On-bead screening of a library to detect host–guest complexation by an aniline reporter

Abstract

We developed a new labeling reagent and a color assay system in water to detect binding between target molecules and library members on beads, which is free of label-induced artifacts that can cause misleading results.

---

Combinatorial chemistry is a powerful tool for rapidly discovering desired molecules such as new drugs and materials.¹ The “one-bead-one-compound” (OBOC) combinatorial library method can easily generate thousands to millions of compounds using a “split-mix” synthesis procedure,^2–4 which can be assayed against a wide variety of labeled targets to discover their ligands^2–8 or synthetic receptors.^9–14 The on-bead color assay for the OBOC library is a very rapid screening method and can be used to easily assess qualitative binding abilities. Wennemers and Still, however, reported that many simple dyes bind certain peptides in water with significant selectivity.¹⁵ Great care should thus be taken to avoid label-induced artifacts.^9,11,13–15 Recently, we reported that the addition of a non-ionic surfactant, Triton-X, can suppress the binding interaction.¹⁶ It is possible, however, that the additive also affects target–ligand interactions. An additive-free screening environment is ideal for estimating real interactions. Trinder reagents¹⁷ such as ALPS (1)¹⁸ are chromogenic reagents for colorimetric determination of hydrogen peroxide activity. The reagents (hydrogen donors) are oxidatively condensed with 4-aminoantipyrine (2, 4-AA) in the presence of H₂O₂ and peroxidase (shown in Scheme 1). This system is widely used for diagnostic assays and biochemical examinations because hydrogen peroxide is produced by enzymatic oxidation of substrates such as glucose. We focused on the Trinder reagent for labeling targets, since it is expected not to bind certain peptides because the reagent is small and less polar than typical dyes and fluorophores.

Scheme 1 Trinder reaction.

At first, some aniline linked carboxylic acids 4–10 (Fig. 1) were treated with 4-AA in the presence of H₂O₂ and horseradish peroxidase (HRP) in water at pH 6.86 to explore the highly sensitive aniline derivatives. The absorption maxima and molar absorption coefficients of the chromogens are shown in Table 1. The chromogen from N-phenylglycine (4) had less than half the intensity of that from the commercially available Trinder reagent ALPS (1). The color intensity of chromogens was decreased by the introduction of methoxy groups on the aromatic ring¹⁹ (compounds 5 and 6). Unexpectedly, the intensity was drastically decreased when a spacer group was connected to 4 by an amide linkage (compound 7). On the other hand, the intensity of the chromogen from 4-(ethylphenylamino)butyric acid (8) was 1.6 times higher than that from 4. Although the introduction of a spacer group decreased the intensity, the molar absorption coefficient of the chromogen remained at approximately 1800 (compound 9). The intensity of the chromogen from the corresponding methyl ester 10 was five times higher than that from 9. The reason for the drastic change in absorbance is unclear. As shown in Fig. 2, absorbance of chromogen from methyl ester 10 reached a maximum within 5 min and then gradually decreased.

Fig. 1 Hydrogen donors in the Trinder reaction.

Table 1 Wavelengths of maximum absorption (λ_max) and molar absorption coefficients (ε) of the chromogens from hydrogen donors 1 and 4–10^a

Compound	λ max/nm	ε
a The data were recorded in pH 6.86 phosphate buffer.
1	560	14499
4	534	6879
5	526	2509
6	582	2104
7	530	71
8	558	11042
9	562	1829
10	560	9861

---

Fig. 2 Time dependence of color development of the chromogen from 10. The absorbance was recorded at 560 nm.

Next, the carboxylic acid 9 was loaded on Novasyn™ TG resin²⁰ to investigate the coloration on a solid support. Resin 11 immediately displayed a striking purple color following treatment with 4-AA, H₂O₂, and HRP in pH 6.86 buffer (Fig. 3). The color of the control resin 12 was invariant with the same treatment.

Fig. 3 Coloration by the Trinder reaction on a solid support 11 (A) and control experiment (B, resin 12).

A previous study indicated that the resin-bound peptidocalix[4]arene 13b binds dye-labeled tripeptide 14a.¹⁶ Based on this result, tripeptide 16a, which had the same peptide sequence as 14a, was synthesized. The mixture of resin-bound peptidocalix[4]arene 13b and N-acetylaminomethylated resin (control resin) 12 was incubated with a labeled peptide 16a in pH 6.86 phosphate buffer. After agitation for 12 h, the beads were rinsed several times with the buffer, and subsequently treated with 4-AA, HRP, and H₂O₂. The mixture was incubated at 38 °C for 5 min and inspected under a low-power microscope. Large beads stained purple and small beads had no color, which indicated that the present method can be applied to an on-bead binding assay. The detection limit for 16a was less than 13 µmol L⁻¹ (Fig. 4).

Fig. 4 Detection of host (13b, large beads)–guest (16a) complexation on beads by the Trinder reaction in various guest concentrations with control resin (12, small beads). Guest concentration: (A) 1.3 mmol L⁻¹, (B) 130 µmol L⁻¹, (C) 13 µmol L⁻¹.

Prior to screening of the library, we confirmed that the labeling group did not bind any library members. No colored beads were observed in the screening of a peptidocalix[4]arene library 13a, consisting of 1000 members, even with 1.3 mmol L⁻¹ of 17 (Fig. 5A). In contrast , almost 10% of the beads were stained in the screening of 13a with corresponding dye-labeled compound 15 (5 µmol L⁻¹, Fig. 5B). Thus, the present method avoids misleading results caused by label-induced artifacts. Application to an actual screening of 13a with an aniline-labeled tripeptide 16a is shown in Fig. 6. There were only a few colored beads. All colored beads were isolated and decoded to identify their amino acid sequences. The sequences on the beads mainly consisted of two or three Glu (Table 2). The preference for an acidic amino acid was consistent with that obtained for dye-labeled tripeptide 14a, which has the same peptide sequence as 16a.¹⁶

Fig. 5 Screening of the library 13a for binding to 17 (1.3 mmol L⁻¹) by the Trinder reaction (A) and for binding to 15 (5 µmol L⁻¹) by color assay (B) in pH 6.86 phosphate buffer.

