Open Collections

UBC Faculty Research and Publications

Evolution, substrate specificity and subfamily classification of glycoside hydrolase family 5 (GH5) Aspeborg, Henrik; Coutinho, Pedro M; Wang, Yang; Brumer, Harry; Henrissat, Bernard Sep 20, 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12862_2012_Article_2161.pdf [ 551.33kB ]
JSON: 52383-1.0074600.json
JSON-LD: 52383-1.0074600-ld.json
RDF/XML (Pretty): 52383-1.0074600-rdf.xml
RDF/JSON: 52383-1.0074600-rdf.json
Turtle: 52383-1.0074600-turtle.txt
N-Triples: 52383-1.0074600-rdf-ntriples.txt
Original Record: 52383-1.0074600-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessEvolution, substrate specificity and subfamilyclassification of glycoside hydrolase family 5 (GH5)Henrik Aspeborg1†, Pedro M Coutinho3†, Yang Wang1, Harry Brumer III1,2 and Bernard Henrissat3*AbstractBackground: The large Glycoside Hydrolase family 5 (GH5) groups together a wide range of enzymes acting onβ-linked oligo- and polysaccharides, and glycoconjugates from a large spectrum of organisms. The long andcomplex evolution of this family of enzymes and its broad sequence diversity limits functional prediction. With theobjective of improving the differentiation of enzyme specificities in a knowledge-based context, and to obtain newevolutionary insights, we present here a new, robust subfamily classification of family GH5.Results: About 80% of the current sequences were assigned into 51 subfamilies in a global analysis of all publiclyavailable GH5 sequences and associated biochemical data. Examination of subfamilies with catalytically-activemembers revealed that one third are monospecific (containing a single enzyme activity), although new functionsmay be discovered with biochemical characterization in the future. Furthermore, twenty subfamilies presently haveno characterization whatsoever and many others have only limited structural and biochemical data. Mapping offunctional knowledge onto the GH5 phylogenetic tree revealed that the sequence space of this historical andindustrially important family is far from well dispersed, highlighting targets in need of further study. The analysisalso uncovered a number of GH5 proteins which have lost their catalytic machinery, indicating evolution towardsnovel functions.Conclusion: Overall, the subfamily division of GH5 provides an actively curated resource for large-scale proteinsequence annotation for glycogenomics; the subfamily assignments are openly accessible via theCarbohydrate-Active Enzyme database at Protein evolution, Enzyme evolution, Functional prediction, Glycogenomics, Glycoside hydrolase family5, Phylogenetic analysis, Subfamily classificationBackgroundCarbohydrates, in the form of mono-, di-, oligo-, andpolysaccharides, as well as glycoconjugates, play funda-mental roles in all forms of life [1]. Beyond their role inenergy storage, carbohydrates are central to diverse bio-logical processes such as host-pathogen interactions, sig-nal transduction, inflammation, intracellular trafficking,diseases, and differentiation/development. Not least, asstructural components of terrestrial biomass, carbohy-drates comprise approximately 75% of the carbon fixedannually by primary production [2]. Sugar-rich plant cellwalls, seeds, and tubers thus represent a renewablematerial with significant potential to address energy andmaterial needs.A striking feature of carbohydrates is their remarkablestructural complexity, due to a rich diversity of mono-saccharide building blocks, and the possibility of numer-ous stereo- and regiospecific linkages [3], which give riseto both simple linear and complex, highly branchedmolecules [1]. A decade of investments in genomics andproteomics has greatly improved our interpretation ofthe molecular language of the cell, but deciphering thecomplex carbohydrate-based information in the biomo-lecular landscape is still in its infancy. Indeed, glycomicshas been identified both as “the last frontier of molecu-lar and cellular biology” [4] as well as an “emerging tech-nology that will change the world” [5].Functional analysis of glycans and glycoconjugates iscomplicated by the fact that they are not direct genetic* Correspondence:†Equal contributors3Architecture et Fonction des Macromolécules Biologiques, Aix-MarseilleUniversité, CNRS, UMR 7257, 163 Avenue de Luminy, Marseille 13288, FranceFull list of author information is available at the end of the article© 2012 Aspeborg et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (, which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.Aspeborg et al. BMC Evolutionary Biology 2012, 12:186, but are instead synthesized, recognized, modi-fied, and degraded by a plethora of carbohydrate-activeenzymes (CAZymes) and binding proteins. In the syn-thetic direction, phosphosugar-dependent glycosyltrans-ferases (GTs) catalyze the formation of glycosidiclinkages, whereas their breakdown is mediated by glyco-side hydrolases (GHs) and polysaccharide lyases (PLs),with the assistance of carbohydrate esterases (CEs). Thestructural diversity of carbohydrates is reflected in anabundance of CAZyme-encoding genes, which comprise1-3% of the genome of most organisms [6]. Expandingand harnessing knowledge of the complexity of the“CAZome” is thus essential to understanding the com-plexity of the glycome.The protein sequence-based classification of CAZymeswas initiated in 1991 as a complement to the long-standing Enzyme Commission (EC) number system [7],which is based solely on enzyme activities [8]. Given theprevalence of convergent evolution of enzymes thatcleave glycosidic bonds, as well as the demonstrablecatalytic promiscuity of individual enzymes, sequence-based classification has proven to be a robust way tounify information on enzyme structure, specificity, andmechanism, which provides enormous predictive power[9]. Initially motivated by a need to delineate cellulases(EC into distinct structural families [10], the firstincarnation of the GH family classification, as such,comprised 35 GH families [8]. The number of familiesincreased steadily with the growing interest in Glycobiol-ogy so that, as of August 2012, 130 sequence-based fam-ilies of GHs have been defined in the continuouslyupdated CAZy database [11].Presently, one of the largest GH families is GH5, his-torically known as “cellulase family A” as it was the firstcellulase family described [10]. GH5 exemplifies a familywith a large variety of specificities: it currently containsclose to 20 experimentally determined enzyme activitiesdenoted with an EC number. The abundance of GH5enzymes in different ecological niches has been high-lighted by their frequent identification in metagenomesof diverse microbial communities [12-14], as well as thegenomes of individual organisms [11]. As with otherCAZyme families [15], GH5 members are commonlyfound to be encoded as parts of multi-modular polypep-tide chains containing other catalytic, substrate-binding,and functionally unidentified or yet to be describedmodules.Within the large GH5 family, a discernible diversity ofsequences was observed soon after its creation. The firstfive subfamilies of GH5 (A1-A5) were identified as earlyas 1990 [16]. Subfamily A6 was introduced in 1997 [17]and the following year eukaryotic and prokaryotic β-mannanases were assigned to A7 and A8, respectively[18]. Subsequently, subfamily A9 was introduced in astudy, which notably also suggested the merger of A5and A6 [19]. Finally, A10 was the most recently definedGH5 subfamily [20], while new subfamilies that pres-ently lack a unique identifier have also been suggested[21,22]. Family GH5 belongs to clan GH-A, which pres-ently groups 19 GH families to form the largest set ofevolutionarily related GH families described in CAZythus far (a clan is a group of families that arise from acommon but very distant ancestor; despite weak se-quence similarity, clan members share conserved proteinfold and catalytic machinery).Families such as GH5 were originally defined with avery small number of sequences. With the accumulationof an increasing body of sequence data, the relationshipbetween the original families has sometimes changedenough to merit reexamination of family membership.Very recently, detailed three-dimensional structural ana-lysis led to the reclassification of several GH5 sequencesinto family GH30 based on the organization of second-ary structural elements around the conserved (β/α)8 foldof the catalytic module [23].Given the continuing expansion in sequence numbersand the partial GH5/GH30 reclassification, it is clearthat a global re-analysis of the subfamily division of GH5is now needed. The rapid accumulation of genomic datain the past decade revealed a complex and varied se-quence space, with the consequence that a substantialportion of GH5 family members are currently notassigned to any subfamily. This situation will only be-come worse as the rate of (meta)genomic sequencingcontinues to increase with phenomenal rapidity. Further,this flood of data will cause an increasing reliance oncomputer-based annotation, which necessarily requires arobust framework to produce meaningful functional pre-dictions. The division of CAZyme families into subfam-ilies based on phylogenetic analysis has been applied asa successful approach to meet this challenge: Subfamilyclassification of GH13, GH30 and all of the PL familieshas demonstrated that the majority of the defined sub-families were monospecific, thus indicating a signifi-cantly better