Advertisement

Functional diversity of three tandem C-terminal carbohydrate-binding modules of a β-mannanase

Open AccessPublished:April 06, 2021DOI:https://doi.org/10.1016/j.jbc.2021.100638
      Carbohydrate active enzymes, such as those involved in plant cell wall and storage polysaccharide biosynthesis and deconstruction, often contain repeating noncatalytic carbohydrate-binding modules (CBMs) to compensate for low-affinity binding typical of protein–carbohydrate interactions. The bacterium Saccharophagus degradans produces an endo-β-mannanase of glycoside hydrolase family 5 subfamily 8 with three phylogenetically distinct family 10 CBMs located C-terminally from the catalytic domain (SdGH5_8-CBM10x3). However, the functional roles and cooperativity of these CBM domains in polysaccharide binding are not clear. To learn more, we studied the full-length enzyme, three stepwise CBM family 10 (CBM10) truncations, and GFP fusions of the individual CBM10s and all three domains together by pull-down assays, affinity gel electrophoresis, and activity assays. Only the C-terminal CBM10-3 was found to bind strongly to microcrystalline cellulose (dissociation constant, Kd = 1.48 μM). CBM10-3 and CBM10-2 bound galactomannan with similar affinity (Kd = 0.2–0.4 mg/ml), but CBM10-1 had 20-fold lower affinity for this substrate. CBM10 truncations barely affected specific activity on carob galactomannan and konjac glucomannan. Full-length SdGH5_8-CBM10x3 was twofold more active on the highly galactose-decorated viscous guar gum galactomannan and crystalline ivory nut mannan at high enzyme concentrations, but the specific activity was fourfold to ninefold reduced at low enzyme and substrate concentrations compared with the enzyme lacking CBM10-2 and CBM10-3. Comparison of activity and binding data for the different enzyme forms indicates unproductive and productive polysaccharide binding to occur. We conclude that the C-terminal-most CBM10-3 secures firm binding, with contribution from CBM10-2, which with CBM10-1 also provides spatial flexibility.

      Keywords

      Abbreviations:

