Synthesis and binding studies of Alzheimer ligands on solid support

Article abstract of DOI:10.1021/jo061918x

(Chemical Equation Presented) Aminopyrazole derivatives constitute the first class of nonpeptidic rationally designed β-sheet ligands. Here we describe a double solid-phase protocol for both synthesis and affinity testing. The presented solid-phase synthe

Full text of DOI:10.1021/jo061918x

Synthesis and Binding Studies of Alzheimer Ligands on Solid
Support
Petra Rzepecki, Nina Geib, Manuel Peifer, Frank Biesemeier, and Thomas Schrader*
Fachbereich Chemie, UniVersita¨t Marburg, Hans-Meerwein-Strasse, 35032 Marburg, Germany
schradet@staff.uni-marburg.de
ReceiVed September 18, 2006
Aminopyrazole derivatives constitute the first class of nonpeptidic rationally designed â-sheet ligands.
Here we describe a double solid-phase protocol for both synthesis and affinity testing. The presented
solid-phase synthesis of four types of hybrid compounds relies on the Fmoc strategy and circumvents
subsequent HPLC purification by precipitating the final product from organic solution in pure form.
Hexa- and octapeptide pendants with internal di- and tetrapeptide bridges are now amenable in high
yields to combinatorial synthesis of compound libraries for high-throughput screening purposes. Solid-
phase peptide synthesis (SPPS) on an acid-resistant PAM allows us, after PMB deprotection, to subject
the free aminopyrazole binding sites in an immobilized state to on-bead assays with fluorescence-labeled
peptides. From the fluorescence emission intensity decrease, individual binding constants can be calculated
via reference curves by simple application of the law of mass action. Gratifyingly, host/guest complexation
can be monitored quantitatively even for those ligands, which are almost insoluble in water.
Introduction
mainly cures symptoms, e.g., through administration of acetyl-
choline esterase inhibitors,2 often combined with NMDA
receptor antagonists, phytotherapeutics,3 and antiinflammatory
drugs.4 In the 90s of the past century, the major thrust of
pharmaceutical research involved the development of â- and
γ-secretase inhibitors; however, new problems arose in the form
of low BBB permeability (â) and interference with Notch
signaling (γ).5 Probably the most promising approach has
Neurodegenerative diseases such as Alzheimer’s, Parkinson’s,
and Creutzfeldt-Jacob challenge modern medicinesbecause a
remedy has not yet been found. Patients suffering from
Alzheimer’s disease alone constitute an ever growing percentage
of the western population and amount to >2 million in the U.S.
The underlying cause is well characterized by the term protein-
misfolding diseases: for widely unknown reasons, native
proteins adopt a â-sheet rich abnormal protein conformation
which acts as a seed to nucleate protein misfolding and
subsequent complex formation, until large insoluble aggregates
are formed in the human brain.1 Medical treatment until today
(2) Acetylcholinesterase inhibitors against AD: Kratzsch, T. Neurotrans-
mitter 2002, 3, 52-55.
(3) Phytotherapeutics against AD: (a) Perry, G.; Cash, A. D.; Smith,
M. A. J. Biomed. Biotechnol. 2002, 2, 120-123. (b) Aliev, G.; Obrenovich,
M. E.; Smith, M. A.; Perry, G. J. Biomed. Biotechnol. 2003, 3, 162-163.
(4) Ibuprofen against AD: (a) Lim, G. P.; Yang, F.; Chu, T.; Chen, P.;
Beech, W.; Teter, B.; Tran, T.; Ubeda, O.; Hsiao Ashe, K.; Frautschy, S.
A.; Cole, G. M. J. Neurosci. 2000, 20, 5709-5714. (b) Mackenzie, I. R.
A. Neurology 2000, 54, 732.
(1) (a) Lansbury, P. T., Jr. Acc. Chem. Res. 1996, 29, 317-321. (b)
Cohen, F. E.; Kelly, J. W. Nature 2003, 426, 905-909. (c) Sacchettini, J.
C.; Kelly, J. W. Nat. ReV. Drug DiscoVry 2002, 1, 267-275.
10.1021/jo061918x CCC: $37.00 © 2007 American Chemical Society
Published on Web 04/12/2007
3614
J. Org. Chem. 2007, 72, 3614-3624

Synthesis of Alzheimer Ligands on Solid Support
TABLE 1. New Aminopyrazole and Peptide Derivatives with Their Corresponding Yields Obtained from Solution or Solid-Phase Synthesisa
entry
molecule
type
synthesis
composition
overall yield (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
8a
8b
8c
2
2
2
2
1
3
3
3
3
3
4
solution (sequ.)
solution (sequ.)
solution (sequ.)
solution (sequ.)
SPSS (block)
SPSS (sequ.)
SPSS (sequ.)
SPSS (sequ.)
solution (sequ.)
solution (sequ.)
solution (block)
SPSS (sequ.)
glycole-Pz-K-A-Pz-OMe
glycole-Pz-A-K-Pz-OMe
glycole-Pz-F-K-Pz-OMe
glycole-Pz-A-D-Pz-OMe
Ac-Pz-Pz-K-F-F-OH
glycole-Pz-K-L-V-F-Pz-OH
glycole-Pz-F-F-K-K-Pz-G-OH
glycole-Pz-K-F-F-K-Pz-G-OH
glycole-Pz-F-F-K-V-Pz-OMe
glycole-Pz-F-F-K(Dans)-V-Pz-OMe
Fmoc-K8-Pz-Pz-Pz-OMe
KKLVFF
11
4
18
6
8d
10a
13a
13b
13c
13d
13e
16
17a
17b
17c
17d
3
75
80
90
2
1
27
95
80
75
25
peptide
peptide
peptide
peptide
SPSS (sequ.)
SPSS (sequ.)
solution (sequ.)
KKLVFFAK
KKLVFF-dansyl
dansyl-glycole-AKLVFF-OMe
a n.a. ) not available; n.d. ) not determined; block ) fragment coupling of preformed oligomeric building blocks; sequ. ) strictly linear nonconvergent
elongation protocol.
SCHEME 1. Type 1-4 Aminopyrazole-Peptide Hybrid Compounds Which Were Synthesized and Affinity Tested on Solid
Supporta
a Purple: hydrophilic glycole tail. Blue: peptidic part.
reached clinical trials now; it centers around passive or active
antibody immunization.6
Some years ago, we embarked on a program to design, from
modeling studies, a heterocyclic rigid template which is capable
of recognizing the â-strand conformation of amyloid structures.
The key fragment, a 3-aminopyrazole, binds to every available
hydrogen bond donor and acceptor at the top and bottom face
of a peptide, as soon as it resides in an extended conformation.7
In extension of the concept, this central building block was
converted into a heterocyclic amino acid and combined with
proteinogenic amino acids to hybrid ligands.8 These retain
perfect â-sheet complementarity but introduce, at the same time,
an element of sequence selectivity. The in vitro interaction of
(5) â- and γ-secretase inhibitors against AD: (a) Yamada, K.; Ren, X.;
Nabeshima, T. Jpn. J. Pharmacol. 1999, 80, 9-14. (b) Soto, C. Mol. Med.
Today 1999, 5, 343-350.
(6) Active or passive immunization against AD: (a) Bard, F.; Cannon,
C.; Barbour, R.; Burke, R. L.; Games, D.; Grajeda, H.; Guido, T.; Hu, K.;
Huang, J.; Johnson-Wood, K.; Khan, K.; Kholodenko, D.; Lee, M.;
Lieberburg, I.; Motter, R.; Nguyen, M.; Soriano, F.; Vasquez, N.; Weiss,
K.; Welch, B.; Seubert, P.; Schenk, D.; Yednock, T. Nat. Med. 2000, 6,
916-919. (b) Pepys, M. B.; Herbert, J.; Hutchinson, W. L.; Tennent, G.
A.; Lachmann, H. J.; Gallimore, J. R.; Lovat, L. B.; Bartfai, T.; Alanine,
A.; Hertel, C.; Hoffmann, T.; Jakob-Roetne, R.; Norcross, R. D.; Kemp, J.
A.; Yamamura, K.; Suzuki, M.; Taylor, G. W.; Murray, S.; Thompson, D.;
Purvis, A.; Kolstoe, S.; Wood, S. P.; Hawkins, P. N. Nature 2002, 417,
254-259.