correlation of substrate specificity betweensequences at the subfamily level than the family level[23-25]. Significantly, the division into subfamilies allowsthe identification of currently uncharacterized subfam-ilies that can subsequently be analyzed biochemicallyand structurally to potentially unveil new activities.Hence, we present here an improved, robust subfamilyclassification for GH5 by employing a large-scale ana-lysis of all publicly available sequences. Our intention isthat the introduction of this additional hierarchical levelacross this important GH family will serve to guide en-zyme discovery, structure-function analysis, and biocata-lyst improvement in post-genomic efforts. Not least,many enzyme activities relevant to biomass analysis andAspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 2 of 16 are found in GH5 (e.g., cellulases, manna-nases, xylanases, galactanases, and xyloglucanases), asare enzymes with biomedical applications [26]. Signifi-cantly, the present analysis unveiled a large number ofsparsely or incompletely characterized subfamilies thatmay still hide a number of unsuspected activities andsingular structural features.Results and discussionOur bioinformatics approach allowed the division ofclose to 2300 GH5 catalytic modules into 51 distinctsubfamilies, as shown in the global phylogenetic tree(Figure 1 and Additional file 1: Figure S1); subfamily in-formation is summarized in Table 1. Subfamily namingfollows the procedure devised for GH13, where the fam-ily number is followed by an Arabic numeral thatreflects the order of creation [24]: GH5_1 to GH5_53.This series is essentially continuous, with a few excep-tions due to historical reasons: All of the previouslydescribed subfamilies (A1-A10) have been re-identifiedin the current investigation except for A3 and A4, whichare merged into a single subfamily GH5_4 and A5 andA6 which are unified in subfamily GH5_5 (Figure 1). Tomaintain consistency with earlier literature, the re-identified historical subfamilies have retained the ori-ginal Arabic numeral. For example, the subfamilyformerly known as A2 is hereby designated GH5_2. Theabsence of subfamilies GH5_3 and GH5_6 reflects thetwo fusion events involving the historical subfamiliesdescribed above [19].In addition to the new and historical designations, thetaxonomical range of the included sequences, experi-mentally determined enzyme activities and representa-tive 3-D structures are presented in Table 1 for eachsubfamily. Notably, all of the 33 enzymes with a solved3-D structure have been assigned to a subfamily, result-ing in thirteen individual subfamilies out of 51 with atleast one structural representative. Genes that encodeGH5 enzymes are present in most organisms rangingfrom Archaea and Eubacteria to Eukaryotes, e.g. fungi4253752383936944453314234950153247 4622 48 1243282934275226213584012016115124537304113191842101731Figure 1 Phylogenetic tree of family GH5. In this circular phylogram, the branches corresponding to subfamilies 1–53 are shown in color andthe subfamily numbers are indicated next to the exterior color circle. The branches corresponding to sequences not included into subfamilies arein black. A detailed version of this tree is found in Additional file 1: Figure S1.Aspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 3 of 16 1 Newly defined subfamilies within glycoside hydrolase family GH5New GH5subfamilyHistoricalsubfamilyNumber ofsequencesPresent taxonomicaldistributionECnumberRepresentative PDBstructureGH5_1 A1b 133 Archaea Bacteria Eukaryota 2ZUN3.2.1.733.2.1.91GH5_2 A2b 245 Bacteria Eukaryota 2A3H3.2.1.132GH5_4 A3 + A4b 160 Bacteria Eukaryota 2JEQ3.2.1.1513.2.1.733.2.1.8GH5_5 A5 + A6b,c,e 123 Bacteria Eukaryota IGZJGH5_7 A7d 133 Archaea Bacteria Eukaryota IRH93.2.1. 782.4.1.-GH5_8 A8d 71 Bacteria Eukaryota 2WHLGH5_9 A9e 107 Eukaryota 3N9K3.2.1.753.2.1.21GH5_10 A10e,f 19 Bacteria Eukaryota (Metazoa) 2C0HGH5_11 19 Eukaryota (Plants;Fungi) NDGH5_12 42 Bacteria Eukaryota 59 Bacteria NDGH5_14 15 Eukaryota (Plants) 10 Eukaryota (Fungi) 10 Eukaryota (Fungi) 5 Bacteria 24 Bacteria NDGH5_19 23 Archaea NDGH5_20 17 Eukaryota (Stramenopiles) NDGH5_21 10 Bacteria 12 Bacteria Eukaryota 5 Eukaryota (Fungi) 5 Eukaryota (Fungi) NDGH5_25 16 Bacteria 3MMW3.2.1.78GH5_26 17 Bacteria 5 Eukaryota (Metazoa;Fungi) 8 Bacteria 2OSXGH5_29 5 Bacteria 5 Eukaryota (Fungi) NDAspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 4 of 16 plants. From an anthropocentric perspective, a GH5member is notably lacking in the human genome. Exam-ples of metazoan GH5 genes are also notably scarce andare limited to nematodes, mollusks, and arthropods,likely resulting from horizontal transfer. For example,several independent horizontal events of transfer of cel-lulase and xylanase genes from Bacteria to nematodeshave been described [27] and the transfer of a bacterialβ-mannanase to an insect was recently documented[28]. The taxonomical range at the subfamily level is, nat-urally, more restricted. A few smaller subfamilies are cur-rently specific to certain types of organisms (Table 1). Forexample, eight subfamilies (GH5_15, GH5_16, GH5_23,GH5_24, GH5_30, GH5_31, GH5_49 and GH5_50) con-tain only fungal sequences. Subfamily GH5_14 containsonly plant members, whereas members of GH5_20Table 1 Newly defined subfamilies within glycoside hydrolase family GH5 (Continued)GH5_31 5 Eukaryota (Fungi) 3.2.1.-gGH5_32 5 Eukaryota (Plants; Stramenopiles) NDGH5_33 9 Eukaryota (Stramenopiles) NDGH5_34 5 Bacteria 3.2.1.-g 2Y8KGH5_35 5 Bacteria NDGH5_36 23 Bacteria IVJZ3.2.1.78GH5_37 19 Bacteria ICEN3.2.1.733.2.1.74GH5_38 10 Bacteria 3.2.1.-hGH5_39 7 Bacteria 8 Bacteria NDGH5_41 14 Bacteria NDGH5_42 9 Bacteria NDGH5_43 11 Bacteria NDGH5_44 24 Bacteria NDGH5_45 5 Bacteria NDGH5_46 15 Bacteria 3.2.1.-hGH5_47 6 Bacteria NDGH5_48 18 Bacteria 3.2.1.-hGH5_49 20 Eukaryota (Fungi) NDGH5_50 7 Eukaryota (Fungi) NDGH5_51 5 Bacteria Eukaryota (Fungi) NDGH5_52 6 Bacteria 6 Bacteria Activity not determined yet.aExperimentally determined.bsee [16].csee [17].dsee [18].esee [19].fsee [20].gEC numbers not yet defined.hActive enzyme(s) present but with unclear EC number(s).* See for most recent information.Historical names for some subfamilies are provided, along with the taxonomical range, the characterization level and structural information from representativestructures. Known enzyme activities in family GH5 are provided using the following Enzyme Classification (EC) numbers and corresponding activities: –endo-β-1,4-glucanase or cellulase; – endo-β-1,4-xylanase; – β-glucosidase; – β-mannosidase; – β-glucocerebrosidase; –glucan β-1,3-glucosidase; – licheninase; – cellodextrinase; – glucan endo-β-1,6-glucosidase; – mannan endo-β-1,4-mannosidase orendo-β-1,4-mannanase; – cellulose β-1,4-cellobiosidase or cellobiohydrolase; – endoglycoceramidase; – chitosanase; – β-primeverosidase; – xyloglucan-specific endo-β-1,4-glucanase; – endo-β-1,6-galactanase; – hesperidin 6-O-α-L-rhamnosyl-β-glucosidase;3.2.1.- – undefined EC numbers for β-1,3-mannanase/β-1,3-glucomannanase or for arabinoxylan-specific β-xylanase or still unclear EC numbers depending on thesubfamily (see notes and text); 2.4.1.- – β-mannan transglycosidase.Aspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 5 of 16 exclusively from Stramenopiles. These limited taxo-nomic distributions may represent biological (catalytic)specialization, if not biased by a still incomplete genomesequencing of organisms.The definition of subfamilies was restricted to phylo-genetic clades with five or more members from differentorganisms available in the public protein databases(GenBank and UniProt) in order to capture sufficient di-versity for a robust subfamily definition. Using thesecriteria, the overall success rate of subfamily groupingwas of approximately 80%, i.e., about 20% of the ana-lyzed GH5 sequences could not be classified into sub-families having at least five public members. Thesequences that have not yet been assigned to subfamilieswill likely define new subfamilies as the pool of availablesequences continues to increase. In the future, thesesubfamilies will be gradually released with new identi-fiers when they have been sufficiently populated.The_fusca|AAZ54938|8| Str_therm|BAK26781|8| Str_gallo|CBI12654|8 Pae_polym|CCC85403|8 Rum_albus|ADU23074|8 Bac_circu|BAA25878|8| Spi_therm|AEJ61551|8 Mah_austr|AEE96309|8 Vib_MA-138|BAA25188|8| Cel_japon|AAO31760|8| Cal_besci|ACM60954|8  /  Cal_besci|ACM60954|1_2 Cal_besci|ACM60953|8| Cal_sacch|AAA71887|8| Cal_cellu|AAF22274|8| Clo_cellu|ADL52789|7  /  Clo_cellu|ADL52789|7_2 Cal_sacch|ABP66692|1| Cla_michi|CAA44467|1| Sal_tropi|ABP56852|2 Ver_bacte|EDY84333|34 Ace_cellu|EFL62381|34 Cop_ciner|EAU93630|30 Xan_oryza|AAW77290 Cel_japon|ACE86060 Can_Midic|AEI88911|48 Clo_therm|ABN52700|dist a b c d Figure 2 Examples of modular GH5 proteins. (a) Diverse modular arrangements of putative monofunctional modular enzymes from subfamilyGH5_8. (b) Same for putative bifunctional GH5 enzymes containing a subfamily GH5_8 module. (c) Other putative bifunctional enzymescontaining at least a single GH5 module. (d) Selected examples of proteins containing GH5 modules having lost one or more catalytic residues.For a given protein, each GH5 module is identified by a number of fields separated by “|” indicating: (i) the organism, with 3 letters for the genreand either 5 letters for the species or full strain code; (ii) the GenBank protein accession; (iii) if attributed, the subfamily number