      AGE (affinity gel electrophoresis), BMCC (bacterial microcrystalline cellulose), CBM (carbohydrate-binding module), CBM10 (CBM family 10), CD (catalytic domain), CGM-lv (low-viscosity carob galactomannan), DNS (3,5-dinitrosalicylic acid), GG (guar gum), GH (glycoside hydrolase), GH5_8 (GH family 5 subfamily 8), HEC (hydroxyethyl cellulose), INM (ivory nut mannan), KGM (konjac glucomannan), LPMO (lytic polysaccharide monooxygenase), PASC (phosphoric acid–swollen cellulose)
      Interactions between proteins and carbohydrates play a vital role in life and have through evolution been optimized to match the environments where they take place. Affinities of protein–carbohydrate complexes range from millimolar to nanomolar. Low-affinity binding (millimolar range) is considered a key factor in dynamic systems, such as plant cell wall synthesis and degradation, microbe–host interplay, and synthesis and mobilization of storage polysaccharides. The carbohydrate active enzymes catalyzing these different reactions very often contain one or more noncatalytic carbohydrate-binding modules (CBMs) that facilitate the formation of enzyme–substrate complexes and can specifically bind with the substrate polysaccharides or with different polysaccharides located nearby, such as in the plant cell wall. It is common that polysaccharide hydrolases contain several CBMs, but it is difficult to understand the function of these individual CBMs because of their interplay with each other, substrate, other polysaccharides, and the catalytic domain (CD), and perhaps even more than one CD is present (
      • Conway J.M.
      • Crosby J.R.
      • Hren A.P.
      • Southerland R.T.
      • Lee L.L.
      • Lunin V.V.
      • Alahuhta P.
      • Himmel M.E.
      • Bomble Y.J.
      • Adams M.W.W.
      • Kelly R.M.
      Novel multidomain, multifunctional glycoside hydrolases from highly lignocellulolytic Caldicellulosiruptor species.
      ). Certain insights are established, for example, on enzymes in cellulosomes with domains of the polyspecific CBM32 family interacting with galactose, lactose, polygalacturonic acid, and N-acetyllactosamine (
      • Low K.E.
      • Smith S.P.
      • Abbott D.W.
      • Boraston A.B.
      The glycoconjugate-degrading enzymes of Clostridium perfringens: Tailored catalysts for breaching the intestinal mucus barrier.
      ). There are many examples where a functional role has been depicted for specific domains, while being unknown for others in multidomain architectures (
      • Krska D.
      • Larsbrink J.
      Investigation of a thermostable multi-domain xylanase-glucuronoyl esterase enzyme from Caldicellulosiruptor kristjanssonii incorporating multiple carbohydrate-binding modules.
      ). Clearly, there are relatively few three-dimensional structures available of the more complex multidomain enzymes because of a number of challenges. One being the recombinant production of multimodular proteins containing interdomain linker regions prone to limited proteolysis, another is the dynamic flexibility inherent to the multimodular domain architectures (
      • Low K.E.
      • Smith S.P.
      • Abbott D.W.
      • Boraston A.B.
      The glycoconjugate-degrading enzymes of Clostridium perfringens: Tailored catalysts for breaching the intestinal mucus barrier.
      ,
      • Krska D.
      • Larsbrink J.
      Investigation of a thermostable multi-domain xylanase-glucuronoyl esterase enzyme from Caldicellulosiruptor kristjanssonii incorporating multiple carbohydrate-binding modules.
      ).
      Mannan polysaccharides (Fig. 1) are found in cell walls of most plants and as storage polysaccharides in seeds and tubers (
      • Ebringerová A.
      Structural diversity and application potential of hemicelluloses.
      ). The four prominent types of β-mannans include galactomannans having a β-1,4 mannose backbone with different extents of α-1,6 galactose decoration, glucomannans with an essentially linear backbone of glucosyl and mannosyl units, which in the galactoglucomannans are substituted by α-1,6-galactose (
      • Dey P.M.
      Biochemistry of plant galactomannans.
      ,
      • McCleary B.V.
      • Clark A.H.
      • Dea I.C.M.
      • Rees D.A.
      The fine structures of carob and guar galactomannans.
      ); essentially linear β-mannans are found in plant stems and as highly packed storage polysaccharides (
      • Hori R.
      • Sugiyama J.
      • Wada M.
      The thermal expansion of mannan I obtained from ivory nuts.
      ,
      • Heux L.
      • Hägglund P.
      • Putaux J.L.
      • Chanzy H.
      Structural aspects in semicrystalline samples of the mannan II family.
      ) (Fig. 1). Some mannans are acetylated to varying degree depending on the type and botanical source (
      • Xu C.
      • Leppänen A.S.
      • Eklund P.
      • Holmlund P.
      • Sjöholm R.
      • Sundberg K.
      • Willför S.
      Acetylation and characterization of spruce (Picea abies) galactoglucomannans.
      ).
      Figure thumbnail gr1
      Figure 1Structures of the included polysaccharides. Mannose (Man), galactose (Gal), and glucose (Glc) ratios are given for galactomannans and glucomannan.
      Mannans are depolymerized by endo-β-mannanases found in glycoside hydrolase (GH) families GH5, GH26, GH113, and GH134 (
      • Lombard V.
      • Golaconda Ramulu H.
      • Drula E.
      • Coutinho P.M.
      • Henrissat B.
      The carbohydrate-active enzymes database (CAZy) in 2013.
      ), some of which also contain CBMs facilitating contact with the substrates. Several applications are reported of endo-β-mannanases, for example, for production of oligosaccharides and in saccharification of plant biomass and lignocellulose degradation, where CBMs are of potential importance (
      • Morrill J.
      • Månberger A.
      • Rosengren A.
      • Naidjonoka P.
      • von Freiesleben P.
      • Krogh K.B.R.M.
      • Bergquist K.E.
      • Nylander T.
      • Karlsson E.N.
      • Adlercreutz P.
      • Stålbrand H.
      β-Mannanase-catalyzed synthesis of alkyl mannooligosides.
      ,
      • Von Freiesleben P.
      • Spodsberg N.
      • Stenbæk A.
      • Stålbrand H.
      • Krogh K.B.R.M.
      • Meyer A.S.
      Boosting of enzymatic softwood saccharification by fungal GH5 and GH26 endomannanases.
      ,
      • Wei X.
      • Fu X.
      • Xiao M.
      • Liu Z.
      • Zhang L.
      • Mou H.
      Dietary galactosyl and mannosyl carbohydrates: In-vitro assessment of prebiotic effects.
      ,
      • Zhao D.
      • Zhang X.
      • Wang Y.
      • Na J.
      • Ping W.
      • Ge J.
      Purification, biochemical and secondary structural characterisation of β-mannanase from Lactobacillus casei HDS-01 and juice clarification potential.
      ).
      CBMs are classified into 88 families based on amino acid sequence similarity (CAZy, database of Carbohydrate Active enZYmes: http://cazy.org) (
      • Lombard V.
      • Golaconda Ramulu H.
      • Drula E.
      • Coutinho P.M.
      • Henrissat B.
      The carbohydrate-active enzymes database (CAZy) in 2013.
      ). CBM family 10 (CBM10), despite the very small size of 29 to 42 amino acids, presents a diversity of carbohydrate-binding specificities and affinity ranges (Table 1) (
      • Gill J.
      • Rixon J.E.
      • Bolam D.N.
      • McQueen-Mason S.
      • Simpson P.J.
      • Williamson M.P.
      • Hazlewood G.P.
      • Gilbert H.J.
      The type II and X cellulose-binding domains of Pseudomonas xylanase A potentiate catalytic activity against complex substrates by a common mechanism.
      ,
      • Ponyi T.
      • Szabó L.
      • Nagy T.
      • Orosz L.
      • Simpson P.J.
      • Williamson M.P.
      • Gilbert H.J.
      Trp22, Trp24, and Tyr8 play a pivotal role in the binding of the family 10 cellulose-binding module from Pseudomonas xylanase A to insoluble ligands.
      ,
      • Hogg D.
      • Pell G.
      • Dupree P.
      • Goubet F.
      • Marti S.M.
      • Gilbert H.J.
      The modular architecture of Cellvibrio japonicus mannanases in glycoside hydrolase families 5 and 26 points to differences in their role in mannan degradation.
      ,
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      ,
      • Attia M.
      • Stepper J.
      • Davies G.J.
      • Brumer H.
      Functional and structural characterization of a potent GH74 endo-xyloglucanase from the soil saprophyte Cellvibrio japonicus unravels the first step of xyloglucan degradation.
      ,
      • Crouch L.I.
      • Labourel A.
      • Walton P.H.
      • Davies G.J.
      • Gilbert H.J.
      The contribution of non-catalytic carbohydrate binding modules to the activity of lytic polysaccharide monooxygenases.
      ,
      • Pires V.M.R.
      • Pereira P.M.M.
      • Brás J.L.A.
      • Correia M.
      • Cardoso V.
      • Bule P.
      • Alves V.D.
      • Najmudin S.
      • Venditto I.
      • Ferreira L.M.A.
      • Romão M.J.
      • Carvalho A.L.
      • Fontes C.M.G.A.
      • Prazeres D.M.
      Stability and ligand promiscuity of type A carbohydrate-binding modules are illustrated by the structure of Spirochaeta thermophila StCBM64C.
      ). CBM10s are thus associated with different CDs in plant cell wall polysaccharide-degrading enzymes, including cellulases (
      • Hall J.
      • Black G.W.
      • Ferreira L.M.
      • Millward-Sadler S.J.
      • Ali B.R.
      • Hazlewood G.P.
      • Gilbert H.J.
      The non-catalytic cellulose-binding domain of a novel cellulase from Pseudomonas fluorescens subsp. cellulosa is important for the efficient hydrolysis of Avicel.
      ,
      • Gilbert H.J.
      • Hall J.
      • Hazlewood G.P.
      • Ferreira L.M.A.
      The N-terminal region of an endoglucanase from Pseudomonas fluorescens subspecies cellulosa constitutes a cellulose-binding domain that is distinct from the catalytic centre.
      ,
      • Gardner J.G.
      • Crouch L.
      • Labourel A.
      • Forsberg Z.
      • Bukhman Y.V.
      • Vaaje-Kolstad G.
      • Gilbert H.J.
      • Keating D.H.
      Systems biology defines the biological significance of redox-active proteins during cellulose degradation in an aerobic bacterium.
      ,
      • Ferreira L.M.A.
      • Hazlewood G.P.
      • Barker P.J.
      • Gilbert H.J.
      The cellodextrinase from Pseudomonas fluorescens subsp. cellulosa consists of multiple functional domains.
      ,
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      ,
      • Watson B.J.
      • Zhang H.
      • Longmire A.G.
      • Young H.M.
      • Hutcheson S.W.
      Processive endoglucanases mediate degradation of cellulose by Saccharophagus degradans.
      ,
      • Ekborg N.A.
      • Morrill W.
      • Burgoyne A.M.
      • Li L.
      • Distel D.L.
      CelAB, a multifunctional cellulase encoded by Teredinibacter turnerae T7902T, a culturable symbiont isolated from the wood-boring marine bivalve Lyrodus pedicellatus.
      ), xylanases (
      • Gill J.
      • Rixon J.E.
      • Bolam D.N.
      • McQueen-Mason S.
      • Simpson P.J.
      • Williamson M.P.
      • Hazlewood G.P.
      • Gilbert H.J.
      The type II and X cellulose-binding domains of Pseudomonas xylanase A potentiate catalytic activity against complex substrates by a common mechanism.
      ,
      • Attia M.
      • Stepper J.
      • Davies G.J.
      • Brumer H.
      Functional and structural characterization of a potent GH74 endo-xyloglucanase from the soil saprophyte Cellvibrio japonicus unravels the first step of xyloglucan degradation.
      ,
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      ,
      • Millward-Sadler S.J.
      • Davidson K.
      • Hazlewood G.P.
      • Black G.W.
      • Gilbert H.J.
      • Clarke J.H.
      Novel cellulose-binding domains, NodB homologues and conserved modular architecture in xylanases from the aerobic soil bacteria Pseudomonas fluorescens subsp. cellulosa and Cellvibrio mixtus.
      ), mannanases (
      • Hogg D.
      • Pell G.
      • Dupree P.
      • Goubet F.
      • Marti S.M.
      • Gilbert H.J.
      The modular architecture of Cellvibrio japonicus mannanases in glycoside hydrolase families 5 and 26 points to differences in their role in mannan degradation.
      ,
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      ,
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      ,
      • Tamaru Y.
      • Araki T.
      • Morishita T.
      • Kimura T.
      • Sakka K.
      • Ohmiya K.
      Cloning, DNA sequencing, and expression of the β-1,4-mannanase gene from a marine bacterium, Vibrio sp. strain MA-138.
      ,
      • Kim D.Y.
      • Ham S.J.
      • Lee H.J.
      • Cho H.Y.
      • Kim J.H.
      • Kim Y.J.
      • Shin D.H.
      • Rhee Y.H.
      • Son K.H.
      • Park H.Y.
      Cloning and characterization of a modular GH5 β-1,4-mannanase with high specific activity from the fibrolytic bacterium Cellulosimicrobium sp. strain HY-13.
      ,
      • Jin Y.
      • Petricevic M.
      • John A.
      • Raich L.
      • Jenkins H.
      • Portela De Souza L.
      • Cuskin F.
      • Gilbert H.J.
      • Rovira C.
      • Goddard-Borger E.D.
      • Williams S.J.
      • Davies G.J.
      A β-mannanase with a lysozyme-like fold and a novel molecular catalytic mechanism.
      ,
      • Sakai K.
      • Kimoto S.
      • Shinzawa Y.
      • Minezawa M.
      • Suzuki K.
      • Jindou S.
      • Kato M.
      • Shimizu M.
      Characterization of pH-tolerant and thermostable GH 134 β-1,4-mannanase SsGH134 possessing carbohydrate binding module 10 from Streptomyces sp. NRRL B-24484.
      ,
      • Shi P.
      • Yuan T.
      • Zhao J.
      • Huang H.
      • Luo H.
      • Meng K.
      • Wang Y.
      • Yao B.
      Genetic and biochemical characterization of a protease-resistant mesophilic β-mannanase from Streptomyces sp. S27.
      ,
      • Arcand N.
      • Kluepfel D.
      • Paradis F.W.
      • Morosoli R.
      • Shareck F.
      β-Mannanase of Streptomyces lividans 66: Cloning and DNA sequence of the manA gene and characterization of the enzyme.
      ), and lytic polysaccharide monooxygenases (LPMOs) (
      • Crouch L.I.
      • Labourel A.
      • Walton P.H.
      • Davies G.J.
      • Gilbert H.J.
      The contribution of non-catalytic carbohydrate binding modules to the activity of lytic polysaccharide monooxygenases.
      ). Typically, these enzymes have more than one CBM10 or contain also CBMs from other families, CBM2s being particularly common (Table 1).
      Table 1Summary of characteristics of enzymes having one or more CBM10(s)
      OrganismGenBank accession no.Modular organizationCBM10 bindingEnzymatic activityBinding-site residuesDisulphide bridgesRef.
      C. japonicusACE85176CBM10-GH26ND■◆DWF2(
      • Hogg D.
      • Pell G.
      • Dupree P.
      • Goubet F.
      • Marti S.M.
      • Gilbert H.J.
      The modular architecture of Cellvibrio japonicus mannanases in glycoside hydrolase families 5 and 26 points to differences in their role in mannan degradation.
      )
      S. degradans (SdGH5CBM10-1)ABD79918GH5_8-CBM10-CBM10-CBM10■(Kd = 3.7 g/l) [◯•]■□◆SWY2
      C. japonicus (CjAA10CBM10)ACE84760AA10-CBM10◯(Kd = 7.5–17.3 μM)◯[•]YWN2(
      • Crouch L.I.
      • Labourel A.
      • Walton P.H.
      • Davies G.J.
      • Gilbert H.J.
      The contribution of non-catalytic carbohydrate binding modules to the activity of lytic polysaccharide monooxygenases.
      ,
      • Gardner J.G.
      • Crouch L.
      • Labourel A.
      • Forsberg Z.
      • Bukhman Y.V.
      • Vaaje-Kolstad G.
      • Gilbert H.J.
      • Keating D.H.
      Systems biology defines the biological significance of redox-active proteins during cellulose degradation in an aerobic bacterium.
      )
      A. bacteriumAIF91534GH5_8-CBM10-CBM10-CBM10NDYWY2(
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      )
      C. japonicusACE82655CBM5-CBM10-CBM35-GH5_7ND■□◆YWW2(
      • Hogg D.
      • Pell G.
      • Dupree P.
      • Goubet F.
      • Marti S.M.
      • Gilbert H.J.
      The modular architecture of Cellvibrio japonicus mannanases in glycoside hydrolase families 5 and 26 points to differences in their role in mannan degradation.
      )
      C. japonicusACE82688CBM2-CBM10-GH45•▲YWY2(
      • Gilbert H.J.
      • Hall J.
      • Hazlewood G.P.
      • Ferreira L.M.A.
      The N-terminal region of an endoglucanase from Pseudomonas fluorescens subspecies cellulosa constitutes a cellulose-binding domain that is distinct from the catalytic centre.
      )
      C. mixtusCAA88761GH11-CBM60-CE4-CBM10ND▲[•]WWW2(
      • Millward-Sadler S.J.
      • Davidson K.
      • Hazlewood G.P.
      • Black G.W.
      • Gilbert H.J.
      • Clarke J.H.
      Novel cellulose-binding domains, NodB homologues and conserved modular architecture in xylanases from the aerobic soil bacteria Pseudomonas fluorescens subsp. cellulosa and Cellvibrio mixtus.
      )
      A. bacteriumAIF91529CBM10-CBM10-GH10_4NDYWW2(
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      )
      C. japonicusACE84673GH5_8-CBM10-CBM10◯□[■•]■◆[□•]YWW2(
      • Hogg D.
      • Pell G.
      • Dupree P.
      • Goubet F.
      • Marti S.M.
      • Gilbert H.J.
      The modular architecture of Cellvibrio japonicus mannanases in glycoside hydrolase families 5 and 26 points to differences in their role in mannan degradation.
      )
      C. japonicusACE85978CBM2-CBM10-GH6ND◯[•▲]YWW2(
      • Gardner J.G.
      • Crouch L.
      • Labourel A.
      • Forsberg Z.
      • Bukhman Y.V.
      • Vaaje-Kolstad G.
      • Gilbert H.J.
      • Keating D.H.
      Systems biology defines the biological significance of redox-active proteins during cellulose degradation in an aerobic bacterium.
      )
      C. japonicusACE82870CBM2-CBM10-GH5_53◯(CBM2+10?)•[◯▲]YWY2(
      • Ferreira L.M.A.
      • Hazlewood G.P.
      • Barker P.J.
      • Gilbert H.J.
      The cellodextrinase from Pseudomonas fluorescens subsp. cellulosa consists of multiple functional domains.
      )
      C. japonicus (CjGH10CBM10)ACE85439CBM2-CBM10-GH10◯(Kd = 3.9 μM) [•]•▲YWW2(
      • Gill J.
      • Rixon J.E.
      • Bolam D.N.
      • McQueen-Mason S.
      • Simpson P.J.
      • Williamson M.P.
      • Hazlewood G.P.
      • Gilbert H.J.
      The type II and X cellulose-binding domains of Pseudomonas xylanase A potentiate catalytic activity against complex substrates by a common mechanism.
      ,
      • Ponyi T.
      • Szabó L.
      • Nagy T.
      • Orosz L.
      • Simpson P.J.
      • Williamson M.P.
      • Gilbert H.J.
      Trp22, Trp24, and Tyr8 play a pivotal role in the binding of the family 10 cellulose-binding module from Pseudomonas xylanase A to insoluble ligands.
      ,
      • Pires V.M.R.
      • Pereira P.M.M.
      • Brás J.L.A.
      • Correia M.
      • Cardoso V.
      • Bule P.
      • Alves V.D.
      • Najmudin S.
      • Venditto I.
      • Ferreira L.M.A.
      • Romão M.J.
      • Carvalho A.L.
      • Fontes C.M.G.A.
      • Prazeres D.M.
      Stability and ligand promiscuity of type A carbohydrate-binding modules are illustrated by the structure of Spirochaeta thermophila StCBM64C.
      ,
      • Raghothama S.
      • Simpson P.J.
      • Szabó L.
      • Nagy T.
      • Gilbert H.J.
      • Williamson M.P.
      Solution structure of the CBM10 cellulose binding module from Pseudomonas xylanase A.
      )
      C. japonicusACE84076GH5_2-CBM10-CBM2◯[▲](CBM2+10)◯•▲[■]YWW2(
      • Hall J.
      • Black G.W.
      • Ferreira L.M.
      • Millward-Sadler S.J.
      • Ali B.R.
      • Hazlewood G.P.
      • Gilbert H.J.
      The non-catalytic cellulose-binding domain of a novel cellulase from Pseudomonas fluorescens subsp. cellulosa is important for the efficient hydrolysis of Avicel.
      ,
      • Gardner J.G.
      • Crouch L.
      • Labourel A.
      • Forsberg Z.
      • Bukhman Y.V.
      • Vaaje-Kolstad G.
      • Gilbert H.J.
      • Keating D.H.
      Systems biology defines the biological significance of redox-active proteins during cellulose degradation in an aerobic bacterium.
      )
      C. japonicusAAO31760GH5_8-CBM10-CBM2◯□[■•]■◆[□•]YWW2(
      • Hogg D.
      • Pell G.
      • Dupree P.
      • Goubet F.
      • Marti S.M.
      • Gilbert H.J.
      The modular architecture of Cellvibrio japonicus mannanases in glycoside hydrolase families 5 and 26 points to differences in their role in mannan degradation.
      )
      C. japonicus (CjGH74CBM10)ACE84745GH74-CBM10-CBM2◯(Kd = 1.5 μM)[▲]▲•[■◆]YWW2(
      • Attia M.
      • Stepper J.
      • Davies G.J.
      • Brumer H.
      Functional and structural characterization of a potent GH74 endo-xyloglucanase from the soil saprophyte Cellvibrio japonicus unravels the first step of xyloglucan degradation.
      )
      C. japonicus (CjGH9CBM10)ACE85757GH9-CBM10-CBM2NDYWW2(
      • Duan C.-J.
      • Huang M.-Y.
      • Pang H.
      • Zhao J.
      • Wu C.-X.
      • Feng J.-X.
      Characterization of a novel theme C glycoside hydrolase family 9 cellulase and its CBM-chimeric enzymes.
      )
      S. degradans (SdGH5CBM10-3)ABD79918GH5_8-CBM10-CBM10-CBM10■(Kd < 0.13 g/l)