(7) (a) Schrader, T.; Kirsten, C. J. Chem. Soc., Chem. Commun. 1996,
2089. (b) Schrader, T.; Kirsten, C. N. J. Am. Chem. Soc. 1997, 119, 12061-
12068. (c) Rzepecki, P.; Wehner, M.; Molt, O.; Zadmard, R.; Schrader, T.
Synthesis 2003, 1815-1826.
(8) (a) Cˇernovska´, K.; Kemter, M.; Gallmeier, H.; Rzepecki, P.; Schrader,
T.; Ko¨nig, B. Org. Biomol. Chem. 2004, 2, 1603-1611. (b) Rzepecki, P.;
Gallmeier, H.; Geib, N.; Cˇ ernovska´, K.; Ko¨nig, B.; Schrader, T. J. Org.
Chem. 2004, 69, 5168-5178.
J. Org. Chem, Vol. 72, No. 10, 2007 3615

Rzepecki et al.
FIGURE 1. Conventional solution synthesis of type 2 hybrid compounds 8 starting from PMB-protected aminopyrazole carboxylate 1.
FIGURE 2. Automated synthesis of type 1 hybrid compounds starting from a C-terminal peptide.
such hybrid ligands with peptides taken from Aâ or the whole
amyloid peptide(1-42) (aggregation assays with isolated Aâ),9
however, furnished results conflicting with in vivo experiments
(Aâ lesion protocols with living nerve cells), which spurred
further investigations.
To understand the diverging experimental results, it hence
seemed necessary to develop efficient synthetic and screening
protocols for a larger number of aminopyrazole hybrid mol-
ecules, in other words, to create an avenue for lead optimization
in the frame of this model system. Specifically, we envisioned
to facilitate combinatorial screening by a double solid-phase
protocol for both synthesis and affinity testing, which is
described here.
FIGURE 3. Large-scale synthesis of the central PMB- and Fmoc-
protected aminopyrazole carboxylic acid 12 from 3-nitropyrazole-5-
carboxylate for incorporation into the automated SSPS protocol.
tide bridge.10 In principle, all these ligands can be synthesized
from preformed building blocks in solution. However, the
multiple synthesis of similar sequences is extremely labor
Results and Discussion
Solid-Phase Synthesis. Four types of hybrid compounds were
envisaged (Scheme 1), carrying their peptidic fragment on the
N-terminal or the C-terminal end or in between two flanking
pyrazole units. Modeling studies restricted the central peptide
unit to contain an even number of amino acids to maintain
perfect â-sheet complementarity, allowing for a di- or tetrapep-
(9) (a) Schrader, T.; Riesner, D.; Nagel-Steger, L.; Aschermann, K.;
Kirsten, C.; Rzepecki, P.; Molt, O.; Zadmard, R.; Wehner, M. Patent DE
102 21 052.7 of 10/05/2002. (b) Rzepecki, P.; Nagel-Steger, L.; Feuerstein,
S.; Linne, U.; Molt, O.; Zadmard, R.; Aschermann, K.; Wehner, M.;
Schrader, T.; Riesner, D. J. Biol. Chem. 2004, 279, 47479-47505.
3616 J. Org. Chem., Vol. 72, No. 10, 2007

Synthesis of Alzheimer Ligands on Solid Support
FIGURE 4. Automated synthesis of type 3 hybrid compounds starting from the new PMB- and Fmoc-protected aminopyrazole carboxylate 12.
Note the final double acid cleavage steps for consecutive peptide release as well as Boc and PMB removal.
intensive. We therefore developed an efficient solid-phase
synthesis protocol relying on the Fmoc strategy and circumvent-
ing subsequent HPLC purification by precipitating the final
product from organic solution in pure form (Table 1).
For example, type 2 hybrid compounds with dipeptide bridges
were prepared in the following conventional way: the PMB-
protected aminopyrazole ester 1 was elongated with a Z-
protected amino acid by HCTU/Cl-HOBt coupling, followed
by hydrogenolytic Z-removal. These two steps were repeated,
and the resulting dipeptide aminopyrazole 5 was coupled with
an acylaminopyrazole carboxylic acid 6, activated with the
Mukaiyama reagent (1-methyl-2-chloropyridinium iodide)11 to
yield, after double PMB deprotection with hot TFA, the final
product 8 (Figure 1). To increase water solubility, the acyl
protecting group at the N-terminus of 8 was chosen to carry an
oligoethyleneglycole unit. Overall yields for this six-step
procedure varied between 6% and 18%.
FIGURE 5. Automated synthesis of Dansyl-labeled Aâ fragment 17c
starting from Dansylethylenediamine-preloaded Wang resin.
By contrast, most type 1, 3, and 4 hybrids were synthesized
on solid support. Wang resin was used as such or preloaded
with a C-terminal glycine. Fmoc amino acids were coupled with
TBTU and HOBt in DMF12 and deprotected with piperidine.
Aminopyrazole carboxylic acids were coupled with HCTU and
Cl-HOBt or with Mukaiyama’s reagent (vide supra).13 Release
from the resin with concomitant ꢀ-amino Boc removal was
affected with a cleavage cocktail containing TFA (95%), water
(2.5%), and TIPS (2.5%). Only the final PMB deprotection
required harsher conditions, i.e., prolonged heating with TFA
at 70 °C. The hybrid compounds could be precipitated from
cold diethyl ether and were thus obtained, after washing, in
spectroscopically pure form.
Compounds 10 and 11 (type 1) were produced in almost
quantitative yield, albeit in a 97:3 ratio; obviously, the first
coupling step proceeded with incomplete conversion. Here, the
three C-terminal amino acids were coupled sequentially, fol-
lowed by covalent attachment of a preformed dimeric aminopy-
razole building block (Figure 2).
The most economic synthesis yielded three prototypes 13a-c
of an octapeptide hybrid ligand. To this end, 3-nitropyrazole-
5-carboxylic acid was converted into the Fmoc building block
unit in a five-step sequence: esterification and PMB protection
were followed by hydrogenolytic release of the amine func-
tionality and subsequent Fmoc protection. Mild ester dealky-
lation with LiI finally furnished the free acid 12 (Figure 3).14 It
was coupled to the resin with HCTU and Cl-HOBt and
elongated in a 4-fold repetition with a tetrapeptide bridge,
followed by the N-terminal oligoethyleneglycole-protected
aminopyrazole carboxylic acid. Double acid treatment effected
cleavage from the resin, and complete deprotection of the hybrid
compounds, which were again purified by precipitation from
(10) MacroModel 7.2, Force-field: Amber*, OPLS-AA, GBA water
solvation model, 2000 steps molecular mechanics, followed by 2000 steps
Monte-Carlo simulations, and subsequent MD calculation, 300 K, 10 ps.
(11) (a) Bald, E.; Saigo, K.; Mukayima, T. Chem. Lett. 1975, 1163. (b)
Mukaiyama, T. Angew. Chem. 1979, 91, 798. (c) Li, P.; Xu, J. C.
Tetrahedron 2000, 56, 8119.
(12) Carpino, L. A.; Imazumi, H.; El-Faham, A.; Ferrer, F. J.; Zhang,
C.; Lee, Y.; Foxman, B. M.; Henklein, P.; Hanay, C.; Mu¨gge, C.; Wenschuh,
H.; Klose, J.; Beyermann, M. Angew. Chem., Int. Ed. 2002, 41, 441-445.
(13) HCTU: (a) Davis, A. M.; Teague, S. J. Angew. Chem., Int. Ed 1999,
38, 737. (b) Sabatino, G.; Mulinacci, B.; Alcaro, M. C.; Chelli, M.; Rovero,
P.; Papini, A. M. Lett. Pept. Sci. 2002, 9, 119. (c) Marder, O.; Shvo, Y.;
Albericio, F. Chim. Oggi-Chem. Today 2002, 20, 37.
(14) (a) Karaman, R.; Goldblum, A.; Breuer, E. J. Chem. Soc., Perkin
Trans. 1989, 765. (b) Krawczyk, H. Synth. Commun. 1997, 27, 3151-
3161.
J. Org. Chem, Vol. 72, No. 10, 2007 3617

Rzepecki et al.
FIGURE 6. Left: Boc-protected 3-aminopyrazole-5-carboxylic acid for the automated synthesis of hybrid ligands on PAM resin. Right: aminolytic
release of the deprotected ligand (TM ) target molecule) from the PAM resin for analytical purposes.