or otherinformation; (iv) EC numbers if available. These individual tags are analogous to what is found in Additional file 1: Figure S1. The module typesand other protein segments present are: GHx_y – glycoside hydrolase family x subfamily y (pink); CEx – carbohydrate esterase module of family x(light brown); Cip21 – chitin-binding protein type 21 module with putative carbohydrate oxidative cleaving activity, formerly CBM33 (dark gray);CBMx – carbohydrate binding modules of family x (light green); FN3 – fibronectin type III modules (dark green); DOC – cellulosomal dockerinmodules (light violet); EXPN – expansin modules (dark purple); signal peptides (purple); transmembrane segments (yellow); linkers (light blue);other regions (light grey).Aspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 6 of 16 to the GH13 subfamily classification [24], theGH5 subdivision has resulted in both a higher numberof subfamilies and a higher number of uncharacterizedsubfamilies, suggesting that family GH5 is comparativelyless well explored.To further refine the global subfamily analysis, max-imum likelihood phylogenetic analysis has been per-formed on each subfamily (Additional file 1: Figures S1:GH5_1 – GH5_53). In addition to database accessionnumbers, information about substrate specificity (as pro-vided by EC numbers) has been included, and 3-Dstructures highlighted in each subfamily tree. The 51 sub-families (numbered GH5_1 to GH5_53 as explained else-where) have been categorized based on available enzymeactivity data, to aid their individual descriptions, below.Thus, the first group, comprised of “Enzymatically-ActiveSubfamilies”, contains subfamilies with at least one mem-ber whose enzyme activity has been shown. The extent ofthe documented characterization varies substantiallywithin the group, from simple information obtained fromenzymatic assays insufficient to assign a particular ECnumber (these enzymes are denoted with a star inAdditional file 1: Figure S1), to detailed enzyme specificityand kinetics studies. Among the better characterized sub-families, monospecific subfamilies are distinguished by thepresence of only one EC number for one or more mem-bers, whereas in polyspecific subfamilies, two or more en-zymatic activities have been observed in differentmembers. “Uncharacterized Subfamilies” comprise thesecond major group; these subfamilies currently lackdocumented enzymatic activity altogether.In this context, it is worth noting that the majority ofthe large, well characterized subfamilies were polyspeci-fic. These highly populated subfamilies were also thefirst ones to be identified and described, and often, des-pite observed polyspecificity, one particular activity pre-dominates. For example, 22 of 24 characterized enzymesin GH5_1 are reported to be endo-glucanases, whereasone protein is a documented as a cellobiohydrolase, andanother was described as displaying licheninase activity.It is, however, difficult to draw far-reaching conclusionsbased on these observations. On one hand, it may bethat the acquisition of a new specificity within a subfam-ily is a rare event; alternatively, the observation of one orfew activities in specific subfamilies may simply be aconsequence of differences in the range of substratestested experimentally.Another significant aspect of many GH5 family pro-teins is that their protein sequence may include add-itional modules with different functions, and inparticular CBMs [29]. A great variety of modular struc-tures may be found throughout the family and in a num-ber of individual subfamilies. An analysis of all thecomplexity of modular structures found in the familygoes beyond the objectives of this study, and someaspects of this diversity are illustrated in Figure 2. Forinstance, many members of subfamily GH5_8 are modu-lar and reveal two major trends: (i) the addition of oneor of multiple CBMs (see Figure 2a) is more commonand may be associated not only to the nature of themain substrate of the corresponding catalytic domain,particularly in complex substrates; and (ii) the combina-tions with other catalytic modules to form bifunctionalenzymes (see Figure 2b but also 2c), are more rare butparticular useful to reveal interacting or synergistic en-zyme activities of some catalytic modules. Numerousmodular arrangements can also be found in other largesubfamilies like GH5_1, GH5_2 and GH5_7. CBMs canbe located on the N- or C-terminal side of the GH5module (as illustrated in GH5_8 in Figure 2a). The com-bination of catalytic domains may target different tissuecomponents. Some may, for instance, target celluloseand cellulose associated substrates (as in Figure 2b) butbifunctional enzymes likely targeting hemicellulose mayalso be found (several examples in Figure 2c).Subfamilies with identified active enzymesMonospecific subfamiliesA number of subfamilies exhibit a single activity amongtheir characterized enzyme members. Multiple individualexamples within a subfamily improve the degree of con-fidence regarding subfamily monospecificity, while sub-families with only a single characterized representativemay be subject to reinterpretation in the future as thebreadth of biochemical data increases.GH5_5 endo-β-1,4-glucanases (EC The largestsubfamily that contains only a single EC number isGH5_5, which is primarily composed of secreted bacter-ial and fungal enzymes (Additional file 1: Figure S1:GH5_5). All investigated enzymes in GH5_5 displayendo-β-1,4-glucanase activity (EC One crystalstructure has been determined for a Thermoascus auran-tiacus endo-glucanase [19]. In fungi, about half of theGH5 proteins harbor a CBM1 module at the N- or C-terminus, which is compatible with an active role on cel-lulose. No modular proteins are found among the bac-terial members of the subfamily.GH5_8, GH5_10 and GH5_17: endo-β-1,4-mannanase(EC Members of the large GH5_8 subfamilyare all extracellular mannan endo-β-1,4-mannosidases(EC according to available biochemical charac-terization, and this subfamily was historically describedas the bacterial mannanase subfamily A8 [18]. Structuralanalysis has highlighted distinctive features of alkaline β-mannanases [30]. The subfamily now contains a singleeukaryotic enzyme from the beetle HypothenemusAspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 7 of 16, resulting from a horizontal gene transfer frombacteria [28].In the closely related but distinct subfamilies GH5_10and GH5_17, only extracellular enzymes with endo-β-1,4-mannanase activity (EC have equally beenreported. Metazoan sequences, including sequences ori-ginating from mollusks and arthropods, as well as bac-terial sequences compose the GH5_10 subfamily,whereas subfamily GH5_17 only harbors bacterialenzymes. Currently, none of the bacterial GH5_10 mem-bers have a documented enzyme activity, although aFibrobacter succinogenes enzyme is active on AZCL-galactomannan [31]. The bacterial enzymes in GH5_17are all cellulosomal components from the genusClostridium.GH5_14 : (plant) exo-β-1,3-glucosidase (EC GH5_14 comprises exclusively plant enzymesand relies on a single functional characterization atpresent. Recombinant expression of the rice OsGH5BGshowed that the enzyme had glucan β-1,3-glucosidaseactivity (EC [22]. A unique feature of GH5_14is the fascin-like module inserted after β-strand 1. Fascinis a human actin-binding protein, but the function of theplant fascin-like domain is unknown. Members ofGH5_14 are well represented throughout the plant king-dom, but most interestingly, a representative is absent inthe leading plant model organism Arabidopsis thaliana.GH5_15 : (fungal) (endo-)β-1,6-glucanases (EC GH5_15 is a small but well characterized sub-family composed of secreted fungal enzymes. The identi-fied β-1,6-glucanase activity (EC is importantfor the mycoparasitic activity and probably cell wall re-cycling by some fungi. The GH5_15 phylogenetic treedisplays two major clades (Additional file 1: Figure S1:GH5_15). The largest clade is formed by enzymes issuedfrom fungi from the class of Sordariomycetes. Eurotio-mycetes are found the second subgroup.GH5_16 : (fungal) endo-β-1,6-galactanase (EC is another example of a monospecific subfamily ofsecreted enzymes where only a single fully sequenced bio-chemically characterized β-1,6-galactanase (EC iscurrently known [32], although partial N-terminal sequenceof an Aspergillus enzyme, closely related to sequences ofother members of the subfamily, yields the same activity[33]. Several other enzymes with β-1,6-galactanase activityhave been moved from GH5 to GH30_5 [23], but subfamilyGH5_16 clearly remains within family GH5. The knownβ-1,6-galactanase (EC is involved in larch woodarabinogalactan degradation [32].GH5_21 : (bacteroidetes) endo-β-1,4-xylanase (EC3.2.1.8) Xylanase (EC activity has been recentlyestablished [34] for a number of xylanolytic bacteroi-detes enzymes belonging to this subfamily. TheseGH5_21 endo-xylanases integrate xylan utilization geneclusters found in Prevotella and Bacteroides species andare all apparently secreted. Significant differences intheir mode of action have been observed, despite the in-clusion in the same subfamily. Different GH5_21enzymes were shown to release different products fromwheat arabinoxylan [34].GH5_22 GH5_31, GH5_34, GH5_39, and GH5_53 :single β-glycanase characterizations In addition tosubfamilies GH5_14 and GH5_16 described above, fiveother subfamilies are distinguished by harboring only asingle experimentally characterized enzyme. Endo-β-1,4-glucanase activity (EC has been determined insubfamilies GH5_22 and GH5_39. GH5_31 is a smallsubfamily currently restricted to secreted fungal pro-teins. A β-1,3-(gluco)mannanase activity (EC 3.2.1.