Design and synthesis of a novel DNA-encoded chemical library using Diels-Alder cycloadditions

Article abstract of DOI:10.1016/j.bmcl.2008.07.038

DNA-encoded chemical libraries are increasingly being employed for the identification of binding molecules to protein targets of pharmaceutical relevance. Here, we describe the synthesis and characterization of a DNA-encoded chemical library, consisting of 4000 compounds generated by Diels-Alder cycloaddition reactions. The compounds were encoded with unique DNA fragments which were generated through a stepwise assembly process and serve as amplifiable bar codes for the identification and relative quantification of library members.

Full text of DOI:10.1016/j.bmcl.2008.07.038

Bioorganic & Medicinal Chemistry Letters 18 (2008) 5926–5931
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of a novel DNA-encoded chemical library
using Diels-Alder cycloadditions q
Fabian Buller a, Luca Mannocci a, Yixin Zhang a, Christoph E. Dumelin b, Jörg Scheuermann a, Dario Neri a,
*
a Institute of Pharmaceutical Sciences, Department of Chemistry and Applied Biosciences, ETH Zurich, Wolfgang-Pauli-Strasse 10, CH-8093 Zurich, Switzerland
b Philochem AG, c/o ETH Zurich, Wolfgang-Pauli-Strasse 10, CH-8093 Zurich, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
DNA-encoded chemical libraries are increasingly being employed for the identification of binding mole-
cules to protein targets of pharmaceutical relevance. Here, we describe the synthesis and characterization
of a DNA-encoded chemical library, consisting of 4000 compounds generated by Diels-Alder cycloaddi-
tion reactions. The compounds were encoded with unique DNA fragments which were generated through
a stepwise assembly process and serve as amplifiable bar codes for the identification and relative quan-
tification of library members.
Received 16 June 2008
Accepted 10 July 2008
Available online 15 July 2008
Keywords:
DNA-encoded chemical library
Diels-Alder cycloaddition
Streptavidin
Ó 2008 Elsevier Ltd. All rights reserved.
Albumin
High-throughput sequencing
The isolation of small organic molecules, capable of specific
binding to proteins, is a central problem in chemistry, biology,
and medicine. Whenever the target proteins of interest display a
measurable enzymatic activity or have labeled ligands which can
be used in displacement assays, large libraries of chemical
compounds can be screened individually (one molecule at a time)
in order to identify novel ligands. These hits are then selected for
further optimization by medicinal chemistry.1 Such high-through-
put screening approaches, which are routinely used by the phar-
maceutical industry and increasingly also by academic research
centers, may be expensive both in terms of target protein require-
ments and library-associated costs (e.g., synthesis, quality controls,
library management, screening assays, robotics). Thus, practical
screening campaigns are limited to a few hundred thousand
compounds at best. Alternative screening procedures have been
proposed, such as the incubation of the target protein with a mix-
ture of ꢀ100 compounds, followed by the separation of molecules
capable of protein interaction via gel filtration.2 While these
approaches are interesting in terms of reducing screening costs
and times as well as for the identification of ligands to proteins
for which biochemical screening methods are not available, they
may nonetheless be limited by compound solubility, mass spectro-
metric detection of binding compounds, and library sizes which
can realistically be screened.