SCHEME 2. Schematic Illustration of Solid-Phase
Screening Assay on Peptide Affinity
to the peptides remains minute; evaluation of the resulting almost
linear binding isotherms did not proceed to convergence in most
cases. Only an NMR titration in water furnished distinct
chemical shift changes with some of the larger type 2 and 3
hybrid compounds. However, in the presence of 10 mM
phosphate buffer, these changes disappeared completely. This
may well originate from the buffer competition overriding the
weak pyrazole peptide interaction. In the absence of buffer, it
was noted that the pH of the solution shifted toward lower values
(e.g., from 7.0 to 6.7), most likely due to the addition of the
slightly acidic lysine-containing hybrid ligand. Unfortunately,
the diagnostic C-terminal NH proton is also sensitive toward
such a pH change, as demonstrated by a titration with small
amounts of TFA.17 We therefore decided to use our new
automated synthesis protocol on an acid-resistant PAM resin,
so that after PMB deprotection the free aminopyrazole binding
sites can be subjected in an immobilized state to on-bead assays
with labeled peptides.18
diethyl ether. Spectroscopically pure products were obtained in
75-90% total yields over 13 steps on solid support, without
the need for HPLC purification (Figure 4).15
Solid-Phase Binding Studies. The underlying idea of this
approach focuses on the automated synthesis of aminopyrazole
hybrid molecules on the solid phase, their subsequent equilibra-
tion with fluorescence-labeled Aâ fragments, and the calculation
of individual binding constants from reference curves by simple
application of the law of mass action.19 If this concept proves
valid, it can be extended to â-sheet ligand libraries and used
for high-throughput screening (Scheme 2).20 The inverse ap-
proach, i.e., immobilization of peptide libraries and treatment
with a particular fluorescence-labeled aminopyrazole oligomer,
allows us to determine putative sequence selectivities.
A Boc approach was selected because in most resins used
with the Fmoc strategy the bond between the linker and the
substrate is too acid labile. To be able to cleave off small
amounts of product for analytical purposes, a â-alanine func-
tionalized PAM resin was chosen.21 Aminolysis with primary
amines at elevated temperatures releases the product from the
solid support, so that the conventional HF cleavage can be
avoided, which is normally applied during Boc strategy. Boc
protection of the PMB-protected pyrazole amino acid was
impossible with Boc2O but achieved in high yield with Boc-
HOBt.22 Subsequent LiOH saponification furnished the key
building block 24, which was used for SPPS (Figure 6).
Type 4 compounds with more than two aminopyrazole nuclei
are normally notoriously insoluble in water but turn smoothly
water soluble on introduction of one or two lysines. The most
critical part is the repetitive coupling of aminopyrazole units to
each other because the aromatic amine is hardly nucleophilic.
Thus, in spite of multiple coupling steps, total yields remained
moderate in this case. For optimal aminopyrazole condensations,
a combination of HCTU/Cl-HOBt with Hu¨nig’s base was used
for all coupling steps.
Finally, the C-terminally Dansyl-labeled KKLVFF peptide
17c was also prepared on a Dansyl-preloaded resin (Dansyl
NovaTag resin), with a short connecting ethylenediamine spacer
(Figure 5).16 For the largest hybrid ligands with a tetrapeptide
bridge, a representative octapeptide fragment was truncated from
Aâ and synthesized in a nonfluorescent form (H-KKLVF-
FAK-OH).
We conclude that, although shorter versions of aminopyrazole
hybrid ligands may be synthesized by time-consuming multistep
protocols in solution, yields are up to 10-fold higher (up to 90%)
with automated solid-phase peptide synthesis (SPPS) procedures,
and precipitation from ethereal solution even circumvents HPLC
purification. Hence, especially type 2 and 3 hybrid compounds
are now amenable to combinatorial synthesis of compound
libraries for high-throughput screening purposes. It should
certainly be taken into account that the amino acid reactants
are used in a 4-5-fold excess for SPPS, so that yields are only
high compared to the immobilized growing molecule.
Unfortunately, a direct determination of binding constants
between the new hybrid ligands and small Aâ peptide fragments
proved futile by fluorescence techniques. The emission intensity
change of the weak pyrazole chromophore on hydrogen bonding
(17) We cannot rule out that Ka values determined via NMR titrations
in pure water reflect true complex stabilities, but these are always connected
with the uncertainty of a potential influence of a minor pH change. The
diagnostic NH proton is located next to the free C-terminal carboxylate
group; this in turn may be partly protonated by the approaching alkylam-
monium side chain of the ligand’s lysine residue.
(18) A solution-based library for â-peptides: Murray, J. K.; Farooqi,
B.; Sadowsky, J. D.; Scalf, M.; Freund, W. A.; Smith, L. M.; Chen, J.;
Gellman, S. H. J. Am. Chem. Soc. 2005, 127, 13271-13280.
(19) (a) Schmuck, C.; Heil, M. ChemBioChem 2003, 4, 1232-1238.
(b) Wennemers, H.; Conza, M.; Nold, M.; Krattiger, P. Chem.-Eur. J. 2001,
7, 3342-3347.
(20) Yingyongnarongkul, B.; How, S.-E.; Diaz-Mochon, J. J.; Muzerelle,
M.; Bradley, M. Comb. Chem. High Throughput Screening 2003, 6, 577-
587.
(15) (a) Similar problems arise during the synthesis of â-peptide
oligomers: Porter, E. A.; Weisblum, B.; Gellman, S. H. J. Am. Chem. Soc.
2005, 127, 11516-11529. (b) Murray, J. K.; Gellman, S. H. Org. Lett.
2005, 7, 1517-1520.
(16) Dansyl NovaTag resin (Novabiochem): average loading ) 0.59
mmol/g.
(21) PAM resin: Mitchel, A. R.; Kent, S. B. H.; Engelhard, M.;
Merrifield, R. B. J. Org. Chem. 1978, 43, 2845.
3618 J. Org. Chem., Vol. 72, No. 10, 2007

Synthesis of Alzheimer Ligands on Solid Support
FIGURE 7. Top: Structures of immobilized aminopyrazole ligands 25-27, synthesized on the PAM resin. Bottom: Dansyl-labeled AKLVFF
peptide 17d, synthesized for the on-bead assays.
Three representative oligomeric aminopyrazole compounds
were synthesized: trimeric 25, type 2 hybrid 26, and type 4
hybrid 27 (Figure 7). The general procedure involved Boc
removal from the â-alanine linker, followed by repetitive double
coupling with Boc-protected pyrazole amino acid 24 or protei-
nogenic amino acids; lysine side chains were orthogonally
ꢀ-Fmoc protected. To ensure complete conversion, 24 was
preactivated with HATU/HOAt under sonication, and each
coupling step was carried out under argon in the dark.23 An
Nf31 color test proved problematic, most likely due to the
lowered nucleophilicity of the aromatic amine.24 Final deblock-
ing was affected with hot TFA. Gravimetric analyses docu-
mented negligible weight losses due to unwanted linker cleavage
during the whole solid support synthesis. A small amount of
the immobilized ligand was subsequently cleaved off the resin
with N,N-dimethylpropylamine and examined by HPLC-MS
(Scheme 1 in the Supporting Information).25 It revealed that in
the case of type 2 hybrid 26 a byproduct lacking the N-terminal
aminopyrazole was also formed. However, the same status was
found before PMB removal; so, the last coupling step must have
been incomplete (single coupling only), and no TFA cleavage
of the internal amide bond had occurred.
The fluorescent peptide substrate was synthesized in a
convergent approach from a dansyl unit with a hydrophilic
oligoethyleneglycole spacer,26 and the aggregation initiator
sequence AKLVFF. Purification via semipreparative HPLC (RP-
18) afforded 17d as a water-, ethanol-, and chloroform-soluble
material (Figure 7).
Binding studies were carried out in four different media:
chloroform, ethanol, water/acetonitrile 1:1, and phosphate buffer/
acetonitrile 1:1. In these solvents, beads with immobilized
aminopyrazole ligands were incubated for 20 h, at ambient
temperature and in the dark, with the fluorescence-labeled
AKLVFF fragment from Aâ (40 µM labeled peptide solution
over 1 mg of resin and ∼ peptide/ligand ratio 1:100). The
starting value was obtained from a sample without beads,
whereas a linear reference curve for absolute peptide concentra-
tions was generated from a series of peptide samples (Scheme
3). After filtration of the beads from the peptide solution, a
marked decrease in fluorescence intensity was detected (Figure
8). This was not the case if unmodified pure resin was incubated
at identical conditions with labeled peptide; hence, nonspecific
aggregation can be excluded. From the law of mass action, Kass
) ([R - S])/([R][S]), the association constants Kass can be
directly calculated.27 They are summarized in Table 2. Because
of the spatial separation of aminopyrazole ligands on the resin,
exclusive 1:1 complex formation is assumed.