-)ahas been reported for an enzyme from Paecilomyces lila-cinus [35]. In the small subfamily GH5_34, composed ofextracellular modular proteins from bacterial origin,there is a single enzymatically and structurally character-ized enzyme from Clostridium thermocellum. Notably,although this is the first reported enzyme with arabinox-ylanase activity (EC 3.2.1.-)b it was designated CtXyl5Ain spite of its inability to attack unsubstituted xylans[36]. Finally, the small subfamily GH5_53 is a modularextracellular subfamily that contains a single character-ized cellodextrinase (EC [37].GH5_27 GH5_28 and GH5_29 : endo-glycosylcerami-dases (EC GH5_27, GH5_28 and GH5_29are subfamilies exclusively containing extracellular endo-glycosylceramidases. Subfamily GH5_27 is formed ofsequences of eukaryotic origin while the small subfam-ilies GH5_28 and GH5_29 are bacterial. Interestingly,the first subfamily is found among the four subfamiliesthat contain metazoan GH5 enzymes. All but one of theenzymes found in GH5_28 are from Actinobacteria. Thecrystal structure of a Rhodococcus endo-glycoceramidaserevealed an active site channel atypical for GH5enzymes, which explains the unusual substrate for thistype of enzymes [38]. All the GH5_29 sequencesreported here are from the genus Rhodococcus. Thecharacterized bacterial enzymes of GH5_28 hydrolyzeganglio- and lacto-series glycosphingolipids. In contrast,the only GH5_29 enzyme investigated is not capable ofhydrolyzing these substrates. Instead, this enzyme showsactivity against 6-gala series glycosphingolipids andthe designation oligogalactosyl-N-acylsphingosine 1,1’-β-galacto-hydrolase has been proposed [39].Aspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 8 of 16 : cellodextrinases (EC Two enzymesin the small intracellular bacterial GH5_52 subfamilyexhibited cellodextrinase activity (EC Inaddition two enzymes isolated from the cow rumen havebeen shown to have hydrolytic activity on carboxymethylcellulose (CMC) agar [14].Polyspecific subfamiliesSome GH5 subfamilies group together a panel of activ-ities and are described as polyspecific subfamilies. Theapparent plasticity of these subfamilies suggests thatonly few subtle changes could be sufficient to switchfrom one activity to the other. More likely, many of theenzymes present on the subfamily level are polyspecificto some extent and therefore should have more than asingle EC attributed.GH5_1 and GH5_2 : β-1,4-glucan cleaving enzymesExtracellular enzymes from archaea, bacteria and uncul-tured symbiotic protists are represented in subfamilyGH5_1 (Additional file 1: Figure S1: GH5_1). The activ-ity observed for most characterized GH5_1 enzymes isendo-β-1,4-glucanase activity (EC The apparentexception in subfamily GH5_1 is an exo-acting cellobio-hydrolase activity (EC from Clostridium ther-mocellum [40]. However, one should note that thebiochemical distinction of exo- versus endo-acting cellu-lases is particularly difficult to establish experimentally.Interestingly, an enzyme from Ruminococcus albus ableto cleave CMC and glucomannan but particularly activeon lichenin was recently described [41]. Many proteinsin this subfamily are modular (data not shown), a featureshared with many members of subfamily GH5_8 asdescribed previously.Subfamily GH5_2 is currently the largest in familyGH5. This subfamily of extracellular enzymes, many ofwhich are multimodular, contains a large number ofcharacterized members that display endo-β-1,4-gluca-nase activity (EC These endo-glucanases are dis-tributed across the subfamily tree and are found in everymajor clade Additional file 1: Figure S1: GH5_2). Oneendo-glucanase from Fibrobacter succinogenes S85 inthis subfamily was reported to be active both on CMCand oat spelt xylan [42], but xylanase activity has notbeen observed in other members thus far. Interestingly,one representative of this subfamily has been reported asa chitosanase (EC with transglycosylation ac-tivity [43]. A bifunctional cellulase/chitosanase has alsobeen identified in Bacillus sp. NBL420 [44] in a differentclade. Significantly, a closely related N-terminal se-quence of a bifunctional cellulase/chitosanase from Myx-obacter sp. AL-1 [45], suggests that a specific subgroupbearing both activities may exist. As for subfamilyGH5_1, many members of this subfamily are multimod-ular, having both CBMs and cellulosomal-like dockerins.GH5_4 : (xyloglucan-specific) endo-β-1,4-glucanases(EC and EC, licheninases (EC3.2.1.73), and xylanases (EC Subfamily GH5_4members are typically extracellular bacterial enzymes, al-though some members come from ciliates and fungi,predominantly rumen organisms. In total, four enzymeactivities have been reported for subfamily GH5_4. Thusfar, GH5_4 is the only subfamily containing enzymeswith reported xyloglucanase activity (EC [46].Interestingly, the xyloglucanases are found in two differ-ent clades of the GH5_4 subfamily tree suggesting thatthe switch of enzyme activity inside this subfamily oc-curred at different times. (Additional file 1: Figure S1:GH5_4). Licheninases (EC have also beendescribed in this subfamily. Significantly, endo-β-1,4-xylanase activities (EC have been reported for afew enzymes, but always in conjunction with other activ-ities. For example, the xylan degrading specific activityof Clostridium cellulovorans EngB and EngD are both ofapproximately 14% of their respective specific activitieson lichenan [47]. Such features suggest an important de-gree of enzyme promiscuity given the structural similar-ity of the β-linked substrates. However, the mostcommonly reported EC number for representatives ofGH5_4 is EC Except for a few fungal pathogenmembers of the subfamily that bear a CBM1 module, noother known CBM is present, in sharp contrast to whathas been found for extracellular subfamilies GH5_1 andGH5_2.GH5_7 : β-1,4-mannan-cleaving enzymes (EC EC In subfamily GH5_7, previouslynamed A7 (originally comprised of only eukaryotic man-nanases but now also containing archaeal and bacterialmembers) [18], the three reported enzyme activities areassociated to the degradation or to the modification ofβ-mannan-containing polysaccharides. Virtually all exa-mined GH5_7 enzymes possess endo-β-1,4-mannanaseactivity (EC One exception is the tomato pro-tein LeMan4a, which in addition to hydrolytic activity,can act in vitro as a mannan transglycosylase (EC 2.4.1-)[48]. Recently, mannan transglycosylase activity was alsoreported for two fungal GH5_7 enzymes [49]. It is notimpossible that further GH5_7 enzymes may reveal atransglycosylase activity in the future, since the distinc-tion between hydrolytic and transglycosylase activity isdictated by the tendency of the glycosyl-enzyme inter-mediate to be intercepted by a water molecule or asaccharide molecule, respectively. Another interestingcase among GH5_7 members is the β-mannosidaseCmMan5A, which is able to release mannose from theAspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 9 of 16 end of manno-oligosaccharides and -poly-saccharides (EC and is thus exo- and not endo-acting. This difference was explained by the length ofthree loops which modify the active center accessibility[50]. Furthermore, this subfamily has both extracellularand intracellular enzymes. Many GH5_7 extracellularenzymes from fungi contain a CBM. In many bacteria,CBMs from different families are also found appendedto GH5_7 catalytic modules (data not shown).GH5_9 : fungal cell wall modifying enzymes This sub-family contains only sequences of fungal origin puta-tively found in different cell locations: some aresecreted, several present a GPI-anchor, others have sin-gle transmembrane segments and yet others appear tobe intracellular. The activity exo-β-1,3-glucanase (EC3.2.1.58) has been described for most of the character-ized examples, all found among the apparently secretedenzymes. Structural investigation of the Candida albi-cans Exg protein provides a clue to the evolution ofthese exo-hydrolases; the structure reveals an active sitepocket shaped for cleavage of β-1,3- but not β-1,4 glyco-sidic linkages [51]. Surprisingly, the periplasmic Exg1protein from Schizosaccharomyces pombe was demon-strated to be an endo-β-1,6-glucanase (EC,while the membrane-anchored protein Exg2 protein wasshown to produce cell wall material when over-expressed[52]. Two of the characterized GH5_9 exo-β-1,3-glu-canases have been also described as β-glucosidases(EC [53].GH5_12 : β-glucosylceramidases (EC and(flavonoid) β-glucosidase (EC Previouslyknown to contain a flavonoid β-glucosidase (EC yeast [53], the panorama of specificities in this sub-family recently expanded to include several fungal β-glucosylceramidases (EC [54]. The subfamily isgrouped into a fungal clade, including three Cryptococ-cus sequences (one now characterized), and a cladedominated by bacterial enzymes (Additional file 1:Figure S1: GH5_12).GH5_23 : fungal β-diglycosidases (EC andEC Subfamily GH5_23 is composed of secretedfungal proteins, two of which have been characterized as β-diglycosidases that break down plant diglycoconjugated fla-vonoids. Deglycosylation of these compounds most ofteninvolves the sequential action of two β-glycosidases in con-trast to the one-step hydrolytic release of the disaccharidemoiety from the aglycone by β-diglycosidases [55]. Thecharacterized enzymes are a hesperidin 6-O-α-L-rhamno-syl-β-glucosidase (EC from Stilbella fimetariafrom which a partial sequence has been obtained [56] and areported β-primeverosidase (EC from Penicilliummulticolor TS-5 [57]. This subfamily is present in the gen-era Aspergillus and Penicillium known for interaction withplants.GH5_25 : endo-β-1,4-glycanases (EC and EC3.2.1.78) Most enzymes found in subfamily GH5_25 arederived from thermophiles. Interestingly, characterizedenzymes in this subfamily represent examples of GH5enzymes with multiple activities. For instance, Cel5Afrom Thermotoga maritima exhibits activity on both β-mannan-based and β-glucan-based polymers. Analysesof the TmCel5A structure have highlighted features im-portant both for the nature of the duality and the ther-mostability [58,59].GH5_26 : endo-β-1,4-glycanases (EC and EC3.2.1.73) GH5_26 is a small subfamily with a majority ofsequences from uncultured microorganisms. The dom-inating activity found is endo-β-1,4-glucanase (EC3.2.1.4), but one enzyme has also high activity againstlichenan (EC [60].GH5_36 endo-b-1,4-glycanases (EC and EC3.2.1.78) Several bacterial phyla are represented in sub-family GH5_36. One enzyme has a demonstrated endo-β-1,4-mannanase activity (EC, whereas a secondenzyme exhibits licheninase activity (EC inaddition to β-mannanase activity [61,62]; a 3-D structureis available for the latter enzyme.GH5_37 : endo-β-1,3/4-glycanases (EC and EC3.2.1.73) + cellodextrinase (EC Three differ-ent activities are found in subfamily GH5_37, whichconsists of sequences of bacterial origin encoding intra-cellular proteins. The majority of the characterizedenzymes are cellulases (EC, but there are alsoexamples of licheninase activity (EC [41], andcellodextrinase activity (EC [63].GH5_38 and GH5_46 : bacterial enzymes active onmodel plant cell wall (PCW) compounds In subfamilyGH5_38, three enzymes isolated from rumen metage-nomic projects have been partially characterized [12,14].To these, we may add a Prevotella ruminicola 23 enzymedescribed as cellulase (PRU_1856) and shown to be ac-tive against CMC, Avicel, and lichenan [64]. Another en-zyme discovered in the microbial community of the cowrumen is active on CMC and belongs to subfamilyGH5_46 [14]. This is the single evidence of activity inthis subfamily.GH5_48 : bacterial enzymes active on chitin and chit-osan derivatives Several bacterial phyla are currentlyrepresented in subfamily GH5_48, mostly composed ofAspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 10 of 16 and membrane-anchored proteins of un-known function. A single member, a partial sequence(not shown in trees because incomplete) coding for aprotein from Pseudomonas putida P3(4)), over 98% iden-tical to locus PPS_1333 (GenBank accession AEJ11908)from Pseudomonas putida S16, has been described as abifunctional enzyme as it was active on preparativeforms of chitin and colloidal chitosan and on themodel compounds pNP-β-N-acetylglucosaminide and 4-methylumbelliferyl-N-acetyl β-D-glucosaminide [65]. Gi-ven that chitosanases are already present in family GH5,related activities are not unexpected. Interestingly, onesequence in GH5_48 seems to lack the catalytic machin-ery (Additional file 1: Figure S1).Subfamilies lacking experimental characterizationA large number of subfamilies still require evidence ofenzyme activity (Table 1). A large characterization effortis still needed to identify the hidden activities of thesesubfamilies. For a number of subfamilies, particularly forthose containing bacterial enzymes, direct hints of theactivities to be tested can be obtained from the add-itional modules present, such as CBMs, but also fromoperon-like organizations. But the contribution of otherTable 2 Characterized carbohydrate-active enzymes of family GH5 not yet classified into subfamiliesDescription EC Assay Organism Accessions ModularstructureTaxonomic classβ-mannanase A (ManA;CelA) Caldanaerobiuspolysaccharolyticus KM-THCJAAD09354 GH5 CBM16CBM16B-Firmicutes_Clostridiaendo-β-1,4-glucanase D (CelD;CelCCD; EGCCD; Ccel_0840) Clostridium cellulolyticum H10 [B] BAA14354ACL75216GH5 CBM11DOCB-Firmicutes_Clostridiaendo-β-1,4-glucanase/b-1,3:1,4-glucanase H (CelH) thermocellum NCIB10682AAA23225 GH26 GH5CBM11 DOCB-Firmicutes_Clostridiacellulase (EBI-244) Desulfurococcaceae archaeonEBI-244AEB53062 GH5 A-Crenarchaeotaendo-β-1,4-glucanase 3 (Cel3;Cel-3; Eg3; Fisuc_2230; FSU_2772) Fibrobacter succinogenes subsp.succinogenes S85AAA24893ACX75816ADL25000GH5 B-Fibrobacteres_AcidobacteriagroupFisuc_2933/FSU_0196 AGM Fibrobacter succinogenes subsp.succinogenes S85ACX76513ADL26912GH5 CBM4 B-Fibrobacteres_AcidobacteriagroupFisuc_1523/FSU_2005 MUC Fibrobacter succinogenes subsp.succinogenes S85ACX75120ADL26743GH5 B-Fibrobacteres_Acidobacteriagroupendoglucanase (CelA; lpg1918) AHEC Legionella pneumophila subsp.pneumophila str. Philadelphia 1AAU27988 GH5 B-Gammaproteobacteriaendo-β-1,4-glucanase 5B(Sde_2490) Saccharophagus degradans 2-40 ABD81750 CBM6 GH5 B-Gammaproteobacteriaendo-β-1,4-glucanase 5E(Sde_2929) Saccharophagus degradans 2-40 ABD82186 CBM6 CBM6GH5B-Gammaproteobacteriaendo-β-1,6-glucanase (Exg3;SPBC2D10.05) Schizosaccharomyces pombe972 h-CAA21163NP_596224GH5 E-Fungiendo-β-1,4-glucanase (Cel5G) uncultured bacterium ADD71777 GH5 B-environmental samplesSARM_0034/694713_55880/TW39 LICCMCuncultured organism ADX05705 GH5 U-unclassified sequencesSARM_0047/1057205_158590/TW-15CMC uncultured organism ADX05718 GH5 U-unclassified sequencesSARM_0086/0_06533/TW-18 PCW uncultured organism ADX05761 GH5 U-unclassified sequencesβ-glucanase (RR.06; RR.06-1; BglC) BBGCel5LICunidentified microorganism CAJ19140 GH5 U-unclassified sequencesβ-glucanase (RR.10; RR.10-1) BBG unidentified microorganism CAJ19146 U-unclassified sequencesThe active enzymes are tagged by their Enzyme Classification (EC) number (see Table 1) or by the significant positively assayed substrates that are: AEHC - AZCL-HE cellulose; AGM - AZCL-galactomannan; BBG – barley β-glucan; Cel5 – cellopentose; CMC - carboxymethyl cellulose; LIC –lichenan; MUC - 4-methylumbelliferyl-β-D-cellobioside; PCW – plant cell wall.Aspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 11 of 16 indirect approaches like transcriptomics and func-tional metagenomics using innovative sets of conditionsand enlarged classes of substrates are equally potent toprovide clues to enzyme function. The uncharacterizedsubfamilies are distributed across the GH5 phylogenetictree. However, in one of the three major clades, a largeblock of uncharacterized subfamilies (GH5_13, GH5_18,GH5_19, GH5_30, GH5_41 and GH5_42) is located be-tween characterized subfamilies containing β-mannan-acting enzymes (GH5_7, GH5_10, GH17, and GH5_31)(Figure 1). It is tempting to speculate that at least someof these uncharacterized subfamilies target substratesrelated to β-mannan.Unclassified GH5 proteinsRoughly 20% of the analyzed GH5 sequences were notassigned to subfamilies, although some of these proteinshave been characterized to different levels. The mainreason for the inability to assign them to subfamilies wasthe lack of sufficiently related sequences to define a sub-family with at least five members. This is likely tochange in the future due to a predicable increase in thenumber of available sequences and these unclassifiedsequences represent a pool out of which many moresubfamilies will emerge. A total of 17 characterizedenzymes were identified in this diverse set of non-classified sequences (see Table 2). Interestingly, twomain groups arise: (i) post-genomic enzyme characteri-zations, and (ii) enzymes originating from functionalmetagenomics-based discovery and characterization.Here, the cytoplasmic endo-β-1,6-glucanase Exg3 fromthe model organism Schizosaccharomyces pombe [52]is a representative of the former efforts focused on theidentification of fundamental activities. On the otherhand, the archael multidomain hyperthermophilic cellu-lase EBI-244 screened for its ability to degrade CMCat high temperature represents the latter category ofefforts [66].Non-catalytic GH5 subfamilies and GH5 modulesInterestingly, the analysis subjacent to the subfamilyclassification revealed proteins that are likely to be cata-lytically inactive (or perhaps possess alternate mechan-isms and activities to the canonical GHs), due toincomplete catalytic machinery (see Figure 2d). GH5_30is the only subfamily where all sequences are lacking theessential amino acids for GH activity. In addition, somemembers of other groups also appear to have lost theircatalytic machinery, such as a Xanthomonas-specificsubgroup that appears to rapidly emerge from subfamilyGH5_1 (Additional file 1: Figure S1). All three membersof this subgroup present an architecture where the in-active GH5 module is appended to an expansin moduleand an adjacent CBM63 module at the C-terminus.Although no catalytic chemical activity has been identi-fied for expansin modules, it is significant that theknockout of CelXoB renders Xanthomonas oryzae pv.oryzae KACC 10331 avirulent [67]. Another emergingnon-catalytic GH5 subfamily is closely related to sub-family GH5_8. All three members of this new subgroupare lipoproteins that combine the apparently catalyticallyinactive GH5 module with a large C-terminal extension.These two subgroups share long branches in the com-mon GH5 tree suggesting rapid evolution. Interestingly,a single member of subfamily GH5_48 has lost the cata-lytic machinery. Besides this loss, it is however still