Abstract
Background
The discovery of novel biocatalysts for the sustainable valorization of complex biomass feedstocks remains a significant challenge. Domain-centric exploration of characterized CAZyme families offers a promising but underexplored strategy for identifying enzymes with unusual architectures and potentially expanded substrate specificities.
Results
Systematic analysis of archaeal glycoside hydrolase family 18 (GH18) chitinases using the CANDy domain annotation pipeline led to the identification of TcChi from Thermococcus chitonophagus, a multidomain enzyme combining a GH12 and a GH18 catalytic domain alongside two carbohydrate-binding modules. Given that T. chitonophagus also encodes dedicated standalone cellulases and chitinases, we hypothesized that this multidomain assembly may have evolved a broader functional range than either composing domain alone. Biochemical assays of truncated constructs confirmed this hypothesis: the GH18 domain hydrolyzed chitin, chitosan, and β-1,3-glucan, marking the first report of β-1,3-glucanase activity (EC 3.2.1.58) in a GH18 chitinase, while the GH12 domain exhibited strong cellulase activity alongside unexpected chitosanase activity (EC 3.2.1.132), extending the known functional range of this family. Both domains demonstrated high thermostability consistent with the hyperthermophilic origin of T. chitonophagus.
Conclusions
TcChi is a thermostable, multifunctional biocatalyst capable of degrading chitin, chitosan, cellulose, and β-1,3-glucan from a single protein scaffold, making it a promising candidate for consolidated biomass deconstruction and waste valorization. These findings also demonstrate that domain-centric analysis of CAZyme families is an effective strategy for uncovering hidden functional diversity in well-characterized enzyme families.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s13068-026-02793-z.
Keywords: Biocatalysis, Carbohydrates, Glycoside hydrolases, Protein domains, Chitinases, Cellulases
Introduction
The push toward a circular economy has intensified the focus on deriving valuable chemicals from renewable resources, including chitin waste [1]. Degrading the abundant β-1,4-N-acetylglucosamine (GlcNAc) polymer into smaller chitin oligosaccharides (CHOS) creates opportunities across multiple sectors, including agriculture, medicine, and food production [2]. CHOS have been shown to enhance plant resistance to fungal infections, act as anti-inflammatory agents, and serve as prebiotics [3–7]. These represent only a fraction of their potential applications, as CHOS and their deacetylated analogs chitosan oligosaccharides (COS) also exhibit anti-inflammatory, antibacterial, and antioxidant properties, along with roles in food preservation and immune protection [8].
A sustainable and specific approach to chitin hydrolysis employs biocatalysts, namely chitinases (EC 3.2.1.14), which cleave the β-1,4-linkages between GlcNAc units. Chitinases are widespread across bacteria, fungi, insects, and humans [9]. The Gram-negative soil bacterium Serratia marcescens is well known for its chitinolytic system, comprising the chitinases SmChiA, SmChiB, and SmChiC [10]. SmChiA and SmChiB function as exochitinases, degrading chitin in a processive manner from the reducing and non-reducing ends, respectively [11]. In contrast, SmChiC is an endochitinase that cleaves chitin non-processively [12]. Despite their effectiveness, a chitinolytic cocktail including these enzymes, together with a β-N-acetylhexosaminidase and a lytic polysaccharide monooxygenase, requires high enzyme loadings (~ 15 mg g⁻1 chitin) to achieve substantial conversion (~ 40% for mushroom-derived and 70–75% for crustacean-derived chitin) [13, 14]. Consequently, although enzymatic hydrolysis is more environmentally benign than chemical methods, further optimization remains necessary. In particular, the identification or development of alternative or more efficient chitinases could enhance process efficiency while reducing overall enzyme demand [15].
Identifying novel biocatalysts with improved characteristics is inherently challenging and requires thoughtful and informed strategies [16]. A common starting point involves searching for homologous enzymes, either at the sequence or structural level, which are likely to exhibit similar substrate specificities [17, 18]. However, while close homologs may preserve functional features, they do not necessarily offer improvements in properties such as catalytic efficiency, thermostability, or heterologous expression. Tools like KcatNet, TOME and SoluProt can assist in selecting candidates that are more likely to function efficiently in desired conditions or express solubly in hosts [19–21].
Beyond homology-based methods, deeper insights into enzyme mechanism and sequence diversity can support more rational and targeted discovery strategies [22]. Phylogenetic analyses, for example, can help visualize evolutionary relationships and highlight unexplored clades within enzyme families, offering opportunities to tap into underrepresented diversity [23]. In addition, methods such as coevolutionary analysis and correlated mutation mapping can uncover residue networks responsible for functional specificity, potentially revealing enzymes with novel or enhanced activities [24, 25]. These more data-driven approaches offer a pathway not just to identify improved variants of known enzymes, but to discover entirely new biocatalysts with unique or previously unknown characteristics.
In this study, we adopt a domain-centric approach (i.e., focusing on the modular domain architecture of enzymes rather than overall sequence or organismal taxonomy) to identify and characterize novel archaeal chitinases from glycoside hydrolase family 18 (GH18). Specifically, we employ the carbohydrate-active enzyme domain analysis pipeline (CANDy), a high-throughput framework that integrates domain annotations from multiple InterPro member databases such as Pfam, SMART and NCBIFam with phylogenetic analysis and a protein domain co-occurrence network to systematically map domain modularity across CAZy families [26]. GH18 is a particularly suitable target for such an analysis, as its members are known to display a remarkably diverse domain architecture, frequently occurring in combination with carbohydrate-binding modules, ancillary domains, and, in some cases, additional catalytic domains from other GH families [27]. This architectural diversity suggests that domain modularity may be an important driver of functional diversification within GH18, and that systematic mapping of these architectures may reveal enzymes with broad or unconventional substrate specificities that would be overlooked by sequence-based approaches alone [28]. By applying this framework to archaeal GH18 chitinases, we aim to uncover enzymes with unconventional architectures that may confer enhanced performance under industrially relevant conditions, such as elevated temperatures and complex biomass substrates, thereby expanding the functional diversity of known GH18 chitinases and contributing to the more efficient and sustainable valorization of complex biomass waste streams.
Results and discussion
In silico exploration of archaeal GH18 chitinases
Protein domains in archaeal GH18 domains
CANDy was employed to identify novel biocatalysts among archaeal chitinases, a small but relatively well-characterized subset of family GH18. Of the approximately 60,000 sequences in this family, only 111 unique archaeal sequences are present. After clustering sequences at 95% identity to remove redundancy, 83 representative sequences remained, including 7 that have been experimentally characterized.
Ten distinct domain types were identified among archaeal GH18 chitinases (Figure S1). The most common domain types included the GH18 catalytic domain (hereafter referred to as the catalytic domain), carbohydrate-binding modules (CBMs), and polycystic kidney disease (PKD) domains (Fig. 1, Table S2). In general, CBMs are responsible for carbohydrate recognition and binding, and in this enzyme family they are specific to chitin. CANDy annotations revealed distantly related CBMs from families 5 and 12, both of which are known to contain chitin-binding domains [29, 30]. Additionally, nine annotations were associated with CBM family 2 (CBM2), which includes both chitin-binding and cellulose-binding domains [31]. Nevertheless, as not all CBMs identified in this study could be assigned to a specific family by CANDy, they were all grouped under a general ‘CBM’ category.
Fig. 1.
