Boltz-2 tested on a heat-stable anti-Salmonella dairy enzyme
Gdansk researchers compared Boltz-2, Chai-1 and AlphaFold3 models of a four-module enzyme built to kill Salmonella in milk.
Matic Broz, PhD3 min
- Authors
- Aleksandra M. Kocot, Wojciech Rusinek, Aleksandra Majkowska, Dennis Grimon, et al.
- Published in
- Journal of Dairy Science · 2026-11 · University of Gdansk
- ProteinIQ in this study
- The paper links ProteinIQ's Boltz-2 page as the source of one of three structure predictors, and Chai-1 produced the final model.
- Tools used
- Boltz 2
The research
Milk is a good growth medium for bacteria, and Salmonella is one of the contaminants dairies worry about most. Heat treatment handles most of it, but a protein that survives pasteurization and keeps killing bacteria in the finished product would add a second line of defense. Phage endolysins, the enzymes viruses use to burst their host cells, are natural candidates.
Aleksandra Kocot, Magdalena Plotka and colleagues at the University of Gdansk and Ghent University built a chimeric enzyme called MLE-19 from four parts. A cecropin A peptide sits at the front, followed by a flexible linker, a cell wall binding domain from a Pseudomonas phage endolysin, and the catalytic domain of Ph2119, a heat-stable endolysin from a Thermus phage. A His tag at the end allowed purification.
The study then asked a practical question: does the food itself change how well the enzyme works? The team tested milk, yogurt and cottage cheese against Salmonella enterica.
ProteinIQ in action
Before running the dairy experiments, the team wanted a three-dimensional model of MLE-19. Nobody has solved a structure for a four-module chimera like this, so they predicted one from the amino acid sequences of the individual fragments, using a simplified version of the protocol from their earlier work on a related enzyme, MLE-15.
They compared three predictors: the AlphaFold3 server, the Chai-1 server and Boltz-2. The Methods section gives the ProteinIQ Boltz-2 page as the address for Boltz-2 and says it was run with default parameters through a command-line interface. The paper does not describe a ProteinIQ job, so this story treats ProteinIQ as the cited source rather than a place where the authors ran the prediction.
Each model was superimposed in ChimeraX 1.10 on three crystal structures: cecropin A from Aedes aegypti (PDB 2MMM), the binding domain of the phiKZ endolysin (3BKV) and the Ph2119 catalytic module (6SU5). The team scored how well each prediction reproduced the secondary structure of residues 2 to 37, 56 to 137 and 143 to 296.
Chai-1, run with its MSA and template options, gave the best match and supplied the model shown in the paper. Boltz-2 and AlphaFold3 served as cross-checks. Comparing several structure prediction models against known fragments is a sensible way to pick a model when the full chimera has no reference structure.
Key findings
The wet-lab results carry the paper. Nanoscale differential scanning fluorimetry gave a melting temperature of 57.78 °C for the binding domain alone and 103.56 °C for Ph2119. The fused MLE-19 melted at 101.66 °C, so attaching a mesophilic binding domain barely cost any heat stability.
Heating the enzyme at 80 °C for 10 minutes, a pasteurization condition, improved its effect on Salmonella: the heated enzyme cut counts by 1.14 log units against 0.45 for the unheated one. The minimum inhibitory concentration was 100 µg/mL, and 0.5 mM EDTA lowered it fourfold to 25 µg/mL. Significant drops in bacterial load appeared at 12.5 µg/mL for free-swimming cells and 200 µg/mL for biofilms.
The food matrix mattered. Milk gave the largest reductions, up to 3 log units at 4 °C and 20 °C, while cottage cheese gave the smallest. The authors suggest the enzyme moves more freely through liquid milk and reaches more cells than it can in a solid curd.
