Bacterial Enzyme Breakthrough Paves the Way for Greener Drugs and Food Ingredients

Chemists have long relied on brute force to turn alcohols into the more reactive compounds that serve as building blocks for pharmaceuticals, food ingredients, and flavours and fragrances. Traditional oxidation chemistry typically demands harsh oxidising agents, elevated temperatures, and solvent systems that generate toxic waste streams, an environmental burden that the fine-chemical and pharmaceutical industries have been under growing pressure to eliminate. A research team at the Instituto de Tecnologia Química e Biológica António Xavier of NOVA University Lisbon (ITQB NOVA) in Portugal now reports a milestone that could shift much of this chemistry into water, at room temperature, powered by nothing more than oxygen from the air. In a study published in Chem & Bio Engineering and selected as an ACS Editors’ Choice article, the group describes the first galactose oxidase ever characterised in bacteria, an enzyme they named PsGalOx, and shows how it can be engineered into a biocatalyst that is dramatically more efficient than its natural starting point.

Galactose oxidases are not new to science, but until now they have been the exclusive province of fungi. These copper-containing enzymes catalyse the oxidation of primary alcohols and simple sugars under remarkably mild conditions, using molecular oxygen as the terminal electron acceptor and releasing only hydrogen peroxide as a by-product. That combination of mild operating conditions, an abundant and benign oxidant, and clean chemistry makes galactose oxidases highly attractive candidates for industrial biocatalysis, where replacing metal-oxo reagents and stoichiometric oxidants with enzymatic transformations can slash energy consumption and waste generation. The problem has been that the best-characterised galactose oxidases come from fungi, and their properties do not always map neatly onto the substrates and process conditions that industry would prefer. Expanding the family to bacteria promised access to new catalytic behaviours, but bacterial versions of the enzyme had remained elusive or intractable.

The ITQB NOVA team, led by Lígia Martins, head of the Microbial & Enzyme Technology laboratory, encountered exactly that intractability. For several years, the bacterial enzyme proved exceptionally difficult to produce in a soluble form, which is a prerequisite for any serious biochemical characterisation. Proteins that aggregate into insoluble clumps inside their production host cannot be purified, crystallised, or assayed in a meaningful way, and many potentially valuable enzymes have been abandoned for precisely this reason. The turning point, according to Martins, came with the emergence of AlphaFold, the artificial intelligence system developed for protein structure prediction. The AI-generated structural models helped the researchers understand the enzyme’s unusual architecture and, crucially, identified specific changes that could improve its production in soluble form. This is a striking example of how computational structure prediction has become a practical tool for solving real experimental bottlenecks rather than merely an academic exercise in modelling.

Once soluble protein was obtained, the structural analysis revealed a feature that distinguishes the bacterial enzymes from their fungal counterparts. The best-characterised fungal galactose oxidases carry a single catalytic module, but the bacterial versions described in the study possess, in addition to the catalytic module, two near-identical sugar-binding modules. Carbohydrate-binding modules are accessory domains commonly found in enzymes that act on sugars, where they assist by tethering the catalytic domain to its substrate. Because such modules often look interchangeable in sequence databases, a common simplification in the field has been to treat them as functionally equivalent. The Portuguese team’s findings challenge that assumption directly. Tomás Frazão, one of the paper’s first authors and a PhD student at ITQB NOVA, explains that although the duplicated modules share approximately 80 percent sequence identity, they have evolved distinct functions: one enhances binding to sugars, whereas the other primarily contributes to protein stability.

This functional divergence within a pair of highly similar domains carries broader implications for enzyme engineering. If two modules that look almost identical have been shaped by evolution to perform different roles, then swapping them, deleting them, or assuming their redundancy is a risky shortcut. For biotechnologists, the lesson is that the accessory architecture of an enzyme can encode hidden determinants of both activity and manufacturability, and that careful dissection of these regions can unlock improvements that a naive engineering campaign would miss. It also raises an evolutionary question that the team intends to pursue: why would a bacterium maintain two copies of a binding module with divergent specialisations, and what does that tell us about how substrate specificity and protein robustness are balanced in natural enzyme families?