Phylogenetic tree of 83 archaeal GH18 chitinases annotated with predicted domain architectures. The tree depicts the evolutionary relationships among archaeal GH18 sequences, with domain architecture annotations derived from CANDy. Characterized enzymes are highlighted in red. The chitinase from Thermococcus kodakarensis KOD1, which contains two catalytic domains with different substrate specificities, is marked with a black dot. The chitinase from Thermococcus chitonophagus (TcChi), further characterized in this study, is indicated by an arrow. Other T. chitonophagus enzymes are marked with a ‘ + ’. GH glycoside hydrolase, CBM carbohydrate binding module, PKD polycystic kidney disease, CARD caspase recruitment domains, TAT twin-arginine translocation
The third most frequently occurring domain was the PKD domain, present in approximately 65% of the analyzed sequences. This domain, originally discovered in polycystin-1, a cell-surface glycoprotein implicated in polycystic kidney disease, adopts an immunoglobulin-like fold [32]. Although it remains unclear whether the PKD domain itself mediates adhesive interactions, polycystin-1 is involved in protein–protein and protein–carbohydrate interactions [33]. In Archaea, PKD domains are found in surface-layer proteins, which contribute to cellular protection in extreme environments [34]. As these domains are widely associated with the catalytic domains, this suggests that they may exhibit a similar stabilizing role in the enzyme complex [35].
Several other immunoglobulin-like domains were detected and classified under the broader immunoglobulin-like superfamily due to their structural homology. In addition to these more common domains, several less frequently occurring domains were identified including dockerin domains, caspase recruitment domains (CARD), and one concanavalin A-like lectin/glucanase domain (Table S1).
Domain architectures in archaeal GH18 chitinases
The presence of diverse domain types in archaeal GH18 chitinases contributes to substantial structural heterogeneity, resulting in 25 distinct domain architectures predicted within this subset (Fig. 1, Table S2). Among these, the most prevalent configuration consists of a CBM, PKD domain and catalytic domain, collectively representing 38% of all sequences (Figure S1). This is followed by architectures composed solely of a single catalytic domain (14%) and a more complex CBM + CBM + PKD + catalytic domain arrangement, which accounts for approximately 13% of the observed architectures.
Among the seven biochemically characterized enzymes identified in this archaeal subset, five exhibit the most common CBM + PKD + catalytic domain architecture (Fig. 1, Table S2) [36, 37]. The remaining two characterized proteins possess more unique configurations. One was predicted to contain an immunoglobulin-like domain followed by CBM + catalytic domain + CBM + CBM + catalytic domain, while the other exhibited a tandem catalytic domain arrangement flanking a CBM (i.e., catalytic domain + CBM + catalytic domain) (Table S2) [38, 39].
All characterized enzymes in this dataset have been classified as strict chitin hydrolases (EC 3.2.1.14), suggesting conserved catalytic function despite domain variability. A particularly noteworthy example is the chitinase from Thermococcus kodakaraensis KOD1 (TkChi), which features two GH18 catalytic domains with distinct substrate specificities (Fig. 1). The N-terminal domain functions as an exochitinase, releasing diacetylchitobiose units from the non-reducing end (EC 3.2.1.200), while the C-terminal domain acts as an endochitinase (EC 3.2.1.14) [39]. This dual-function arrangement implies a synergistic interplay between domains, potentially enhancing the enzyme’s overall efficiency in degrading chitin polymers.
Characterization of a multidomain chitinase from Thermococcus chitonophagus
Selection of TcChi
Located near the TkChi in the phylogenetic tree is an enzyme with a particularly interesting domain architecture (Fig. 1). This enzyme, originating from Thermococcus chitonophagus (GenBank ID: ASJ17009.1) and hereafter referred to as TcChi, is predicted to contain the following domain arrangement: a concanavalin A-like lectin/glucanase domain, followed by two CBMs, and a C-terminal catalytic domain (Figs. 2, S2). This catalytic domain is annotated as a member of family GH18, suggesting that TcChi likely functions as a chitinase.
Fig. 2.
Schematic representation of Thermococcus chitonophagus chitinase (TcChi) and the truncated constructs analyzed in this study. The full-length enzyme contains a GH12 catalytic domain (pink), two carbohydrate-binding modules (CBMs; yellow), and a GH18 catalytic domain (blue). Domain boundaries were defined based on annotations obtained using the CANDy tool
The N-terminal domain, however, is predicted to belong to the concanavalin A-like lectin/glucanase superfamily. This structurally diverse superfamily encompasses enzymes from multiple GH families, including GH7, GH16, and GH32, and is known to feature a conserved β-sandwich fold. To further investigate the functional identity of this N-terminal domain, a BLAST search was performed using the domain boundaries defined by CANDy (Fig. 2). The query sequence returned its highest similarity (49.4%) to a glycoside hydrolase from Thermococcus sp. AM4 (GenBank ID: WP_052306407.1), which belongs to GH12. Analysis of this family confirms the presence of TcChi (Figure S3), suggesting that the N-terminal domain may exhibit catalytic activity consistent with enzymes of family GH12, which includes cellulases (EC 3.2.1.4), licheninases (EC 3.2.1.73), or xyloglucan endo-β-1,4-glucosidases (EC 3.2.1.151).
Further analysis using InterPro confirmed that both CBMs in TcChi belong to CBM2, a family typically associated with binding to cellulose, chitin, or xylan. Together with the presence of GH12 and GH18 catalytic domains, this modular architecture suggests that TcChi may act on multiple polysaccharide substrates, particularly cellulose and chitin, and could therefore contribute to the degradation of complex biomass composed of mixed carbohydrate components. To evaluate whether such multifunctionality is unique or part of a broader enzymatic strategy, the genomic context of T. chitonophagus was examined.
This analysis revealed that the genome encodes a repertoire of enzymes with distinct, substrate-specific domain architectures. Notably, two additional GH18 enzymes are present: one features a CBM2 + catalytic domain configuration, which represents the first chitinase ever identified in Archaea and served as inspiration to name T. chitonophagus accordingly (GenBank ID: CUX784951) [40, 41]. The second chitinase exhibits a more complex architecture composed of CBM2 + immunoglobulin-like domain + catalytic domain + CBM2 + CBM2 + catalytic domain (GenBank ID: BAU88507.1). Here, both catalytic domains are predicted to belong to family GH18, indicative of chitinase activity. Moreover, a GH12 enzyme with two N-terminal CBM2s is also encoded in the genome, suggesting its involvement in cellulose degradation (GenBank ID: CUX77597.1). These enzymes, alongside others from families GH1, GH5, GH35, GH38, and GH57, reflect a broad enzymatic toolkit and suggest that T. chitonophagus possesses a diverse arsenal of specialized glycoside hydrolases.
Given that the genome of T. chitonophagus encodes dedicated enzymes for both chitin and cellulose degradation, TcChi is unlikely to serve merely as a redundant hydrolase. Instead, its unique multidomain architecture suggests a broader role in processing complex or heterogeneous polysaccharide substrates. TcChi may therefore function as a multifunctional biocatalyst that enhances the organism’s capacity to exploit structurally diverse biomass in dynamic environmental conditions.
Exploring substrate specificities of TcChi
To evaluate whether TcChi indeed shows this promiscuity, an enzymatic screening was performed on nine structurally distinct polysaccharides, differing both in their constituent monosaccharides and glycosidic linkages (Table 1). Only the catalytic domains of TcChi were tested in this assay: TcChiΔ234 comprising the GH12 catalytic domain, and TcChiΔ123 containing the GH18 catalytic domain. The full-length TcChi, as well as constructs TcChiΔ1 and TcChiΔ4, exhibited minimal or no soluble expression in E. coli (Figure S4) and were therefore excluded from further analysis.
Table 1.
