Zarah Forsberg

Multidomain lytic polysaccharide monooxygenases (LPMOs) are widely distributed in human pathogenic bacteria and increasingly recognized for promoting host interactions and virulence. Recent work has shown that immunizing mice with CbpD, a trimodular LPMO from Pseudomonas aeruginosa , provides protection against lethal P. aeruginosa infection. In several human intestinal pathogenic bacteria, LPMOs occur as tetra- and pentamodular proteins and their ability to interact with the host and facilitate virulence has primarily been ascribed to their ability to bind to N -acetylglucosamine containing glycans. Recently, AlphaFold predictions and substrate-binding studies of the Vibrio cholerae colonization factor GbpA suggested the presence of a copper-binding site in its non-catalytic third domain. In light of the recent demonstration that MUC2, the primary intestinal mucin, harbors two conserved and distinct copper binding sites for Cu(II) and Cu(I), we hypothesized that the auxiliary non-catalytic copper-binding domains of intestinal LPMOs evolved as a response to the intestinal environment. Here, we examine GbpA and homologous multidomain LPMOs from food-borne intestinal disease-causing Gram-positive bacteria from the genera Bacillus and Listeria . By combining mutagenesis with biochemical, spectroscopic, and computational approaches, we demonstrate that these multidomain LPMOs contain conserved, yet structurally distinct copper(I)-binding motifs on their non-catalytic third domain. A comprehensive review and reassessment of the literature spanning the past two decades reveals consistent links between the copper-binding ability of these LPMOs and host-pathogen interactions. Thus, our findings link these LPMOs to copper management in the intestine, offering an explanation for why their unique architecture is preserved across different bacterial lineages causing intestinal disease.

Lytic polysaccharide monooxygenases (LPMOs) catalyze oxidative cleavage of glycosidic bonds and are important members of Nature’s enzymatic machinery for the deconstruction of recalcitrant polysaccharides. While chitin-active LPMOs are typically found in microorganisms that also produce chitinases, Lactiplantibacillus plantarum, a commensal lactic acid bacterium, lacks the chitinolytic machinery yet encodes a single-domain LPMO, hereinafter called LpLPMO10A. Here, we characterize LpLPMO10A produced both homologously and heterologously and demonstrate through structural modeling and sequence comparisons that it groups with canonical chitin-active LPMOs. Biochemical assays show that LpLPMO10A catalyzes oxidative cleavage of chitin, producing C1-oxidized products. Analysis of productive (chitin peroxygenase reaction) and nonproductive (ascorbate peroxidase reaction) H₂O₂-consumption, together with additional functional assays, showed that LpLPMO10A is very similar to SmLPMO10A from Serratia marcescens, with a known role in chitin degradation. Moreover, LpLPMO10A was more efficient in degrading chitin than BlLPMO10A from Bacillus licheniformis, which harbors a chitinolytic gene repertoire including chitinases. For all three LPMOs, these activities showed only minor pH-dependence in the pH 5 – 7 range, meaning that LpLPMO10A does not seem to have a particular potential to act at acidic pH. This lack of pH dependence is shared with fungal LPMOs despite the presence of different titratable groups in the second coordination sphere of the copper. While our results reveal functional variation among chitin-active bacterial LPMOs, for example regarding their ability to withstand oxidative damage, they do not identify distinct functional features that clarify the biological role of LpLPMO10A. Taken together, our findings highlight LpLPMO10A as an evolutionarily intriguing LPMO, namely a chitin-active enzyme retained in a nonchitinolytic bacterium.

Chitosan is a polysaccharide-based biopolymer with unique properties and multiple applications in agricultural and medical industries. Chitin deacetylases (CDAs) can potentially deacetylate chitinous molecules, producing chitosans and partially acetylated chitooligosaccharides (COS). Here, we describe the expression and characterization of a CDA from the filamentous fungus Absidia coerulea, AcCDA1, active toward colloidal chitin, partially deacetylated chitin, chitosan with different degrees of acetylation, and COS. When incubated with COS with degrees of polymerization 3-6, the enzyme rapidly produced fully deacetylated COS, which is a relatively rare property among known CDAs. The mode of action of AcCDA1 toward penta-N-acetyl chitopentaose ((GlcNAc)5) was studied by mass spectrometry, using a novel approach for labeling the reducing N-acetyl-D-glucosamine unit of the COS with chitooligosaccharide oxidase (FgChitO) from Fusarium graminearum. Kinetic studies with COS revealed high enzyme efficiency with kcat/Km values reaching 18.8 mM-1s-1 for (GlcNAc)5. The enzyme removed 12 % of the acetyl groups in colloidal chitin whereas it drastically reduced the degree of acetylation of various chitosans. A comparative study with three other CDAs showed that AcCDA1 is highly efficient enzyme and revealed both similarities and differences. These findings show that various CDAs can achieve different deacetylation targets during the enzymatic processing of chitinous materials.