The use of DNA fragments as amplifiable ‘bar-codes’ for the
identification of chemical compounds in a library represents an
attractive avenue for the synthesis and screening of large combina-
torial libraries.3–7 Several strategies can be considered for the con-
struction of DNA-encoded chemical libraries, for example the use
of self-assembling chemical libraries where each of the two com-
plementary DNA strands carries a different chemical moiety (‘dual
pharmacophore chemical libraries’)8–10 or the use of individual
compounds covalently attached to unique DNA fragments. In turn,
these ‘single pharmacophore chemical libraries’ can be generated
by a variety of synthetic approaches, including the transfer of
chemical moieties from assisting complementary oligonucleo-
tides,4,11 the oligonucleotide-assisted separation of DNA deriva-
tives followed by chemical reactions,5 the stepwise alternated
growth of chemical structures and encoding DNA fragments,12 or
simply the direct coupling of reactive chemical moieties to differ-
ent amino-tagged oligonucleotides.13,14
In general, the construction of DNA-encoded chemical libraries
is facilitated by the ability to perform chemical reactions at high
yields and with good purities, using synthetic methodologies
which work well in water and preserve the integrity of DNA. While
until now most libraries are relying on amide bond forming reac-
tions,4,8,12,15 other reactions could be considered, such as reductive
amination and ether bond formation.
q
Manuscript for a special Symposium-in-Print (SIP) issue in honor of the recipient
of the 2008 Young Investigator Award in Bioorganic and Medicinal Chemistry,
Benjamin F. Cravatt.
* Corresponding author. Tel.: +41 44 6337401.
E-mail address: neri@pharma.ethz.ch (D. Neri).
In this Letter, we describe the construction of a DNA-encoded
chemical library, consisting of 4000 compounds attached to indi-
vidual double-stranded DNA fragments carrying unique DNA se-
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.07.038

F. Buller et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5926–5931
5927
quences that allow the unambiguous identification of the dis-
played chemical moieties. The synthesis of the library was
achieved using a split-and-pool procedure, which featured the fol-
lowing sequential steps: (i) conjugation of different dienes to dis-
tinct aminomodified synthetic oligonucleotides; (ii) pool and
split; (iii) Diels-Alder reaction with suitable maleimide derivatives;
(iv) encoding of the cycloaddition reaction by hybridization of par-
tially complementary oligonucleotides followed by Klenow-medi-
ated DNA polymerization, yielding the final compounds in a
double-stranded DNA format. The purity of the intermediate steps
in the library synthesis was extensively investigated using HPLC
and mass spectrometry, while an assessment of library perfor-
mance was provided by the decoding of selections performed
against two model target proteins.
cules to two well-characterized target proteins: albumin and
streptavidin.
Bicyclic Diels-Alder derivatives were obtained synthetically by
cycloaddition from 2,4-hexadienes and maleimides as dienophiles
(Fig. 1A). In order to evaluate the feasibility of library synthesis
based on this experimental strategy, we synthesized a small proto-
type library based on N-Acetyl-O-2(E),4(E)-hexadienyl-
L-hydroxy-
proline, which was then coupled to synthetic 42-mer
a
oligonucleotide, carrying a primary amino group connected with
a C12-linker at the 50 extremity (Fig. 1A). The resulting DNA-conju-
gate was HPLC purified and submitted in parallel to 54 different
cycloaddition reactions with maleimide derivatives, which were
either commercially available or generated in situ from the corre-
sponding primary amines and maleic anhydride (Fig. 1A, supple-
mentary methods). HPLC and ESI-MS analysis of the reaction
confirmed the identity of all 54 products (supplementary Table
1), with an average yield of 64%. In a final synthetic step, the oligo-
nucleotide conjugates were hybridized to partially complementary
DNA-fragments, which carried a unique sequence for the identifi-
Attracted by the mild conditions at which certain Diels-Alder
cycloaddition reactions can be performed in water,16–18 we
investigated the possibility to generate Diels-Alder products,
which are chemically coupled to DNA. These DNA-compound
conjugates were then used for the selection of binding mole-
Figure 1. (A) Reaction scheme of the 54-compound prototype DNA-encoded chemical library using Diels-Alder cycloadditions between a 2(E),4(E)-hexadienyl moiety and
maleimide derivatives. (a) Maleic anhydride/DMF. (b) Ac2O or HOAc, 80 °C. (c) EDC/sulfo-NHS mediated coupling in DMSO/H2O 75 mM TEA-Cl, pH 10. (d) Diels-Alder
cycloaddition in DMF/acetate buffer, pH 4.7. (e) Klenow-Polymerase-mediated encoding of the obtained Diels-Alder structure upon hybridization with a partially
complementary oligonucleotide carrying a unique sequence. (B) Schematic representation of the strategy used for library synthesis and encoding of the 4000 compound
library. (a) Coupling of 20 carboxylic acids containing a 2(E),4(E)-hexadienyl moiety to 20 unique 50-amino-modified oligonucleotides. (b) Pool and split of the 20-compound
sub-library into 200 vessels. (c) 200 Diels-Alder cycloadditions with maleimides and subsequent Klenow-Polymerase encoding. (d) Pooling of the 200 reactions to yield a
DNA-encoded library containing 4000 members.