Overview of the polysaccharide substrates used to assess the substrate specificity of TcChiΔ234 and TcChiΔ123, along with their corresponding chemical structures
| Substrate | Chemical structure | TcChiΔ234 | TcChiΔ123 |
|---|---|---|---|
| Fungal chitin (⍺) | [β-D-GlcNAc-(1 → 4)-]n | − | + + |
| Squid chitin (β) | [β-D-GlcNAc-(1 → 4)-]n | − | + + + |
| Chitosan | [β-D-GlcN-(1 → 4)-]n | + | + + |
| Avicel | [β-D-Glc-(1 → 4)-]n | + + | − |
| CMC | [β-D-Glc-(1 → 4)-]n | + + + | − |
| Xyloglucan | ⍺-Xyl-(1 → 6)-[β-D-Glc-(1 → 4)-]n | − | − |
| Lichenan | [β-D-Glc-(1 → 4)-]n-[β-D-Glc-(1 → 3)-]m | + + | + |
| β-1,3-Glucan | [β-D-Glc-(1 → 3)-]n | − | + + |
| Galactomannan | ⍺-Gal-(1 → 6)-[β-D-Glc-(1 → 4)-]n | − | − |
Hydrolysis reactions were carried out for 24 h, and the released monosaccharides were analyzed using HPAEC-PAD. Hydrolytic activity was scored based on monosaccharide release: “ + ” indicates the lowest detected amount (0.39 nC*min for TcChiΔ234; 1.17 nC*min for TcChiΔ123), “ + + + ” indicates the highest measured amount (6.15 nC*min for TcChiΔ234; 6.45 nC*min for TcChiΔ123), and “−” denotes no detectable hydrolysis. Note that this table is intended as a qualitative substrate screening overview; quantitative apparent turnover numbers for the most active substrate–enzyme combinations are reported in Table 2. Tc Thermococcus chitonophagus, HPAEC-PAD high-performance anion-exchange chromatography–pulsed amperometric detection, ⍺ ⍺-chitin featuring antiparallel chitin chains, β β-chitin featuring antiparallel chitin chains, CMC carboxymethyl cellulose
As expected, TcChiΔ123 exhibited clear chitinase activity, releasing N-acetylglucosamine (GlcNAc) after 24 h of incubation with both fungal and squid-derived chitin. Notably, the hydrolytic yield was higher for squid-derived chitin, suggesting a possible preference for β-chitin. In addition, measurable hydrolytic activity was observed on chitosan, the partially deacetylated derivative of chitin, further confirming the enzyme’s chitinolytic capability. More unexpected, however, was the observed activity of TcChiΔ123 on lichenan and β-1,3-glucan, glucose homopolysaccharides composed of mixed β-1,3 and β-1,4, or exclusively β-1,3 glycosidic linkages, respectively. No activity was detected on cellulose-based substrates such as Avicel and carboxymethyl cellulose (CMC), both of which consist exclusively of β-1,4 linkages. This suggest that TcChiΔ123 may exhibit substrate promiscuity toward β-1,3 glucosidic linkages, while showing little to no activity on β-1,4 linked regioisomers.
To further investigate the mode of action of TcChiΔ123 on β-1,3-glucan and chitin, hydrolysis experiments were performed on defined laminarioligosaccharides and chitooligosaccharides (CHOS), with results presented in Figures S6 and S7, respectively. Hydrolysis of laminarioligosaccharides revealed high activity on laminaribiose and laminaritriose, and moderate activity on laminaritetraose, laminaripentaose, and laminarihexaose, with glucose as the primary detected product across all substrates. The efficient hydrolysis of laminaribiose indicates an exolytic mode of action. For CHOS hydrolysis, no activity was detected on chitobiose, confirming that TcChiΔ123 does not function as a chitobiase. High activity was observed on all other CHOS tested up to DP6. Hydrolysis of chitotetraose yielded an approximately equimolar ratio of GlcNAc and chitobiose, suggesting productive binding spanning the − 1 and + 3 subsites. Chitopentaose and chitohexaose similarly yielded GlcNAc and chitobiose as the primary products. Together, these results indicate that TcChiΔ123 acts as an endochitinase that preferentially cleaves internal glycosidic bonds, releasing chitobiose as the predominant product alongside GlcNAc.
It is at the least remarkable that a single enzyme is capable of binding and hydrolyzing two distinct polysaccharides with different regiochemistry as well as monosaccharide composition. To exclude the possibility of experimental artifacts, several independent controls were performed. Both TcChiΔ12 and TcChiΔ123 were expressed and purified on separate occasions, consistently yielding active preparations with reproducible activity profiles on β-1,3-glucan, as illustrated by the time-course chromatogram presented in Figure S5. To exclude contaminating β-1,3-glucanase activity originating from the E. coli expression host, the GH12-containing constructs TcChiΔ234 and TcChiΔ34, produced under identical conditions, were tested on β-1,3-glucan and showed no detectable activity after 24 h of incubation. Additionally, SmChiA, SmChiB, and SmChiC from Serratia marcescens were assayed on β-1,3-glucan and likewise showed no activity, confirming that β-1,3-glucanase activity is unique to the GH18 domain of TcChi. Activity on lichenan, a mixed-linkage β-1,3/β-1,4-glucan discussed in the previous paragraph, further confirm the genuine and reproducible β-1,3-glucan hydrolysis capacity of TcChiΔ123 and TcChiΔ12.
The assignment of TcChi's catalytic domain to family GH18 is justified by its conserved catalytic motif (DXDXE; DFDIE), although several structural traits commonly found in other GH18 members are missing. Indeed, the GH18 catalytic domain of TcChi resembles that of endochitinase SmChiC (Figure S8), as it consists of a canonical TIM-barrel fold and lacks the typical chitinase insertion domain as well as the extended roof-forming loop [12]. In addition, TcChiΔ123 also lacks a β-hairpin observed in SmChiC. The structural basis for the observed broad substrate specificity was further investigated through molecular docking of pentaacetyl-chitopentaose, laminaripentaose, and cellopentaose into the active site of TcChiΔ123 (Figure S9). Both pentaacetyl-chitopentaose and laminaripentaose adopted similar binding poses, with Tyr10 providing aromatic stacking against the − 1 sugar ring and the scissile bond positioned at 4.2 Å and 5.0 Å from Glu114, respectively. In contrast, the scissile bond of cellopentaose, which serves as a negative control, was located 5.7 Å from Glu114, with the − 1 ring shifted approximately one sugar unit along the binding cleft, disrupting the Tyr10 stacking interaction. These results suggest that β-1,3-glucan can be accommodated in the active site. Full elucidation of the structural basis for this activity remains an important avenue for future work.
TcChiΔ234, containing a catalytic domain from the GH12 family, typically associated with cellulase, licheninase, and xyloglucanase activities, showed high activity on CMC, indicating strong cellulase activity. Moderate activity was observed on lichenan, while no activity was detected on xyloglucan. This profile suggests a primary specificity toward unsubstituted or minimally branched β-1,4-linked glucans, as the lack of activity on xyloglucan likely reflects steric hindrance imposed by its heavily substituted backbone. Interestingly, TcChiΔ234 also showed measurable activity on chitosan. This observation is consistent with reports of chitosanase activity as a side activity of retaining cellulases from GH families 5 and 7, and can be rationalized by the structural similarity between chitosan and cellulose, both being β-1,4-linked glucan-based polymers differing only at the C2 position [42]. Since the retaining double-displacement mechanism employed by GH12 does not rely on participation of the C2 substituent, the active site can readily accommodate chitosan without a fundamental mechanistic shift. Together, these results demonstrate that TcChi, through its dual promiscuous catalytic domains, is capable of acting on a broad range of structurally diverse polysaccharides, supporting its role as a multifunctional biocatalyst.
pH and temperature optimum of TcChi234 and TcChi123
To enable the further characterization of TcChi, the optimal pH and temperature conditions were determined for each catalytic domain. TcChiΔ123 was assayed using 4-methylumbelliferyl β-D-N,N′,N′′-triacetylchitotrioside ([GlcNAc]₃-MU) as a substrate, while TcChiΔ234 was tested with 4-nitrophenyl β-D-cellobioside ([Glc]₂-NP).