simi-lar to other members of the subfamily. Whether this is asnapshot of the early steps of a new arising function or asequencing error it is premature to say. Finally, the onlydistant GH5 member having lost its catalytic acid–basethat has been functionally characterized is the putativecarbohydrate biosensor Rsi24C-GH5 from Clostridiumthermocellum ATCC 27405 [68]. Although its catalyticactivity was lost, the extracellular GH5 module wasshown to interact with crystalline cellulose so that a rec-ognition signal could be conveyed by to its intracellularN-terminal anti-σ factor. The losses of the catalyticmachinery here described convey that family GH5sequences are also subject to recurring evolution thatleads to novel functions. This type of evolutionary eventhas been described previously in other GH families. Forinstance, amino acid transporters derived from ancestralα-amylases are found in family GH13 [24], inactivatedchitinases evolved into xylanase inhibitors in familyGH18 [69], and mammalian lactalbumins, which arerelated to GH22 lysozymes, are all well-known examplesof the recent evolution of glycosidases to acquire novelfunctionalities [70].ConclusionsWhen the first five historical subfamilies of GH5 wereestablished in 1990, the total number of GH5 proteinsequences was 21 [16]. More than 20 years later (August2012), this number has increased approximately 150times to exceed 3,200. The practical difficulties of hand-ling such large datasets notwithstanding, this abundanceof sequences is both a boon and a bane for phylogeneticanalysis and functional prediction.Assigning proteins to a large GH-family, like GH5,which harbors multiple specificities and activities, doesnot unlock the full potential of sequence-based classifi-cation. Thus, one aim of the present investigation was toobtain an improved correlation between proteinsequences and catalytic specificity by refining a finerhierarchical level, the subfamily, for GH5 members. Upto 80 percent of the existing GH5 sequences were segre-gated into 51 subfamilies. Of these subfamilies, a total of31 contained at least one member characterized to someAspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 12 of 16, whereas 20 lacked enzymatically-characterizedmembers altogether. Out of the 31 subfamilies charac-terized to some extent, 17 were monospecific and elevenwere polyspecific (containing two or more enzyme activ-ities). Nonetheless, one activity typically predominateswithin polyspecific subfamilies. Interestingly, both endo-and exo-acting enzymes have been observed in the samesubfamily, e.g. GH5_1, GH5_7 and GH5_9, illustratingthat (if real) the two types of activities reflect details ofthe three-dimensional structures. As a consequence ofthe canonical double displacement mechanism employedby GH5 enzymes, which involves the formation of a co-valent glycosyl-enzyme intermediate, GH5 members canpotentially catalyze transglycosylation in addition to, orinstead of, hydrolysis [71]. Although the amount of bio-chemical data is presently limited, we observed that sub-family classification in GH5 does not appear to correlatewith transglycosylation activity, thus indicating that thisproperty is also a consequence of subtle protein struc-tural details.This effort provides a first comprehensive view of thecoverage and distribution of the curated set of 400 ex-perimentally characterized enzymes in the GH5 familyand couples this information with an extensively up-dated sequence-based GH5 subfamily division. In par-ticular, it provides insights into the evolution of GH5proteins, and the classification results can be used to as-sist in candidate protein selection for enzyme discoveryand bioprospecting projects. For instance, both themembers from the twenty defined subfamilies lackingfunctional characterization, as well as the numerousphylogenetic outliers, provide a vast number of interest-ing targets for future studies. In particular, although asignificant amount of tertiary structural data is alreadyavailable for GH5, the present work highlights that alarge number subfamilies would benefit from a 3-Dstructure for at least one subfamily member. Moreover,the data presented here, and available at the CAZy data-base [11] as a community resource, will serve as a guidefor protein engineering approaches exploiting the diverseactivities found within the GH5 family.Finally, in the present climate in which sequence datais literally flooding public databases, incorrect proteinfunction annotations are too easily propagated by auto-mated computer-based prediction methods, therebyjeopardizing the usefulness of these annotations. Increas-ingly rapid sequence accumulation is worsening the sce-nario. This problem is particularly illustrated by theGH5_11 subfamily of plant and fungal proteins that areannotated as cellulases in public databases, including thewidely-used Arabidopsis Information Resource [72], des-pite a complete lack of experimental support for any oneof its members. Such excesses of over-annotation equallyaffect the presumed non-catalytically active proteins andsubfamilies. For example, subfamily GH5_30 further ex-emplifies the pitfalls of automated (mis-)annotation: sev-eral members are publicly annotated as mannanases,although they lack the conserved catalytic machinery ofthe family. To avoid such error propagation, we stronglyadvocate designating all predicted enzymes as “GH5_n”(where n is the subfamily number) until an activityhas been rigorously demonstrated by biochemicalexperimentation.The GH5 subfamily classification presented here pro-vides a framework to sort family members into meaning-ful, predictive categories. By taking a conservativeapproach to protein annotation, this method offers arigorous strategy to avoid misleading functional predic-tion in large-scale genomic sequencing projects. Whilstthe subfamilies described herein generally act on a singlesubstrate (seventeen monospecific subfamilies wereidentified), it is important to stress that precise details ofglycoside hydrolase function, such as the extent of endo-vs. exo- modes of cleavage or the transglycosylation-to-hydrolysis ratio is unlikely to be predictable from se-quence alone. We therefore recommend that such over-reaching predictions be altogether abandoned in gen-omic sequence annotation. To aid and advance globalefforts in de novo sequence annotation, the GH5 sub-family classification scheme is now publicly available atthe CAZy database [11].MethodsThe GH5 subfamilies were defined using the methodsdescribed for the subfamily classification of GH13 and allPL families [24,25], which are briefly summarized here.After an initial removal of obviously incomplete and/or er-roneous sequences, a total of 2347 full length GH5 cata-lytic module sequences were retrieved from the CAZydatabase (October 2011) and subdivided into two sets.One set of 414 sequences contained all GH5 modulesfrom biochemically characterized and sequences positivelytested in activity tests against a variety of substrates. Thesecond set was composed of 1957 non-characterizedsequences The latter subset was clustered at 75% identityusing UCLUST4.0, a part of the USEARCH 4.0 package[73], and was reduced to 971 sequences. When combinedwith the 414 GH5 module sequences from a biochemicallycharacterized and positively assayed subset we obtained atotal of 1385 sequences. These sequences were alignedusing MUSCLE 3.7 [74] in two steps. An initial alignmentwas performed and its quality visually inspected so thatthe remaining incomplete and problem sequences wereidentified and edited or removed. This procedure ensuredthat the GH5 module boundaries were clear and that thesequences were trimmed if required, and that a majorityof the residues constituting the catalytic site was presentor that the alignment was not ambiguous. A final set ofAspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 13 of 16 remaining sequences were then realigned using thesame procedure. The eliminated, often fragmentary,sequences were used to complement biochemical activityinformation when relevant at later stages.The resulting multisequence alignment of familyGH5 catalytic domain sequences was used to infer anapproximate-maximum-likelihood phylogenetic tree withFASTTREE 2.1 [75], a program adapted to the analysis oflarge sequence sets, using the Whelan Goldman model ofamino acid evolution, the gamma option to rescale thebranch lengths and compute a Gamma20-based likeli-hood, a total of four rounds of minimum-evolution moves,and options to make the maximum-likelihood nearest-neighbor interchanges more exhaustive. The identifica-tion and tagging of subfamilies followed a multi-stepprocedure. First, the tree was analyzed to tag the differ-ent sequences and nodes corresponding to the first 10“historical” subfamilies (A1 to A10, described in theIntroduction). These steps were performed to ensurecontinuity in subfamily definitions. For the remainingsequences in the tree, distinct nodes correspondingpotential subfamilies were visually identified and theirconsistency checked using a procedure similar to thatdescribed [24]. Within each of these groups, differentstarting sequences were selected and manual BLAST2queries [76] performed against all the sequences found inCAZy in order to identify self-contained ensembles andestablish subfamily limits. This analysis ensured that theeach subfamily that was retained was singular and that theremoval of sequences by the initial UCLUST filtering pro-cedure and