Â鶹ÒùÔº

An international study led by researchers from the Department of Medicine and Life Sciences (MELIS) of Pompeu Fabra University and Stanford University (California) has designed a system to identify highly selective peptides with high therapeutic potential.

The method is based on the use of biologically- and chemically-modified bacteriophages to screen, with a high degree of precision, up to 1 billion peptides simultaneously. This allows identifying those that can successfully distinguish very similar proteins that play a significant role in the development of cancer and diabetes, respectively, and that are currently being treated with non-specific drugs.

The experimental study, published in the , presents the new technique, based on phage display, to specifically recognize the interaction between two proteins: protease and substrate.

The new technique has made it possible to very precisely identify peptides that can distinguish very similar proteins—to 70% homology—such as fibroblast activation protein α (FAPα) and dipeptidyl peptidase 4 (DPP4). These proteins play an important role in cancer and type-2 diabetes, respectively.

FAPα is a protein with high therapeutic potential that is overexpressed in 90% of carcinomas and, when present at high levels, is associated with a poor prognosis. The structure of this factor is very similar to DPP4, a protein on which drugs act to regulate type 2-diabetes, so that there is a cross-reactivity between FAPα inhibitor drugs that recognize DPP4 non-specifically.

To selectively distinguish the two proteins, the research team has made two modifications to the aforementioned technique. It has added a macrocyclic peptide and a fluorescent element to the library of substrates for analysis. "The macrocyclic peptide, due to its ring shape, reduces the flexibility of the substrate, which helps us minimize non-specific binding with other proteases," explains Marta Barniol-Xicota, head of the Biological Chemistry Group at MELIS-UPF and co-leader of the study. And, the element of fluorescence allows the identification of the substrate protein of interest in real time, even if the test is done with live elements.

Thus, they have managed to reduce cross-reactivity between proteases and substrates, keeping only those with strong and specific interactions. "With this technique we achieve more precise recognition of the target proteins than can be achieved by some drugs that are being administered," Barniol-Xicota adds. An example of this is a type-2 diabetes drug approved by the United States Food and Drug Administration (FDA) that acts on DPP4 in humans, but also on the homologous protein in bacteria. This could alter the microbiome of the people who take it.

According to Barniol-Xicota, the peptides the team has designed can be used in vivo –in studies with cells and organisms– for diagnostic purposes, since "the circular structures are more rigid and, therefore, more resistant to degradation." Moreover, thanks to its high selectivity, the technique is more sensitive because "by reducing non-specific reactions, it is not necessary to have a large amount of protease to identify the protein with which it is associated."

Hence, these new selective peptides could be used to identify prognostic biomarkers for some cancers or detect new therapeutic targets for other diseases. Furthermore, thanks to the ability to emit fluorescence, this new technique could have applications in medicine in fluorescence-guided surgeries.