TcChiΔ123 exhibited a temperature optimum of 85 °C and maintained high activity across a broad pH range, with peak activity at pH 6.0 (Fig. 3). These characteristics are consistent with previously described GH18 chitinases from thermophilic Archaea. For example, the chitinase from Pyrococcus furiosus displays optimal activity at 90 °C within a pH range of 6.0–7.5 [38]. Similarly, the bifunctional chitinase from Thermococcus kodakaraensis, containing two GH18 domains, exhibits temperature optima of 85 °C and 90 °C, with corresponding pH optima of 4.5 and a broader range between 4.5 and 8.0 [39].
Fig. 3.
pH and temperature optimum for TcChiΔ234 (pink) and TcChiΔ123 (blue). Optima were determined using 4-nitrophenyl β-D-cellobioside for TcChiΔ234 and 4-methylumbelliferyl β-D-N,N′,N′′-triacetylchitotrioside for TcChiΔ123. Error bars represent the standard deviation on the mean of three biological replicates. Tc Thermococcus chitonophagus
The GH12 domain TcChiΔ234 demonstrated a similar pH optimum, with peak activity at pH 5.5 (Fig. 3). At the upper end of the temperature testing range (95 °C), the enzyme had not yet reached its temperature optimum, suggesting exceptional thermostability (Fig. 3). This observation is in agreement with a previously characterized P. furiosus endoglucanase from the GH12 family, which showed a temperature optimum of 100 °C and a pH optimum of 6.0 [43].
The remarkable thermostability of both TcChi catalytic domains likely reflects the extremophilic nature of Thermococcus chitonophagus (also known as Pyrococcus chitonophagus), a hyperthermophilic archaeon isolated from a deep-sea hydrothermal vent [41]. Although the full-length TcChi protein could not be solubly expressed in E. coli, preventing direct assessment of its complete domain assembly, the thermostability of the individual GH12 and GH18 domains strongly suggests that the intact enzyme is likewise highly thermophilic.
Apparent turnover of TcChi toward various substrates
To evaluate the catalytic efficiency of both catalytic domains, TcChiΔ234 and TcChiΔ123, as well as their neighboring CBM-containing constructs (TcChiΔ34 and TcChiΔ12), turnover numbers were measured at optimal conditions on the four substrates that exhibited the highest activity during initial screening: β-1,3-glucan, squid chitin, chitosan, and CMC. Both TcChiΔ123 and TcChiΔ12 exhibited comparable apparent turnover numbers of 26–27 min⁻1 on squid chitin at 85 °C, which is markedly higher than of well-characterized chitinases from Serratia marcescens measured at 37 °C (Table 2). In particular SmChiA, SmChiB, and SmChiC have considerably lower turnover numbers of 3.6 ± 0.2 min⁻1, 2.2 ± 0.3 min⁻1, and 5.1 ± 0.2 min⁻1, respectively, corresponding to a 5- to 12-fold lower catalytic efficiency compared to TcChi, although this difference should be interpreted in the context of the lower assay temperatures used for S. marcescens. While diacetylchitobiose remains the predominant hydrolysis product, TcChiΔ123 and TcChiΔ12 also generate noticeably higher levels of GlcNAc and triacetylchitotriose (Figure S10), illustrating their different cleavage preferences. Moreover, none of the tested S. marcescens chitinases exhibit activity on β-glucan, highlighting a unique β-1,3-glucanase activity (EC 3.2.1.58) in TcChiΔ123 that has not previously been reported for any GH18 chitinase.
Table 2.
Apparent turnover of TcChiΔ234, TcChiΔ123, and their CBM-containing counterparts (TcChiΔ34 and TcChiΔ12) on four polysaccharide substrates
| Substrate | Apparent turnover (min−1) | |||
|---|---|---|---|---|
| TcChiΔ234 | TcChiΔ34 | TcChiΔ12 | TcChiΔ123 | |
| β-1,3-Glucan | n.d | n.d | 9.8 ± 0.5 | 2.1 ± 0.1 |
| Squid chitin | n.d | n.d | 25.8 ± 0.4 | 27 ± 1 |
| Chitosan | 0.9 ± 0.2 | 0.07 ± 0.02 | 0.53 ± 0.01 | 0.56 ± 0.07 |
| CMC | 10 ± 1 | 2.1 ± 0.7 | n.d | n.d |
Reactions were performed under optimal temperature and pH conditions. Apparent turnover numbers were calculated based on the initial rate of monosaccharide release, quantified via HPAEC-PAD, and normalized to enzyme concentration (V0/[E]). Values represent the mean of three independent replicates. Tc Thermococcus chitonophagus, CBM carbohydrate-binding module, HPAEC-PAD high-performance anion-exchange chromatography-pulsed amperometric detection, CMC carboxymethyl cellulose
n.d. Activity below detection limit for reactions at 15 mg/mL substrate and 15 mg enzyme/g substrate at optimal conditions, monitored for 180 min
TcChiΔ234 and TcChiΔ34 were further confirmed to function as cellulases based on their high activity toward CMC. In addition, both constructs exhibited measurable chitosanase activity (EC 3.2.1.132). While chitosan degradation by non-specific cellulases has previously been reported for enzymes from GH families GH5, GH7, and GH8, this represents the first observation of a bifunctional GH12 cellulase demonstrating activity toward chitosan [42]. This finding expands the known substrate range of GH12 enzymes and highlights the unique catalytic versatility of TcChi.
Interestingly, the addition of the adjacent CBM to the GH18 chitinase domain in TcChiΔ12 did not enhance activity on chitin or chitosan, but it resulted in a fivefold increase in activity on β-1,3-glucan. This suggests that the CBM contributes to β-1,3-glucan binding, a property not yet attributed to members of the CBM2 family. These findings indicate a potentially novel binding behavior, warranting further investigation into the substrate specificity of this CBM. In contrast, the CBM adjacent to the GH12 catalytic domain in TcChiΔ34 led to decreased activity on both chitosan and CMC compared to the standalone catalytic construct TcChiΔ234. This reduction may result from steric hindrance imposed by the CBM, or from a binding specificity that does not complement these substrates. Moreover, as TcChiΔ34 is an artificial construct lacking two additional domains present in full-length TcChi, it is possible that the truncated protein suffers from improper folding or aggregation, potentially leading to a misinterpretation of CBM function. To clarify the binding preferences of both CBMs and their contributions to substrate recognition, affinity gel electrophoresis on a panel of polysaccharides should be performed.
Conclusion
In this study, we identified and characterized a previously unreported multidomain enzyme (TcChi) from Thermococcus chitonophagus using the CANDy tool to explore domain architectures within already well-characterized archaeal GH18 chitinases. TcChi was predicted to comprise a GH12 catalytic domain, two CBMs, and a GH18 catalytic domain—representing a rare fusion of cellulase and chitinase activities within a single enzyme. While similar architectures have been computationally predicted, this is the first biochemical characterization of an enzyme combining both cellulolytic and chitinolytic catalytic domains [44].
