Study on cyclodipeptide synthases (CDPSs), a new family of tRNA-dependent peptide bond-forming enzymes – CyDiPepS

The multidisciplinary approach involved conventional methods in molecular biology and protein biochemistry. It also used structural biology (crystallography) to determine the 3D structures of CDPSs. In addition, aminoacylated tRNAs and their selected analogues were produced by a semi-synthetic method involving organic synthesis of substituted dinucleotides.

This project allowed understanding how CDPSs work. We elucidated the catalytic mechanism used by these enzymes, and identified CDPS residues involved in the interaction with each of the two substrates. The project also allowed identifying then characterizing new CDPS members in terms of numbers, origin and cyclodipeptides produced.
The basic knowledge acquired in the proposal will allow using these enzymes in biotechnological processes to produce new bioactive compounds.

The basic knowledge to be acquired in the proposal will finally allow the design and construction of engineered CDPSs to give the opportunity to synthesize new bioactive compounds.

For now, the project has led to five publications and two to three additional manuscripts will be submitted for publication in the near future.
The partners characterized the CDPSs, especially AlbC from S. noursei. They showed that CDPSs are structurally related to class-I aaRSs, by solving the crystal structure of AlbC. They also described the formation of an aminoacyl-AlbC intermediate in the course of the catalysis (Sauguet et al. (2011) Nucl. Acids Res. 39, 4475; Belin et al. (2012) Nat. Prod. Rep. 29, 961). They also demonstrated that the two aa-tRNA substrates of AlbC are accommodated in two different binding sites, and they determined some key specificity determinants of the CDPS for each of its two substrates (Santarem et al. (2014) Bioorg. Med. Chem. Lett. 4, 3231 ; Moutiez et al. (2014) Nucl. Acids Res. 42, 7247). The partners identified new CDPS members (Seguin et al. (2011) Chem. Biol. 18, 1362; to be published), and showed their large distribution in organism, and the repertoire of the produced cyclodipeptides.

Cyclodipeptide synthases (CDPSs) belong to a new family of enzymes, recently discovered by two partners of this proposal (Gondry et al., Nature Chem. Biol. 2009, 5, 414). These enzymes use two amino acid loaded transfer RNAs (aa-tRNAs) to catalyze the formation of the two peptide bonds of a cyclodipeptide, which constitutes a scaffold for many natural products endowed with useful biological activities. The CDPSs are amazing enzymes that are unrelated not only to nonribosomal peptide synthetases (NRPSs), the other enzyme family synthesizing cyclodipeptides, but also to all other functionally characterized proteins. Moreover, they constitute a novel example of the versatility of aa-tRNAs in donating activated amino acids to different processes in the cell (Von Dohren, News and Views in Nature Chem Biol, 2009, 5, 374). Indeed, aa-tRNAs are central biomolecules essentially devoted to the delivering of amino acids for ribosomal synthesis of proteins, but some enzymes also use these molecules as substrates. Among these enzymes, the aminoacyl-tRNA protein transferases and Fem aminoacyl transferases, which belong to the same structural GCN5-related N-acetyltransferase (GNAT) protein superfamily, also catalyze the formation of peptide bonds. However, they differ from CDPSs because they use only one aa-tRNA to transfer the activated amino acid to a peptide or a protein acceptor.

In this context, our project aims at understanding how CDPSs recognize their aa-tRNA substrates and elucidating how they use the double activation input resulting of the involvement of two aa-tRNAs for the formation of two peptide bonds. Our study will benefit from results that we recently obtained. Thus, partner 1 of this project solved the three dimensional structure of one member of the CDPS family. These structural data, besides the fact that they allow us to confirm the originality of CDPSs that do not have a GNAT fold, revealed a structural similarity between CDPSs and another family of enzymes, which has been the subject of many structural and engineering studies. Exploiting this similarity gives us considerable strengths to achieve our objectives. Furthermore, our study will be performed using various experimental approaches due to the complementary expertises of the four partners involved in the project. Thus, partner 1 will carry out the enzymatic characterization of CDPSs and will perform all structural biology experiments. Partners 2 and 3 will support the design and synthesis of aa-tRNA substrates as well as of any relevant aa-tRNA analogues. As they are used to working with such molecules, their contribution is particularly valuable to achieve the project. Partner 4 will be responsible for searching new members of the CDPS family and identifying the cyclodipeptides produced by these new CDPSs. These data will contribute to a better understanding of the substrate specificity of CDPSs, and provide additional information on residues involved in catalysis. In conclusion, the project will undoubtedly allow characterizing and documenting a new family of enzymes that proves quite original.