(22) (a) Boc-HOBt: Singh, J.; Fox, R.; Wong, M.; Kissick, T. P.; Moniot,
J. L.; Gougoutas, J. Z.; Malley, M. F.; Kocy, O. J. Org. Chem. 1988, 53,
205. (b) Paquette, L. A. Encyclopedia of Reagents for Organic Synthesis;
John Wiley & Sons: New York, 1995.
(23) (a) HATU: Carpino, L. A.; Imazumi, H.; El-Faham, A.; Ferrer, F.
J.; Zhang, C.; Lee, Y.; Foxman, B. M.; Henklein, P.; Hanay, C.; Mu¨gge,
C.; Wenschuh, H.; Klose, J.; Beyermann, M.; Bienert, M. Angew. Chem.,
Int. Ed. 2002, 114, 458. (b) Speicher, A.; Klaus, T.; Eicher, T. J. Prakt.
Chem./Chem.-Ztg. 1998, 340, 581.
For all aminopyrazole ligands, binding remains relatively
weak in unpolar solution, where hydrogen bonds play a crucial
role (Ka < 200 M-1). On transition to ethanol, affinity to the
Dansyl-labeled hexapeptide is completely lost. However, water
(24) Nf31 color test: Madder, A.; Farcy, N.; Hosten, N. G. C.; De
Muynck, H.; De Clercq, P. J.; Barry, J.; Davis, A. P. Eur. J. Org. Chem.
1999, 2787.
(25) Nucleodur C-18 column, 40 °C; gradient elution with acetonitrile/
water and 0.1% TFA.
(26) Boumrah, D.; Campbell, M. M.; Fenner, S.; Kinsman, R. G.
Tetrahedron 1997, 53, 6977.
(27) Definition of R and S terms: [S] unbound substrate; [R-S] complex;
R [free receptor]. For details see Experimental Section and Supporting
Information.
J. Org. Chem, Vol. 72, No. 10, 2007 3619

Rzepecki et al.
SCHEME 3. Fluorescence Spectra of Peptide 17d Solution after Equilibration with Beads Carrying Immobilized Ligands in
Acetonitrile/Buffer ) 1:1 (Top, 800 µM Sodium Phosphate; Bottom, 10 mM Sodium Phosphate)a
a A ) reference; B,C ) solutions incubated with unmodified PAM resin; D,E ) solutions incubated with PAM resin carrying trimeric 25; F,G )
solutions incubated with PAM resin carrying type 2 hybrid 26; right, corresponding reference curves.
TABLE 2. Binding Constants for Hybrid Molecules 25-27, All Calculated from On-Bead Assays with Fluorescent Peptide 17d in Buffered
Acetonitrilea
Ka [M-1
on-bead
]
1 mM
buffer
10 mM
buffer
Ka [M-1
]
free solution
aminopyrazole
composition
25
25
26
27
trimer (20% impure)
trimer (pure)
PzKKPz
200
500
660
260
180
430
620
n.d.
90
120
130
n.d.
n.a.
n.a.
1700
130
PzPzKVF
a n.a. ) not available; n.d. ) not determined. The experimental error (i.e., the Ka difference between two independent measurements) was calculated at
3-11%.
calculated association constant (∼200 M-1), highlighting the
importance of examining pure materials in this assay. In this
respect, the value for hybrid 26 must probably be doubled and
then compares favorably with the binding constant determined
in free solution by Watergate NMR titrations (∼1700 M-1).28
In 800 mL of phosphate buffer at pH ) 7.0, the peptide affinity
only slightly decreases by ∼5%; however, raising the buffer to
10 mM produces a competitive medium with Ka values around
FIGURE 8. (a) Fluorescent PAM resin beads after equilibration with
the Dansyl-labeled KLVFF peptide. (b) Fluorescence intensity change
of the peptide solution before (left) and after (right) binding to the
immobilized aminopyrazole hybrid ligand.
100 M-1
.
We conclude that the above-described on-bead assay consti-
tutes a valuable new tool to estimate Alzheimer ligand/Aâ
peptide affinities in various solvents, by performing only one
single new measurement. Gratifyingly, host/guest complexation
can be monitored quantitatively even for those ligands, which
proves to be superior even to less polar chloroform, documenting
the obvious importance of additional hydrophobic interactions,
most likely exerted by the LVFF side chains (Ka > 500 M-1).
A resin carrying trimeric 25 with ∼20% impurities furnished a
markedly lower intensity shift than its counterpart with pure
25. Intriguingly, this eventually leads to a distinctly lowered
(28) (a) Piotto, M.; Saudek, V.; Sklenar, V. J. Biomol. NMR 1992, 2,
661-665. (b) Sklenar, V.; Piotto, M.; Leppik, R.; Saudek, V. J. Magn.
Res. (A) 1993, 102, 241-245.
3620 J. Org. Chem., Vol. 72, No. 10, 2007

Synthesis of Alzheimer Ligands on Solid Support
1.00 equiv) of N-benzyloxycarbony-(S)-alaninyl-5-amino-2-(4-
methoxybenzyl)-2H-pyrazole-3-carboxylic acid methyl ester (2a)
gave 178 mg (536 µmol, 85%) of 2b as a colorless solid. 1H NMR
(300 MHz, CDCl3): δ [ppm] ) 1.42 (d, 3J ) 7.1 Hz, 3H), 3.74 (s,
are almost insoluble in water (e.g., 25). This paves the way
toward a combinatorial ligand screening as well as toward a
deeper understanding of energetic contributions from different
structural ligand or guest features. In the case of 25-27, e.g.,
the hybrid ligand seems to be superior to the other two
structures, perhaps because the dipeptide bridge allows for an
induced fit toward perfect â-sheet complementarity.
3
3H), 3.83 (m, 4H), 5.58 (s, 2H), 6.80 (d, J ) 8.7 Hz, 2H), 7.17
(d, 3J ) 8.7 Hz, 2H), 7.26 (s, 1H), 9.97 (brs, 1H). MS (ESI): m/z
) 333 (M + H)+. HRMS: calcd for C16H21O4N4, 333.1557; found,
333.1560. Rf 0.12 in ethyl acetate.
In the future, we intend to attach the Dansyl label via an
extended oligoethyleneglycol to our best aminopyrazole hybrid
ligands and equilibrate these fluorescent ligands with small
libraries of immobilized peptides, preferably truncated overlap-
ping versions of Aâ1-42. This will allow us to identify the hot
spot(s) on Aâ for aminopyrazole binding. Alternatively, a series
of ESI-MS measurements of free ligand libraries will help to
identify the best binders among them, thus a 1:1 complex could
very recently be detected between the above-discussed hybrid
peptide 10b (∼inverted 27) and Aâ1-40 (data not shown).
General Procedure C for Coupling of Z-Protected Amino
Acids with Hybrid Peptides (Amino Acid-Aminopyrazole
Methyl Carboxylate) at Their Free N-Termini. N-Benzyloxy-
carbonyl-(S)-amino acid (1.00 mmol, 1.00 equiv) was dissolved
under argon in a 3:1 mixture of dry dichloromethane and DMF
(5.0 mL) and treated with HCTU (414 mg, 1.00 mmol, 1.00 equiv),
Cl-HOBt (424 mg, 2.50 mmol, 2.50 equiv), and 2,6-lutidine (349
µL, 3.00 mmol, 3.00 equiv). For preactivation, this mixture was
stirred in an ice bath for 30 min. In a separate flask, the respective
hybrid peptide (1.00 mmol, 1.00 equiv) was dissolved under argon
in the same solvent mixture (2.0 mL) and treated with 2,6-lutidine
(116 µL, 1.00 mmol, 1.00 equiv). Both solutions were combined,
and the resulting mixture was stirred for 3.5 h. The organic phase
was washed three times with 1 M HCl, then three times with
saturated aq NaHCO3, and finally three times with saturated aq
NaCl. After drying over MgSO4 and filtration, the organic solvent
was removed in vacuo. The crude product was subsequently purified
by chromatography over silica gel eluting with dichloromethane/
methanol (35:1). After evaporation of the solvent, the hybrid product
was isolated as a colorless solid.