Given the extensive glycoside hydrolase repertoire in T. chitonophagus, the specific biological role of such a multifunctional enzyme was unclear. We hypothesized that TcChi functions as a multifunctional biocatalyst capable of acting on diverse polysaccharide substrates. To evaluate this, the individual catalytic domains TcChiΔ123 (GH18) and TcChiΔ234 (GH12) were screened against nine structurally distinct polysaccharides. The results demonstrated that both catalytic domains exhibit notable substrate promiscuity. The GH18 domain not only hydrolyzed chitin but also acted on chitosan and β-1,3-glucan. The latter is particularly surprising, as it represents a glucose-based polymer with different regiochemistry from chitin. This forms the first report of β-1,3-glucanase activity by a GH18 chitinase. The structural basis for this activity remains to be fully elucidated, and future work should aim to clarify this through structural studies and site-directed mutagenesis of key active site residues. Likewise, the GH12 domain showed high activity on CMC, confirming cellulase function, but also displayed activity on chitosan, an observation not previously described for GH12 enzymes.
While one CBM, located upstream of the GH18 domain, appears to contribute to β-1,3-glucan binding, the specific roles and substrate preferences of both CBMs remain to be elucidated. Nonetheless, our findings suggest that TcChi enables T. chitonophagus to degrade a wide range of polysaccharides containing β-1,3-linked glucose and β-1,4-linked glucose, glucosamine, and N-acetylglucosamine units. These include β-1,3-glucan, cellulose, chitin, chitosan and lichenan, supporting the hypothesis that TcChi serves as a versatile biocatalyst in the degradation of complex environmental biomass. It should be noted, however, that all biochemical characterization was performed on truncated constructs, as the full-length TcChi could not be solubly expressed in E. coli. These constructs may therefore not fully recapitulate the native function of the intact enzyme, as potential interdomain cooperativity or CBM contributions may be altered in the absence of the complete domain assembly. Expression in alternative hosts or under different conditions should therefore be explored to enable full characterization of the intact enzyme.
Given its broad substrate specificity and thermostable catalytic domains, TcChi warrants further investigation as a candidate for biotechnological applications, particularly in the valorization or pre-treatment of polysaccharide-rich biomass and industrial waste streams. Future studies should aim to establish substrate conversion yields and detailed time-course product profiles to fully evaluate its practical applicability, as well as determine the enzyme’s thermal stability (e.g., half-life at elevated temperatures) and evaluate its performance under high-salt conditions as previous reports on archaeal chitinases have demonstrated enhanced activity in the presence of salt [39]. These investigations will provide further insights into the enzyme’s robustness and operational range, and contribute to its potential implementation in industrial biocatalytic processes.
Materials and methods
All chemicals were purchased from Merck, unless noted otherwise.
CANDy analysis of archaeal GH18 chitinases
Domain architectures of archaeal GH18 chitinases were analyzed using the latest Google Colab implementation of CANDy. The tool was run with default parameters, except for the following modifications: the clustering cut-off was set to 95%, and MMseqs2 was selected as the clustering algorithm. During the iterative domain-cleaning step, the PFAM and SUPERFAMILY databases were set to priority position 3 and 4, respectively. Additionally, the maximum allowed overlap percentage between two domain annotations was reduced from 31 to 20%.
From an initial dataset of 166 sequences retrieved from CAZy, three entries lacking UniParc references and two partial sequences were excluded from further analysis. After removing redundant identifiers and sequences, and clustering at 95% sequence identity, a total of 83 unique sequences—including seven characterized enzymes—were retained for downstream analysis. Notably, the characterized enzymes AAG19274.1 and CAP13543.1 are listed separately in CAZy but correspond to the same protein sequence. To avoid redundancy, only CAP13543.1 was included in the final dataset. The resulting phylogenetic tree was visualized in iTOL and annotated with the annotation files provided by CANDy [45]. The domain co-occurrence network was visualized in Cytoscape using the yFiles Organic Layout [46].
Plasmids and culture conditions
Lysogeny broth (LB) was used to grow the strains for routine cloning and enzyme expression (10 g/L tryptone, 5 g/L NaCl, and 5 g/L yeast extract). Liquid cultures were grown at 30 °C and shaken at 200 rpm [LS-X (5 cm orbit), Adolf Kühner AG, Switzerland].
The coding sequences of TcChi (UniProt: A0A2Z2NGA0) and TcChi truncations are listed in Table S3, together with the coding sequences of Serratia marcescens chitinases SmChiA, SmChiB and SmChiC. All sequences were codon-optimized for expression in E. coli, synthesized and subcloned into the NheI and XHoI restriction sites of a pET21a vector by Life Technologies (Belgium). The genes contained a C-terminal His6-tag. The plasmids were used for transformation to electrocompetent E. coli BL21(DE3) (Invitrogen, USA) cells.
Protein expression and purification
E. coli BL21(DE3) cells carrying the desired pET21a expression plasmid of TcChi and its variants were used to inoculate 250 mL of LB medium supplemented with 50 μg/mL ampicillin. The culture was incubated at 37 °C with shaking at 200 rpm until the optical density at 600 nm (OD600) reached 0.6. Protein expression was induced by adding isopropyl β-D-1-thiogalactopyranoside (IPTG) to a final concentration of 0.5 mM, and the culture was incubated at 18 °C for an additional 16 h. Following incubation, the cells were harvested by centrifugation, and the resulting cell pellets were frozen overnight at − 20 °C.
To extract and purify the enzymes, the cell pellets were thawed and resuspended in 8 mL of lysis buffer (10 mM imidazole, 0.1 mM phenylmethylsulfonyl fluoride (PMSF), 1 mg/mL lysozyme from chicken egg white, and 50 mM phosphate-buffered saline (PBS); pH 7.4). The suspension was incubated on ice for 30 min. Following incubation, the lysate was sonicated three times for 3 min each (using a Branson Sonifier 450, level 3, 50% duty cycle). The cell extract was then separated from the cell debris by centrifugation at 20,000 g for 1 h at 4 °C. The supernatant was purified using nickel-nitrilotriacetic acid (Ni–NTA) chromatography according to the manufacturer's instructions (HisPur Ni–NTA; Thermo Fisher Scientific, Belgium). The purified enzyme fraction was concentrated, and the buffer was exchanged to 50 mM 2-morpholinoethanesulphonic acid (MES) buffer at pH 6.5 using an Amicon centrifugal filter unit with a 10 or 30 kDa cut-off. Protein concentrations were measured using a NanoDrop ND-1000 spectrophotometer at 280 nm (Thermo Fisher Scientific, Belgium). The molecular weight and molar extinction coefficient of each construct were calculated from the amino acid sequence using Benchling (Benchling Inc., San Francisco, CA, USA). Purified enzyme preparations were stored at − 20 °C.
Purification fractions were analyzed on SDS-PAGE. The purified enzymes were first diluted to 1 mg/mL using 50 mM MES buffer pH 6.5. Pellets were resuspended in 8 mL 50 mM MES buffer pH 6.5. 10 μL of each sample was combined with 20 μL sample buffer and incubated at 100 °C for 10 min. 10 µL of each was loaded on a 12% SDS-PAGE mini-PROTEAN TGX Stain-Free gel (Bio-Rad, Belgium). Gels were stained using Colloidal Coomassie staining solution (Bio-Rad, Belgium).
Substrate screening
15 mg/mL of each polysaccharide substrate: chitin from shrimp shells (Sigma-Aldrich), fungal chitin (prepared as described by Windels et al. [14]), chitosan (Sigma-Aldrich), Avicel (Sigma-Aldrich), carboxymethyl cellulose (CMC; Sigma-Aldrich), xyloglucan (Biosynth), lichenan (Megazyme), galactomannan (Megazyme), or β-1,3-glucan (prepared according to De Doncker et al. [47]) was incubated with 15 mg of either TcChiΔ123 or TcChiΔ234 per gram of substrate [14, 47]. Reactions were performed in 50 mM MES buffer (pH 6.5) at 80 °C with shaking at 800 rpm. After 24 h, samples were collected and inactivated by mixing equal volumes of reaction mixture and acetonitrile. Prior to HPAEC-PAD analysis, samples were diluted 100-fold.