Fig. 6 Screening of the library 13a for binding to 16a (13 µmol L⁻¹) by the Trinder reaction in pH 6.86 phosphate buffer.

Table 2 Peptide sequences of the library members for binding to a guest peptide 16a

Entry	AA1	AA2	AA3	Frequency^a
a Number of beads having the indicated sequences.
1	Glu	Glu	Glu	4
2	Glu	Glu	Tyr	4
3	Glu	Leu	Glu	4
4	Glu	Pro	Glu	3
5	Glu	Glu	Pro	2
6	Ser	Glu	Glu	2
7	Glu	Asn	Glu	1
8	Glu	Glu	Ser	1
9	Glu	Ser	Glu	1
10	Glu	Tyr	Glu	1
11	Tyr	Glu	Glu	1
Total				24

---

Finally, screening of the library 13a for binding with some tripeptides 16 was performed to compare with the results of the conventional dye-labeled method by compounds 14, reported previously.¹⁶ In the screening for tripeptide 14a and 16a, some colored beads were found, in both cases as described above. There were no colored beads found, however, for binding to 16b, which has a different peptide sequence order than 16a. In contrast, in the corresponding screening of 14b by the dye-labeled method, there were faintly colored beads. These were detectable under a fluorescence microscope. Similar results were obtained for peptides 14c and 16c. These observations indicated that the present method is comparable in sensitivity with a color assay, but less sensitive than fluorescence detection (Fig. 7).

Fig. 7 Screening of the library 13a for binding to some tripeptides (compounds shown in parentheses under each picture). A series: detection by the Trinder reaction (13 µmol L⁻¹ of peptides in pH 6.86 phosphate buffer) B and C series: color detection (B) and fluorescence detection (C) (5 µmol L⁻¹ of peptides in pH 6.86 phosphate buffer containing 5% Triton X).

In summary we have developed a novel color assay system in water without additives for detecting host–guest complexation on beads. This assay system is free of label-induced artifacts that might lead to misleading results.

Notes and references

1. K. S. Lam, M. Lebl and V. Krchnak, Chem. Rev., 1997, 97, 411–448.
2. K. S. Lam, R. W. Liu, S. Miyamoto, A. L. Lehman and J. M. Tuscano, Acc. Chem. Res., 2003, 36, 370–377.
3. K. S. Lam and M. Lebl, Methods: A Companion to Methods in Enzymology, 1994, 6, 372–380.
4. K. S. Lam, S. E. Salmon, E. M. Hersh, V. J. Hruby, W. M. Kazmierski and R. J. Knapp, Nature, 1991, 354, 82–84.
5. P. G. Alluri, M. M. Reddy, K. Bachhawat-Sikder, H. J. Olivos and T. Kodadek, J. Am. Chem. Soc., 2003, 125, 13995–14004.
6. M. M. Reddy, K. Bachhawat-Sikder and T. Kodadek, Chem. Biol., 2004, 11, 1127–1137.
7. J. K. Chen, W. S. Lane, A. W. Braner, A. Tanaka and S. L. Schreiber, J. Am. Chem. Soc., 1993, 115, 12591–12592.
8. R. Xu, G. Greiveldinger, L. E. Marenus, A. Cooper and J. A. Ellman, J. Am. Chem. Soc., 1999, 121, 4898–4899.
9. W. C. Still, Acc. Chem. Res., 1996, 29, 155–163.
10. M. Torneiro and W. C. Still, Tetrahedron, 1997, 53, 8739–8750.
11. Y. Cheng, T. Suenaga and W. C. Still, J. Am. Chem. Soc., 1996, 118, 1813–1814.
12. M. Davies, M. Bonnat, F. Guillier, J. D. Kilburn and M. Bradley, J. Org. Chem., 1998, 63, 8696–8703.
13. R. E. Gawley, M. Dukh, C. M. Cardona, S. H. Jannach and D. Greathouse, Org. Lett., 2005, 7, 2953–2956.
14. H. Wennemers, M. C. Nold, M. M. Conza, K. J. Kulicke and M. Neuburger, Chem.–Eur. J., 2003, 9, 442–448.
15. H. Wennemers and W. C. Still, Tetrahedron Lett., 1994, 35, 6413–6416.
16. M. Kubo, E. Nashimoto, T. Tokiyo, Y. Morisaki, M. Kodama and H. Hioki, Tetrahedron Lett., 2006, 47, 1927–1931.
17. D. Barham and P. Trinder, Analyst, 1972, 94, 142–145.
18. K. Tamaoku, Y. Murao, K. Akiura and Y. Ohkura, Anal. Chim. Acta, 1982, 136, 121–127.
19. Similar spectroscopic behavior has been reported. See: K. Tamaoku, K. Ueno, K. Akiura and Y. Ohkura, Chem. Pharm. Bull., 1982, 30, 2492–2497.
20. The amphiphilic NoveSyn™ TG amino resin (130 µm beads, Loading: 0.20–0.30 mmol g⁻¹ resin, purchased from EMD Biosciences) was used in all experiments.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental procedure for the screening of the library for binding to tripeptide 16a by the Trinder reaction. See DOI: 10.1039/b604723h

---