Efficient enzymatic depolymerization of recalcitrant polysaccharides such as chitin and cellulose relies on processive glycoside hydrolases (GHs), whose efficiency can be enhanced through cooperation with lytic polysaccharide monooxygenases (LPMOs). For processive GHs, retained binding during successive catalytic events aids depolymerization of crystalline substrates but comes at the expense of slow dissociation (low k_off) that limits turnover. Here, we engineered a series of mutants of the processive exochitobiohydrolase SmChiB from Serratia marcescens, in which a major determinant of substrate affinity and processivity, Trp220, was replaced by Tyr, Phe, His, Gln, or Ala. Functional analysis showed that these mutants had stepwise reductions in substrate affinity and processivity, with the latter being a signature of an increased k_off. Molecular dynamics simulations confirmed that Trp220 plays an important role in substrate binding. The less processive SmChiB variants, and in particular W220Y, were able to reach wild-type-like performance only at high substrate concentrations. Importantly, the use of LPMOs to decrystallize the substrate and thereby increase its effective concentration, enhanced the performance of the less processive mutants to a much greater extent than for wild-type SmChiB. In some reaction setups, the combination of W220Y with an LPMO yielded twice as much soluble product compared to the wild-type enzyme under identical conditions. Thus, when combined with LPMOs, less processive GHs become more favorable because they are intrinsically more efficient catalysts when acting on noncrystalline substrates. These findings shed light on how the interplay between GHs and LPMOs can be optimized for efficient enzymatic conversion of recalcitrant polysaccharides.

Enzymes known as lytic polysaccharide monooxygenases (LPMOs) are exceptionally powerful small redox enzymes that master the controlled generation and productive use of potentially damaging hydroxyl radicals in what is essentially a H ₂ O ₂ -driven peroxygenase reaction. We have used ancestral sequence reconstruction and enzyme resurrection to unravel evolutionary steps leading to this exceptional catalytic ability. Real-time monitoring of copper reoxidation and amino acid radical formation showed evolutionary improvement of both the capacity to avoid futile turnover of H ₂ O ₂ and the ability to scavenge damaging radicals resulting from such turnover through a hole hopping pathway. Through mutational studies of ancestral LPMOs, we show that adoption of an extant-like conformation of residues in the hole hopping pathway yields improvements in redox robustness to near-extant levels. These results show how selective pressure imposed by the need for generating a highly oxidizing intermediate is a key driver of metalloenzyme evolution, involving large parts of the enzyme, well beyond the catalytic center.

Several bacterial pathogens secrete multidomain enzymes known as lytic polysaccharide monooxygenases (LPMOs) that are important for virulence. One example is the Vibrio cholerae virulence factor GbpA (VcGbpA), in which an N-terminal LPMO domain is followed by two domains of unknown function called GbpA2 and GbpA3, and a C-terminal chitin-binding domain called CBM73. In-depth functional characterization of full-length and truncated variants of VcGbpA and a homologue from V. campbellii (previously V. harveyi, VhGbpA) showed that the catalytic LPMO domains of these proteins exhibit properties similar to natural single-domain LPMOs with established roles in chitin degradation. Interestingly, binding to chitin and efficient degradation of this substrate were affected by the presence of the GbpA2 and GbpA3 domains. Combined with structural predictions and analyses of sequence conservation, our data show that GbpA3 has evolved to interact with the reduced catalytic copper site in the LPMO domain to prevent off-pathway reactions in the absence of substrate. Substrate binding by CBM73 weakens this interaction, enabling the activation of the LPMO only when substrate is present. These observations shed new light into the functionality of these multidomain LPMOs and uncover a novel mechanism for regulating LPMO activity.