Experimental Section
I. Solution-Phase Synthesis Protocols for Type 2 Hybrid
Compounds. For each general procedure, one representative
example is given in detail. For all other related derivatives, please
refer to the Supporting Information material.
General Procedure A for Coupling of Z-Protected Amino
Acids with PMB-Protected Aminopyrazole Methyl Carboxylate
1. N-Benzyloxycarbonyl-(S)-amino acid (1.00 mmol, 1.00 equiv)
was dissolved under argon in a 3:1 mixture of dry dichloromethane
and DMF (5.0 mL) and treated with HCTU (414 mg, 1.00 mmol,
1.00 equiv), Cl-HOBt (424 mg, 2.50 mmol, 2.50 equiv), and 2,6-
lutidine (349 µL, 3.00 mmol, 3.00 equiv). For preactivation, this
mixture was stirred in an ice bath for 30 min. In a separate flask,
5-amino-2-(4-methoxybenzyl)-2H-pyrazole-3-carboxylic acid meth-
yl ester 1 (261 mg, 1.00 mmol, 1.00 equiv) was dissolved under
argon in the same solvent mixture (2.0 mL) and treated with 2,6-
lutidine (116 µL, 1.00 mmol, 1.00 equiv). Both solutions were
combined, and the resulting mixture was stirred for 3.5 h. The
organic phase was washed three times with 1 M HCl, then three
times with saturated aq NaHCO3, and finally three times with
saturated aq NaCl. After drying over MgSO4 and filtration, the
organic solvent was removed in vacuo. The crude product was
subsequently purified by chromatography over silica gel eluting
with dichloromethane/methanol (35:1). After evaporation of the
solvent, the hybrid product was isolated as a colorless solid.
N-Benzyloxycarbonyl-(S)-alaninyl-5-amino-2-(4-methoxyben-
zyl)-2H-pyrazole-3-carboxylic Acid Methyl Ester 2a. A portion
of 223 mg (1.00 mmol, 1.00 equiv) of N-benzyloxycarbonyl-(S)-
N-Benzyloxycarbonyl-N-tert-butyloxycarbonyl-(S)-lysinyl-(S)-
alaninyl-5-amino-2-(4-methoxybenzyl)-2H-pyrazole-3-carboxy-
lic Acid Methyl Ester 4a. A portion of 188 mg (493 µmol, 1.10
equiv) of N-benzyloxycarbonyl-N-tert-butyloxycarbonyl-(S)-lysine
and 149 mg (448 µmol, 1.00 equiv) of (S)-alaninyl-5-amino-2-(4-
methoxybenzyl)-2H-pyrazole-3-carboxylic acid methyl ester (3a)
1
gave 313 mg (487 mmol, 99%) of 4a as a pale yellow solid. H
NMR (300 MHz, CDCl3): δ [ppm] ) 1.23-1.57 (m, 16H), 1.59-
1.68 (m, 1H), 1.86-1.93 (m, 1H), 2.97-3.04 (m, 2H), 3.72 (s,
3H), 3.84 (s, 3H), 4.23 (brs, 1H), 4.72-4.79 (m, 1H), 5.11 (s, 2H),
5.49 (d, 2J ) 8.1 Hz, 1H), 5.59 (d, 2J ) 8.1 Hz, 1H), 6.20 (d, 3J)
3
3
14.7 Hz, 2H), 6.74 (d, J ) 14.6 Hz, 2H), 6.92 (d, J ) 8.1 Hz,
3
1H), 7.11 (d, J ) 8.6 Hz, 2H), 7.25-7.42 (m, 6H), 9.59 (s, 1H).
MS (ESI): m/z ) 595 (M + H)+, 717 (M + Na)+. HRMS: calcd
for C35H46N6O9Na, 717.3218; found 717.3205. Rf 0.49 in ethyl
acetate.
General Procedure D for N-Terminal Z-Deprotection from
Hybrid Peptides with Two Consecutive Amino Acids. The
N-terminally Z-protected hybrid peptide (∼500 µmol, 1.00 equiv)
was dissolved in methanol (25 mL), treated with 5 mol % of Pd/C
(70 mg, 0.033 mmol, 0.07 equiv, Degussa Type E101 NE/W), and
stirred under a hydrogen atmosphere for 16 h. The reaction mixture
was filtered over celite or silica gel, and the organic solvent was
evaporated to dryness. The crude product was purified by chro-
matography over silica gel eluting with dichloromethane/methanol
(10:1). In some cases, a greenish impurity precipitated on dissolving
the purified compound in methanol for an NMR sample. This
procedure was subsequently used for further purification and
explains the suboptimal yields in two cases.
1
alanine gave 308 mg (660 µmol, 66%) of product 2a. H NMR
(300 MHz, CDCl3): δ [ppm] ) 1.43 (d, 3J ) 7.1 Hz, 3H), 3.75 (s,
3H), 3.85 (s, 3H), 4.45 (brs, 1H), 5.11 (d, 2J ) 12.2 Hz, 1H), 5.16
(d, 2J ) 12.2 Hz, 1H), 5.47 (d, 3J ) 6.8 Hz), 5.57 (s, 2H), 6.79 (d,
3
3J ) 8.7 Hz, 2H), 7.17 (d, J ) 8.7 Hz, 2H), 7.25-7.41 (m, 6H),
8.87 (brs, 1H). MS (ESI): m/z ) 489 (M + Na)+. HRMS: calcd
for C24H26O6N4Na, 489.1745; found, 489.1743; Rf 0.48 in ethyl
acetate.
General Procedure B for N-Terminal Z-Deprotection from
Hybrid Peptides with One Amino Acid. The N-terminally
Z-protected hybrid peptide (∼500 µmol, 1.00 equiv) was dissolved
in methanol (25-50 mL), treated with 5 mol % of Pd/C (70 mg,
0.03 mmol, 0.07 equiv, Degussa Type E101 NE/W), and stirred
under a hydrogen atmosphere for 16 h. The reaction mixture was
filtered over celite or silica gel and washed three times with
methanol, and the organic solvent was evaporated to dryness. The
crude product was purified by chromatography over silica gel
eluting with dichloromethane/methanol (10:1).
N-tert-Butyloxycarbonyl-(S)-lysinyl-(S)-alaninyl-5-amino-2-(4-
methoxybenzyl)-2H-pyrazole-3-carboxylic Acid Methyl Ester
(5a). A portion of 293 mg (422 µmol, 1.00 equiv) of N-benzy-
loxycarbonyl-N-tert-butyloxy-carbonyl-(S)-lysinyl-(S)-alaninyl-5-
amino-2-(4-methoxybenzyl)-2H-pyrazole-3-carboxylic acid methyl
ester (4a) gave 176 mg (314 mmol, 74%) of 5a as a colorless solid.
1H NMR (300 MHz, CDCl3): δ [ppm] ) 1.25-1.51 (m, 16H),
1.52-1.76 (m, 1H), 1.76-1.93 (m, 1H), 3.02-3.12 (m, 2H), 3.61-
3.71 (m, 1H), 3.72 (s, 3H), 3.81 (s, 3H), 4.63-4.71 (m, 1H, R-H),
3
(S)-Alaninyl-5-amino-2-(4-methoxybenzyl)-2H-pyrazole-3-
carboxylic Acid Methyl Ester 3a. A portion of 296 mg (635 µmol,
4.86 (brs, 1H, R-H), 5.56 (s, 2H), 6.76 (d, J ) 8.5 Hz, 2H), 7.13
3
(d, J ) 8.5 Hz, 2H), 7.22 (s, 1H), 8.12 (s, 1H), 9.50 (s, 1H). MS
J. Org. Chem, Vol. 72, No. 10, 2007 3621

Rzepecki et al.
(ESI): m/z ) 561 (M + H)+, 583 (M + Na)+. HRMS: calcd for
C27H41N6O7, 561.3031; found, 561.3027. Rf 0.05 in ethyl acetate.
For the related solution-phase synthesis of the type 2 hybrid
peptides Glycole-Pz-K-V-Pz-OMe 8e and Glycole-Pz-K-
K-Pz-OMe 8f, see ref 29.