5928
F. Buller et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5926–5931
cation of individual chemical moieties. Conversion into a double-
stranded DNA fragment was achieved by Klenow polymerization
(Fig. 1A).
To assess whether a DNA-encoded chemical library based on
the bicyclic Diels-Alder scaffold could display chemical moieties
capable of selective interaction with target proteins immobilized
on a solid support, the 54-compound library was panned onto
streptavidin–sepharose and onto albumin–sepharose. Three of
maleimide derivatives) had previously been characterized as suit-
able ligands for these target proteins.13,14,19 In Figure 2 the results
of decoding experiments are depicted, showing the relative abun-
dance of library compounds after capture on streptavidin resin, on
human serum albumin resin, or on empty resin produced from
CNBr-activated sepharose, which was quenched with Tris. The
DNA-encoded chemical library was submitted to an asymmetric
PCR amplification using a fluorescently labeled primer. The result-
ing mixture was then hybridized onto an oligonucleotide micro-
array, carrying oligonucleotides complementary to the individual
the 54 building blocks used for library construction (
D-desthiobio-
tin-, 6-ethoxy-6-oxohexanoyl-, and 4-(p-iodophenyl)butyl-based
Figure 2. (A) Microarray readout after PCR amplification of selections with the 54-compound library against streptavidin, human serum albumin and of a control selection
with no antigen on resin (spotted in quintuplicate). The inset highlights the signal enhancement of compounds 52–54, included as positive controls in the 54-compound
library. (B) Quantitative microarray evaluation: Enrichment of oligonucleotide-compound conjugates given as a ratio of selection versus control selection. (C) Affinity
chromatography of enriched (Streptavidin: conjugates 53 and 54, HSA: conjugates 35, 52) and non-enriched (conjugate 7) conjugates, each hybridized to a radiolabeled
complementary oligonucleotide. The retained radioactivity was monitored during multiple washing steps. (D) Structures of control conjugates 52–54 and of a novel binder to
HSA, conjugate 35.

F. Buller et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5926–5931
5929
codes of the compounds in the library.9,10,13,14 Figure 2A shows the
fluorescence images of the microarrays of three selection experi-
ments, revealing a preferential enrichment of compounds 53 and
54 in streptavidin selections and of compounds 52 and 35 in albu-
min selections. Figure 2B presents a quantitative evaluation of the
spot fluorescence intensities in the decoding microarrays, display-
Figure 3. Synthesis scheme of 2(E),4(E)-hexadienyl derivatives. (A) Synthesis of 2a–m starting from primary amines. I—Na2CO3, THF; II—MeCN. (B) Synthesis of 4a–f starting
from carboxylic acids. I—NaH, THF; II—HATU, DIPEA, DMF.
