.; Pires, J. R.; Landgraf, C.; Oschkinat, H.;
Schneider-Mergener, J. Angew. Chem., Int. Ed. Engl. 2001,
43. Elliott, R. D.; Brockman, R. W.; Montgomery, J. A. J. Med.
Chem. 1986, 29, 1052–1056.
40, 897–900.
2
2
2
2
2
5. Scharn, D.; Wenschuh, H.; Reineke, U.; Schneider-Mergener,
J.; Germeroth, L. J. Comb. Chem. 2000, 2, 361–370.
6. Scharn, D.; Germeroth, L.; Schneider-Mergener, J.;
Wenschuh, H. J. Org. Chem. 2001, 66, 507–513.
44. Jones, D. S.; Barstad, P. A.; Feild, M. J.; Hachmann, J. P.;
Hayag, M. S.; Hill, K. W.; Iverson, G. M.; Livingston, D. A.;
Palanke, M. S.; Tibbetts, A. R.; Yu, L.; Coutts, S. M. J. Med.
Chem. 1995, 38, 2138–2144.
7. Heine, N.; Germeroth, L.; Schneider-Mergener, J.; Wenschuh,
H. Tetrahedron Lett. 2001, 42, 227–230.
45. Ammann, H.; Dupuis, G. Can. J. Chem. 1988, 68, 1651–1655.
46. Cuatrecasas, P.; Wilchek, M.; Anfinsen, C. J. Biol. Chem.
1969, 244, 4316–4329.
8. Ast, T.; Heine, N.; Germeroth, L.; Schneider-Mergener, J.;
Wenschuh, H. Tetrahedron Lett. 1999, 40, 4317–4318.
9. Simon, R. J.; Kania, R. S.; Zuckermann, R. N.; Huebner, V. D.;
Jewell, D. A.; Banville, S.; Ng, S.; Wang, L.; Rosenberg, S.;
Marlowe, C. K.; Spellmeyer, D. C.; Tan, R.; Frankel, A. D.;
Santi, D. V.; Cohen, F. E.; Bartlett, P. A. Proc. Natl. Acad. Sci.
USA 1992, 89, 9367–9371.
47. Mir, I.; Siddiqui, M. T.; Comrie, A. M. J. Pharm. Sci. 1991,
80, 548–550.
48. Drysdale, M. J.; Pritchard, M. C.; Horwell, D. C. J. Med.
Chem. 1992, 35, 2573–2581.
49. Korlipara, V. L.; Takemori, A. E.; Portoghese, P. S. J. Med.
Chem. 1995, 38, 1337–1343.
3
0. Zuckermann, R. N.; Kerr, J. M.; Kent, S. B. H.; Moos, W. H.
J. Am. Chem. Soc. 1992, 114, 10646–10647.
50. Rink, H. Tetrahedron Lett. 1987, 28, 3787.
51. Shoham, M. J. Mol. Biol. 1993, 232, 1169–1175.
52. Eichler, J.; Bienert, M.; Stierandova, A.; Lebl, M. Pept. Res.
1991, 4, 296–307.
3
1. Ostergaard, S.; Holm, A. Mol. Divers. 1997, 3, 17–27.
2. Hoeprich, P. D.; Langton, B. C.; Zhang, J.-W.; Tam, J. P.
J. Biol. Chem. 1989, 264, 19086–19091.
3
53. Feldman, D.; Segal-Lew, D.; Rabinovitz, M. J. Org. Chem.
1991, 56, 7350–7354.
3
3
3. Kramer, A.; Vakalopoulou, E.; Schleuning, W. D.; Schneider-
Mergener, J. Mol. Immunol. 1995, 32, 459–465.
54. BIACORE-X Hardware manuals (BIAapplications Handbook
5’94; BIAevaluation Software Handbook 7’97; BIAtechnology
Handbook 6’94); Biacore AB: Uppsala, Schweden.
4. Schneider-Mergener, J.; Kramer, A.; Reineke, U. In
Combinatorial Libraries—Synthesis, Screening and Appli-