General Procedure E for Coupling of PMB- and Glycole-
Protected Aminopyrazole Methyl Carboxylates with Hybrid
Peptides at Their Free N-Termini. The respective side chain and
nucleus-protected hybrid peptide (∼0.50 mmol, 1.00 equiv) was
dissolved in dry dichloromethane (10 mL) under argon and treated
with 2-(4-methoxybenzyl)-5-{2-[2-(2-methoxy-ethoxy)-ethoxy]-
acetylamino}-2H-pyrazole-3-carboxylic acid (6, 1.05 equiv), 3-chloro-
1-methylpyridinium iodide (Mukaiyama’s reagent, 1.20 equiv), and
diisopropylethylamine (DIEA, 3.00 equiv). The reaction mixture
was stirred at ambient temperature for 16 h. Subsequently, the
organic layer was washed three times with 1 M HCl, three times
with saturated aq NaHCO3, and finally three times with saturated
aq NaCl. After drying over MgSO4 and filtration, the solvent was
evaporated to dryness. The resulting crude product was purified
by chromatography over silica gel eluting with dichloromethane/
methanol 50:1 f 10:1 (gradient elution).
II. Automated Solid-Phase Synthesis Protocols of Type 1-4
Hybrid Compounds: Type 1 Hybrid Compounds Acyl-Pz-
Pz-AA1-AA2-AA3, General Procedure H. The complete
synthesis was performed with the Advanced ChemTech model
Apex396. Nonpreloaded Wang-resin with an average loading of
1.2 mmol/g was used as the polymeric carrier. Prior to the first
coupling step, the resin was swollen in DMF (2 mL) for 2 h. The
critical coupling steps with Fmoc amino acids employed coupling
reagents TBTU and HOBt and DIEA as a base. Each coupling
required 4.0 equiv of Fmoc amino acid, 4.0 equiv of TBTU, 4.0
equiv HOBt, and 8.0 equiv of DIEA. The first amino acid as well
as the last building block 3-{[5-acetylamino-2-(4-methoxybenzyl)-
2H-pyrazole-3-carbonyl]-amino}-2-(4-methoxybenzyl)-2H-pyrazole-
3-carboxylic acid 97 were doubly coupled. Removal of the Fmoc
protecting group was effected with 20% piperidine in DMF (2 ×
10 min). Each coupling and Fmoc deprotection was followed by
washing the resin with dry DMF.
The peptide was cleaved off the resin concomitant with depro-
tection of lysine’s ꢀ-amino-Boc groups via 3-fold treatment with
an acidic cleavage cocktail (95% TFA, 2.5% water, and 2.5% TIPS).
In the first step, 400 µL of this cocktail was added to the reaction
vessel, shaken for 45 min, and filtered off. This procedure was
repeated twice (300 µL, 45 min; 200 µL, 15 min).
By bubbling N2 through the solution, the TFA volume was
reduced to ∼0.3 mL. It was then cooled to 0 °C, and the PMB-
protected hybrid peptide was precipitated with cold diethyl ether.
For complete PMB removal, the crude intermediate was redissolved
in TFA (2.0 mL) and heated for 2 h with stirring to 70 °C. The
solution was cooled in an ice bath and pipetted into cold diethyl
ether for precipitation of the final product, which was centrifuged
off. The colorless solid was taken up three times in diethyl ether,
vortexted, and again centrifuged, to separate impurities.
2-[2-(2-Methoxy-ethoxy)-ethoxy]-acetyl-N-2-(4-methoxyben-
zyl)-2H-pyrazolcarboxyl-N-tert-butyloxycarbonyl-(S)-lysinyl-(S)-
alaninyl-5-amino-2-(4-methoxybenzyl)-2H-pyrazol-3-carbon-
sa1uremethyl Ester 7a. A portion of 176 mg (313 µmol, 1.00 equiv)
of N-tert-butyloxycarbonyl-(S)-lysinyl-(S)-alaninyl-5-amino-2-(4-
methoxybenzyl)-2H-pyrazole-3-carboxylic acid methyl ester (5a)
and 134 mg (330 µmol, 1.05 equiv) of 2-(4-methoxybenzyl)-5-{2-
[2-(2-methoxy-ethoxy)-ethoxy]-acetylamino}-2H-pyrazole-3-car-
boxylic acid (6) gave 195 mg (205 µmol, 62%) of 7a as a colorless
1
3
solid. H NMR (300 MHz, CDCl3): δ [ppm] ) 1.23 (d, J ) 6.6
Hz, 3H), 1.24-1.48 (m, 13H), 1.62-1.77 (m, 1H), 1.77-1.93 (m,
1H), 2.88-3.06 (m, 2H), 3.30 (s, 3H), 3.52-3.71 (m, 8H), 3.71,
3.72 (2s, 6H), 3.83 (s, 3H), 4.11 (d, 2J ) 16.4 Hz, 1H), 4.19 (d, 2J
) 16.2 Hz, 1H), 4.48-4.60 (m, 1H), 4.73-4.98 (m, 2H), 5.46-
5.69 (m, 4H), 6.75, 6.76 (2d, 3J ) 8.8 Hz, 3J′ ) 8.6 Hz, 4H), 7.06,
7.18 (2d, 3J ) 8.8 Hz, 3J′ ) 8.6 Hz, 4H), 7.27, 7.35 (2s, 2H), 8.09
3
(d, J ) 8.1 Hz, 1H), 9.28, 10.13 (2s, 2H). 13C NMR (300 MHz,
5-Acetylamino-2H-pyrazole-3-carbonyl-amino-2H-pyrazole-
3-carbonyl-(S)-lysinyl-(S)-phenylalaninyl-(S)-phenylalanine Tri-
fluoroacetate 10a. The NMR spectrum revealed incomplete
coupling of the first amino acid Phe and formation of major
byproduct 11a, which could be sparated from 10a by HPLC
chromatography (125/4 Nucleodur 100-5 C-18, Macherey and
Nagel). Thus, 48 µg (68 nmol) of 10a and 979 µg (1.77 mmol) of
11a were obtained in pure form (MS detection). MALDI-TOF: m/z
) 701 (10a + H)+, 554 (11a + H)+.
CDCl3): δ [ppm] ) 17.6, 22.06, 27.4, 28.5, 29.7, 48.0, 50.9, 52.2,
52.5, 52.8, 54.1, 57.9, 69.1, 69.3, 69.6, 70.3, 70.8, 77.9, 97,9, 101.7,
112.7, 112.8, 127.5, 127.7, 128.1, 128.1, 128.5, 131.0, 133.6, 143.6,
144.8, 155.0, 158.0, 158.1, 159.0, 167.4, 169.3, 170.5. MS (ESI):
m/z ) 950 (M + H)+, 972 (M + Na)+, 988 (M + K)+. HRMS:
calcd for C46H63N9O13, 972.4438; found, 972.4424. Rf 0.12 in ethyl
acetate.
General Procedure F for Total Deprotection of All Hybrid
Peptides in One Step with Hot TFA. A portion of ∼200 µmol
(1.00 equiv) of the side chain and nucleus-protected complete hybrid
peptide was dissolved in dry trifluoroacetic acid (2.0 mL, HPLC
grade) and stirred under argon for 3 h at 70 °C. Subsequently, the
solution was cooled to 0 °C and treated with ice-cold diethyl ether
(∼25 mL) in a Falcon tube. The precipitating colorless solid was
centrifuged off, washed three times with cold diethyl ether, and
dried.
2-[2-(2-Methoxy-ethoxy)-ethoxy]-acetylaminopyrazolecarbo-
nyl-(S)-lysinyl-(S)-alaninyl-aminopyrazolecarboxylic Acid Meth-
yl Ester Trifluoroacetate 8a. A portion of 185 mg (195 µmol,
1.00 equiv) of 2-[2-(2-methoxy-ethoxy)-ethoxy]-acetyl-N-2-(4-
methoxybenzyl)-2H-pyrazolecarbonyl-N-tert-butyloxycarbonyl-(S)-
lysinyl-(S)-alaninyl-5-amino-2-(4-methoxybenzyl)-2H-pyrazole-3-
carboxylic acid methyl ester (7a) gave 79 mg (83 µmol, 43%) of
8a as a colorless solid. 1H NMR (300 MHz, CD3OD): δ [ppm] )
1.45 (d, 3J ) 7.1 Hz, 3H), 1.47-1.58 (m, 2H), 1.60-1.78 (m, 2H),
1.80-1.89 (m, 1H), 1.77-1.93 (m, 1H), 2.89-3.02 (m, 2H), 3.34
(s, 3H), 3.52-3.79 (m, 8H), 3.89 (s, 3H), 4.18 (s, 2H), 4.43-4.66
(m, 2H), 6.90, 6.95 (2s, 2H). 13C NMR (500 MHz, CDCl3): δ [ppm]
) 17.9, 23.3, 23.6, 28.1, 29.1, 32.6, 32.8, 40.5, 50.8, 52.5, 52.7,
54.2, 59.0, 71.3, 71.3, 72.2, 72.8, 96.8, 100.3, 161.6, 163.6, 170.8,
172.9, 173.7. MS (ESI): m/z ) 610 (M + H)+, 632 (M + Na)+,
648 (M + K)+. HRMS: calcd for C25H40N9O9, 610.29; found,
610.29.