HPAEC-PAD analysis
Hydrolysis products were analyzed using a Dionex ICS-6000 HPAEC-PAD system (Thermo Fisher Scientific) equipped with a CarboPac™ PA20 analytical column (150 mm) and a 30 mm guard column. For reactions involving chitin substrates, elution was performed with 30 mM NaOH at a flow rate of 0.5 mL/min at 30 °C over a 10-min run. For chitosan assays, the method was extended to 30 min to ensure complete separation and detection of hydrolysis products. For all other polysaccharide substrates, the method as outlined in Table 3 was applied at a constant flow of 0.5 mL/min. Monosaccharide products were quantified using standards from Megazyme, and chromatographic peaks were manually integrated. The method provided a LOD of approximately 1 µM and a LOQ in the range of ~ 1–100 µM. For the time-course analysis of β-1,3-glucan hydrolysis by TcChiΔ123, reactions were performed under optimal conditions (85 °C, pH 6.0, 50 mM MES buffer) using 15 mg/mL β-1,3-glucan and an enzyme concentration of 6.39 µM. Samples were taken at regular intervals over a 24-h period and analyzed by HPAEC-PAD using a CarboPac™ PA200 analytical column (250 mm) equipped with a 50 mm guard column (Thermo Fisher Scientific, Belgium). A gradient elution method as outlined in Table 4 was applied at a constant flow rate of 0.5 mL/min. Released monosaccharides were identified by comparison with standard retention times.
Table 3.
HPAEC-PAD method for the analysis of hydrolysis products on polysaccharide substrates
| Time | Eluent A (ultrapure H2O) | Eluent B (100 mM NaOH) | Eluent C (100 mM NaOH, 1 M NaOAc) |
|---|---|---|---|
| 0.0 | 80.0 | 20.0 | 0.0 |
| 4.0 | 80.0 | 20.0 | 0.0 |
| 8.0 | 0.0 | 100.0 | 0.0 |
| 10 .0 | 0.0 | 100.0 | 0.0 |
| 30.0 | 0.0 | 90.0 | 10.0 |
| 30.5 | 0.0 | 50.0 | 50.0 |
| 35.0 | 0.0 | 50.0 | 50.0 |
| 36.0 | 80.0 | 20.0 | 0.0 |
| 41.0 | 80.0 | 20.0 | 0.0 |
NaOH sodium hydroxide, NaOAc sodium acetate, HPAEC-PAD high-performance anion-exchange chromatography-pulsed amperometric detection
Table 4.
Gradient elution method applied for HPAEC-PAD analysis of monosaccharide release during β-1,3-glucan hydrolysis by TcChiΔ123 using a CarboPac™ PA200 column
| Time | Eluent A (ultrapure H2O) | Eluent B (100 mM NaOH) | Eluent C (100 mM NaOH, 1 M NaOAc) |
|---|---|---|---|
| 0.0 | 0.0 | 99.0 | 1.0 |
| 10.0 | 0.0 | 99.0 | 1.0 |
| 69.0 | 0.0 | 60.0 | 40.0 |
| 70 .0 | 0.0 | 99.0 | 1.0 |
| 75.0 | 0.0 | 99.0 | 1.0 |
NaOH sodium hydroxide, NaOAc sodium acetate, HPAEC-PAD high-performance anion-exchange chromatography-pulsed amperometric detection
Optimal conditions and turnover number
The pH optimum of TcChiΔ123 was determined using 50 µM 4-methylumbelliferyl β-D-N,N′,N′′-triacetylchitotrioside ([GlcNAc]₃-MU) with 5 µM enzyme at 80 °C in the following buffer systems: 50 mM acetate (pH 3.5–4.5), 50 mM MES (pH 5.0–6.5), 50 mM 3-(N-morpholino)propanesulfonic acid (pH 7.0–7.5), and 50 mM 2-amino-2-hydroxymethyl-propane-1,3-diol-HCl (pH 8.0–8.5). The temperature profile was assessed at pH 6.0 in 50 mM MES buffer. During the assay, 20 µL aliquots were collected at time points 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, and 4.0 min, and quenched in 80 µL of 0.3 M glycine buffer (pH 10). Fluorescence was measured using a Tecan M200 Pro (MNano +) plate reader (Tecan) with excitation at 360 nm and emission at 460 nm. Product formation was quantified against a 0–50 µM 4-methylumbelliferone standard curve.
The pH optimum of TcChiΔ234 was determined similarly, using 4-nitrophenyl β-D-cellobioside ([Glc]₂-NP) as the substrate. Absorbance was measured at 405 nm and product concentration was calculated using a 0–100 µM 4-nitrophenol standard curve. Temperature profiling for TcChiΔ234 was performed in 50 mM MES buffer at pH 5.5.
Apparent turnover numbers for TcChiΔ123 and TcChiΔ12 were measured at 85 °C in 50 mM MES buffer (pH 6.0), while assays for TcChiΔ234 and TcChiΔ34 were conducted at 95 °C in 50 mM MES buffer (pH 5.5). For comparison, turnover numbers of SmChiA, SmChiB, and SmChiC were determined at 37 °C in 50 mM MES buffer (pH 6.5). All reactions were performed using 15 mg/mL substrate and 15 mg enzyme per gram of substrate, corresponding to molar enzyme concentrations of 4.34 µM (TcChiΔ12), 6.39 µM (TcChiΔ123), 4.91 µM (TcChiΔ34), 7.07 µM (TcChiΔ234), 3.62 µM (SmChiA), 3.97 µM (SmChiB), and 4.31 µM (SmChiC). Apparent turnover numbers were calculated by quantifying the released monosaccharide during hydrolysis and normalizing the initial reaction rates to the enzyme concentration. For variants where no activity was detected within 40 min, the reaction time was extended up to 3 h to confirm inactivity.
Hydrolysis of laminari-oligosaccharides and chitooligosaccharides
To investigate the mode of action of TcChiΔ123 on β-1,3-glucan and chitin, hydrolysis experiments were performed on defined laminarioligosaccharides and chitooligosaccharides (COS). For laminarioligosaccharide hydrolysis, reactions were carried out at a substrate concentration of 0.5 mM and an enzyme concentration of 250 µg/mL under optimal conditions (85 °C, pH 6.0, 50 mM MES buffer). For COS hydrolysis, a substrate concentration of 5 µg/mL was used under identical incubation conditions. In both cases, reactions were stopped by mixing 50 µL of sample with 50 µL of acetonitrile, followed by a 2.5-fold dilution prior to analysis. Hydrolysis products were analyzed by HPAEC-PAD using the gradient method described for polysaccharide chitin substrate analysis, respectively. Oligosaccharide standards ranging from DP1 to DP6 were used to identify and assign the detected peaks.
Molecular docking
Molecular docking was performed to investigate the structural basis of the substrate specificity of TcChiΔ123. The three-dimensional structure was predicted using Boltz-2 of TcChiΔ123 was used as the receptor model [48]. Ligands were prepared as pentasaccharide oligomers of the respective substrates: chitopentaose (β-1,4-linked GlcNAc, cognate substrate), laminaripentaose (β-1,3-linked glucose, promiscuous substrate), and cellopentaose (β-1,4-linked glucose, negative control). All docking simulations were performed using AutoDock Vina, with 25 independent runs per ligand [49]. Resulting poses were clustered based on root mean square deviation, yielding five clusters for chitopentaose and four clusters each for laminaripentaose and cellopentaose. The top-ranked cluster for each ligand was selected for analysis based on binding energy and cluster size. Docking poses were visualized in PyMOL, and relevant inter-atomic distances were measured using the Wizard tool. As all distances were measured in the static docking model, they are expected to represent upper estimates relative to a dynamically relaxed complex.