Lytic polysaccharide monooxygenases (LPMOs) are monocopper enzymes that oxidatively cleave recalcitrant polysaccharides such as cellulose. Like other redox enzymes, LPMOs face challenges in handling reactive oxygen species (ROS) generated at their active site, which must be controlled to prevent damaging off-pathway reactions. In the case of copper catalysts, oxidative damage can become self-reinforcing, as released copper from damaged catalytic centers promotes additional ROS formation via abiotic redox cycling. Here we show that a subgroup of carbohydrate-binding modules (CBM2s), exclusively tethered to cellulose-active LPMOs, has evolved the ability to bind copper. The copper site, found on the opposite side of the CBM from the cellulose-binding surface, confers redox protection through two distinct mechanisms: preferential binding of free Cu(I) and direct interaction with the reduced catalytic copper site of the LPMO. Together, these mechanisms prevent off-pathway reactions in the absence of substrate. The protective role of these copper-binding CBM2s is demonstrated by thorough characterization of the kinetics and redox stability of a series of engineered LPMO variants. Predicted models of CBM2-LPMO interactions revealed an interdomain copper-site resembling the Cu(B) site in particulate methane monooxygenase. Notably, this spatial arrangement enables substrate-dependent regulation of copper site reactivity: the CBM2-LPMO interactions that inhibit damaging redox chemistry in the absence of substrate are relieved upon cellulose binding. This work shows a road toward improved redox stability and tunable reactivity of copper catalysts.

Several bacterial pathogens secrete multi-domain enzymes known as lytic polysaccharide monooxygenases (LPMOs) that are important for virulence. One example is the Vibrio cholerae virulence factor GbpA ( Vc GbpA), in which an N-terminal LPMO domain is followed by two domains of unknown function called GbpA2 and GbpA3, and a C-terminal chitin-binding domain, called CBM73. In-depth functional characterization of full-length and truncated variants of Vc GbpA and a homologue from V. campbellii (previously V. harveyi ; Vh GbpA) showed that the catalytic LPMO domains of these proteins exhibit properties similar to natural single-domain LPMOs with established roles in chitin degradation. Interestingly, binding to chitin and efficient degradation of this substrate were affected by the presence of the GbpA2 and GbpA3 domains. Combined with structural predictions and analyses of sequence conservation, our data suggests that GbpA3 has evolved to interact with the reduced copper site to prevent off-pathway reactions in the absence of substrate. Substrate binding by CBM73 weakens this interaction, enabling the activation of the LPMO only when substrate is present. These observations shed new light into the functionality of these multi-domain LPMOs and uncover a novel mechanism for regulating LPMO activity.

Enzymes known as lytic polysaccharide monooxygenases (LPMOs) are exceptionally powerful small redox enzymes that master the controlled generation and productive use of potentially damaging hydroxyl radicals in what is essentially a H ₂ O ₂ -driven peroxygenase reaction. We have used ancestral sequence reconstruction and enzyme resurrection to unravel evolutionary steps leading to this unprecedented catalytic power. Real-time monitoring of copper re-oxidation and amino acid radical formation showed evolutionary improvement of both the capacity to avoid futile turnover of H ₂ O ₂ and the ability to scavenge damaging radicals resulting from such turnover through a hole hopping pathway. These results show how selective pressure imposed by the need for generating a highly oxidizing intermediate shapes metalloenzymes, involving large parts of the enzyme, well beyond the catalytic center.

Lytic polysaccharide monooxygenases (LPMOs) are monocopper enzymes that hydroxylate recalcitrant polysaccharides such as cellulose. Like other redox enzymes, LPMOs face challenges in handling the reactive oxygen species generated at their active site, which must be controlled to prevent off-pathway reactions that lead to enzyme inactivation. In the case of LPMOs, oxidative damage may be self-reinforcing because free copper released from damaged catalytic centers will promote abiotic redox reactions that generate reactive oxygen species. Here we show that some members of a widely spread family of carbohydrate-binding modules (CBM2s) have evolved the ability to bind copper and that this ability is exclusively found in CBM2s that are appended to LPMOs. We show that the copper site in these CBM2s protects the LPMO from inactivation both by scavenging free copper, preferably Cu(I), and by interacting directly with the reduced catalytic copper site of the LPMO, thus preventing the enzyme from engaging in off-pathway reactions. These effects are demonstrated by studies on the redox stability of a series of engineered LPMO variants as well as AlphaFold3 models of the CBM2-containing enzymes. Interestingly, the copper site and the cellulose-binding surface are located on different sides of these CBM2s, enabling a mode of action in which the CBM inhibits potentially damaging LPMO activity in the absence of substrate, while such inhibition would be relieved upon binding of the CBM2 to cellulose. These findings show that CBMs have biologically relevant functions beyond carbohydrate-binding and reveal a mechanism for substrate-dependent regulation of LPMO reactivity.