For the solution synthesis of Ac-Pz-Pz-K-V-F-OMe 10b,
see ref 29.
Type 3 Hybrid Compounds Glycole-Pz-AA1-AA2-AA3-
AA4-Pz-OH, General Procedure G. The complete synthesis was
performed with the Advanced ChemTech model Apex396. Non-
preloaded Wang-resin with an average loading of 1.2 mmol/g was
used as the polymeric carrier. The critical coupling steps with Fmoc
amino acids employed coupling reagents HCTU and Cl-HOBt and
DIEA as a base. Each coupling required 4.0 equiv of Fmoc amino
acid, 4.0 equiv of HCTU, 4.0 equiv of Cl-HOBt, and 8.0 equiv of
DIEA. Removal of the Fmoc protecting group was effected with
20% piperidine in DMF (2 × 10 min). Each coupling and Fmoc
deprotection was followed by washing the resin with dry DMF.
Prior to the first coupling, the resin was swollen for 2 h, to
optimize the accessability of all styrene groups and circumvent
sterical problems. The first building block N-4-methoxybenzyl-3-
(9-fluorenylmethoxycarbonyl)-aminopyrazole-5-carboxylic acid (12)
was attached to the resin via double coupling (coupling time 2.5
h). Due to the low nucleophilicity of the free aromatic aminopy-
razole amino group, the first amino acid was also introduced via a
double coupling protocol with extended coupling times of ∼2.5 h.
The other three amino acids were coupled under standard conditions
(29) Rzepecki, P.; Schrader, T. J. Am. Chem. Soc. 2005, 127, 3016-
3025.
3622 J. Org. Chem., Vol. 72, No. 10, 2007

Synthesis of Alzheimer Ligands on Solid Support
within 1 h. Finally, another double coupling (2.5 h) elongated the
growing hybrid molecule by the 2-(4-methoxybenzyl)-5-{2-[2-(2-
methoxy-ethoxy)-ethoxy]-acetylamino}-2H-pyrazole-3-carboxyl-
ic acid fragment (6). The peptide was cleaved off the resin
concomitant with deprotection of lysine’s ꢀ-amino-Boc groups via
3-fold treatment with an acidic cleavage cocktail (95% TFA, 2.5%
water, and 2.5% TIPS). In the first step, 400 µL of this cocktail
was added to the reaction vessel, shaken for 45 min, and filtered
off. This procedure was repeated twice (300 µL, 45 min; 200 µL,
15 min).
By bubbling N2 through the solution, the TFA volume was
reduced to ∼0.3 mL. It was then cooled to 0 °C, and the PMB-
protected hybrid peptide was precipitated with cold diethyl ether.
For complete PMB removal, the crude intermediate was redissolved
in TFA (2.0 mL) and heated for 2 h with stirring to 70 °C. The
solution was cooled in an ice bath and pipetted into cold diethyl
ether for precipitation of the final product, which was centrifuged
off. The colorless solid was taken up three times in diethyl ether,
vortexted, and again centrifuged, to separate impurities.
Fmoc amino acid, 3.0 equiv of HBTU, 6.0 equiv of HOBt, and 6.0
equiv of DIEA. Coupling times were ∼1 h; prior to each coupling
step and after each Fmoc deprotection, the resin was washed with
DMF. The Fmoc group was mildly removed with 20% piperidine
in DMF (2 × 10 min). Peptide release from the resin was affected
with the cleavage cocktail (95% TFA, 2.5% water, and 2.5% TIS).
The mixture was cooled to 0 °C, and the PMB-protected hybrid
1
peptide was precipitated with cold diethyl ether. H NMR (500
MHz, CD3OD): δ [ppm] ) 0.77 (d, 3J ) 6.6 Hz, 3H), 0.83 (d, 3J
3
3
) 6.6 Hz, 3H), 0.88 (d, J ) 6.4 Hz, 3H), 0.93 (d, J ) 6.4 Hz,
3
3H), 1.38 (d, 3H, J ) 7.0 Hz), 1.43-2.07 (m, 22H), 2.78-3.19
(m, 10H), 3.89-3.99 (m, 1H), 4.06-4.17 (m, 1H), 4.28-4.46 (m,
4H), 4.52-4.68 (m, 2H), 7.13-7.33 (m, 10H). MS (ESI): m/z )
980 (M + H)+. HRMS: calcd for C50H82N11O9, 980.6291; found,
980.6292.
(c) [5-Dimethylamino-naphthyl-1-sulfonic acid]-(S)-lysinyl-
(S)-lysinyl-(S)-leucinyl-(S)-valinyl-(S)-phenylalaninyl-(S)-pheny-
lalaninyl-(2-amino-ethyl)-amide Trifluoroacetate 17c. The syn-
thesis was performed on the Advanced ChemTech model Apex396.
Dansyl NovaTag resin with an average loading of 0.59 mmol/g
was used as the polymeric carrier. The synthesis followed a very
similar protocol as that for 17a. All amino acids were coupled twice.
Coupling reagents were HCTU/Cl-HOBt. Peptide release from the
resin was affected with the cleavage cocktail (95% TFA, 2.5%
water, and 2.5% TIS). The mixture was cooled to 0 °C, and the
PMB-protected hybrid peptide was precipitated with cold diethyl
ether. 1H NMR (500 MHz, CD3OD): δ [ppm] ) 0.79 (d, 3J ) 6.6
2-[2-(2-Methoxy-ethoxy)-ethoxy]-acetylaminopyrazolecarbo-
nyl-(S)-lysinyl-(S)-leucinyl-(S)-valinyl-(S)-phenylalaninyl-ami-
nopyrazolecarboxylic Acid Trifluoroacetate 13a. Overall yield
75%. 1H NMR (500 MHz, CD3OD): δ [ppm] ) 0.78 (d, 3J ) 6.6
3
3
Hz, 3H), 0.82 (d, J ) 6.6 Hz, 3H), 0.88 (d, J ) 6.4 Hz, 3H),
3
0.93 (d, J ) 6.4 Hz, 3H), 1.41-1.73 (m, 10H), 2.81-2.96 (m,
2H), 2.83-3.23 (m, 2H), 3.35 (s, 3H), 3-54-3.81 (m, 8H), 4.18
(s, 2H), 4.38-4.51 (m, 1H), 4.52-4.59 (m, 1H, R-H), 4.59-4.68
(m, 2H, R-H’s), 6.88 (brs, 2H), 7.06-7.28 (m, 5H). 13C NMR (500
MHz, CDCl3): δ [ppm] ) 18.7, 19.8, 21.9, 23.5, 25.9, 28.1, 32.2,
32.6, 39.1, 40.6, 41.6, 59.1, 71.3, 71.4, 72.2, 72.8, 127.8, 129.5,
129.5, 130.3, 130.4, 138.2, 170.8, 173.0. MS (ESI): m/z ) 922
(M + K)+. HRMS: calcd for C41H61N11O11, 922.4752; found,
922.4757.
Type 4 Hybrid Compounds AA1-AA2-Pz-Pz-Pz-OMe,
General Procedure I. The synthesis followed a very similar
protocol to the above-described type 3 hybrid peptides. In this case
glycine-preloaded Wang-resin was used, with an average loading
of 0.8 mmol/g. All pyrazole building blocks and the first lysine
were doubly coupled for 2.5 h.
