This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
A Novel Family of Lectins Evolutionarily Related to ClassV Chitinases: An Example of Neofunctionalizationin Legumes1[W][OA]
Els J.M. Van Damme*, Raphael Culerrier, Annick Barre, Richard Alvarez,Pierre Rouge, and Willy J. Peumans
Department of Molecular Biotechnology, Laboratory of Biochemistry and Glycobiology, Ghent University,9000 Gent, Belgium (E.J.M.V.D., W.J.P.); Surfaces Cellulaires et Signalisation chez les Vegetaux, Unite Mixtede Recherche, Centre National de la Recherche Scientifique, Universite Paul Sabatier 5546, Pole deBiotechnologies Vegetales, 31326 Castanet-Tolosan, France (R.C., A.B., P.R.); and Department ofBiochemistry and Molecular Biology, University of Oklahoma, Health Sciences Center,Oklahoma City, Oklahoma 73104 (R.A.)
A lectin has been identified in black locust (Robinia pseudoacacia) bark that shares approximately 50% sequence identity withplant class V chitinases but is essentially devoid of chitinase activity. Specificity studies indicated that the black locustchitinase-related agglutinin (RobpsCRA) preferentially binds to high-mannose N-glycans comprising the proximal pentasac-charide core structure. Closely related orthologs of RobpsCRA could be identified in the legumes Glycine max, Medicagotruncatula, and Lotus japonicus but in no other plant species, suggesting that this novel lectin family most probably evolved inan ancient legume species or possibly an earlier ancestor. This identification of RobpsCRA not only illustrates neofunction-alization in plants, but also provides firm evidence that plants are capable of developing a sugar-binding domain from anexisting structural scaffold with a different activity and accordingly sheds new light on the molecular evolution of plant lectins.
Flowering plants express a whole battery of carbo-hydrate-binding proteins commonly known as lectinsor agglutinins. Despite the apparent heterogeneity inmolecular structure and sugar specificity, virtually allknown plant lectins can be classified into seven fam-ilies of structurally and evolutionarily related proteins(Van Damme et al., 1998, 2004). Taking into account theobvious differences in both the overall fold and struc-ture of the carbohydrate-binding sites, it seems likelythat each of these seven sugar-binding domains is thefinal result of a unique evolutionary pathway. Severalmodern plant lectins belong to protein families with anobvious prokaryotic origin. For example, proteins shar-ing reasonable sequence similarity with plant lectinscomprising a ricin B domain (cd00161; pfam00652) or
GNA domains (cd00028; pfam01453) have been iden-tified in bacteria as well as in various nonplant eukary-otes (for a quick overview, see the National Center forBiotechnology Information [NCBI] conserved domains[http://www.ncbi.nlm.nih.gov/Structure/cdd] and thePfam Protein Families database [http://www.sanger.ac.uk/Software/Pfam]). Other plant lectins have nocounterparts in prokaryotes but are clearly related tohomologous proteins or protein domains found inanimals, fungi, or some lower eukaryotes. Heveindomains (cd00035; pfam00187), for instance, are notconfined to plants but are quite common in fungi,indicating that this carbohydrate-binding unit wasalready present in an early common eukaryotic ances-tor. The same applies to the legume lectin domain(pfam00139), which is classified in the same proteinsuperfamily as the animal and fungal vesicular inte-gral membrane protein 36 (VIP36) and the endoplasmicreticulum-Golgi-intermediate compartment 53-kD pro-tein (ERGIC-53). For jacalin-related lectins (pfam01419),the situation is less clear. Recent reports claimed thatthe zymogen granule membrane protein 16 found inmouse, rat, and a few other vertebrates, as well as thelectin from the mushroom Grifola frondosa, belong tothe jacalin family (Nagata et al., 2005). However, theresidual sequence identities are low, suggesting that,even if a common ancestral domain occurred in anearly eukaryote, parallel evolution eventually led todistantly related modern animal and fungal homologsof the plant jacalins. No homologous proteins or corre-sponding genes have hitherto been identified outside
1 This work was supported by the Fund for Scientific Research-Flanders (project no. G.0201.04) and the Research Council of GhentUniversity. The glycan array analysis was conducted by the Protein-Glycan Interaction Core H of the Consortium for Functional Glyco-mics, funded by the National Institute of General Medical Sciences(grant no. GM62116).