      ◯(Kd = 1.5 μM)[•]
      ■□◆YWW2
      A. bacteriumAIF191551CBM10-CBM10-CBM60-GH10_4NDYWW2(
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      )
      A. bacteriumAIF91559GH11-CBM60-CE4-CBM10NDYWW2(
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      )
      C. japonicusACE84179GH11-CBM60-CE4-CBM10◯[▲]▲[•]WWW2(
      • Millward-Sadler S.J.
      • Davidson K.
      • Hazlewood G.P.
      • Black G.W.
      • Gilbert H.J.
      • Clarke J.H.
      Novel cellulose-binding domains, NodB homologues and conserved modular architecture in xylanases from the aerobic soil bacteria Pseudomonas fluorescens subsp. cellulosa and Cellvibrio mixtus.
      )
      A. bacteriumAIF91534GH5_8-CBM10-CBM10-CBM10NDYFW2(
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      )
      A. bacteriumAIF91551CBM10-CBM10-CBM60-GH10_4NDYWY2(
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      )
      A. bacteriumAIF91529CBM10-CBM10-GH10_4NDYWY2(
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      )
      A. bacteriumAIF91557CBM2-CBM10-GH5_53ND◯•YWW2(
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      )
      A. bacteriumAIF91536GH5_2-CBM10-CBM2NDYWW2(
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      )
      A. bacteriumAIF91534GH5_8-CBM10-CBM10-CBM10NDYWY2(
      • O’Connor R.M.
      • Fung J.M.
      • Sharp K.H.
      • Benner J.S.
      • McClung C.
      • Cushing S.
      • Lamkin E.R.
      • Fomenkov A.I.
      • Henrissat B.
      • Londer Y.Y.
      • Scholz M.B.
      • Posfai J.
      • Malfatti S.
      • Tringe S.G.
      • Woyke T.
      • et al.
      Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
      )
      S. degradansABD81F896CBM2-CBM10-GH5_4◯(Low)◯•YWW2(
      • Pires V.M.R.
      • Pereira P.M.M.
      • Brás J.L.A.
      • Correia M.
      • Cardoso V.
      • Bule P.
      • Alves V.D.
      • Najmudin S.
      • Venditto I.
      • Ferreira L.M.A.
      • Romão M.J.
      • Carvalho A.L.
      • Fontes C.M.G.A.
      • Prazeres D.M.
      Stability and ligand promiscuity of type A carbohydrate-binding modules are illustrated by the structure of Spirochaeta thermophila StCBM64C.
      ,
      • Watson B.J.
      • Zhang H.
      • Longmire A.G.
      • Young H.M.
      • Hutcheson S.W.
      Processive endoglucanases mediate degradation of cellulose by Saccharophagus degradans.
      )
      T. turneraeABS72374GH5_2-CBM5-CBM10-GH6_4◯(Low), chitin•▲[◯]YWW2(
      • Pires V.M.R.
      • Pereira P.M.M.
      • Brás J.L.A.
      • Correia M.
      • Cardoso V.
      • Bule P.
      • Alves V.D.
      • Najmudin S.
      • Venditto I.
      • Ferreira L.M.A.
      • Romão M.J.
      • Carvalho A.L.
      • Fontes C.M.G.A.
      • Prazeres D.M.
      Stability and ligand promiscuity of type A carbohydrate-binding modules are illustrated by the structure of Spirochaeta thermophila StCBM64C.
      ,
      • Ekborg N.A.
      • Morrill W.
      • Burgoyne A.M.
      • Li L.
      • Distel D.L.
      CelAB, a multifunctional cellulase encoded by Teredinibacter turnerae T7902T, a culturable symbiont isolated from the wood-boring marine bivalve Lyrodus pedicellatus.
      )
      S. degradans (SdGH5CBM10-2)ABD79918GH5_8-CBM10-CBM10-CBM10■(Kd = 0.4 g/l) [◯•]■□◆TWW1
      B. animalis (BaGH5CBM10)ACS46797GH5_8-CBM10■(Kd = 0.31 g/l)[□◯•]■□◆[◯]TWW1(
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      )
      VibrioBAA25188GH5_8-CBM10-CBM10ND-WW1(
      • Tamaru Y.
      • Araki T.
      • Morishita T.
      • Kimura T.
      • Sakka K.
      • Ohmiya K.
      Cloning, DNA sequencing, and expression of the β-1,4-mannanase gene from a marine bacterium, Vibrio sp. strain MA-138.
      )
      C. japonicusACE84673GH5_8-CBM10-CBM10◯□[■•]■◆[□•]AWW1(
      • Hogg D.
      • Pell G.
      • Dupree P.
      • Goubet F.
      • Marti S.M.
      • Gilbert H.J.
      The modular architecture of Cellvibrio japonicus mannanases in glycoside hydrolase families 5 and 26 points to differences in their role in mannan degradation.
      )
      VibrioBAA25188GH5_8-CBM10-CBM10ND-WW0(
      • Tamaru Y.
      • Araki T.
      • Morishita T.
      • Kimura T.
      • Sakka K.
      • Ohmiya K.
      Cloning, DNA sequencing, and expression of the β-1,4-mannanase gene from a marine bacterium, Vibrio sp. strain MA-138.
      )
      Cellulosi-microbiumAEE43708GH5_8-CBM10-CBM10◯□, chitosan, chitin■□AWW1(
      • Kim D.Y.
      • Ham S.J.
      • Lee H.J.
      • Cho H.Y.
      • Kim J.H.
      • Kim Y.J.
      • Shin D.H.
      • Rhee Y.H.
      • Son K.H.
      • Park H.Y.
      Cloning and characterization of a modular GH5 β-1,4-mannanase with high specific activity from the fibrolytic bacterium Cellulosimicrobium sp. strain HY-13.
      )
      Cellulosi-microbiumAEE43708GH5_8-CBM10-CBM10◯□, chitosan, chitin■□AWW1(
      • Kim D.Y.
      • Ham S.J.
      • Lee H.J.
      • Cho H.Y.
      • Kim J.H.
      • Kim Y.J.
      • Shin D.H.
      • Rhee Y.H.
      • Son K.H.
      • Park H.Y.
      Cloning and characterization of a modular GH5 β-1,4-mannanase with high specific activity from the fibrolytic bacterium Cellulosimicrobium sp. strain HY-13.
      )
      StreptomycesWP_030268297CBM10-GH134[□◯▲, chitin]■◆AWW1(
      • Jin Y.
      • Petricevic M.
      • John A.
      • Raich L.
      • Jenkins H.
      • Portela De Souza L.
      • Cuskin F.
      • Gilbert H.J.
      • Rovira C.
      • Goddard-Borger E.D.
      • Williams S.J.
      • Davies G.J.
      A β-mannanase with a lysozyme-like fold and a novel molecular catalytic mechanism.
      ,
      • Sakai K.
      • Kimoto S.
      • Shinzawa Y.
      • Minezawa M.
      • Suzuki K.
      • Jindou S.
      • Kato M.
      • Shimizu M.
      Characterization of pH-tolerant and thermostable GH 134 β-1,4-mannanase SsGH134 possessing carbohydrate binding module 10 from Streptomyces sp. NRRL B-24484.
      )
      StreptomycesADK91085GH5_8-CBM10ND■◆[•▲]-WW1(
      • Shi P.
      • Yuan T.
      • Zhao J.
      • Huang H.
      • Luo H.
      • Meng K.
      • Wang Y.
      • Yao B.
      Genetic and biochemical characterization of a protease-resistant mesophilic β-mannanase from Streptomyces sp. S27.
      )
      St. lividansAAA26710GH5_8-CBM10ND■□◆-WW1(
      • Arcand N.
      • Kluepfel D.
      • Paradis F.W.
      • Morosoli R.
      • Shareck F.
      β-Mannanase of Streptomyces lividans 66: Cloning and DNA sequence of the manA gene and characterization of the enzyme.
      ,
      • Kumagai Y.
      • Kawakami K.
      • Uraji M.
      • Hatanaka T.
      Binding of bivalent ions to actinomycete mannanase is accompanied by conformational change and is a key factor in its thermal stability.
      )
      ND, not determined.
      The order of the proteins follows the grouping seen in the phylogenetic tree (Fig. 3). When an enzyme has more than one CBM10, if possible, the information regarding carbohydrate binding is given for the particular CBM10 (underlined). Biochemical information is given using the following symbols for the polysaccharides: ■, soluble galactomannan; □, insoluble mannan (INM or β-mannan); ◆, soluble glucomannan; •, soluble cellulose (HEC, CMC); ◯, insoluble cellulose (Avicel, BMCC, PASC, filter paper, bagasse); ▲, other soluble polysaccharides (including xyloglycans, xylan, β-glucans, and lichenan). [ ] indicates that the given polysaccharide has been shown not to bind or being a substrate.
      CBM10 is a type A surface-binding module with a characteristic “planar” face containing aromatic residues capable of binding to regularly organized surfaces, for example, of the crystalline polysaccharides, cellulose and chitin (
      • Boraston A.B.
      • Bolam D.N.
      • Gilbert H.J.
      • Davies G.J.
      Carbohydrate-binding modules: Fine-tuning polysaccharide recognition.
      ). In the case of CBM10, a typical binding motif has three regularly spaced aromatic side chains, tyrosine, tryptophan, and tryptophan, initially identified in a CBM10 from xylanase A from Cellvibrio japonicus (CjGH10CBM10) (
      • Ponyi T.
      • Szabó L.
      • Nagy T.
      • Orosz L.
      • Simpson P.J.
      • Williamson M.P.
      • Gilbert H.J.
      Trp22, Trp24, and Tyr8 play a pivotal role in the binding of the family 10 cellulose-binding module from Pseudomonas xylanase A to insoluble ligands.
      ,
      • Raghothama S.
      • Simpson P.J.
      • Szabó L.
      • Nagy T.
      • Gilbert H.J.
      • Williamson M.P.
      Solution structure of the CBM10 cellulose binding module from Pseudomonas xylanase A.
      ) (Table 1). These residues were pivotal for binding to microcrystalline cellulose (Avicel). Furthermore, CjGH10CBM10 bound to bacterial microcrystalline cellulose (BMCC) and phosphoric acid–swollen cellulose (PASC) with a dissociation constant (Kd) of 0.2 to 0.3 μM but not to hydroxyethyl cellulose (HEC) (
      • Gill J.
      • Rixon J.E.
      • Bolam D.N.
      • McQueen-Mason S.
      • Simpson P.J.
      • Williamson M.P.
      • Hazlewood G.P.
      • Gilbert H.J.
      The type II and X cellulose-binding domains of Pseudomonas xylanase A potentiate catalytic activity against complex substrates by a common mechanism.
      ). It moreover bound differentially to plant cell walls depending upon cell type, tissue, and plant taxon (
      • Blake A.W.
      • McCartney L.
      • Flint J.E.
      • Bolam D.N.
      • Boraston A.B.
      • Gilbert H.J.
      • Knox J.P.
      Understanding the biological rationale for the diversity of cellulose-directed carbohydrate-binding modules in prokaryotic enzymes.
      ). CjGH10CBM10 is the only structure determined CBM10 (Protein Data Bank ID: 1E8R) (
      • Raghothama S.
      • Simpson P.J.
      • Szabó L.
      • Nagy T.
      • Gilbert H.J.
      • Williamson M.P.
      Solution structure of the CBM10 cellulose binding module from Pseudomonas xylanase A.
      ). Another cellulose-specific C. japonicus CBM10 (CjAA10CBM10) from LPMO (of auxiliary activities family 10; AA10A) showed a similar polysaccharide-binding preference and bound with Kd = 7.5 to 17.3 μM to disordered (PASC) as well as to crystalline (Avicel and BMCC) cellulose (
      • Crouch L.I.
      • Labourel A.
      • Walton P.H.
      • Davies G.J.
      • Gilbert H.J.
      The contribution of non-catalytic carbohydrate binding modules to the activity of lytic polysaccharide monooxygenases.
      ), which were also substrates of the LPMO (
      • Gardner J.G.
      • Crouch L.
      • Labourel A.
      • Forsberg Z.
      • Bukhman Y.V.
      • Vaaje-Kolstad G.
      • Gilbert H.J.
      • Keating D.H.
      Systems biology defines the biological significance of redox-active proteins during cellulose degradation in an aerobic bacterium.
      ). Finally, CBM10 (CjGH74CBM10) of the C. japonicus GH74 endo-xyloglucanase that also contains a CBM2, when analyzed as a GFP fusion did not bind to soluble tamarind seed xyloglucan and barley 1,3;1,4-β-glucan, both of which are substrates of the GH74 endo-xyloglucanase, or to HEC (Table 1), but bound with high affinity (Kd = 1.5 μM) to Avicel, which is not a substrate (
      • Attia M.
      • Stepper J.
      • Davies G.J.
      • Brumer H.
      Functional and structural characterization of a potent GH74 endo-xyloglucanase from the soil saprophyte Cellvibrio japonicus unravels the first step of xyloglucan degradation.
      ).
      The single CBM10 from a mannanase of GH family 5 subfamily 8 (GH5_8) from the probiotic bacterium Bifidobacterium animalis subsp. lactis Bl-04 (BaGH5CBM10) was able to bind to low-viscosity carob galactomannan (CGM-lv) with a Kd of 0.31 mg/ml but not to microcrystalline cellulose (Avicel), soluble cellulose (HEC), or insoluble ivory nut mannan (INM) (
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      ) (see Fig. 1 for polysaccharide structures) (Table 1). Notably, even though the single BaGH5CBM10 was critical for binding onto galactomannan, the enzyme kinetic parameters for this substrate did not change when the CBM10 was removed by truncation, whereas the truncated enzyme was not able to bind galactomannan (
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      ). The lack of binding onto the crystalline cellulose and INM was speculated to be due to substitution of an otherwise conserved tyrosine residue from the typical type A planar carbohydrate-binding motif (
      • Ponyi T.
      • Szabó L.
      • Nagy T.
      • Orosz L.
      • Simpson P.J.
      • Williamson M.P.
      • Gilbert H.J.
      Trp22, Trp24, and Tyr8 play a pivotal role in the binding of the family 10 cellulose-binding module from Pseudomonas xylanase A to insoluble ligands.
      ) and also by insertion of five residues, which expand a loop. These characteristics were seen for a phylogenetic subgroup distinct from the classical CBM10s (
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      ).
      A CBM10 from Clostridium cellulolyticum cellulase of GH9 adsorbed to PASC, Avicel, alkaline-pretreated sugarcane bagasse, and filter paper powder (
      • Duan C.-J.
      • Huang M.-Y.
      • Pang H.
      • Zhao J.
      • Wu C.-X.
      • Feng J.-X.
      Characterization of a novel theme C glycoside hydrolase family 9 cellulase and its CBM-chimeric enzymes.
      ). In addition, by comparing 40 type A CBMs (of CBM1, CBM2, CBM3, CBM10, CBM63, CBM64, CBM78, and CBM79) fused with GFP, three CBM10s in two multimodular enzymes from Teredinibacter turnerae (GH5_2-CBM5-CBM10-GH6_4) and the marine bacterium Saccharophagus degradans (CBM2-CBM10-GH5_4) as well as the CjGH10CBM10 mentioned previosuly showed low affinity for Sigmacell cellulose type 20 (20 μm particles) and chitosan compared with the other type A CBMs (
      • Pires V.M.R.
      • Pereira P.M.M.
      • Brás J.L.A.
      • Correia M.
      • Cardoso V.
      • Bule P.
      • Alves V.D.
      • Najmudin S.
      • Venditto I.
      • Ferreira L.M.A.
      • Romão M.J.
      • Carvalho A.L.
      • Fontes C.M.G.A.
      • Prazeres D.M.
      Stability and ligand promiscuity of type A carbohydrate-binding modules are illustrated by the structure of Spirochaeta thermophila StCBM64C.
      ). Finally, CBM10 in a GH134 β-1,4-mannanase from Streptomyces sp. NRRL both inhibited enzyme activity and conferred stability at neutral-to-alkaline pH as well as in the presence of organic solvents and detergents (
      • Sakai K.
      • Kimoto S.
      • Shinzawa Y.
      • Minezawa M.
      • Suzuki K.
      • Jindou S.
      • Kato M.
      • Shimizu M.
      Characterization of pH-tolerant and thermostable GH 134 β-1,4-mannanase SsGH134 possessing carbohydrate binding module 10 from Streptomyces sp. NRRL B-24484.
      ).
      Here, we shall dissect the function of individual CBM10s in a GH5_8 endo-β-mannanase from the marine bacterium and carbohydrate super degrader S. degradans by producing the CBM10 domains individually and together as GFP fusions and the full-length enzyme (SdGH5_8-CBM10x3) as well as its three CBM10 truncations (Fig. 2). The three consecutive CBM10s situated C-terminally to the CD are here named SdGH5CBM10-1, SdGH5CBM10-2, and SdGH5CBM10-3 and shown to belong to three distinct phylogenetic subgroups (Fig. 3). Polysaccharide binding was monitored by affinity gel electrophoresis (AGE) or pull-down assays. Enzyme activity was determined using CGM-lv and high-viscosity CGMs and glucomannan, a highly decorated very viscous guar gum (GG) galactomannan, insoluble crystalline INM, and microcrystalline cellulose (Avicel). The three SdGH5CBM10s showed distinctly different polysaccharide-binding profiles and impact on the enzyme activity and substrate specificity.
      Figure thumbnail gr2
      Figure 2Domain architecture of the native Saccharophagus degradans protein (GenBank accession no.: ABD79918) and the different recombinant full-length and truncated enzyme forms and CBM10 domains produced and characterized. The residues at domain borders are numbered.