Table 1
MS characterization of 20 synthesized diene–oligonucleotide conjugates
Oligonucleotide-
conjugate
2,4-Hexadienyl-derivative
used for conjugation
6 base-pair code
Deconvoluted mass
of diene–oligonucleotide
conjugate
Calculated mass
of diene–oligonucleotide
conjugate
Deconvoluted
mass of Diels-Alder
producta
Calculated mass
of Diels-Alder
product
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
2a
2b
2c
2d
2e
2f
2g
2h
2i
2j
2k
2l
2m
4a
4b
4c
4d
4e
4f
CAACGC
ACGCGC
CACGAA
GGCACA
GAGGAG
CTACCG
CTATAT
TTGTTT
GCTCAT
AAACAT
GGGAGC
CAGAGT
TGCTCT
CTGTGA
CGGATG
GTAGTA
CACTTT
TGGGCG
GTCGCT
TTGCGG
13666.4
13694.0
13610.6
13685.0
13706.8
13663.0
13627.6
13687.8
13629.2
13681.6
13802.8
13658.6
13679.2
13768.6
13741.4
13791.0
13671.8
13784.6
13735.6
13576.6
13661.0
13689.3
13608.0
13680.1
13702.1
13657.0
13621.0
13682.0
13624.0
13675.1
13797.2
13652.1
13673.2
13763.1
13737.1
13786.1
13666.0
13779.3
13730.1
13576.2
13876.0
13883.6
13792.6
13852.0
13794.8
13881.6
13828.8
13863.0
13831.0
13865.2
13913.8
13808.2
13887.8
13930.2
13878.6
13928.8
13898.4
13903.2
13918.8
13770.4b
13867.3
13879.6
13790.5
13846.3
13788.3
13872.3
13822.2
13861.2
13824.4
13858.3
13907.4
13803.5
13882.5
13923.3
13872.3
13922.3
13891.3
13898.5
13914.3
13770.4
4g
a
Diels-Alder cycloaddition with the maleimide product prepared from 2-phenylethylamine and maleic anhydride. Oligonculeotide code: ATC TTA.
Oligonculeotide code: ATG GAG.
b

5930
F. Buller et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5926–5931
ing ratios of fluorescence signals in the selection experiments rel-
ative to the control selection on empty resin. In order to confirm
preferential binding to the target proteins, we radiolabeled the
individual DNA conjugates and measured their ability to be cap-
tured by their cognate proteins, which were immobilized on resin
(Fig. 2C). The retention profiles correlated with the binding affini-
ties of the ligands.13,14 The relative enrichment factors are given in
Figure 2B.
Encouraged by these results, we decided to synthesize a 4000
member library, based on a two-step split-and-pool strategy, as
depicted in Figure 1B. Initially, twenty 2,4-hexadiene derivatives
carrying a carboxylic acid moiety were synthesized (see scheme
in Fig. 3 and supplementary methods) and coupled to amino-
tagged oligonucleotides carrying distinctive six base codes. The
resulting derivatives were HPLC purified and characterized by
ESI-MS (Table 1). In all cases, yields were greater than 50% as as-
sessed by HPLC. A representative HPLC profile and deconvoluted
ESI-MS spectrum of a diene conjugate are shown in Figure 4A
and B.
suggest that endo diastereomers may be preferentially formed.16
When DNA-compound conjugates preferentially bind to a target
protein of interest, individual compounds can be resynthesized in
the absence of DNA and stereoisomers can be analyzed. It is con-
ceivable that the bicyclic derivatives may be further optimized
for medicinal chemistry applications through suitable modification
strategies. As to the drug-likeliness of the Diels-Alder bicyclic scaf-
fold, it was reassuring for us that a number of lead pharmaceutical
compounds based on the Diels-Alder reaction with maleimides are
currently in preclinical and clinical development.20,21
The DNA-encoded library presented in this article represents
one of the libraries which is currently being developed in our lab-
oratory.8,12–14 While in principle DNA-encoded chemistry could
yield libraries of huge dimensionalities, if multiple reaction steps
were used within split-and-pool strategies, we have initially
focused on the construction of individual libraries containing
In order to assess the ability of the diene conjugates to effi-
ciently undergo Diels-Alder cycloaddition with maleimide deriva-
tives, a model reaction was performed for all 20 intermediates
with 2-phenylethylamine and maleic anhydride. A representative
deconvoluted ESI-MS profile for the crude single-pot reaction mix-
ture after ethanol precipitation is displayed in Figure 4C, revealing
an excellent purity. All calculated and experimental mass values
for the 20 model reactions can be found in Table 1.