The author responsible for distribution of materials integral to thefindings presented in this article in accordance with the policydescribed in the Instructions for Authors (www.plantphysiol.org) is:Els J.M. Van Damme ([email protected]).
[W] The online version of this article contains Web-only data.[OA] Open Access articles can be viewed online without a sub-
662 Plant Physiology, June 2007, Vol. 144, pp. 662–672, www.plantphysiol.org � 2006 American Society of Plant Biologists www.plantphysiol.orgon April 10, 2019 - Published by Downloaded from
the plant kingdom for the amaranthins (pfam07468)and Cucurbitaceae phloem lectins. Although no de-finitive conclusions can be drawn on the basis of cur-rently available data, it is tempting to speculate thatthe latter carbohydrate-binding domains evolved inplants. One of the major problems in unraveling theunderlying mechanisms is the fact that no potentialtemplate protein or peptide can be traced.
Here we report the identification of a novel family ofplant lectins that is structurally and evolutionaryclosely related to proteins that were originally de-scribed as class V chitinases, but, according to thegenerally accepted CAZY classification system (http://afmb.cnrs-mrs.fr/CAZY/index.html), are placed in theglycoside hydrolase 18 (GH18) family (Henrissat andBairoch, 1993), which is an ancient chitinase familyfound in all kingdoms from bacteria to fungi, animals,and plants. Only a relatively small number of plantGH18 chitinases have previously been identified. Mostplant chitinases are classified, indeed, in the GH19family, which so far has been found only in higherplants. GH18 and GH19 plant chitinases not onlydiffer in sequence, but also in hydrolytic mechanismsbecause they operate with retention and inversion,respectively, of the anomeric configuration (Iseli et al.,1996). According to available sequence data, none ofthe plant GH18 enzymes comprises a putative chitin-binding domain in addition to the canonical catalyticdomain. Cloning and characterization of the purifiedprotein revealed that the bark of black locust (Robiniapseudoacacia) contains a lectin that shares high sequenceidentity with class V chitinases but is essentially de-void of chitinase activity. Closely related expressedorthologs and/or corresponding genes were found inseveral other legumes but could not be identifiedoutside the family Fabaceae, indicating that the novellectin might have arisen in an evolutionary recent pastin an ancestor of modern legumes. The identificationof the novel black locust agglutinin provides evidencethat plants are capable of developing a domain withspecific sugar-binding activity from a structural scaf-fold found in an existing protein and, accordingly,provides a well-defined example of neofunctionaliza-tion. Similar conversion of a chitinase into a lectinhas been reported in mammalian systems. However,although both plants and mammals used a homologouschitinase as a structural scaffold, there are importantdifferences in the conversion of a carbohydrate-modifying into a carbohydrate-binding protein.