      Results and discussion

      Phylogenetic analysis of CBM10s

      The phylogenetic analysis of CBM10 reveals a number of different subgroups (Fig. 3), which have characteristic features with regard to (i) the presence of three aromatic residues (tyrosine, tryptophan, and tryptophan) seen in a canonical type A CBM10 (
      • Ponyi T.
      • Szabó L.
      • Nagy T.
      • Orosz L.
      • Simpson P.J.
      • Williamson M.P.
      • Gilbert H.J.
      Trp22, Trp24, and Tyr8 play a pivotal role in the binding of the family 10 cellulose-binding module from Pseudomonas xylanase A to insoluble ligands.
      ), (ii) the presence of cysteine residues potentially forming one of the two observed disulphide bonds, and (iii) the presence of a five to six residues insert between the tyrosine and the two tryptophans of the planar-binding motif (Figs. 4 and S1). For example, the BaGH5CBM10 in the Bifidobacterium mannanase, which was needed for binding to galactomannan, but did not confer binding to cellulose or INM, had the following distinct features: (i) substitution of the first tyrosine (to a threonine) of the three aromatic residues critical for Avicel binding as demonstrated for C. japonicus CjGH10CBM10 (
      • Ponyi T.
      • Szabó L.
      • Nagy T.
      • Orosz L.
      • Simpson P.J.
      • Williamson M.P.
      • Gilbert H.J.
      Trp22, Trp24, and Tyr8 play a pivotal role in the binding of the family 10 cellulose-binding module from Pseudomonas xylanase A to insoluble ligands.
      ), (ii) lack of two cysteines that form one of the two disulphide bridges, and (iii) a five-residue long insertion (Figs. 4 and S1).
      Figure thumbnail gr3
      Figure 3Phylogenetic tree including CBM10(s) isolated from full-length enzyme sequences (i.e., without CD and other modules). CBM10s originating from characterized enzymes are labeled (see for detailed information). See for the multiple sequence alignment used to generate the phylogenetic tree.
      Figure thumbnail gr4
      Figure 4Comparison of characterized CBM10s. A, structure-based multiple sequence alignment including characterized CBM10s. Cysteines involved in disulphide bridges in the CjGH10CBM10 structure (Protein Data Bank ID: 1E8R) are denoted by a “C,” whereas binding residues of CjGH10CBM10 (Y8, W22, and W24) are indicated by asterisks. The insertion differentiating the subgroup containing SdGH5CBM10-2 from the other regions of the phylogenetic tree is indicated by a dotted line. BD, superimposition of the structure of CjGH10CBM10 (orange; Protein Data Bank ID: 1E8R) and homology models of the three SdGH5CBM10s (B. SdGH5CBM10-1, green; C. SdGH5CBM10-2, blue; D. SdGH5CBM10-3, purple) with binding residues shown as sticks. The numbering is according to CjGH10CBM10, and the equivalent residue of the SdGH5CBM10s is given. The region of the insertion in SdGH5CBM10-2 (C) is encircled.
      To further investigate functional characteristics of CBM10s from different regions of the phylogenetic tree, we selected a putative GH5_8 mannanase from the marine bacterium and carbohydrate super degrader S. degradans having three C-terminal CBM10s (SdGH5_8-CBM10x3; GenBank accession no.: ABD79918) (Fig. 2). These CBM10s were hypothesized to display distinct functions as they belong to three different phylogenetic subgroups (Fig. 3).
      Notably, the innermost SdGH5CBM10-1 linked to the CD of the S. degradans mannanase (Fig. 2) was found in the phylogenetic tree at a distance of characterized CBM10s. The middle domain, SdGH5CBM10-2, however belonged to the same subgroup as BaGH5CBM10 from the Bifidobacterium mannanase, although they were not closely related. The C-terminal SdGH5CBM10-3 belongs to the same subgroup as the previously well-characterized C. japonicus CBM10s from xylanase A of GH10 (CjGH10CBM10) and of the GH74 endoglucanase active on xyloglucan and mixed-linked 1,3;1,4-β-glucan (CjGH74CBM10) (
      • Attia M.
      • Stepper J.
      • Davies G.J.
      • Brumer H.
      Functional and structural characterization of a potent GH74 endo-xyloglucanase from the soil saprophyte Cellvibrio japonicus unravels the first step of xyloglucan degradation.
      ) (Fig. 3).

      Effect of CBM10s on specificity and activity of GH5_8 endo-β-mannanase

      The full-length enzyme, SdGH5_8-CBM10x3, and its three stepwisely truncated forms lacking one, two, or all three CBM10s were recombinantly produced in Escherichia coli (Fig. 2). For production of the individual CBM10s and all three CBM10s together, the GFP fusion strategy reported previously to enable production of CBM10 from the GH74 endo-xyloglucanase of C. japonicus (CjGH74CBM10) (
      • Attia M.
      • Stepper J.
      • Davies G.J.
      • Brumer H.
      Functional and structural characterization of a potent GH74 endo-xyloglucanase from the soil saprophyte Cellvibrio japonicus unravels the first step of xyloglucan degradation.
      ) was adopted (Fig. 2). All recombinant enzyme forms and the CBM10–GFP fusions were obtained in reasonably good yields of 0.7 to 7.5 mg/g cells after purification. Notably, some of the multimodular forms, in particular the full-length SdGH5_8-CBM10x3, undergo significant proteolytic cleavage in the 21- to 30-residue long serine–glycine rich linker regions during production and purification, which is reflected in the lower yields.
      The full-length SdGH5_8-CBM10x3 and its three truncated forms were active on all tested mannan polysaccharides but not on microcrystalline cellulose (Avicel) (Table 2). The substrate specificity of SdGH5_8-CBM10x3 is similar to that of other characterized GH5_8 mannanases but with specific activities at the higher end (
      • Hogg D.
      • Pell G.
      • Dupree P.
      • Goubet F.
      • Marti S.M.
      • Gilbert H.J.
      The modular architecture of Cellvibrio japonicus mannanases in glycoside hydrolase families 5 and 26 points to differences in their role in mannan degradation.
      ,
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      ). Removal of the C-terminal CBM10 (SdGH5CBM10-3) did not essentially affect the specific activity toward the good substrates, that is, CGM-lv and high-viscosity CGM (locust bean gum) and konjac glucomannan (KGM) (Table 2; Fig. 1). Remarkably, however, removal of the C-terminal SdGH5CBM10-3 resulted in doubling, respectively, almost quadrupling of the specific activity (Table 2) on the poor substrates; highly viscous GG galactomannan with on average every second mannosyl backbone unit substituted by galactose (as opposed to one out of four in CGM) and the insoluble crystalline INM (Fig. 1). Removal of both the middle (SdGH5CBM10-2) and the C-terminal (SdGH5CBM10-3) domains reduced activity by 36% to 40% for CGM-lv and high-viscosity CGMs and maintained activity for KGM compared with the full-length enzyme (Table 2). Remarkably, the activity of SdGH5_8-CBM10x1 with only the innermost CBM10, increased fourfold and ninefold on the poor substrates GG and INM, respectively, compared with the full-length enzyme (Table 2). Finally, removing all three CBM10s essentially recovered activity of the full-length enzyme on CGMs and KGM. This finding resembles the lack of effect on activity for CGM-lv by removing the single C-terminal CBM10 from the Bifidobacterium GH5 mannanase (
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      ). Again on the poor substrates GG and INM, the CD alone (SdGH5_8) showed twofold and sixfold higher activity than full-length enzyme. Overall, this suggests that the presence of only SdGH5CBM10-1 slightly hinders CGMs from interacting fully effectively with the CD. The presence of the C-terminal and middle CBM10 seems to prevent optimal productive interaction between the CD and the poor substrates INM and the highly substituted guar GG as reflected in the CD alone (SdGH5_8) and the enzyme form with only the first CBM10 (SdGH5_8-CBM10x1) having higher specific activities at low enzyme concentrations.
      Table 2Specific activity of full-length SdGH5_8-CBM10x3 and the three truncated forms on soluble (CGM-lv, high-viscosity CGM, KGM, and GG) and insoluble (INM) mannose-containing polysaccharides and the insoluble Avicel (see Fig. 1 for polysaccharide structures)
      SubstratesUnitsSdGH5_8-CBM10x3SdGH5_8-CBM10x2SdGH5_8-CBM10x1SdGH5_8 (CD alone)
      CGM-lvU/mg1972 ± 802285 ± 1501539 ± 882906 ± 53
      (μmol/s)/μmol protein
      Specific activity calculated based on molecular weight.
      1729 ± 701763 ± 1161046 ± 601690 ± 31
      Relative (%)
      Specific activity relative to full-length SdGH5_8-CBM10x3 calculated based on the specific activity given in (μmol/s)/μmol protein.
      1001026098
      CGM (high viscosity)U/mg2212 ± 782755 ± 1091814 ± 2923151 ± 304
      (μmol/s)/μmol protein
      Specific activity calculated based on molecular weight.
      1939 ± 692126 ± 841234 ± 1981833 ± 177
      Relative (%)
      Specific activity relative to full-length SdGH5_8-CBM10x3 calculated based on the specific activity given in (μmol/s)/μmol protein.
      1001106495
      KGMU/mg3544 ± 1103617 ± 1125136 ± 1284556 ± 108
      (μmol/s)/μmol protein
      Specific activity calculated based on molecular weight.
      3107 ± 972791 ± 873493 ± 872650 ± 63
      Relative (%)
      Specific activity relative to full-length SdGH5_8-CBM10x3 calculated based on the specific activity given in (μmol/s)/μmol protein.
      1009011285
      GGU/mg40 ± 824 ± 1200 ± 33108 ± 25
      (μmol/s)/μmol protein
      Specific activity calculated based on molecular weight.
      35 ± 719 ± 1136 ± 2363 ± 14
      Relative (%)
      Specific activity relative to full-length SdGH5_8-CBM10x3 calculated based on the specific activity given in (μmol/s)/μmol protein.
      10054389180
      INMU/mg9 ± 138 ± 3106 ± 681 ± 7
      (μmol/s)/μmol protein
      Specific activity calculated based on molecular weight.
      8 ± 129 ± 272 ± 447 ± 4
      Relative (%)
      Specific activity relative to full-length SdGH5_8-CBM10x3 calculated based on the specific activity given in (μmol/s)/μmol protein.
      100363900588
      AvicelU/mgND
      No activity detected.
      ND
      No activity detected.
      ND
      No activity detected.
      ND
      No activity detected.
      ND, not determined.
      a Specific activity calculated based on molecular weight.
      b Specific activity relative to full-length SdGH5_8-CBM10x3 calculated based on the specific activity given in (μmol/s)/μmol protein.
      c No activity detected.
      The kinetic parameters for full-length SdGH5_8-CBM10x3 and truncated forms toward CGM-lv are in agreement with the findings on specific activity (Table 3). Thus, Km of SdGH5_8-CBM10x1 was threefold higher than of the full-length enzyme, whereas the turnover number increased by 1.8-fold, resulting in 36% lower catalytic efficiency (kcat/Km). Notably, the catalytic efficiency of CD alone is doubled and 1.4-fold higher than of the SdGH5_8-CBM10x1 and full-length enzyme, respectively. The Km of CD alone is low and similar to that of full-length SdGH5_8-CBM10x3, whereas the turnover number is 68% higher. The lowest turnover number is seen for full-length enzyme, whereas SdGH5_8-CBM10x1 has the highest turnover number (Table 3).
      Table 3Kinetic parameters for full-length SdGH5_8-CBM10x3 and truncated forms on CGM-lv
      Enzyme formkcat (s−1)Km (mg/ml)kcat/Km (ml [mg s]−1)
      SdGH5_8-CBM10x32333 ± 552.1 ± 0.11096 ± 71
      SdGH5_8-CBM10x23589 ± 2784.2 ± 0.7855 ± 149
      SdGH5_8-CBM10x14253 ± 3376.1 ± 0.9701 ± 114
      SdGH5_83440 ± 752.4 ± 0.11413 ± 82
      A negative effect on activity of the presence of a CBM10 was previously observed for the GH134 β-1,4-mannanase that contains a single CBM10 and for which the catalytic efficiency of the CD alone on galactomannan was 2.6-fold higher than of the full-length enzyme, whereas Km was 1.4-fold lower (
      • Sakai K.
      • Kimoto S.
      • Shinzawa Y.
      • Minezawa M.
      • Suzuki K.
      • Jindou S.
      • Kato M.
      • Shimizu M.
      Characterization of pH-tolerant and thermostable GH 134 β-1,4-mannanase SsGH134 possessing carbohydrate binding module 10 from Streptomyces sp. NRRL B-24484.
      ), suggesting that the CBM10 is associated with nonproductive binding. This CBM10 did not confer the binding to microcrystalline cellulose, β-mannan, chitin, or xylan as the CD alone can bind to these polysaccharides (
      • Sakai K.
      • Kimoto S.
      • Shinzawa Y.
      • Minezawa M.
      • Suzuki K.
      • Jindou S.
      • Kato M.
      • Shimizu M.
      Characterization of pH-tolerant and thermostable GH 134 β-1,4-mannanase SsGH134 possessing carbohydrate binding module 10 from Streptomyces sp. NRRL B-24484.
      ). Although it is not generally seen, there are examples from other CBM families where CBMs have an inhibitory effect on enzyme activity. Thus, removal of a cellulose-binding CBM3 from an endo-β-1,4-glucanase from Bacillus subtilis resulted in both increased kcat and catalytic efficiency and also improved thermal stability (
      • Wang Y.
      • Yuan H.
      • Wang J.
      • Yu Z.
      Truncation of the cellulose binding domain improved thermal stability of endo-β-1,4-glucanase from Bacillus subtilis JA18.
      ). Furthermore, when a CBM5 was removed from a Pectobacterium chrystantemi endoglucanase, Km and Vmax were reduced by 4.5-fold and increased by 2.5-fold, respectively (
      • Park S.R.
      • Cho S.J.
      • Kim M.K.
      • Ryu S.K.
      • Lim W.J.
      • An C.L.
      • Hong S.Y.
      • Kim J.H.
      • Kim H.
      • Yun H.D.
      Activity enhancement of Cel5Z from Pectobacterium chrysanthemi PY35 by removing C-terminal region.
      ).
      The kinetics analysis on the insoluble crystalline INM showed the C-terminal CBM10-3 of full-length SdGH5_8-CBM10x3 from S. degradans to be needed for maximum activity, whereas the three stepwisely CBM10 truncated enzyme forms showed similarly reduced activity relative to the full-length enzyme (Fig. 5A). In the case of GG, the full-length enzyme showed the highest activity, whereas the CD alone in fact was the least active form (Fig. 5B). These results on the highly viscous and densely decorated guar GG indicated that the C-terminal CBM10-3 together with CBM10-1 play a role for activity at high-substrate concentrations. Thus, removal of CBM10-3 resulted in some reduction in activity, whereas removal of CBM10-2 as well did not make a difference, as the SdGH5_8-CBM10x1 and SdGH5_8-CBM10x2 showed the same activity. But the removal of CBM10-1 led to a further reduced activity. Although full substrate saturation was not possible, calculations applying the Michaelis–Menten equation on activity of the full-length SdGH5_8-CBM10x3 on INM indicated Km = 35 ± 14 mg/ml and kcat = 186 ± 53 s−1 (kcat/Km = 5.3 ml/[mg · s]). The Km is 1.3-fold higher as compared with a GH5_8 mannanase with one CBM10 from Streptomyces lividans (Km = 26.5 ± 11.5 mg/ml), whereas the kcat of that enzyme was 69 ± 1.2 s−1 resulting in a catalytic efficiency of 2.6 ml/(mg s) (
      • Kumagai Y.
      • Kawakami K.
      • Uraji M.
      • Hatanaka T.
      Binding of bivalent ions to actinomycete mannanase is accompanied by conformational change and is a key factor in its thermal stability.
      ); hence, SdGH5_8-CBM10x3 is twofold more efficient. Furthermore, the kinetics on INM of the S. lividans mannanase was not affected by removal of the CBM10 (
      • Kumagai Y.
      • Kawakami K.
      • Uraji M.
      • Hatanaka T.
      Binding of bivalent ions to actinomycete mannanase is accompanied by conformational change and is a key factor in its thermal stability.
      ). This CBM10 is in a phylogenetic subgroup with SdGH5CBM10-2; hence, the lack of effect on activity on ivory nut is in line with our findings (Fig. 5A).
      Figure thumbnail gr5
      Figure 5Kinetics of the full-length enzyme SdGH5_8-CBM10x3. The three stepwise truncated forms on ivory nut mannan (INM) (A) and guar gum galactomannan (GG) (B). SdGH5_8-CBM10x3 (■), SdGH5_8-CBM10x2 (•), SdGH5_8-CBM10x1 (▲), and SdGH5_8 (◆). The Michaelis–Menten fit to the SdGH5_8-CBM10x3 data is shown as dashed line in A.
      To understand the reason behind the differences observed between the results of the enzyme kinetics on INM and guar GG (Fig. 5) and the specific activity analyses (Table 2), one should have the differences in assay setup in mind. The specific activities were determined as is typical for mannanases at readily manageable low substrate concentrations to avoid highly viscous solutions and maintained (low) enzyme concentration with end point measurement implying substrate-dependent incubation time (
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      ) (Table 2). The kinetics analysis contrarily was done at up to fourfold higher substrate and 10-fold to 30-fold higher enzyme concentration. The action on the insoluble INM should probably be considered as heterogenous catalysis as found in case of cellulases acting on crystalline cellulose, characterized by the presence of different categories of binding sites and productive enzyme attack sites (
      • Kari J.
      • Andersen M.
      • Borch K.
      • Westh P.
      An inverse Michaelis-Menten approach for interfacial enzyme kinetics.
      ,
      • Kari J.
      • Olsen J.P.
      • Jensen K.
      • Badino S.F.
      • Krogh K.B.R.M.
      • Borch K.
      • Westh P.
      Sabatier principle for interfacial (heterogeneous) enzyme catalysis.
      ). A number of crystalline forms are reported for INM (
      • Hori R.
      • Sugiyama J.
      • Wada M.
      The thermal expansion of mannan I obtained from ivory nuts.
      ,
      • Heux L.
      • Hägglund P.
      • Putaux J.L.
      • Chanzy H.
      Structural aspects in semicrystalline samples of the mannan II family.
      ), but details on actual mannanase degradation of these forms have not been reported. It is possible that most strongly binding, albeit not productively binding, substrate sites are interacting with full-length enzyme via the C-terminal CBM10-3, which deprives the CD access to susceptible substrate bonds. At the higher enzyme concentration used for the kinetics analysis (Fig. 5A), this is in part overcome. For the other poor substrate guar GG, the substrate is assumed to contain enzyme target bonds placed in different structural contexts and having different preferences for the productive enzyme attack to occur (
      • McCleary B.V.
      • Clark A.H.
      • Dea I.C.M.
      • Rees D.A.
      The fine structures of carob and guar galactomannans.
      ). Although a GH26 mannanase from B. subtilis is reported with preference for the guar GG over less substituted galactomannans (
      • Liu H.X.
      • Gong J.S.
      • Li H.
      • Lu Z.M.
      • Li H.
      • Qian J.Y.
      • Xu Z.H.
      • Shi J.S.
      Biochemical characterization and cloning of an endo-1,4-β-mannanase from Bacillus subtilis YH12 with unusually broad substrate profile.
      ), this is not common, and we assume that initial hydrolysis on GG is slow and may open up for more accessible substrate regions possibly influenced by the fine structural preferences of the CBM10-3 and the CD. This type of action may be observed as a degradation involving apparent cooperative events also indicated by the sigmoidal curvature of the kinetic data (Fig. 5B).