We then pooled the 20 diene DNA-conjugates, split them into
200 aliquots, and performed cycloaddition reactions with 200
maleimides or maleimide-generating reagents (Fig.1B, supplemen-
tary methods). As a final step, the second code was introduced to
each DNA-encoded compound by hybridization with partially
complementary oligonucleotides and subsequent Klenow poly-
merization (Fig. 1B), in perfect analogy with the strategy described
in the previous section. Figure 5A depicts in more detail the two-
step encoding strategy employed for the library construction.
The 200 reactions were pooled in 20 vials and analyzed by TBE/
polyacrylamide gel electrophoresis, confirming the purity and
efficiency of the Klenow encoding procedure (Fig. 5B). Finally, the
library was pooled in a single vial and aliquoted to a final concen-
tration of 100 nM (total DNA content). For selection experiments, a
total DNA concentration of 10 nM was used, corresponding to a
2.5 pM concentration of individual library members. The integrity
of library encoding was confirmed by PCR amplification of the
library with primers 50-GCCTCCCTCGCGCCATCAGGGAGCTTGTGA
ATTCTGG-30 and 50-GCCTTGCCAGCCCGCTCAGGTAGTCGGATCCGAC
CAC-30, complementary to the extremities of the DNA fragments in
the library, followed by subcloning and sequencing. Furthermore,
the suitability of the library for selection experiments was con-
firmed by a capture experiment on streptavidin resin, which
revealed a striking preferential enrichment of library members
containing a D-desthiobiotin moiety as building block B (Fig. 1B,
incorporated as a positive control in the library construction, data
not shown).
A Diels-Alder cycloaddition reaction represented the central
step in our library synthesis, yielding bicyclic derivatives with good
purities and yields. Since only a limited number of suitable malei-
mide derivatives were commercially available, we developed a
one-pot reaction featuring the in situ formation of maleimides
starting from maleic anhydride and primary amines. After dilution
of the original reaction mixture with water and pH adjustment, the
2,4-hexadiene–oligonucleotide conjugates were added. This
approach proved to be compatible with a variety of primary
amines and 2,4-hexadiene derivatives (supplementary Table 2).
We could, however, not analyze the stereochemistry of the bicyclic
moiety when coupled to the DNA. Investigations in solution
Figure 4. Example of the purification and MS-characterization of diene–oligonu-
cleotide conjugates. (A) HPLC profile after coupling of 4c to an 50-aminomodifided
oligonucleotide (conjugate 3) (solid line: Abs260 dashed line: Abs280). (B) LC-ESI-MS
characterization of conjugate 3 after deconvolution of the spectra. (C) LC-ESI-MS
characterization of the Diels-Alder product of conjugate 3 and the maleimide
prepared from 2-phenylethylamine and maleic anhydride. (D) Structure schemes.

F. Buller et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5926–5931
5931
Figure 5. (A) Schematic representation of the Klenow encoding of an oligonucleotide-compound conjugate AxBy. x, forward primer site, compatible with primers for 454-
high-throughput sequencing. y, annealing site for the partially complementary oligonucleotides carrying code 2. z, reverse primer site, compatible with primers for 454-high-
throughput sequencing. (B) 20% polyacrylamide-TBE gel analysis of the encoding of 50 Diels-Alder reactions. M, 10 bp Marker; –, single-stranded 42-mer prior to encoding;
1–5, sub-pools after encoding of 10 reactions (68 mers). Staining was performed with Sybr green I.
