Polyphosphate: an ancient polymer with an unfinished story
Polyphosphates (polyP) are linear chains of inorganic phosphate residues joined by high-energy phosphoanhydride bonds — the same bond chemistry that powers ATP. Chains range from a few phosphate units to several hundred.
PolyP is found in every form of life, from bacteria and yeast to plants and animals, where it regulates enzyme activity, protein function, and signalling. Yet despite this universality, how polyP is made, controlled, and used in higher organisms remains largely unresolved.
Our lab works on that gap. We study how polyP modifies proteins, what those modifications do, and how the enzymes that handle phosphate can be targeted.
A polyphosphate structure
A new class of ionic post-translational modification
Protein phosphorylation — the addition of a single phosphate group — is among the best-studied post-translational modifications (PTMs) in biology. Protein polyphosphate modification (PPM) is its far less familiar relative, and our work has helped define it.
We have characterized two forms of this modification:
Histidine-polyphosphate modification (HPM). We discovered that proteins carrying histidine repeats bind polyP with high affinity. The interaction is non-covalent, yet it substantially alters protein behaviour — a combination that makes it a genuinely unusual regulatory mechanism.
Lysine-polyphosphate modification (KPM). Lysine-rich clusters undergo a comparable modification driven by ionic interaction. Attachment of polyP to lysine residues was long assumed to be covalent; our evidence indicates that it is likely not.
Together these define PPM as a novel class of ionic PTMs — modifications that regulate proteins through non-conventional association rather than covalent chemistry. This reframes what a post-translational modification can be.
Screening reveals widespread histidine-polyphosphate modification
Proteins with consecutive histidine or lysine residues undergo modification by polyP that produces a distinct electrophoretic mobility shift on NuPAGE gels — giving us a tractable readout for a screen.
Using this approach, we identified ~100 human, yeast and bacterial proteins containing histidine-rich and lysine-rich clusters that undergo HPM, establishing the modification as widespread rather than idiosyncratic.
- Schematic of the MBP-Snf1(1–65) construct.
- Arginine replacement NuPAGE test.
- Polyanion NuPAGE test. DS, dextran sulfate.
What polyP binding actually does to a protein
The functional consequences are substantial. In our work, polyP modification:
Disrupts phase separation. Pre-treatment with polyP alters the phase-separation behaviour of the DYRK1A intrinsically disordered region.
Impairs kinase activity. HPM inhibits the phosphorylation activity of the human protein kinase DYRK1A.
Inhibits transcription factor function. The transcription factor MafB is likewise suppressed.
That a non-covalent, charge-driven interaction can shut down a kinase and a transcription factor points to HPM as a regulatory mechanism in its own right.
- NuPAGE analysis of samples.
- Kinase activity of MBP-DYRK1A on glutathione-S-transferase-CTD substrate detected via western blotting against phosphorylated CTD.
- Phase separation of WT or ΔHis DYRK1A IDR-EGFP pre-treated with polyP.
From molecules to cells
Using human and yeast cell models, we are extending these findings into cellular context — asking where polyP modification occurs, what it controls, and how it shapes protein localization, interaction, and activity in a living cell.
Schematic of lysine and histidine polyphosphate (polyP) modification and its effects onprotein function.
AI protein and drug design
Alongside the polyP work, we are embracing the latest advances in artificial intelligence to transform protein engineering and drug discovery. By combining AI-guided protein design with experimental validation, we seek to create novel biomolecules and antivirulence therapeutics that specifically target pathogen virulence, opening new avenues for combating antimicrobial resistance.
Simplified inhibitor structure-based drug design pathway. The structure-based drug design process includes protein structure determination or accurate protein model development.
Protein-ligand interactions may be screened with docking simulations and leads are optimized.