      Polysaccharide-binding analysis

      The affinities for soluble CGM-lv, HEC, insoluble microcrystalline cellulose (Avicel), and INM were assessed by retardation in AGE and pull-down assays, respectively. Neither the different enzyme forms nor the CBM10s individually or all three CBM10s together were able to bind HEC in AGE. By contrast, both full-length enzyme SdGH5_8-CBM10x3 and the three CBM10s together (SdGH5CBM10x3) showed high affinity for CGM-lv (Table 4). Although each of the three SdGH5CBM10s when tested individually were able to bind CGM-lv with reasonable affinity, the strong binding depended on the C-terminal domain (SdGH5CBM10-3) but was also clearly a combined effect of all the three CBM10s. The full-length enzyme and the three CBM10s together thus bound very strongly with Kd <0.13 mg/ml, whereas neither of the truncated enzyme forms lacking one, two, or all three CBM10 domains showed significant retardation in AGE. Both the C-terminal CBM10-3 and the middle CBM10-2 domains, however, bound reasonably well to galactomannan, the data suggesting that they together secures the full-length enzyme binding. This is opposed to the analysis of the innermost SdGH5CBM10-1_GFP, which showed ninefold to 18-fold reduced affinity compared with the other two CBM10s having Kd of 0.4 and 0.2 mg/ml, respectively (Table 4). The second domain (SdGH5CBM10-2_GFP), which has an insert of five amino acid residues and lacks a potentially stabilizing disulphide bridge similar to BaGH5CBM10 from the Bifidobacterium mannanase (Figs. 4 and S1), had a Kd similar to that of BaGH5CBM10 (Kd = 0.3 g l−1) (
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      ) that it also resembles phylogenetically (Fig. 3). But surprisingly, SdGH5CBM10-2_GFP had ninefold higher affinity than the SdGH5CBM10-1_GFP (Table 4). A possible explanation is that SdGH5CBM10-1, although it probably contains the two stabilizing disulphide bridges, has a serine replacing the first of the three aromatic binding residues and a tyrosine replacing the tryptophan at the position of the third binding site residue (Table 1; Fig. 4). It belongs to a phylogenetic region shared by few CBM10s. Notably, the LPMO CjAA10CBM10 also has the third tryptophan replaced (by asparagine) but still has good affinity for insoluble cellulose; however, it does have the conserved tyrosine as the first of the three binding residues (Table 1; Fig. 4).
      Table 4Binding to CGM-lv analyzed by AGE
      Protein formKd (mg/ml)
      SdGH5_8-CBM10x3<0.125
      Saturation was reached already at the lowest in-gel galactomannan concentration.
      SdGH5_8-CBM10x2No binding
      SdGH5_8-CBM10x1No binding
      SdGH5_8 (no CBMs)No binding
      SdGH5CBM10x3_GFP<0.125
      Saturation was reached already at the lowest in-gel galactomannan concentration.
      SdGH5CBM10-1_GFP3.7
      SdGH5CBM10-2_GFP0.4
      SdGH5CBM10-3_GFP0.2
      GFPNo binding
      a Saturation was reached already at the lowest in-gel galactomannan concentration.
      A qualitative pull down by insoluble INM of the three individual SdGH5CBM10 and the combined three domain GFP fusions showed that SdGH5CBM10-3_GFP and SdGH5CBM10x3_GFP interacted strongly with INM, whereas the amount of protein bound in the case of both SdGH5CBM10-1_GFP and SdGH5CBM10-2_GFP was in the same range as the amount of GFP alone (Fig. 6). This is in line with the kinetic assay showing that when the C-terminal CBM10-3 was removed, the activity was reduced (Fig. 5A). Moreover, the qualitative pull down by Avicel showed only weak interaction with SdGH5CBM10-1_GFP and SdGH5CBM10-2_GFP, whereas the three CBM10s together (SdGH5CBM10x3_GFP) and the C-terminal domain SdGH5CBM10-3_GFP bound to Avicel (Fig. 6). Quantitative binding isotherms gave Kd for SdGH5CBM10x3_GFP and SdGH5CBM10-3_GFP of 1.04 ± 0.16 and 1.48 ± 0.12 μM, respectively (Fig. 7). The maximum amount of bound SdGH5CBM10x3_GFP and SdGH5CBM10-3_GFP was 0.13 ± 0.00 and 0.27 ± 0.04 μmol/g Avicel, respectively. The binding of SdGH5CBM10-1_GFP and SdGH5CBM10-2_GFP (and GFP alone) was too weak to obtain a Kd. Hence, SdGH5CBM10-3 is responsible for Avicel binding, in accordance with findings for other members of the same phylogenetic subgroup (Fig. 3), which had essentially the same affinity as reported also for CjGH74CBM10-GFP of GH74 endo-xyloglucanase (
      • Attia M.
      • Stepper J.
      • Davies G.J.
      • Brumer H.
      Functional and structural characterization of a potent GH74 endo-xyloglucanase from the soil saprophyte Cellvibrio japonicus unravels the first step of xyloglucan degradation.
      ). Notably, the first and second CBM10s in the S. degradens mannanase belong to different phylogenetic subgroups and lacked two and one, respectively, of the three residues in the planar aromatic binding motif (
      • Ponyi T.
      • Szabó L.
      • Nagy T.
      • Orosz L.
      • Simpson P.J.
      • Williamson M.P.
      • Gilbert H.J.
      Trp22, Trp24, and Tyr8 play a pivotal role in the binding of the family 10 cellulose-binding module from Pseudomonas xylanase A to insoluble ligands.
      ) (Figs. 3 and 4). Previously, it was suggested that the absence of a potentially stabilizing disulphide bond explained a similar lack of Avicel binding for the Bifidobacterium full-length mannanase with one CBM10 (
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      ), but this cannot explain why SdGH5CBM10-1 is not binding to Avicel, as it has the four cysteines predicted to form the two stabilizing disulphides (Fig. 4). While SdGH5CBM10-3 has a Kd for Avicel in the same range as the rest of the characterized CBM10s from the same subgroup in the phylogenetic tree (Fig. 3), it differs by showing ability to bind soluble galactomannan in distinction also from the two other CBM10s from this subgroup tested for galactomannan binding (
      • Hogg D.
      • Pell G.
      • Dupree P.
      • Goubet F.
      • Marti S.M.
      • Gilbert H.J.
      The modular architecture of Cellvibrio japonicus mannanases in glycoside hydrolase families 5 and 26 points to differences in their role in mannan degradation.
      ).
      Figure thumbnail gr6
      Figure 6Qualitative pull-down assay with SdGH5CBM10–GFP fusions of each of the three domains SdGH5CBM10-1, SdGH5CBM10-2, and SdGH5CBM10-3 and all three domains combined SdGH5CBM10x3_GFP by the insoluble polysaccharides microcrystalline cellulose (Avicel) and ivory nut mannan (INM), respectively. SDS-PAGE of samples from supernatants (SN) and pellets (P) as well as the protein stocks used for the pull-down assay. GFP alone is included as a control (right panel). Molecular mass values (kDa) of marker proteins are indicated.
      Figure thumbnail gr7
      Figure 7Quantitative pull-down assay with microcrystalline cellulose (Avicel). Binding isotherms of the C-terminal domain, SdGH5CBM10-3_GFP (A) and all three domains together, SdGH5CBM10x3_GFP (B). The filled and open symbols represent replicates from different days. No significant binding was observed of the first and the middle domains, SdGH5CBM10-1_GFP, SdGH5CBM10-2_GFP, or of GFP alone (data not shown).

      Functional diversity of CBM10s

      Based on the present results, the C-terminal CBM10, SdGH5CBM10-3, is proposed to anchor the secreted enzyme to mannans and to crystalline cellulose, which seems valuable for a free-living marine bacterium as S. degradans. S. degradans is a so-called super degrader and can utilize complex polysaccharides of algal, higher plant, fungal, and animal origin as sole carbon and energy source (
      • Hutcheson S.W.
      • Zhang H.
      • Suvorov M.
      Carbohydrase systems of Saccharophagus degradans degrading marine complex polysaccharides.
      ). Avicel is neither a substrate for SdGH5_8-CBM10x3 not for the C. japonicus GH74 endo-xyloglucanase although the CBM10s of both enzymes bind well to Avicel (
      • Attia M.
      • Stepper J.
      • Davies G.J.
      • Brumer H.
      Functional and structural characterization of a potent GH74 endo-xyloglucanase from the soil saprophyte Cellvibrio japonicus unravels the first step of xyloglucan degradation.
      ). It is well known that CBMs can interact with polysaccharides like cellulose in the plant cell wall or other biomass, rather than soluble branched polysaccharides such as galactomannan (
      • Hervé C.
      • Rogowski A.
      • Blake A.W.
      • Marcus S.E.
      • Gilbert H.J.
      • Knox J.P.
      Carbohydrate-binding modules promote the enzymatic deconstruction of intact plant cell walls by targeting and proximity effects.
      ). However, SdGH5CBM10-1 and SdGH5CBM10-2 seem to have evolved to become dynamic domains securing flexible searching for an enzyme attack site on a substrate surface or gel for the CD, while being firmly anchored via the terminal SdGH5CBM10-3. Notably, in particular, SdGH5CBM10-2 has reasonable affinity for galactomannan and may potentiate binding to the actual substrate polysaccharide molecule in a structurally complex plant cell wall environment. CBM10 family thus includes members with promiscuous functions binding to a diversity of polysaccharides, such as soluble galactomannan, insoluble crystalline mannans, and cellulose with good affinity (e.g. SdGH5CBM10-3), whereas some bind soluble polysaccharide substrates with good affinity (e.g. SdGH5CBM10-2) or with low affinity (e.g. SdGH5CBM10-1). Members of the same phylogenetic subgroup as SdGH5CBM10-3-like CBM10s probably anchor full-length enzymes firmly to substrate or spatially adjacent polysaccharides, whereas CBM10s from phylogenetically different subgroups may still show binding to certain polysaccharides and enable the CD to flexibly survey the vicinity for susceptible glycosidic bonds while yet others may be protein domains serving as linkers or spacers not engaged in direct binding onto a crystalline cellulose via a classical CBM type A platform. In nature, multimodular enzymes characteristically participate in heterogenous catalysis on insoluble substrates to which firm binding is mandatory. However, binding is not the same as productive binding, and the density of functional and nonfunctional binding attack sites depends on the material and enzyme as illustrated in details for cellulose degradation by cellulases (
      • Kari J.
      • Andersen M.
      • Borch K.
      • Westh P.
      An inverse Michaelis-Menten approach for interfacial enzyme kinetics.
      ,
      • Kari J.
      • Olsen J.P.
      • Jensen K.
      • Badino S.F.
      • Krogh K.B.R.M.
      • Borch K.
      • Westh P.
      Sabatier principle for interfacial (heterogeneous) enzyme catalysis.
      ).

      Conclusion and perspectives

      The multimodular endo-β-mannanase of family GH5_8 secreted by S. degradans is able to bind to polysaccharide substrates and substrate containing biological material by aid of three consecutive C-terminal CBM10 domains. Clearly, the most C-terminal CBM10-3 secures the attachment both to crystalline cellulose (Avicel), galactomannan, and also insoluble seed storage mannan. However, the middle domain CBM10-2 shows also affinity with a 10-fold lower Kd than the Km for the good substrate galactomannan and may facilitate its interactions with the CD although the truncated enzyme lacking the C-terminal CBM10 lost ability to bind to the galactomannan. The CBM10-2 therefore may also serve as a linker that provides flexibility in polysaccharide interaction for the C-terminal domain and hence for the CD to interact productively with substrate. The role of the innermost CBM10-1 is more enigmatic, but at low enzyme and substrate concentrations, it was able to secure highest activity of the CD toward the two poor substrates (INM and guar GG) also relative to the activity of the CD alone or with two, respectively, three CBM10s. The CBM10-1 is in a small separate phylogenetic subgroup for which a possible hallmark and biological role, beyond this mentioned stabilization, remains to be identified.

      Experimental procedures

      Bioinformatics

      CBM10 sequences were retrieved from the CAZy database (www.cazy.org; (
      • Lombard V.
      • Golaconda Ramulu H.
      • Drula E.
      • Coutinho P.M.
      • Henrissat B.
      The carbohydrate-active enzymes database (CAZy) in 2013.
      )) and subjected to multiple sequence alignment using MUltiple Sequence Comparison by Log-Expectation tool (
      • Edgar R.C.
      Muscle: Multiple sequence alignment with high accuracy and high throughput.
      ). A phylogenetic tree was constructed using BIONJ (
      • Gascuel O.
      Bionj: An improved version of the NJ algorithm based on a simple model of sequence data.
      ) and visualized using Dendroscope, version 3.6.3 (https://uni-tuebingen.de/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/algorithms-in-bioinformatics/software/dendroscope/) (
      • Huson D.H.
      • Scornavacca C.
      Dendroscope 3: An interactive tool for rooted phylogenetic trees and networks.
      ). A structure-based alignment of selected characterized CBM10s was generated using the PROMALS3D Web server (
      • Pei J.
      • Tang M.
      • Grishin N.V.
      PROMALS3D web server for accurate multiple protein sequence and structure alignments.
      ). Homology models of the three SdGH5CBM10s were made using HHpred (https://toolkit.tuebingen.mpg.de/tools/hhpred) (
      • Zimmermann L.
      • Stephens A.
      • Nam S.-Z.
      • Rau D.
      • Kübler J.
      • Lozajic M.
      • Gabler F.
      • Söding J.
      • Lupas A.N.
      • Alva V.
      A completely reimplemented MPI bioinformatics toolkit with a new HHpred server at its core.
      ) with the structure of CjGH10CBM10 (Protein Data Bank ID: 1E8R) as template.

      Carbohydrates

      CGM-lv, INM, and KGM were from Megazyme; high-viscosity CGM (locust bean gum), GG, microcrystalline cellulose (Avicel), and HEC were from Sigma–Aldrich. See Figure 1 for the polysaccharide structures.

      Gene constructs

      The genes encoding the S. degradans β-mannanase SdGH5_8-CBM10x3 (GenBank accession no.: ABD79918) without the predicted signal peptide (amino acids 1–21) and the three CBM10s SdGH5CBM10x3 fused with GFP, and GFP (AGT98536) were purchased (GeneArt; Thermo Fisher Scientific), and cloned into the pET28a(+) vector using the XhoI and NheI restriction sites resulting in an N-terminal cleavable His tag. The three C-terminally truncated forms SdGH5_8-CBM10x2, SdGH5_8-CBM10x1, and SdGH5_8 (Fig. 2) were constructed by introducing stop codons using Q5 Site-Directed Mutagenesis Kit (New England Biolabs), pET28a-SdGH5_8-CBM10x3 as template and mutagenesis primers (Table S1) designed according to the manufacturers' instructions. Each of the three SdGH5CBM10s were obtained as SdGH5CBM10 to 1_GFP, SdGH5CBM10 to 2_GFP, and SdGH5CBM10-3_GFP by mutagenesis using Q5 Site-Directed Mutagenesis Kit (New England Biolabs), pET28a-SdGH5CBM10x3_GFP as template and mutagenesis primers. To get SdGH5CBM10-1_GFP, both SdGH5CBM10-2 and SdGH5CBM10-3 were deleted in one step. A one-step deletion of SdGH5CBM10-1 and SdGH5CBM10-2 was done to get SdGH5CBM10-3-GFP. Finally, SdGH5CBM10-2_GFP was obtained in two steps; first deleting CBM10-1 and then CBM10-3. The gene constructs were confirmed by sequencing.

      Protein production

      All full-length and variant-encoding plasmids were transformed into E. coli. BL21, screened on LB agar including 50 μg/ml kanamycin, and starter cultures (10 ml) were made by inoculating LB medium, 50 μg/ml kanamycin with a single colony and incubated (37 °C, overnight), which are used to inoculate 750 ml LB medium containing 10 mM glucose and 50 μg/ml kanamycin in shake flasks. Cultures were propagated (30 °C, 160 rpm) to an absorbance of 0.6 at 600 nm followed by decreasing the temperature to 16 °C and induction of expression by a final concentration of 0.1 mM IPTG. Cells were harvested (4000g, 20 min, 4 °C) after approximately 20 h and stored at –20 °C until protein purification.