1000–10,000 compounds, which featured a maximum of two syn-
thetic steps, and concentrated on reactions, which proceed with
excellent yields. Care was always taken to characterize library con-
struction intermediates by HPLC and LC-ESI-MS, and to validate the
last synthetic step with a number of model compounds. Ultimately,
a direct analytical characterization of the individual components in
a large library is not feasible.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.07.038.
References and notes
In principle, the library could be expanded by the use of alterna-
tive diene or maleimide intermediates, or by further modification
of a suitable functional group of moiety A or moiety B (Fig. 1B).
In this case, a third code for the last reaction step can be introduced
in a sequential manner.12
1. Drews, J. Science 2000, 287, 1960.
2. Muckenschnabel, I.; Falchetto, R.; Mayr, L. M.; Filipuzzi, I. Anal. Biochem. 2004,
324, 241.
3. Brenner, S.; Lerner, R. A. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 5381.
4. Gartner, Z. J.; Tse, B. N.; Grubina, R.; Doyon, J. B.; Snyder, T. M.; Liu, D. R. Science
2004, 305, 1601.
5. Halpin, D. R.; Harbury, P. B. PLoS Biol. 2004, 2, 1022.
6. Melkko, S.; Dumelin, C. E.; Scheuermann, J.; Neri, D. Drug Discov. Today 2007,
12, 465.
7. Scheuermann, J.; Dumelin, C. E.; Melkko, S.; Neri, D. J. Biotechnol. 2006, 126,
568.
8. Melkko, S.; Scheuermann, J.; Dumelin, C. E.; Neri, D. Nat. Biotechnol. 2004, 22,
568.
9. Melkko, S.; Zhang, Y.; Dumelin, C. E.; Scheuermann, J.; Neri, D. Angew. Chem.,
Int. Ed. 2007, 46, 4671.
10. Scheuermann, J.; Dumelin, C. E.; Melkko, S.; Zhang, Y.; Mannocci, L.; Jaggi, M.;
Sobek, J.; Neri, D. Bioconjug. Chem. 2008, 19, 778.
11. Li, X.; Liu, D. R. Angew. Chem., Int. Ed. 2004, 43, 4848.
12. Mannocci, L.; Zhang, Y.; Scheuermann, J.; Leimbacher, M.; De Bellis, G.; Rizzi,
E.; Dumelin, C. E.; Melkko, S.; Neri, D. 2008, submitted for publication.
13. Dumelin, C. E.; Scheuermann, J.; Melkko, S.; Neri, D. Bioconjug. Chem. 2006, 17,
366.
14. Dumelin, C. E.; Trussel, S.; Buller, F.; Trachsel, E.; Bootz, F.; Zhang, Y.; Mannocci,
L.; Beck, S. C.; Drumea-Mirancea, M.; Seeliger, M. W.; Baltes, C.; Muggler, T.;
Kranz, F.; Rudin, M.; Melkko, S.; Scheuermann, J.; Neri, D.Angew. Chem., Int. Ed.
2008, 47, 3196.
15. Halpin, D. R.; Lee, J. A.; Wrenn, S. J.; Harbury, P. B. PLoS Biol. 2004, 2, E175.
16. de Araujo, A. D.; Palomo, J. M.; Cramer, J.; Seitz, O.; Alexandrov, K.; Waldmann,
H. Chemistry 2006, 12, 6095.
17. Hill, K. W.; Taunton-Rigby, J.; Carter, J. D.; Kropp, E.; Vagle, K.; Pieken, W.;
McGee, D. P.; Husar, G. M.; Leuck, M.; Anziano, D. J.; Sebesta, D. P. J. Org. Chem.
2001, 66, 5352.