Statement for broader audience
Turning natural waste into valuable products is one of the key challenges in building a more sustainable future. Many forms of biomass, such as shellfish waste or agricultural residues, contain tough structural materials that are difficult to break down efficiently. Finding better biological tools to convert these materials into useful compounds is therefore an important scientific goal. In this study, we discovered a remarkable enzyme from a heat-loving microorganism that thrives in extreme environments. This enzyme can break down multiple types of complex natural polymers, including materials related to chitin (found in shells and fungi) and cellulose (found in plant cell walls). Even more interestingly, it combines different catalytic functions within a single protein and remains highly stable at very high temperatures. Because of its versatility and robustness, this enzyme could help improve sustainable processes that convert biological waste into valuable products, such as bio-based chemicals, materials, or fuels.
Supplementary Information
Author contributions
A. W. contributed to conceptualization, investigation, writing—the original draft, software, methodology and visualization. S.D. was involved in supervision, conceptualization, writing—review and editing and methodology. T.D. did funding acquisition, conceptualization, writing—review and editing, supervision and methodology. All authors reviewed the manuscript.
Funding
This research was supported by FWO Bioeconomy Research Project ‘MyCOS’ (GOG3422N) and funded by the European Union—NextGenerationEU. We would also like to thank the UGent Core Facility ‘HTS for Synthetic Biology’ for training, support and access to the instrument park funded by FWO, BOF and UGent.
Data availability
The data that support the findings of this study are openly available at https://zenodo.org/records/18835535?token=eyJhbGciOiJIUzUxMiJ9.eyJpZCI6IjVjMWNlNmUyLWVjMDYtNDAwNi1iNzcwLTk2Njk1NTcwNzEzZCIsImRhdGEiOnt9LCJyYW5kb20iOiI3N2ZkNjY4NmY2NGYwMDQ0NjY4YjRhMDc0NTY3NDFkOCJ9.HdUeI7gJYkE3bjXKvgT2kGLwFSgboCDfpecdAnAoXhRu5QFtHqIx27g_gBt, reference number https://doi.org/10.5281/zenodo.18835535.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Romero-Hernández O, Romero S. Maximizing the value of waste: from waste management to the circular economy. Thunderbird Int Bus Rev. 2018;60(5):757–64. 10.1002/tie.21968. [DOI] [Google Scholar]
- 2.Kim SK. Chitin, chitosan, oligosaccharides and their derivatives: biological activities and applications. Boca Raton: CRC Press; 2010. [Google Scholar]
- 3.Yin H, Du Y, Dong Z. Chitin oligosaccharide and chitosan oligosaccharide: two similar but different plant elicitors. Front Plant Sci. 2016. 10.3389/fpls.2016.00522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zhang R, Zhao Q, Yi Z, et al. Chitin oligosaccharides for the food industry: production and applications. Syst Microbiol Biomanuf. 2022. 10.1007/s43393-022-00127-2. [DOI] [Google Scholar]
- 5.Ganan M, Lorentzen SB, Agger JW, et al. Antifungal activity of well-defined chito-oligosaccharide preparations against medically relevant yeasts. PLoS ONE. 2019;14(1):e0210208. 10.1371/journal.pone.0210208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sánchez Á, Mengibar M, Fernández M, Alemany S, Heras A, Acosta N. Influence of preparation methods of chitooligosaccharides on their physicochemical properties and their anti-inflammatory effects in mice and in RAW264.7 macrophages. Mar Drugs. 2018;16(11):430. 10.3390/md16110430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zheng J, Cheng G, Li Q, et al. Chitin oligosaccharide modulates gut microbiota and attenuates high-fat-diet-induced metabolic syndrome in mice. Mar Drugs. 2018;16(2):66. 10.3390/md16020066. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zhao L. Oligosaccharides of chitin and chitosan: bio-manufacture and applications. Berlin: Springer Nature; 2019. [Google Scholar]
- 9.Adrangi S, Faramarzi MA. From bacteria to human: a journey into the world of chitinases. Biotechnol Adv. 2013;31(8):1786–95. 10.1016/j.biotechadv.2013.09.012. [DOI] [PubMed] [Google Scholar]
- 10.Demeester W, De Paepe B, Guidi C, De Mey M. Unravelling the regulatory network behind chitin degradation in Serratia marcescens. Biol Rev. 2025. 10.1111/brv.70020. [DOI] [PubMed] [Google Scholar]
- 11.Brurberg MB, Nes IF, Eijsink VGH. Comparative studies of chitinases A and B from Serratia marcescens. Microbiology (Reading). 1996;142(7):1581–9. 10.1099/13500872-142-7-1581. [DOI] [PubMed] [Google Scholar]
- 12.Horn SJ, Sørlie M, Vaaje-Kolstad G, et al. Comparative studies of chitinases A, B and C from Serratia marcescens. Biocatal Biotransform. 2005;24(1/2):39–53. 10.1080/10242420500518482. [DOI] [Google Scholar]
- 13.Mekasha S, Byman IR, Lynch C, et al. Development of enzyme cocktails for complete saccharification of chitin using mono-component enzymes from Serratia marcescens. Process Biochem. 2017;56:132–8. 10.1016/j.procbio.2017.02.021. [DOI] [Google Scholar]
- 14.Windels A, Declerck L, Snoeck S, et al. Bioconversion of mushroom chitin-rich waste into valuable chitin oligosaccharides using a combined approach of biocatalysis and precision fermentation. J Agric Food Chem. 2025;73(16):9769–81. 10.1021/acs.jafc.5c00928. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rossino G, Robescu MS, Licastro E, et al. Biocatalysis: a smart and green tool for the preparation of chiral drugs. Chirality. 2022;34(11):1403–18. 10.1002/chir.23498. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kaul P, Asano Y. Strategies for discovery and improvement of enzyme function: state of the art and opportunities. Microb Biotechnol. 2012;5(1):18–33. 10.1111/j.1751-7915.2011.00280.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.van Kempen M, Kim SS, Tumescheit C, et al. Fast and accurate protein structure search with Foldseek. bioRxiv. 2023. 10.1101/2022.02.07.479398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215(3):403–10. 10.1016/S0022-2836(05)80360-2. [DOI] [PubMed] [Google Scholar]
- 19.Li G, Rabe KS, Nielsen J, Engqvist MKM. Machine learning applied to predicting microorganism growth temperatures and enzyme catalytic optima. ACS Synth Biol. 2019;8(6):1411–20. 10.1021/acssynbio.9b00099. [DOI] [PubMed] [Google Scholar]
- 20.Hon J, Marusiak M, Martinek T, et al. SoluProt: prediction of soluble protein expression in Escherichia coli. Bioinformatics. 2021;37(1):23–8. 10.1093/bioinformatics/btaa1102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pan T, Cui X, Koh HY, et al. KcatNet: a geometric deep learning framework for genome-wide prediction of enzyme catalytic efficiency. bioRxiv. 2025. 10.1101/2025.03.09.642294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Helbert W, Poulet L, Drouillard S, et al. Discovery of novel carbohydrate-active enzymes through the rational exploration of the protein sequences space. Proc Natl Acad Sci U S A. 2019;116(13):6063–8. 10.1073/pnas.1815791116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Salemi M, Vandamme AM. The phylogenetic handbook: a practical approach to DNA and protein phylogeny. Cambridge: Cambridge University Press; 2003. [Google Scholar]
