Researchers have used linguistic analysis to create new antimicrobial agents.
Researchers in the US have used methods borrowed from linguistics in the hunt for new antimicrobial agents. Synthetic antimicrobial peptides (AMPs), based on natural AMPs that we use to fight infection, bear the promise of sidestepping the spread of antibiotic resistance. But synthetic AMPs with the wrong peptide sequence could end up breeding bacteria that are resistant to our natural immune defences.
Peptide sequences follow grammar-like rules - ordering peptides differently, like rearranging the words in a sentence, will give different meanings depending on the grammatical structure. Gregory Stephanopoulos and colleagues at the Massachusetts Institute of Technology, Cambridge, US, have synthesised AMPs with the same grammar as natural AMPs, but with different sequences.
Using methods from linguistics to understand the structure, folding and diversity of proteins has been mooted for ten years.1,2 But this has not led to the creation of new protein sequences. The researchers have now derived a set of grammatical rules for the sequences of natural AMPs and used these to synthesise new peptides.3
The researchers set up a collection of over 700 different grammars. Each grammar is a set of rules that specify which combination of amino acid words can be lined up to form a peptide sentence. Limited to a length of 10 words, each of these grammars specifies the use of one particular word in certain positions, while allowing a choice of several different words in other positions.
Stephanopoulos’ group compiled a list of all 20-word sentences that obey at least one of the 10-word grammars in every single 10-word stretch. From this list, the researchers eliminated all peptides that had significant sequence similarity to known, natural AMPs. This left them with a collection of peptides that were different from natural AMPs, but followed the same grammatical rules. Comparing the antimicrobial effects of these peptides with those of scrambled controls containing the same words in the wrong order, the researchers found that their grammatical rules could predict antimicrobial activity.
Yechiel Shai, who recently presented a new class of ultrasmall antimicrobial lipopeptides, welcomed the addition to the arsenal. ’This is an interesting study which increases the arsenal of antimicrobial peptides available,’ Shai told Chemistry World. ’It would be interesting to find out whether such an approach can be exploited for the discovery of potent short peptides with improved properties over natural ones to be used for therapy.’
Combining the features of two types of antimicrobial in a minimalist design has generated an efficient low-budget antibiotic.
A linguistic approach could revolutionise the analysis and annotation of complex proteome data, an Italian protein expert has argued.
FEBS Lett.,390et al,Nature443, DOI: 10.1038/nature05233