In contrast to other library screening methodologies with mix-
tures of compounds, for which identification procedures become
more and more problematic with increasing library sizes, DNA
encoding enables the use of individual library members at very
low concentrations, provided that the library is captured with a
biotinylated target protein at concentrations greater than Kd, or
on resins coated with protein at high density (>1 mg/ml), in full
analogy to peptide and antibody phage display libraries, in which
rare binders can be efficiently recovered and identified.22,23 The ad-
vent of high-throughput sequencing technologies, such as 454
sequencing,24 opens the possibility of profiling the relative abun-
dance of individual library members before and after a single pan-
ning step, thus providing direct information about the relative
enrichment rate for all compounds in the library.12
In short, we have described the design, synthesis, and character-
ization of a DNA-encoded chemical library, containing 4000 com-
pounds individually coupled to DNA fragments, which serve as
amplifiable identification bar codes. The use of a Diels-Alder strat-
egy with 2,4-hexadiene modified oligonucleotides and a variety of
different maleimides proved to yield cycloaddition products in
good yields and purities, as characterized by HPLC and LC-ESI-
MS. The library is compatible with decoding strategies based on
ultra-high-throughput sequencing technologies (e.g., 454 sequenc-
ing, Roche), which enables the relative quantification of individual
library members before and after selection against target proteins
of interest.
18. Lindstrom, U. M. Chem. Rev. 2002, 102, 2751.
19. Hirsch, J. D.; Eslamizar, L.; Filanoski, B. J.; Malekzadeh, N.; Haugland, R. P.;
Beechem, J. M. Anal. Biochem. 2002, 308, 343.
20. Lee, J. Y.; Hancock, A. A.; Warner, R. B.; Brune, M. E.; Meyer, M. D.; DeBernardis,
J. F. Pharmacology 1998, 56, 17.
21. Rickels, K.; Derivan, A.; Kunz, N.; Pallay, A.; Schweizer, E.
Psychopharmacol. 1996, 16, 212.
22. Silacci, M.; Brack, S.; Schirru, G.; Marlind, J.; Ettorre, A.; Merlo, A.; Viti, F.; Neri,
D. Proteomics 2005, 5, 2340.
J. Clin.
23. Winter, G.; Griffiths, A. D.; Hawkins, R. E.; Hoogenboom, H. R. Annu. Rev.
Immunol. 1994, 12, 433.
Acknowledgments
24. Margulies, M.; Egholm, M.; Altman, W. E.; Attiya, S.; Bader, J. S.; Bemben,
L. A.; Berka, J.; Braverman, M. S.; Chen, Y. J.; Chen, Z.; Dewell, S. B.; Du, L.;
Fierro, J. M.; Gomes, X. V.; Godwin, B. C.; He, W.; Helgesen, S.; Ho, C. H.;
Irzyk, G. P.; Jando, S. C.; Alenquer, M. L.; Jarvie, T. P.; Jirage, K. B.; Kim, J.
B.; Knight, J. R.; Lanza, J. R.; Leamon, J. H.; Lefkowitz, S. M.; Lei, M.; Li, J.;
Lohman, K. L.; Lu, H.; Makhijani, V. B.; McDade, K. E.; McKenna, M. P.;
Myers, E. W.; Nickerson, E.; Nobile, J. R.; Plant, R.; Puc, B. P.; Ronan, M. T.;
Roth, G. T.; Sarkis, G. J.; Simons, J. F.; Simpson, J. W.; Srinivasan, M.;
Tartaro, K. R.; Tomasz, A.; Vogt, K. A.; Volkmer, G. A.; Wang, S. H.; Wang,
Y.; Weiner, M. P.; Yu, P.; Begley, R. F.; Rothberg, J. M. Nature 2005, 437,
376.
Financial support from the ETH Zurich, the Swiss National Sci-
ence Foundation, Philochem AG, KTI (Grant No. 8868.1 PFDS-LS),
and the Gebert-Rüf Foundation is gratefully acknowledged. The
authors are grateful to the Functional Genomics Center Zurich for
help in DNA-encoded chemical library decoding through oligonu-
cleotide microarrays, and to Dr. Samu Melkko (Philochem AG) for
reviewing the manuscript.