- 24.De Beul E, Jongbloet A, Franceus J, Desmet T. Discovery of a kojibiose hydrolase by analysis of specificity-determining correlated positions in glycoside hydrolase family 65. Molecules. 2021;26(20):20. 10.3390/molecules26206321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kim D, Noh MH, Park M, et al. Enzyme activity engineering based on sequence co-evolution analysis. Metab Eng. 2022;74:49–60. 10.1016/j.ymben.2022.09.001. [DOI] [PubMed] [Google Scholar]
- 26.Windels A, Franceus J, Pleiss J, Desmet T. CANDy: automated analysis of domain architectures in carbohydrate-active enzymes. PLoS ONE. 2024;19(7):e0306410. 10.1371/journal.pone.0306410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Huang QS, Xie XL, Liang G, et al. The GH18 family of chitinases: their domain architectures, functions and evolutions. Glycobiology. 2012;22(1):23–34. 10.1093/glycob/cwr092. [DOI] [PubMed] [Google Scholar]
- 28.Pulsford SB, Dalton JAT, Smith OB, et al. The modular evolution of chitinases is governed by coevolution of auxiliary and catalytic domains. bioRxiv. 2026. 10.64898/2026.01.15.699799. [DOI] [Google Scholar]
- 29.Ikegami T, Okada T, Hashimoto M, Seino S, Watanabe T, Shirakawa M. Solution structure of the chitin-binding domain of Bacillus circulans WL-12 Chitinase A1*. J Biol Chem. 2000;275(18):13654–61. 10.1074/jbc.275.18.13654. [DOI] [PubMed] [Google Scholar]
- 30.Brun E, Moriaud F, Gans P, Blackledge MJ, Barras F, Marion D. Solution structure of the cellulose-binding domain of the endoglucanase Z secreted by Erwinia chrysanthemi. Biochemistry. 1997;36(51):16074–86. 10.1021/bi9718494. [DOI] [PubMed] [Google Scholar]
- 31.Honda S, Kunii T, Nohara K, et al. Characterization of a Bacillus thuringiensis chitinase that binds to cellulose and chitin. AMB Express. 2017;7(1):51. 10.1186/s13568-017-0352-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.The International Polycystic Kidney Disease Consortium. Polycystic kidney disease: the complete structure of the PKD1 gene and its protein. Cell. 1995;81(2):289–98. 10.1016/0092-8674(95)90339-9. [DOI] [PubMed] [Google Scholar]
- 33.Wilson PD. Polycystin: new aspects of structure, function, and regulation. J Am Soc Nephrol. 2001;12(4):834. 10.1681/ASN.V124834. [DOI] [PubMed] [Google Scholar]
- 34.Jing H, Takagi J, Liu J-H, et al. Archaeal surface layer proteins contain beta propeller, PKD, and beta helix domains and are related to metazoan cell surface proteins. Struct Lond Engl 1993. 2002;10(10):1453–64. 10.1016/s0969-2126(02)00840-7. [DOI] [PubMed] [Google Scholar]
- 35.Ma L, Xu M, Forman JR, Clarke J, Oberhauser AF. Naturally occurring mutations alter the stability of polycystin-1 polycystic kidney disease (PKD) domains. J Biol Chem. 2009;284(47):32942–9. 10.1074/jbc.M109.021832. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Hou J, Han J, Cai L, et al. Characterization of genes for chitin catabolism in Haloferax mediterranei. Appl Microbiol Biotechnol. 2014;98(3):1185–94. 10.1007/s00253-013-4969-8. [DOI] [PubMed] [Google Scholar]
- 37.García-Fraga B, da Silva AF, López-Seijas J, Sieiro C. Functional expression and characterization of a chitinase from the marine archaeon Halobacterium salinarum CECT 395 in Escherichia coli. Appl Microbiol Biotechnol. 2014;98(5):2133–43. 10.1007/s00253-013-5124-2. [DOI] [PubMed] [Google Scholar]
- 38.Oku T, Ishikawa K. Analysis of the hyperthermophilic chitinase from Pyrococcus furiosus: activity toward crystalline chitin. Biosci Biotechnol Biochem. 2006;70(7):1696–701. 10.1271/bbb.60031. [DOI] [PubMed] [Google Scholar]
- 39.Tanaka T, Fukui T, Imanaka T. Different cleavage specificities of the dual catalytic domains in chitinase from the hyperthermophilic archaeon Thermococcus kodakaraensis KOD1. J Biol Chem. 2001;276(38):35629–35. 10.1074/jbc.M105919200. [DOI] [PubMed] [Google Scholar]
- 40.Andronopoulou E, Vorgias CE. Purification and characterization of a new hyperthermostable, allosamidin-insensitive and denaturation-resistant chitinase from the hyperthermophilic archaeon Thermococcus chitonophagus. Extremophiles. 2003;7(1):43–53. 10.1007/s00792-002-0294-3. [DOI] [PubMed] [Google Scholar]
- 41.Huber R, Stöhr J, Hohenhaus S, et al. Thermococcus chitonophagus sp. nov., a novel, chitin-degrading, hyperthermophilic archaeum from a deep-sea hydrothermal vent environment. Arch Microbiol. 1995;164(4):255–64. 10.1007/BF02529959. [DOI] [Google Scholar]
- 42.Xia W, Liu P, Liu J. Advance in chitosan hydrolysis by non-specific cellulases. Bioresour Technol. 2008;99(15):6751–62. 10.1016/j.biortech.2008.01.011. [DOI] [PubMed] [Google Scholar]
- 43.Bauer MW, Driskill LE, Callen W, Snead MA, Mathur EJ, Kelly RM. An endoglucanase, EglA, from the hyperthermophilic archaeon Pyrococcus furiosus hydrolyzes β-1,4 bonds in mixed-linkage (1→3),(1→4)-β-d-glucans and cellulose. J Bacteriol. 1999;181(1):284–90. 10.1128/jb.181.1.284-290.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Talamantes D, Biabini N, Dang H, Abdoun K, Berlemont R. Natural diversity of cellulases, xylanases, and chitinases in bacteria. Biotechnol Biofuels. 2016;9(1):133. 10.1186/s13068-016-0538-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Letunic I, Bork P. Interactive tree of life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res. 2021;49(W1):W293–6. 10.1093/nar/gkab301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Shannon P, Markiel A, Ozier O, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13(11):2498–504. 10.1101/gr.1239303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.De Doncker M, Vleminckx S, Franceus J, Vercauteren R, Desmet T. Converting bulk sugars into functional fibers: discovery and application of a thermostable β-1,3-oligoglucan phosphorylase. J Agric Food Chem. 2024;72(18):10497–505. 10.1021/acs.jafc.4c01990. [DOI] [PubMed] [Google Scholar]
- 48.Passaro S, Corso G, Wohlwend J, et al. Boltz-2: towards accurate and efficient binding affinity prediction. bioRxiv. 2025. 10.1101/2025.06.14.659707. [DOI] [Google Scholar]
- 49.Eberhardt J, Santos-Martins D, Tillack AF, Forli S. AutoDock Vina 1.2.0: new docking methods, expanded force field, and Python bindings. J Chem Inf Model. 2021;61(8):3891–8. 10.1021/acs.jcim.1c00203. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are openly available at https://zenodo.org/records/18835535?token=eyJhbGciOiJIUzUxMiJ9.eyJpZCI6IjVjMWNlNmUyLWVjMDYtNDAwNi1iNzcwLTk2Njk1NTcwNzEzZCIsImRhdGEiOnt9LCJyYW5kb20iOiI3N2ZkNjY4NmY2NGYwMDQ0NjY4YjRhMDc0NTY3NDFkOCJ9.HdUeI7gJYkE3bjXKvgT2kGLwFSgboCDfpecdAnAoXhRu5QFtHqIx27g_gBt, reference number https://doi.org/10.5281/zenodo.18835535.