      Protein purification

      The proteins were purified in two steps. Cells were resuspended in HisTrap equilibration buffer (10 mM Hepes, pH 7.4, 10 mM imidazole, 0.5 M NaCl, and 10% glycerol), added two cOmplete Mini, EDTA-free, protease inhibitor tablets (Roche Diagnostics GmbH), lysed using a high-pressure homogenizer at 1 bar, added 3 μl Benzonase Nuclease (Sigma–Aldrich), and centrifuged (40,000g, 4 °C, 30 min). The filtered (0.45 μm) supernatant was loaded onto a 5 ml HisTrap HP column (GE Healthcare) pre-equilibrated with HisTrap equilibration buffer at a flow of 3 ml/min and eluted by a linear gradient from 2.5% to 100% HisTrap elution buffer (10 mM Hepes, pH 7.4, 400 mM imidazole, 0.5 M NaCl, and 10% glycerol) in 20 column volumes. Protein-containing fractions (based on absorbance at 280 nm) were pooled, concentrated to 4 ml (10 kDa Amicon Ultra centrifugal filters; Merck-Millipore) and further purified by gel filtration (Superdex 16/60 75 equilibrated with 10 mM Hepes, pH 7.0, 150 mM NaCl, and 10% glycerol) at a flow rate of 0.5 ml/min. Fractions containing recombinant protein of correct size as based on SDS-PAGE were pooled and concentrated as above to 2 to 3 mg/ml. The absorbance at 280 nm was measured, and the protein concentration was determined using theoretical extinction coefficients.

      Substrate specificity

      The specific activity was determined essentially as previously described (
      • Morrill J.
      • Kulcinskaja E.
      • Sulewska A.M.
      • Lahtinen S.
      • Stålbrand H.
      • Svensson B.
      • Abou Hachem M.
      The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
      ,
      • Dilokpimol A.
      • Nakai H.
      • Gotfredsen C.H.
      • Baumann M.J.
      • Nakai N.
      • Abou Hachem M.
      • Svensson B.
      Recombinant production and characterisation of two related GH5 endo-β-1,4-mannanases from Aspergillus nidulans FGSC A4 showing distinctly different transglycosylation capacity.
      ) using a reducing end 3,5-dinitrosalicylic acid (DNS) assay. Briefly, full-length and truncated enzyme forms (0.4–9.1 nM) were incubated with low- and high-viscosity galactomannans (2.5 mg/ml, 10 min), KGM (2.5 mg/ml, 10 min), GG (2.5 mg/ml, 2.5 h), INM (5 mg/ml, 45 min), or Avicel (10 mg/ml, 3 h) under standard activity assay conditions (400 μl assay volume, 40 mM sodium phosphate citrate pH 6.0, 0.005% [v/v] Triton X-100, 37 °C) in triplicates. The reactions were stopped by addition of 600 μl DNS reagents followed by heat treatment at 95 °C for 15 min. After 15 min in ice water, the samples were centrifuged at 20,000g for 12 min. Finally, the absorbance was measured at 540 nm.

      Kinetic analysis

      The kinetics on CGM-lv, GG, and INM were determined using the aforementioned DNS assay with a range of substrate concentrations (0.45–9 mg/ml CGM-lv; 1.25–7.5 mg/ml GG; and 2.5–20 mg/ml INM) and enzyme (full-length and truncated forms: 1.5–2.8 nM for CGM-lv, 13–25 nM for GG, and 99–179 nM for INM) in 2000 μl and withdrawing 400 μl aliquots (200 μl in the case of INM) at 3, 6, 9, and 12 min. The kinetic assays were done in duplicates. In the case of the CGM-lv and INM data, the Michaelis–Menten model was fitted to the initial velocity data using GraphPad Prism 6 (GraphPad Software Inc).

      Pull-down assay

      Qualitative screening of binding of the individual CBM10s separately and the triple CBM10 to microcrystalline cellulose (Avicel) and insoluble crystalline INM was done by a pull-down assay, where the samples were analyzed by SDS-PAGE: 10 mg polysaccharide (prewashed three times) was mixed with 200 μl 0.5 mg/ml protein in assay buffer (50 mM phosphate buffer, pH 7.0). GFP was used as a control in the assay. The samples were incubated at 4 °C for 1 h with gentle agitation and then centrifuged (20,000g, 10 min, 4 °C). Supernatants were transferred to fresh tubes and centrifuged (20,000g, 10 min, 4 °C), before 4 μl supernatant was heat treated in the presence of SDS-loading buffer and applied on the SDS-PAGE. The pellets from the pull-down assay were washed with 250 μl assay buffer and pelleted again as aforementioned, resuspended in 200 μl assay buffer, added 50 μl SDS-loading buffer, boiled for 10 min, and applied (4 μl) on the SDS-PAGE. For each protein, a sample of the 0.5 mg/ml protein stock used for the pull-down assay was treated as the supernatant sample and included on the gel.
      Quantitative pull-down assay was done with microcrystalline cellulose (Avicel) utilizing the presence of GFP for quantification (
      • Attia M.
      • Stepper J.
      • Davies G.J.
      • Brumer H.
      Functional and structural characterization of a potent GH74 endo-xyloglucanase from the soil saprophyte Cellvibrio japonicus unravels the first step of xyloglucan degradation.
      ). The total assay volume (750 μl) contained 10 mg/ml Avicel (washed with assay buffer thrice) and the different SdGH5CBM10-GFP forms in ten different concentrations (10–100 μg/ml) in 50 mM phosphate pH 7.0 and 0.1 mg/ml bovine serum albumin. GFP (10–150 μg/ml) served as a negative control. The mixtures were rotated end over end (4 °C, 1 h) and centrifuged (20,000g, 10 min, 4 °C). Supernatants were transferred to fresh tubes, centrifuged (20,000g, 10 min, 4 °C), and the concentration of unbound protein was determined by fluorescence (FP-8500 Spectrofluorometer equipped with an FMP-825 plate reader; JASCO Corporation) with excitation at 450 and emission at 510 nm using a standard curve made with each protein form (1–1250 nM) in assay buffer. The amount of bound protein was calculated by subtracting the unbound from the total protein, and the dissociation constant, Kd, was obtained by fitting Equation 1 to binding isotherm data (GraphPad Prism, version 6).
      [PC]=[FP][PC]maxKd
      (1)


      PC and FP are the bound and unbound protein concentrations, respectively. The pull-down assays were done in duplicates on two different days.

      AGE

      Qualitative screening of binding of all protein forms and GFP (as control) was done by AGE with 2.5 mg/ml CGM-lv or HEC in 12% acrylamide gels (
      • Cockburn D.
      • Wilkens C.
      • Svensson B.
      Chapter 9: Affinity electrophoresis for analysis of catalytic module-carbohydrate interactions.
      ). Protein (10 μl 0.4 mg/ml) in loading buffer was applied to polysaccharide-containing and control (without polysaccharide) gels, respectively. NativeMark unstained protein standard (Invitrogen) was included on all gels for normalization. The gels were run (20 h at 4 °C and 45 V) and stained using InstantBlue (Expedeon). Quantitative protein binding analysis to CGM-lv was done from migration and retardation in gels without (control) or with 0.125, 0.25, 0.5, 1, 1.5, and 2.5 mg/ml CGM-lv, respectively. Kd was determined from retardation distances according to the method (
      • Cockburn D.
      • Wilkens C.
      • Svensson B.
      Chapter 9: Affinity electrophoresis for analysis of catalytic module-carbohydrate interactions.
      ) using a Takeo–Nakamura plot (
      • Takeo K.
      • Nakamura S.
      Dissociation constants of glucan phosphorylases of rabbit tissues studied by polyacrylamide gel disc electrophoresis.
      ) (Equation 2):
      1Rmi=1Rmo(1+cKd)
      (2)


      Rmi and Rmo are migration distances of sample relative to a reference protein in the presence and absence of polysaccharide, respectively; c is the polysaccharide concentration. The relative migration (Equation 3) defines the retardation of migration by the polysaccharide compared with the control.
      Rm=RmiRmo
      (3)


      Data availability

      All the data are contained within the article.

      Supporting information

      This article contains supporting information.

      Conflict of interest

      The authors declare that they have no conflicts of interest with the contents of this article.

      Acknowledgments

      We thank Karina Jansen (Technical University of Denmark) for technical assistance. We also thank Prof Jens Preben Morth (Technical University of Denmark) for introduction to the JASCO FP-8500 Fluorescence Spectrometer. This work was supported by Independent Research Fund Denmark∣Natural Sciences (FNU) by a grant to the project “Roles and structural determinants of low-affinity carbohydrate-protein interactions” (grant no. 6108-00476B).

      Author contributions

      M. S. M. and B. S. conceptualized the work and designed the experiments. M. S. M. and B. H. performed phylogenetic analysis. M. S. M. and S. E. B. performed the experiments. M. S. M. and B. S. analyzed the data. M. S. M. wrote the article with input from B. S. B. S. acquired the funding.