Automated Solid-Phase Synthesis Protocols for KKLVFF,
Dansyl-KKLVFF, and KKLVFFAK. (a) (S)-Lysinyl-(S)-lysi-
nyl-(S)-leucinyl-(S)-valinyl-(S)-phenylalaninyl-(S)-phenylala-
nine Trifluoroacetate 17a. The complete synthesis was performed
with the Advanced ChemTech model Apex396. Nonpreloaded
Wang-resin with an average loading of 1.2 mmol/g was used as
the polymeric carrier. The first Phe was attached to the resin by a
double coupling step. Coupling reagents were HOBt/HBTU or
HCTU/Cl-HOBt. Release of the peptide from the resin was affected
with the above-decribed cleavage cocktail (95% TFA, 2.5% water,
and 2.5% TIS).
3
3
Hz, 3H), 0.83 (d, J ) 6.7 Hz, 3H), 0.87 (d, J ) 6.4 Hz, 3H),
3
0.92 (d, J ) 6.4 Hz, 3H), 1.35-2.07 (m, 16H), 2.60-3.22 (m,
18H), 3.95 (t, 1H, 3J ) 6.2 Hz), 4.13 (d, 1H, 3J ) 7.4 Hz), 4.34-
3
3
4.49 (m, 3H), 4.57 (d, J ) 5.9 Hz, 1H, Phe R-HR), 4.60 (d, J )
5.9 Hz, 1H, Phe R-Hâ), 7.03-7.23 (m, 10H), 7.27 (d, 3J ) 7.4 Hz,
1H, Naphthyl-H), 7.57 (d, 3J ) 7.6 Hz, 1H, Naphthyl-H), 7.60 (d,
3J ) 7.6 Hz, 1H, Naphtalin-H), 8.18 (d, 3J ) 7.4 Hz, 1H, Naphtalin-
3
3
H), 8.32 (d, J ) 8.5 Hz, 1H, Naphthyl-H), 8.58 (d, J ) 8.5 Hz,
1H, Naphthyl-H). MS (ESI): m/z ) 1057 (M + H)+, 1079 (M +
Na)+. HRMS: calcd for C55H82N11O8S, 1056.6063; found, 1056.6043.
Synthesis of PAM-Immobilized Aminopyrazole Ligands 25-
27 Employing the Boc Strategy. Immobilized 3-{[5-({5-[(5-
Amino-1H-pyrazole-3-carbonyl)-amino]-1H-pyrazole-3-carbonyl}-
amino)-1H-pyrazole-3-carbonyl]-amino}-propionyl-PAM (PAM-
â-Ala-Pz-Pz-Pz) 25. For these ligands, a manual solid-phase
synthesis protocol was employed. PAM resin was used as a polymer
support, with an average loading of 0.8 mmol/g. Coupling of Boc-
protected amino acids was accomplished with HATU/HOAt and
lutidine according to the following method: For each coupling step
[5-tert-butoxycarbonylamino-2-(4-methoxybenzyl)-2H-pyrazol-3-
yl]-carboxylic acid 24 (5.0 equiv),22 HATU (4.5 equiv), HOAt (12.5
equiv), and lutidine (15 equiv) were used in DMF solution. To this
end, the aminopyrazole carboxylic acid (24) was preactivated ∼10
min prior to the peptide coupling in an ultrasound bath with HATU/
HOAt and lutidine in DMF in a brown glass flask. Removal of the
Boc-protecting group occurs by treatment with TFA (1.6 mL),
dichloromethane (400 µL), and triisopropylsilane (100 µL, 1 × 5
min, 1 × 20 min). The resin was washed thoroughly with
dichloromethane before and after each deprotection step. To cleave
the PMB protecting groups on the pyrazole nucleus, the resin was
heated under argon in dry TFA for 2.5 h to 70 °C. The solution
was subsequently filtered, and the resin was washed with dichlo-
romethane and dried in vacuo. The complete protocol is listed in a
summarizing table, found in the Supporting Information.
By bubbling N2 through the solution, the TFA volume was
reduced to ∼0.3 mL. It was then cooled to 0 °C, and the PMB-
protected hybrid peptide was precipitated with cold diethyl ether.
The colorless solid was taken up three times in diethyl ether,
1
vortexted, and again centrifuged, to separate impurities. H NMR
3
(500 MHz, CD3OD): δ [ppm] ) 0.78 (d, J ) 6.6 Hz, 3H), 0.83
(d, 3J ) 6.6 Hz, 3H), 0.88 (d, 3J ) 6.4 Hz, 3H), 0.93 (d, 3J ) 6.4
Hz, 3H), 1.27-2.03 (m, 16H), 2.80-2.98 (m, 4H), 2.98-3.21 (m,
4H), 3.96 (brs, 1H), 4.16 (d, 1H, 3J ) 7.6 Hz), 4.36-4.50 (m, 2H),
3
4.55-4.71 (m, 2H), 7.12-7.30 (m, 10H), 7.88 (d, 1H, J ) 6.4
Hz, amide-H), 8.16, 8.42, 8.66 (3brs, 3H, amide-H). MS (ESI):
m/z ) 782 (M + H)+, 804 (M + Na)+, 820 (M + K)+. HRMS:
calcd for C41H65N8O7, 781.4971; found, 781.4965.
On-Bead Assay: Calculation of Binding Constants. The on-
bead assays were conducted under rigorous light exclusion. Solvents
were dry dichloromethane, acetonitrile (HPLC-grade), bidistilled
water, and sodium phosphate buffer (pH 7.0). A defined amount
of resin was weighed into an Eppendorf tube (870 µg-5.50 mg)
and treated with a defined amount of a 40 µM solution of
fluorescence-labeled peptide 17d (ratio ligand/peptide ) 100:1).
Usually, [the amount of resin with a loading of 0.8 mmol/g in
(b) (S)-Lysinyl-(S)-lysinyl-(S)-leucinyl-(S)-valinyl-(S)-pheny-
lalaninyl-(S)-phenylalaninyl-(S)-alaninyl-(S)-lysine Trifluoro-
acetate 17b. The synthesis was performed on the Advanced
ChemTech model Apex396. Nonpreloaded chlorotrityl-chloride
resin with an average loading of 1.4 mmol/g was used as the
polymeric carrier. Couplings were carried out with 3.0 equiv of
J. Org. Chem, Vol. 72, No. 10, 2007 3623

Rzepecki et al.
milligrams] × 200 ) the amount of the solution in microliters. In
all cases, two to three samples of immobilized 25-27, two samples
of nonfunctionalized Boc-â-alanine PAM resin, and one sample
without a polymer were preincubated overnight under light exclu-
sion at 20 °C. The last mentioned sample served as a reference; it
revealed potential unspecific fluorescence intensity changes because
it was treated under the same conditions as all resin-containing
samples. To achieve an optimal distribution of polymer beads, each
solution was treated with an ultrasound for 30 s under light
exclusion prior to incubation. After 16-18 h the solutions were
pipetted off the resin, and the remaining fluorescence intensity was
measured. The labeled peptide was measured at various concentra-
tions in the appropriate solvent to obtain a reference curve.
Instrumental Values. Excitation wavelength: 310 nm. Measure-
ment range: 450-600 nm. Data pitch: 1 nm. Scanning speed: 1000
nm/min. Response: medium. Sensitivity: medium. Bandwidth-
(Ex): 5 nm. Bandwidth(Em): 10 nm.
From the fluorescence intensities of peptide reference and
solutions after incubation, the association constant can be calculated,
if a 1:1 complex is formed.
[S]: Concentration of unbound substrate which can be determined
from the remaining fluorescence intensity of the solution with a
reference curve.
[R - S]: Complex concentration which is obtained from the
difference between original fluorescence intensity and fluorescence
intensity after incubation, [R - S] ) [S0] - [S].
[R]: Concentration of the free receptor which can be calculated
from the employed amount of receptor (amount of resin, loading,
and correction factor for the molar mass of the bound receptor)
and the concentration of the receptor-substrate complex.
Acknowledgment. Financial support from the Volkswagen
foundation is gratefully acknowledged. We thank Dr. Uwe
Linne, Marburg, for the ESI-MS experiment with Aâ1-40.
Supporting Information Available: Full characterization of
compounds 2b-20. Detailed information about the solid-phase
synthesis protocol for 25-27 on the PAM resin. Scheme 1: HPLC
traces for 25-PMB and free 25. 1H NMR spectra of reported
compounds. This material is available free of charge via the Internet
at http://pubs.acs.org.
[R - S]
[R][S]
Kass
)
JO061918X
3624 J. Org. Chem., Vol. 72, No. 10, 2007