of fragmentary sequences did not introduceany bias. Finally, only subfamilies containing at least fivesequences from different organisms were considered.Subsequently, for each defined subfamily, maximumlikelihood (ML) phylogenetic trees were built by usingPhyML [77], and the reliability of the inferred relation-ships the trees was tested by bootstrap analysis using100 resamplings of the data set.EndnotesaNo EC number is presently available for β-1,3-(gluco)mannanase activity.bNo EC number is presently available for arabinoxyla-nase activity.Additional fileAdditional file 1: Figure S1. Rectangular phylogram view of thephylogenetic tree of family GH5. Branches corresponding to subfamilies1–53 are shown in color and the individual subfamilies have theircorresponding subfamily numbers as indicated in Figure 1. The branchescorresponding to sequences not included into subfamilies are in black.Each individual protein module node is identified by a varying number offields separated by “|” indicating: (i) the organism, with 3 letters for thegenre and either 5 letters for the species or full strain code; (ii) theprotein accession in public databases, typically GenBank; (iii) if attributed,the subfamily number or other information; (iv) if available, EC numbers(node in bold) or a “*” (node in bold and italic) to indicate preciseenzyme characterizations or a simple activity tests, respectively. A suffixlike “_2” may indicate the module position if more than one GH5 moduleis present on peptide. Lower confidence nodes with a SH-like localsupport below 0.7 (varying from low 0 to strong 1) are indicated with ablack dot. Identified sequences without complete catalytic machinery arein red. Individual subfamily trees are also included in this file.Competing interestsThe authors declare they have no competing interests.Authors’ contributionsPedro M. Coutinho performed the global phylogenetic analysis; HenrikAspeborg and Yang Wang performed the phylogenetic analyses of eachsubfamily; Bernard Henrissat and Pedro M. Coutinho collected and curatedall GH5 sequences used in the present analysis; Harry Brumer and BernardHenrissat designed research and analyzed the data. The paper wascollectively written. All authors read and approved the final manuscript.AcknowledgementsWe wish to thank Franz St John for his work on family GH30, whichstimulated us to refine family GH5. Work at UMR 7257 of Aix-MarseilleUniversité was funded by ANR grant BIP:BIP (ANR-10-BINF-03-04). Work in theDivision of Glycoscience, KTH Biotechnology was funded by an IngvarCarlsson Award to H.A. from the Swedish Foundation for Strategic Research.The Swedish Research Council Formas provided additional funding viaCarboMat – The KTH Advanced Carbohydrate Materials Consortium (AFormas Strong Research Environment). H.B. gratefully acknowledges facultysupport from The Michael Smith Laboratories and the University of BritishColumbia.Author details1Division of Glycoscience, School of Biotechnology, KTH - Royal Institute ofTechnology, AlbaNova University Center, Stockholm SE-106 91, Sweden.2Michael Smith Laboratories and Department of Chemistry, University ofBritish Columbia, 2185 East Mall, Vancouver V6T 1Z4, Canada. 3Architecture etFonction des Macromolécules Biologiques, Aix-Marseille Université, CNRS,UMR 7257, 163 Avenue de Luminy, Marseille 13288, France.Received: 22 May 2012 Accepted: 13 September 2012Published: 20 September 2012References1. Varki A: Essentials of glycobiology. 2nd edition. Cold Spring Harbor, N.Y.: ColdSpring Harbor Laboratory Press; 2009.2. Lichtenthaler FW: Carbohydrates as renewable raw materials: a majorchallenge of green chemistry. In Methods and reagents for green chemistry:an introduction. Edited by Tundo P, Perosa A, Zecchini F. Hoboken, NJ: J.Wiley; 2007:23–63.3. Laine RA: A calculation of all possible oligosaccharide isomers bothbranched and linear yields 1.05x10(12) structures for a reducinghexasaccharide: the isomer barrier to development of single-methodsaccharide sequencing or synthesis systems. Glycobiology 1994,4:759–767.4. Sharon N: The conquest of the last frontier of molecular and cell biology.Foreword Biochimie 2001, 83:555–555.5. Anonymous: 10 Emerging Technologies That Will Change the World.Technology Review 2003, 33–51.6. Davies GJ, Gloster TM, Henrissat B: Recent structural insights into theexpanding world of carbohydrate-active enzymes. Curr Opin Struct Biol2005, 15:637–645.7. Nomenclature Committee of the International Union of Biochemistry andMolecular Biology (NC-IUBMB). Henrissat B: A classification of glycosyl hydrolases based on amino-acid-sequence similarities. Biochem J 1991, 280:309–316.9. Davies GJ, Sinnott ML: Sorting the diverse: the sequence-basedclassifications of carbohydrate-active enzymes. Biochem J 2008, 1–5.doi:10.1042/BJ20080382 (online only).Aspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 14 of 16 Henrissat B, Claeyssens M, Tomme P, Lemesle L, Mornon JP: Cellulasefamilies revealed by hydrophobic cluster-analysis. Gene 1989, 81:83–95.11. The Carbohydrate-Active enZYme (CAZy) database. Duan CJ, Xian L, Zhao GC, Feng Y, Pang H, Bai XL, Tang JL, Ma QS, Feng JX:Isolation and partial characterization of novel genes encoding acidiccellulases from metagenomes of buffalo rumens. J Appl Microbiol 2009,107:245–256.13. Elifantz H, Waidner LA, Michelou VK, Cottrell MT, Kirchman DL: Diversityand abundance of glycosyl hydrolase family 5 in the North AtlanticOcean. FEMS Microbiol Ecol 2008, 63:316–327.14. Hess M, Sczyrba A, Egan R, Kim TW, Chokhawala H, Schroth G, Luo SJ, ClarkDS, Chen F, Zhang T, et al: Metagenomic discovery of biomass-degradinggenes and genomes from cow rumen. Science 2011, 331:463–467.15. Cantarel BL, Coutinho PM, Rancurel C, Bernard T, Lombard V, Henrissat B:The Carbohydrate-Active EnZymes database (CAZy): an expert resourcefor Glycogenomics. Nucleic Acids Res 2009, 37:D233–D238.16. Béguin P: Molecular biology of cellulose degradation. Annu Rev Microbiol1990, 44:219–248.17. Lo Leggio L, Parry NJ, VanBeeumen J, Claeyssens M, Bhat MK, Pickersgill RW:Crystallization and preliminary X-ray analysis of the majorendoglucanase from Thermoascus aurantiacus. Acta Crystallogr D BiolCrystallogr 1997, 53:599–604.18. Hilge M, Gloor SM, Rypniewski W, Sauer O, Heightman TD, Zimmermann W,Winterhalter K, Piontek K: High-resolution native and complex structuresof thermostable beta-mannanase from Thermomonospora fusca -substrate specificity in glycosyl hydrolase family 5. Structure Fold Des1998, 6:1433–1444.19. Lo Leggio L, Larsen S: The 1.62 angstrom structure of Thermoascusaurantiacus endoglucanase: completing the structural picture ofsubfamilies in glycoside hydrolase family 5. FEBS Lett 2002, 523:103–108.20. Larsson AM, Anderson L, Xu BZ, Munoz IG, Uson I, Janson JC, Stalbrand H,Stahlberg J: Three-dimensional crystal structure and enzymiccharacterization of beta-mannanase Man5A from blue mussel Mytilusedulis. J Mol Biol 2006, 357:1500–1510.21. Costanzo S, Ospina-Giraldo MD, Deahl KL, Baker CJ, Jones RW: Alternateintron processing of family 5 endoglucanase transcripts from the genusPhytophthora. Curr Genet 2007, 52:115–123.22. Opassiri R, Pomthong B, Akiyama T, Nakphaichit M, Onkoksoong T, CairnsMK, Cairns JRK: A stress-induced rice (Oryza sativa L.) beta-glucosidaserepresents a new subfamily of glycosyl hydrolase family 5 containing afascin-like domain. Biochem J 2007, 408:241–249.23. St John FJ, Gonzalez JM, Pozharski E: Consolidation of glycosyl hydrolasefamily 30: A dual domain 4/7 hydrolase family consisting of twostructurally distinct groups. FEBS Lett 2010, 584:4435–4441.24. Stam MR, Danchin EGJ, Rancurel C, Coutinho PM, Henrissat B: Dividing thelarge glycoside hydrolase family 13 into subfamilies: towards improvedfunctional annotations of alpha-amylase-related proteins. Protein Eng DesSel 2006, 19:555–562.25. Lombard V, Bernard T, Rancurel C, Brumer H, Coutinho PM, Henrissat B: Ahierarchical classification of polysaccharide lyases for glycogenomics.Biochem J 2010, 432:437–444.26. Hancock SM, Rich JR, Caines MEC, Strynadka NCJ, Withers SG: Designerenzymes for glycosphingolipid synthesis by directed evolution. NatChem Biol 2009, 5:508–514.27. Danchin EGJ, Rosso MN, Vieira P, de Almeida-Engler J, Coutinho PM,Henrissat B, Abad P: Multiple lateral gene transfers and duplications havepromoted plant parasitism ability in nematodes. Proc Natl Acad Sci USA2010, 107:17651–17656.28. Acuña R, Padilla BE, Flórez-Ramos CP, Rubio JD, Herrera JC, Benavides P, LeeS-J, Yeats TH, Egan AN, Doyle JJ, et al: Adaptive horizontal transfer of abacterial gene to an invasive insect pest of coffee. Proc Natl Acad U S A2012, 109:4197–4202.29. Boraston AB, Bolam DN, Gilbert HJ, Davies GJ: Carbohydrate-bindingmodules: fine-tuning polysaccharide recognition. Biochem J 2004,382:769–781.30. Zhao YJ, Zhang YH, Cao Y, Qi JX, Mao LW, Xue YF, Gao F, Peng H, WangXW, Gao GF, et al: Structural analysis of alkaline beta-mannanase fromalkaliphilic Bacillus sp N16-5: implications for adaptation to alkalineconditions. PLoS One 2011, 6:e14608.31. Suen G, Weimer PJ, Stevenson DM, Aylward FO, Boyum J, Deneke J,Drinkwater C, Ivanova NN, Mikhailova N, Chertkov O, et al: The completegenome sequence of Fibrobacter succinogenes S85 reveals a cellulolyticand metabolic specialist. PLoS One 2011, 6:e18814.32. Sakamoto T, Taniguchi Y, Suzuki S, Ihara H, Kawasaki H: Characterization ofFusarium oxysporum beta-1,6-galactanase, an enzyme that hydrolyzeslarch wood arabinogalactan. Appl Environ Microbiol 2007, 73:3109–3112.33. Luonteri E, Laine C, Uusitalo S, Teleman A, Siika-aho M, Tenkanen M:Purification and characterization of Aspergillus beta-D-galactanasesacting on beta-1,4- and beta-1,3/6-linked arabinogalactans. CarbohydPolym 2003, 53:155–168.34. Dodd D, Moon YH, Swaminathan K, Mackie RI, Cann IKO: Transcriptomicanalyses of xylan degradation by Prevotella bryantii and insights into energyacquisition by xylanolytic bacteroidetes. J Biol Chem 2010, 285:30261–30273.35. Sugino H, Furuichi S, Murao S, Arai M, Fujii T: Molecular characterization ofa Rhodotorula-lytic enzyme from Paecilomyces lilacinus having beta-1,3-mannanase activity. Biosci Biotechnol Biochem 2004, 68:757–760.36. Correia MAS, Mazumder K, Bras JLA, Firbank SJ, Zhu YP, Lewis RJ, York WS,Fontes CMGA, Gilbert HJ: Structure and function of an arabinoxylan-specific xylanase. J Biol Chem 2011, 286:22510–22520.37. Deboy RT, Mongodin EF, Fouts DE, Tailford LE, Khouri H, Emerson JB,Mohamoud Y, Watkins K, Henrissat B, Gilbert HJ, et al: Insights into plantcell wall degradation from the genome sequence of the soil bacteriumCellvibrio japonicus. J Bacteriol 2008, 190:5455–5463.38. Caines MEC, Vaughan MD, Tarling CA, Hancock SM, Warren RAJ, Withers SG,Strynadka NCJ: Structural and mechanistic analyses of endo-glycoceramidase II, a membrane-associated family 5 glycosidase in the Apoand G(M3) ganglioside-bound forms. J Biol Chem 2007, 282:14300–14308.39. Ishibashi Y, Nakasone T, Kiyohara M, Horibata Y, Sakaguchi K, Hijikata A,Ichinose S, Omori A, Yasui Y, Imamura A, et al: A novel endoglycoceramidasehydrolyzes oligogalactosylceramides to produce galactooligosaccharidesand ceramides. J Biol Chem 2007, 282:11386–11396.40. Zverlov VV, Velikodvorskaya GA, Schwarz WH: A newly describedcellulosomal cellobiohydrolase, CelO, from Clostridium thermocellum:investigation of the exo-mode of hydrolysis, and binding capacity tocrystalline cellulose. Microbiology 2002, 148:247–255.41. Iakiviak M, Mackie RI, Cann IKO: Functional analyses of multiple lichenin-degrading enzymes from the rumen Bacterium Ruminococcus albus 8.Appl Environ Microbiol 2011, 77:7541–7550.42. Cho KK, Kim SC, Woo JH, Bok JD, Choi YJ: Molecular cloning and expressionof a novel family A endoglucanase gene from Fibrobacter succinogenes S85in Escherichia coli. Enzyme Microb Technol 2000, 27:475–481.43. Tanabe T, Morinaga K, Fukamizo T, Mitsutomi M: Novel chitosanase fromStreptomyces griseus HUT 6037 with transglycosylation activity. BiosciBiotechnol Biochem 2003, 67:354–364.44. Hong I-P, Jang H-K, Lee S-Y, Choi S-G: Cloning and characterization of abifunctional cellulase-chitosanase gene from Bacillus licheniformisNBL420. J Microbiol Biotechnol 2003, 13:35–42.45. Pedraza-Reyes M, Gutierrez-Corona F: The bifunctional enzymechitosanase-cellulase produced by the gram-negative microorganismMyxobacter sp. AL-1 is highly similar to Bacillus subtilis endoglucanases.Arch Microbiol 1997, 168:321–327.46. Yaoi K, Nakai T, Kameda Y, Hiyoshi A, Mitsuishi Y: Cloning andcharacterization of two xyloglucanases from Paenibacillus sp strainKM21. Appl Environ Microbiol 2005, 71:7670–7678.47. Foong FCF, Doi RH: Characterization and comparison of Clostridiumcellulovorans endoglucanases-xylanases EngB and EngD hyperexpressedin Escherichia coli. J Bacteriol 1992, 174:1403–1409.48. Schroder R, Wegrzyn TF, Sharma NN, Atkinson RG: LeMAN4 endo-beta-mannanase from ripe tomato fruit can act as a mannan transglycosylaseor hydrolase. Planta 2006, 224:1091–1102.49. Dilokpimol A, Nakai H, Gotfredsen CH, Baumann MJ, Nakai N, Abou HachemM, Svensson B: Recombinant production and characterisation of tworelated GH5 endo-beta-1,4-mannanases from Aspergillus nidulans FGSCA4 showing distinctly different transglycosylation capacity. BiochimBiophys Acta 2011, 1814:1720–1729.50. Dias FMV, Vincent F, Pell G, Prates JAM, Centeno MSJ, Tailford LE, FerreiraLMA, Fontes CMGA, Davies GJ, Gilbert HJ: Insights into the moleculardeterminants of substrate specificity in glycoside hydrolase family 5revealed by the crystal structure and kinetics of Cellvibrio mixtusmannosidase 5A. J Biol Chem 2004, 279:25517–25526.51. Cutfield SM, Davies GJ, Murshudov G, Anderson BF, Moody PCE, Sullivan PA,Cutfield JF: The structure of the exo-beta-(1,3)-glucanase from CandidaAspeborg et al. BMC Evolutionary Biology 2012, 12:186 Page 15 of 16 in native and bound forms: Relationship between a pocket andgroove in family 5 glycosyl hydrolases. J Mol Biol 1999, 294:771–783.52. Duenas-Santero E, Martin-Cuadrado AB, Fontaine T, Latge JP, del Rey F, deAldana CV: Characterization of glycoside hydrolase family 5 proteins inSchizosaccharomyces pombe. Eukaryot Cell 2010, 9:1650–1660.53. Schmidt S, Rainieri S, Witte S, Matern U, Martens S: Identification of aSaccharomyces cerevisiae glucosidase that hydrolyzes flavonoidglucosides. Appl Environ Microbiol 2011, 77:1751–1757.54. Ishibashi Y, Ikeda K, Sakaguchi K, Okino N, Taguchi R, Ito M: Quality controlof fungus-specific glucosylceramide in Cryptococcus neoformans byendoglycoceramidase-related protein 1 (EGCrP1). J Biol Chem 2012,287:368–381.55. Mazzaferro LS, Breccia JD: Functional and biotechnological insights intodiglycosidases. Biocatal Biotransfor 2011, 29:103–112.56. Mazzaferro L, Pinuel L, Minig M, Breccia JD: Extracellular monoenzymedeglycosylation system of 7-O-linked flavonoid beta-rutinosides and itsdisaccharide transglycosylation activity from Stilbella fimetaria. ArchMicrobiol 2010, 192:383–393.57. Tsuruhami K, Mori S, Amarume S, Saruwatari S, Murata T, Hirakake J, SakataK, Usui T: Isolation and characterization of a beta-primeverosidase-likeenzyme from Penicillium multicolor. Biosci Biotechnol Biochem 2006,70:691–698.58. Pereira JH, Chen ZW, McAndrew RP, Sapra R, Chhabra SR, Sale KL, SimmonsBA, Adams PD: Biochemical characterization and crystal structure ofendoglucanase Cel5A from the hyperthermophilic Thermotoga maritima.J Struct Biol 2010, 172:372–379.59. Wu TH, Huang CH, Ko TP, Lai HL, Ma YH, Chen CC, Cheng YS, Liu JR, GuoRT: Diverse substrate recognition mechanism revealed by Thermotogamaritima Cel5A structures in complex with cellotetraose, cellobiose andmannotriose. Biochim Biophys Acta 2011, 1814:1832–1840.60. Voget S, Steele HL, Streit WR: Characterization of a metagenome-derivedhalotolerant cellulase. J Biotechnol 2006, 126:26–36.61. Khan MAS, Akbar M, Kitaoka M, Hayashi K: A unique thermostablelichenase from Thermotoga maritima MSB8 with divergent substratespecificity. Indian J Biotechnol 2007, 6:315–320.62. Han YJ, Dodd D, Hespen CW, Ohene-Adjei S, Schroeder CM, Mackie RI, CannIKO: Comparative analyses of two thermophilic enzymes exhibiting bothbeta-1,4 mannosidic and beta-1,4 glucosidic cleavage activities fromCaldanaerobius polysaccharolyticus. J Bacteriol 2010, 192:4111–4121.63. Wang WY, Thomson JA: Nucleotide sequence of the celA gene encodinga cellodextrinase of Ruminococcus flavefaciens Fd-1. Mol Gen Genet 1990,222:265–269.64. Matsui H, Ogata K, Tajima K, Nagamine T, Nakamura M, Aminov R, Benno Y:Cloning, expression, and characterization of a cellulase gene fromPrevotella ruminicola. Anim Sci J 2001, 72:421–426.65. Singh NA, Shanmugam V: Cloning and characterization of a bifunctionalglycosyl hydrolase from an antagonistic Pseudomonas putida strain P3(4). J Basic Microbiol 2012, 52:340–349.66. Graham JE, Clark ME, Nadler DC, Huffer S, Chokhawala HA, Rowland SE,Blanch HW, Clark DS, Robb FT: Identification and characterization of amultidomain hyperthermophilic cellulase from an archaeal enrichment.Nat Commun 2011, 2:375.67. Temuujin U, Kim JW, Kim JK, Lee BM, Kang HW: Identification of novelpathogenicity-related cellulase genes in Xanthomonas oryzae pv. oryzae.Physiol Mol Plant Pathol 2011, 76:152–157.68. Bahari L, Gilad Y, Borovok I, Kahel-Raifer H, Dassa B, Nataf Y, Shoham Y,Lamed R, Bayer EA: Glycoside hydrolases as components of putativecarbohydrate biosensor proteins in Clostridium thermocellum. J IndMicrobiol Biotechnol 2011, 38:825–832.69. Durand A, Hughes R, Roussel A, Flatman R, Henrissat B, Juge N: Emergenceof a subfamily of xylanase inhibitors within glycoside hydrolase family18. FEBS J 2005, 272:1745–1755.70. Qasba PK, Kumar S: Molecular divergence of lysozymes and alpha-lactalbumin. Crit Rev Biochem Mol Biol 1997, 32:255–306.71. Sinnott ML: Catalytic Mechanisms of Enzymatic Glycosyl Transfer. ChemRev 1990, 90:1171–1202.72. The Arabidopsis Information Resource. Edgar RC: Search and clustering orders of magnitude faster than BLAST.Bioinformatics 2010, 26:2460–2461.74. Edgar RC: MUSCLE: multiple sequence alignment with high accuracy andhigh throughput. Nucleic Acids Res 2004, 32:1792–1797.75. Price MN, Dehal PS, Arkin AP: FastTree 2-approximately maximum-likelihood trees for large alignments. PLoS One 2010, 5:9490.76. Altschul SF, Madden TL, Schaffer AA, Zhang JH, Zhang Z, Miller W, LipmanDJ: Gapped BLAST and PSI-BLAST: a new generation of protein databasesearch programs. Nucleic Acids Res 1997, 25:3389–3402.77. Guindon S, Gascuel O: A simple, fast, and accurate algorithm to estimatelarge phylogenies by maximum likelihood. Syst Biol 2003, 52:696–704.doi:10.1186/1471-2148-12-186Cite this article as: Aspeborg et al.: Evolution, substrate specificity andsubfamily classification of glycoside hydrolase family 5 (GH5). BMCEvolutionary Biology 2012 12:186.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at et al. BMC Evolutionary Biology 2012, 12:186 Page 16 of 16


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items