A modular approach to triazole-containing chemical inducers of dimerisation for yeast three-hybrid screening

Abstract

The yeast three-hybrid (Y3H) approach shows considerable promise for the unbiased identification of novel small molecule-protein interactions. In recent years, it has been successfully used to link a number of bioactive molecules to novel protein binding partners. However despite its potential importance as a protein target identification method, the Y3H technique has not yet been widely adopted, in part due to the challenges associated with the synthesis of the complex chemical inducers of dimerisation (CIDs). The development of a modular approach using potentially "off the shelf" synthetic components was achieved and allowed the synthesis of a family of four triazole-containing CIDs, MTX-Cmpd2.2-2.5. These CIDs were then compared using the Y3H approach with three of them giving a strong positive interaction with a known target of compound 2, TgCDPK1. These results showed that the modular nature of our synthetic strategy may help to overcome the challenges currently encountered with CID synthesis and should contribute to the Y3H approach reaching its full potential as an unbiased target identification strategy.

Figure 1

Schematic representation of the Y3H system used in this study showing the ternary complex between the chemical inducer of dimerization (CID) and the two fusion proteins containing the activation and DNA-binding domains of the transcription factor (AD and DBD, respectively). The CID consists of: (i) methotrexate (shown in purple), which binds to the dihydrofolate reductase DHFR-DBD fusion protein; (ii) a flexible linker unit (black line) and (iii) the bioactive molecule of interest (blue hexagon), which binds to the target protein-AD fusion. Successful formation of the ternary complex results in expression of the reporter genes (e.g., LEU2) that enables the yeast cell to grow in the absence of the amino acid.

Figure 2

Structures of (A) CID MTX-Cmpd2.1 bearing a PEG linker and (B) the erlotinib-based CID as described by Chidley et al. [6].

Scheme 1

Modular approach to MTX-Cmpd2 CIDs enabling components to be mixed and matched as required.

Figure 3

Examples of predicted 3D conformations of (A) MTX-Cmpd2.1 and (B) MTX-Cmpd2.2.

Scheme 2

Synthesis of (A) Compound2 based alkyne 3 and (B) ^tBu-MTX azide 4.

Scheme 3

Hüisgen 1,3-dipolar cycloaddition and deprotection to MTX-Cmpd2.2.

Figure 4

Comparison of the interaction of MTX-Cmpd2.1 and MTX-Cmpd2.2 with a known kinase target of compound 2. LEU2 reporter activation growth assay with yeast expressing a TgCDPK1-AD fusion protein in the presence of MTX-Cmpd2.1 and MTX-Cpmd2.2. Growth was assessed at (A) 48 hour and (B) 96 hour by measuring the OD₆₀₀ of the yeast culture. Samples treated with the DMSO vehicle were used as negative controls. Mean values ± SD are shown (n = 3).

Figure 5

The effect of changing the triazole ring position in the CID linker. (A) Growth assay comparing the ability of MTX-Cmpd2.2 and MTX-Cmpd2.3 to turn on the LEU2 reporter gene in the presence of the TgCDPK1-AD fusion protein as the target. Growth was assessed at 48 and 72 h by measuring the OD₆₀₀ of the yeast culture. Samples treated with the DMSO vehicle were used as negative controls. Mean values ± SD are shown (n ≥ 5) and were compared using paired Student’s t-test. (B) β-Galactosidase assay comparing the ability of MTX-Cmpd2.1, MTX-Cmpd2.2 and MTX-Cmpd2.3 to turn on the LacZ reporter gene in the presence of TgCDPK1-AD as the target ^a.

Figure 6

Effect of CID linker length on the Y3H interaction between compound 2-based CIDs and TgCDPK1. (A) LEU2 reporter activation growth assay with yeast expressing a TgCDPK1-AD fusion protein in the presence of the MTX-Cmpd2 CID series with different linker lengths. Samples treated with the DMSO vehicle were used as negative controls. Mean values ± SD are shown (n ≥ 5) and compared using paired Student’s t-test; (B) LEU2 reporter activation growth assay as in panel A using yeast expressing TgCDPK1-AD; (C) β-Galactosidase assay in the presence of the MTX-Cmpd2 CID series with different linker lengths. Dose response curves with each CID are presented with the lowest dose corresponding to the vehicle (DMSO) treated samples. Mean values ± SD are shown (n ≥ 3).