      Supporting information

      References

        • Conway J.M.
        • Crosby J.R.
        • Hren A.P.
        • Southerland R.T.
        • Lee L.L.
        • Lunin V.V.
        • Alahuhta P.
        • Himmel M.E.
        • Bomble Y.J.
        • Adams M.W.W.
        • Kelly R.M.
        Novel multidomain, multifunctional glycoside hydrolases from highly lignocellulolytic Caldicellulosiruptor species.
        Aiche J. 2018; 64: 4218-4228
        • Low K.E.
        • Smith S.P.
        • Abbott D.W.
        • Boraston A.B.
        The glycoconjugate-degrading enzymes of Clostridium perfringens: Tailored catalysts for breaching the intestinal mucus barrier.
        Glycobiology. 2020; https://doi.org/10.1093/glycob/cwaa050
        • Krska D.
        • Larsbrink J.
        Investigation of a thermostable multi-domain xylanase-glucuronoyl esterase enzyme from Caldicellulosiruptor kristjanssonii incorporating multiple carbohydrate-binding modules.
        Biotechnol. Biofuels. 2020; 13: 1-13
        • Ebringerová A.
        Structural diversity and application potential of hemicelluloses.
        Macromol. Symp. 2006; 232: 1-12
        • Dey P.M.
        Biochemistry of plant galactomannans.
        Adv. Carbohydr. Chem. Biochem. 1978; 35: 341-376
        • McCleary B.V.
        • Clark A.H.
        • Dea I.C.M.
        • Rees D.A.
        The fine structures of carob and guar galactomannans.
        Carbohydr. Res. 1985; 139: 237-260
        • Hori R.
        • Sugiyama J.
        • Wada M.
        The thermal expansion of mannan I obtained from ivory nuts.
        Carbohydr. Polym. 2007; 70: 298-303
        • Heux L.
        • Hägglund P.
        • Putaux J.L.
        • Chanzy H.
        Structural aspects in semicrystalline samples of the mannan II family.
        Biomacromolecules. 2005; 6: 324-332
        • Xu C.
        • Leppänen A.S.
        • Eklund P.
        • Holmlund P.
        • Sjöholm R.
        • Sundberg K.
        • Willför S.
        Acetylation and characterization of spruce (Picea abies) galactoglucomannans.
        Carbohydr. Res. 2010; 345: 810-816
        • Lombard V.
        • Golaconda Ramulu H.
        • Drula E.
        • Coutinho P.M.
        • Henrissat B.
        The carbohydrate-active enzymes database (CAZy) in 2013.
        Nucleic Acids Res. 2014; 42: D490-D495
        • Morrill J.
        • Månberger A.
        • Rosengren A.
        • Naidjonoka P.
        • von Freiesleben P.
        • Krogh K.B.R.M.
        • Bergquist K.E.
        • Nylander T.
        • Karlsson E.N.
        • Adlercreutz P.
        • Stålbrand H.
        β-Mannanase-catalyzed synthesis of alkyl mannooligosides.
        Appl. Microbiol. Biotechnol. 2018; 102: 5149-5163
        • Von Freiesleben P.
        • Spodsberg N.
        • Stenbæk A.
        • Stålbrand H.
        • Krogh K.B.R.M.
        • Meyer A.S.
        Boosting of enzymatic softwood saccharification by fungal GH5 and GH26 endomannanases.
        Biotechnol. Biofuels. 2018; 11: 194
        • Wei X.
        • Fu X.
        • Xiao M.
        • Liu Z.
        • Zhang L.
        • Mou H.
        Dietary galactosyl and mannosyl carbohydrates: In-vitro assessment of prebiotic effects.
        Food Chem. 2020; 329: 127179
        • Zhao D.
        • Zhang X.
        • Wang Y.
        • Na J.
        • Ping W.
        • Ge J.
        Purification, biochemical and secondary structural characterisation of β-mannanase from Lactobacillus casei HDS-01 and juice clarification potential.
        Int. J. Biol. Macromol. 2020; 154: 826-834
        • Gill J.
        • Rixon J.E.
        • Bolam D.N.
        • McQueen-Mason S.
        • Simpson P.J.
        • Williamson M.P.
        • Hazlewood G.P.
        • Gilbert H.J.
        The type II and X cellulose-binding domains of Pseudomonas xylanase A potentiate catalytic activity against complex substrates by a common mechanism.
        Biochem. J. 1999; 342: 473-480
        • Ponyi T.
        • Szabó L.
        • Nagy T.
        • Orosz L.
        • Simpson P.J.
        • Williamson M.P.
        • Gilbert H.J.
        Trp22, Trp24, and Tyr8 play a pivotal role in the binding of the family 10 cellulose-binding module from Pseudomonas xylanase A to insoluble ligands.
        Biochemistry. 2000; 39: 985-991
        • Hogg D.
        • Pell G.
        • Dupree P.
        • Goubet F.
        • Marti S.M.
        • Gilbert H.J.
        The modular architecture of Cellvibrio japonicus mannanases in glycoside hydrolase families 5 and 26 points to differences in their role in mannan degradation.
        Biochemistry. 2003; 1043: 1027-1043
        • Morrill J.
        • Kulcinskaja E.
        • Sulewska A.M.
        • Lahtinen S.
        • Stålbrand H.
        • Svensson B.
        • Abou Hachem M.
        The GH5 1,4-β-mannanase from Bifidobacterium animalis subsp. lactis Bl-04 possesses a low-affinity mannan-binding module and highlights the diversity of mannanolytic enzymes.
        BMC Biochem. 2015; 16: 26
        • Attia M.
        • Stepper J.
        • Davies G.J.
        • Brumer H.
        Functional and structural characterization of a potent GH74 endo-xyloglucanase from the soil saprophyte Cellvibrio japonicus unravels the first step of xyloglucan degradation.
        FEBS J. 2016; 283: 1701-1719
        • Crouch L.I.
        • Labourel A.
        • Walton P.H.
        • Davies G.J.
        • Gilbert H.J.
        The contribution of non-catalytic carbohydrate binding modules to the activity of lytic polysaccharide monooxygenases.
        J. Biol. Chem. 2016; 291: 7439-7449
        • Pires V.M.R.
        • Pereira P.M.M.
        • Brás J.L.A.
        • Correia M.
        • Cardoso V.
        • Bule P.
        • Alves V.D.
        • Najmudin S.
        • Venditto I.
        • Ferreira L.M.A.
        • Romão M.J.
        • Carvalho A.L.
        • Fontes C.M.G.A.
        • Prazeres D.M.
        Stability and ligand promiscuity of type A carbohydrate-binding modules are illustrated by the structure of Spirochaeta thermophila StCBM64C.
        J. Biol. Chem. 2017; 292: 4847-4860
        • Hall J.
        • Black G.W.
        • Ferreira L.M.
        • Millward-Sadler S.J.
        • Ali B.R.
        • Hazlewood G.P.
        • Gilbert H.J.
        The non-catalytic cellulose-binding domain of a novel cellulase from Pseudomonas fluorescens subsp. cellulosa is important for the efficient hydrolysis of Avicel.
        Biochem. J. 1995; 309: 749-756
        • Gilbert H.J.
        • Hall J.
        • Hazlewood G.P.
        • Ferreira L.M.A.
        The N-terminal region of an endoglucanase from Pseudomonas fluorescens subspecies cellulosa constitutes a cellulose-binding domain that is distinct from the catalytic centre.
        Mol. Microbiol. 1990; 4: 759-767
        • Gardner J.G.
        • Crouch L.
        • Labourel A.
        • Forsberg Z.
        • Bukhman Y.V.
        • Vaaje-Kolstad G.
        • Gilbert H.J.
        • Keating D.H.
        Systems biology defines the biological significance of redox-active proteins during cellulose degradation in an aerobic bacterium.
        Mol. Microbiol. 2014; 94: 1121-1133
        • Ferreira L.M.A.
        • Hazlewood G.P.
        • Barker P.J.
        • Gilbert H.J.
        The cellodextrinase from Pseudomonas fluorescens subsp. cellulosa consists of multiple functional domains.
        Biochem. J. 1991; 279: 793-799
        • O’Connor R.M.
        • Fung J.M.
        • Sharp K.H.
        • Benner J.S.
        • McClung C.
        • Cushing S.
        • Lamkin E.R.
        • Fomenkov A.I.
        • Henrissat B.
        • Londer Y.Y.
        • Scholz M.B.
        • Posfai J.
        • Malfatti S.
        • Tringe S.G.
        • Woyke T.
        • et al.
        Gill bacteria enable a novel digestive strategy in a wood-feeding mollusk.
        Proc. Natl. Acad. Sci. 2014; 111: E5096-E5104
        • Watson B.J.
        • Zhang H.
        • Longmire A.G.
        • Young H.M.
        • Hutcheson S.W.
        Processive endoglucanases mediate degradation of cellulose by Saccharophagus degradans.
        J. Bacteriol. 2009; 191: 5697-5705
        • Ekborg N.A.
        • Morrill W.
        • Burgoyne A.M.
        • Li L.
        • Distel D.L.
        CelAB, a multifunctional cellulase encoded by Teredinibacter turnerae T7902T, a culturable symbiont isolated from the wood-boring marine bivalve Lyrodus pedicellatus.
        Appl. Environ. Microbiol. 2007; 73: 7785-7788
        • Millward-Sadler S.J.
        • Davidson K.
        • Hazlewood G.P.
        • Black G.W.
        • Gilbert H.J.
        • Clarke J.H.
        Novel cellulose-binding domains, NodB homologues and conserved modular architecture in xylanases from the aerobic soil bacteria Pseudomonas fluorescens subsp. cellulosa and Cellvibrio mixtus.
        Biochem. J. 1995; 312: 39-48
        • Tamaru Y.
        • Araki T.
        • Morishita T.
        • Kimura T.
        • Sakka K.
        • Ohmiya K.
        Cloning, DNA sequencing, and expression of the β-1,4-mannanase gene from a marine bacterium, Vibrio sp. strain MA-138.
        J. Ferment. Bioeng. 1997; 83: 201-205
        • Kim D.Y.
        • Ham S.J.
        • Lee H.J.
        • Cho H.Y.
        • Kim J.H.
        • Kim Y.J.
        • Shin D.H.
        • Rhee Y.H.
        • Son K.H.
        • Park H.Y.
        Cloning and characterization of a modular GH5 β-1,4-mannanase with high specific activity from the fibrolytic bacterium Cellulosimicrobium sp. strain HY-13.
        Bioresour. Technol. 2011; 102: 9185-9192
        • Jin Y.
        • Petricevic M.
        • John A.
        • Raich L.
        • Jenkins H.
        • Portela De Souza L.
        • Cuskin F.
        • Gilbert H.J.
        • Rovira C.
        • Goddard-Borger E.D.
        • Williams S.J.
        • Davies G.J.
        A β-mannanase with a lysozyme-like fold and a novel molecular catalytic mechanism.
        ACS Cent. Sci. 2016; 2: 896-903
        • Sakai K.
        • Kimoto S.
        • Shinzawa Y.
        • Minezawa M.
        • Suzuki K.
        • Jindou S.
        • Kato M.
        • Shimizu M.
        Characterization of pH-tolerant and thermostable GH 134 β-1,4-mannanase SsGH134 possessing carbohydrate binding module 10 from Streptomyces sp. NRRL B-24484.
        J. Biosci. Bioeng. 2018; 125: 287-294
        • Shi P.
        • Yuan T.
        • Zhao J.
        • Huang H.
        • Luo H.
        • Meng K.
        • Wang Y.
        • Yao B.
        Genetic and biochemical characterization of a protease-resistant mesophilic β-mannanase from Streptomyces sp. S27.
        J. Ind. Microbiol. Biotechnol. 2011; 38: 451-458
        • Arcand N.
        • Kluepfel D.
        • Paradis F.W.
        • Morosoli R.
        • Shareck F.
        β-Mannanase of Streptomyces lividans 66: Cloning and DNA sequence of the manA gene and characterization of the enzyme.
        Biochem. J. 1993; 290: 857-863
        • Boraston A.B.
        • Bolam D.N.
        • Gilbert H.J.
        • Davies G.J.
        Carbohydrate-binding modules: Fine-tuning polysaccharide recognition.
        Biochem. J. 2004; 382: 769-781
        • Raghothama S.
        • Simpson P.J.
        • Szabó L.
        • Nagy T.
        • Gilbert H.J.
        • Williamson M.P.
        Solution structure of the CBM10 cellulose binding module from Pseudomonas xylanase A.
        Biochemistry. 2000; 39: 978-984
        • Blake A.W.
        • McCartney L.
        • Flint J.E.
        • Bolam D.N.
        • Boraston A.B.
        • Gilbert H.J.
        • Knox J.P.
        Understanding the biological rationale for the diversity of cellulose-directed carbohydrate-binding modules in prokaryotic enzymes.
        J. Biol. Chem. 2006; 281: 29321-29329
        • Duan C.-J.
        • Huang M.-Y.
        • Pang H.
        • Zhao J.
        • Wu C.-X.
        • Feng J.-X.
        Characterization of a novel theme C glycoside hydrolase family 9 cellulase and its CBM-chimeric enzymes.
        Appl. Microbiol. Biotechnol. 2017; 101: 5723-5737
        • Wang Y.
        • Yuan H.
        • Wang J.
        • Yu Z.
        Truncation of the cellulose binding domain improved thermal stability of endo-β-1,4-glucanase from Bacillus subtilis JA18.
        Bioresour. Technol. 2009; 100: 345-349
        • Park S.R.
        • Cho S.J.
        • Kim M.K.
        • Ryu S.K.
        • Lim W.J.
        • An C.L.
        • Hong S.Y.
        • Kim J.H.
        • Kim H.
        • Yun H.D.
        Activity enhancement of Cel5Z from Pectobacterium chrysanthemi PY35 by removing C-terminal region.
        Biochem. Biophys. Res. Commun. 2002; 291: 425-430
        • Kumagai Y.
        • Kawakami K.
        • Uraji M.
        • Hatanaka T.
        Binding of bivalent ions to actinomycete mannanase is accompanied by conformational change and is a key factor in its thermal stability.
        Biochim. Biophys. Acta. 2013; 1834: 301-307
        • Kari J.
        • Andersen M.
        • Borch K.
        • Westh P.
        An inverse Michaelis-Menten approach for interfacial enzyme kinetics.
        ACS Catal. 2017; 7: 4904-4914
        • Kari J.
        • Olsen J.P.
        • Jensen K.
        • Badino S.F.
        • Krogh K.B.R.M.
        • Borch K.
        • Westh P.
        Sabatier principle for interfacial (heterogeneous) enzyme catalysis.
        ACS Catal. 2018; 8: 11966-11972
        • Liu H.X.
        • Gong J.S.
        • Li H.
        • Lu Z.M.
        • Li H.
        • Qian J.Y.
        • Xu Z.H.
        • Shi J.S.
        Biochemical characterization and cloning of an endo-1,4-β-mannanase from Bacillus subtilis YH12 with unusually broad substrate profile.
        Process. Biochem. 2015; 50: 712-721
        • Hutcheson S.W.
        • Zhang H.
        • Suvorov M.
        Carbohydrase systems of Saccharophagus degradans degrading marine complex polysaccharides.
        Mar. Drugs. 2011; 9: 645-665
        • Hervé C.
        • Rogowski A.
        • Blake A.W.
        • Marcus S.E.
        • Gilbert H.J.
        • Knox J.P.
        Carbohydrate-binding modules promote the enzymatic deconstruction of intact plant cell walls by targeting and proximity effects.
        Proc. Natl. Acad. Sci. U. S. A. 2010; 107: 15293-15298
        • Edgar R.C.
        Muscle: Multiple sequence alignment with high accuracy and high throughput.
        Nucleic Acids Res. 2004; 32: 1792-1797
        • Gascuel O.
        Bionj: An improved version of the NJ algorithm based on a simple model of sequence data.
        Mol. Biol. Evol. 1997; 14: 685-695
        • Huson D.H.
        • Scornavacca C.
        Dendroscope 3: An interactive tool for rooted phylogenetic trees and networks.
        Syst. Biol. 2012; 61: 1061-1067
        • Pei J.
        • Tang M.
        • Grishin N.V.
        PROMALS3D web server for accurate multiple protein sequence and structure alignments.
        Nucleic Acids Res. 2008; 36: W30-W34
        • Zimmermann L.
        • Stephens A.
        • Nam S.-Z.
        • Rau D.
        • Kübler J.
        • Lozajic M.
        • Gabler F.
        • Söding J.
        • Lupas A.N.
        • Alva V.
        A completely reimplemented MPI bioinformatics toolkit with a new HHpred server at its core.
        J. Mol. Biol. 2018; 430: 2237-2243
        • Dilokpimol A.
        • Nakai H.
        • Gotfredsen C.H.
        • Baumann M.J.
        • Nakai N.
        • Abou Hachem M.
        • Svensson B.
        Recombinant production and characterisation of two related GH5 endo-β-1,4-mannanases from Aspergillus nidulans FGSC A4 showing distinctly different transglycosylation capacity.
        Biochim. Biophys. Acta. 2011; 1814: 1720-1729
        • Cockburn D.
        • Wilkens C.
        • Svensson B.
        Chapter 9: Affinity electrophoresis for analysis of catalytic module-carbohydrate interactions.
        in: Abbott D.W. Lammerts van Bueren A. Protein-Carbohydrate Interactions: Methods and Protocols. Methods in Molecular Biology. 1588. Springer, New York, NY2017: 119-127
        • Takeo K.
        • Nakamura S.
        Dissociation constants of glucan phosphorylases of rabbit tissues studied by polyacrylamide gel disc electrophoresis.
        Arch. Biochem. Biophys. 1972; 153: 1-7