To convert their characterisation into a practical catalyst, the researchers turned to directed evolution, a laboratory method that recreates natural selection on a compressed timescale. The approach involves introducing random mutations into the gene encoding the enzyme, expressing the resulting library in host cells, and screening thousands of bacterial colonies to identify variants with improved performance. It is a powerful but demanding strategy, because the outcome of an evolution campaign is often shaped by what the screen actually measures. In this case, the evolution revealed something unexpected: the highly similar sugar-binding modules recombined into single hybrid modules, and the resulting variants produced more soluble protein. The recombination of duplicated domains into hybrid forms is a fascinating echo of natural evolutionary processes, in which gene duplication and recombination generate new functional combinations, here happening within weeks inside a laboratory flask.

The recombination result also forced the team to confront a subtlety that plagues many enzyme engineering projects. Increased activity measured during an evolution campaign does not necessarily mean that the enzyme itself has become better at speeding up reactions. Much of the apparent improvement in their screens resulted from the cells simply producing more of the enzyme, so that each assay contained more catalyst. Distinguishing genuine catalytic improvement from improved expression is essential, because the two phenotypes call for different engineering strategies and matter differently depending on the application. By recognising this distinction, the researchers were able to target the right property at the right stage of their campaign, and by the end of the process they had developed a variant that was 40 to 1800 times more efficient at processing a range of sugars and alcohols than the starting enzyme. That range reflects the diversity of substrates tested, with improvements varying according to the particular sugar or alcohol being oxidised.

The practical significance of such an enzyme extends across several industries. Oxidised sugars and alcohols are intermediates in the synthesis of active pharmaceutical ingredients, in the production of food ingredients, and in the flavours and fragrances sector, where consumers and regulators increasingly demand natural and sustainably produced compounds. A bacterial galactose oxidase that can be produced in good yield, tolerates engineering, and acts on a broad range of substrates under mild aqueous conditions offers a route to these products that avoids heavy-metal reagents, high temperatures, and toxic waste. The fact that the enzyme comes from bacteria rather than fungi is also strategically relevant, because bacterial expression systems are the workhorses of industrial biotechnology, and an enzyme that is native to a bacterial context may be easier to manufacture at scale than a fungal protein requiring eukaryotic folding machinery.

Looking ahead, Martins and her colleagues have laid out a two-pronged research agenda. The first strand is continued engineering of PsGalOx to act on substrates of practical interest, with the goal of creating a versatile intermediate for food, pharmaceutical, and materials applications. The second strand is exploratory: the team wants to map the evolutionary diversity of bacterial galactose oxidases more broadly, work that Martins says will help the field understand how substrate specificity evolved and identify robust, versatile scaffolds for developing new biocatalysts. In effect, PsGalOx is being positioned as the first member of a toolbox, a foundational catalyst whose descendants could be tailored to individual industrial transformations.

The study was developed in collaboration with several groups at ITQB NOVA, including the Bioorganic Chemistry Lab, the Industry and Medicine Applied Crystallography Lab, and the former Metalloproteins and Molecular Bioenergetics lab, together with the Barcelona-based computational company Zymvol Biomodeling in Spain, reflecting the blend of experimental protein engineering, structural biology, and computational modelling that modern biocatalyst development increasingly demands. Published under the title Engineering a Modular Bacterial Galactose Oxidase through CBM Recombination and Second-Shell Substitutions, the work appeared online on 11 October 2026. For a field that has spent decades working with a single fungal archetype, the arrival of a characterised, engineerable bacterial galactose oxidase widens the evolutionary and functional space available to green chemistry, and it demonstrates how artificial intelligence, directed evolution, and careful biochemical dissection can combine to turn an intractable protein into an industrial candidate.

Subject of Research: Engineering of the first bacterial galactose oxidase biocatalyst for sustainable oxidation of sugars and alcohols

Article Title: ITQB NOVA researchers engineer the first bacterial biocatalyst of a toolbox to produce greener pharma and food ingredients

Article References: ITQB NOVA researchers engineer the first bacterial biocatalyst of a toolbox to produce greener pharma and food ingredients. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: galactose oxidase, biocatalysis, directed evolution, enzyme engineering, AlphaFold, ITQB NOVA, green chemistry, pharmaceuticals, food ingredients, bacterial enzymes, sugar-binding modules, Chem & Bio Engineering