RESULTS
Bark of Black Locust Contains a Lectin Unrelated to
Legume Lectins That Shares Sequence Similarity withTobacco Class V Chitinase
Analysis of black locust bark extracts depleted fromthe legume lectin-type agglutinins black locust barklectin I (RPbAI; Van Damme et al., 1995b) and the self-
aggregatable lectin robiniagrin (previously calledRPbAII; Van Damme et al., 1995b; Ina et al., 2005)revealed the presence of an additional lectin. Prelim-inary experiments indicated that this lectin was asso-ciated with a protein that eluted with a lower apparentmolecular mass (70 kD) than RPbAI/robiniagrin (120kD) from a gel filtration column and consisted almostexclusively of a polypeptide that was at least 5 kDlarger than the RPbAI and robiniagrin subunits. Massspectrometry of the purified lectin yielded a majorpeak of 36,790 D (see Supplemental Fig. S1). Gelfiltration of the purified protein confirmed that thenative lectin eluted with an apparent molecular massof approximately 70 kD. Taking into account that apreviously isolated homolog of class III chitinase frombanana (Musa spp.; which is also classified in theGH18 family) behaved as a monomeric 30-kD proteinwhen run under identical conditions (Peumans et al.,2002), one can reasonably assume that the novel lectinis a homodimer. This conclusion is supported by thehemagglutinating activity of the lectin (because cross-linking of cells requires multivalency). No covalentlybound carbohydrate could be detected by the phenolsulfuric acid method, indicating that the 70-kD lectin isnot glycosylated. N-terminal sequencing revealed thatthe subunit of the 70-kD lectin shares no sequencesimilarity with RPbAI or any other legume lectin, butcan readily be aligned with the N terminus of a class Vchitinase from tobacco (Nicotiana tabacum; Heitz et al.,1994; Melchers et al., 1994; Fig. 1). Edman degradationof a cyanogen bromide cleavage fragment yielded asequence that could be aligned with an internal se-quence of the tobacco class V chitinase, further sup-porting the idea that the 70-kD black locust agglutininis not related to the legume lectins but shares a strikingsequence similarity with class V chitinases.
Molecular Cloning Confirms That the Novel Lectin IsClosely Related to Class V Chitinases
BLAST searches revealed that several other legumesexpress proteins comprising sequences nearly identi-cal to the N-terminal and internal sequences of the70-kD black locust agglutinin. For Medicago truncatula,a complete contig could be assembled that comprisesan open reading frame of 1,095 nucleotides encoding a365-amino acid residue polypeptide (see Supplemen-tal Fig. S2). Removal of a 28-residue signal peptideyields a 337-amino acid protein with an N terminusnearly identical to that of the black locust lectin and aninternal sequence almost identical to the cyanogen bro-mide cleavage product of the lectin. To check whetherthe 70-kD black locust agglutinin is a genuine orthologof the expressed M. truncatula protein, the correspond-ing genomic sequence was cloned. Sequencing of thePCR product confirmed that this fragment contains anopen reading frame encoding a 337-amino acid residuepolypeptide that shares 78.6% and 90% sequence iden-tity and similarity, respectively, with the expressed M.truncatula protein (see Supplemental Fig. S2).
An Inactive Homolog of Class V Chitinases with Lectin Activity
Plant Physiol. Vol. 144, 2007 663 www.plantphysiol.orgon April 10, 2019 - Published by Downloaded from
BLASTp searches using the deduced sequence as aquery revealed that the black locust agglutinin yieldedclass V chitinases from Arabidopsis (Arabidopsis thali-ana; At4g19810) and tobacco (CAA54374; NtChi) asbest matches. The new black locust agglutinin sharesapproximately 54% identity and 80% similarity, re-spectively, with both At4g19810 and CAA54374 (Fig.1), leaving no doubt that it is a structural homolog ofclass V chitinases. Accordingly, the protein will furtherbe referred to as black locust chitinase-related aggluti-nin (RobpsCRA). Although RobpsCRA is undoubtedlya homolog of a class V chitinase, there is apparently amajor difference for what concerns the molecular struc-ture of the native proteins because RobpsCRA is a homo-dimer, whereas all class V chitinases are monomericproteins. This suggests that, unlike class V chitinases,the RobpsCRA subunits contain some structural fea-tures that promote dimerization and hence allowformation of a divalent carbohydrate-binding proteinthat behaves as a genuine agglutinin.
Molecular cloning also yielded additional infor-mation about the biosynthesis and processing of
RobpsCRA. On the analogy of the M. truncatulaortholog, one can reasonably assume that RobpsCRAis synthesized with a signal peptide and follows thesecretory pathway. The calculated molecular mass ofthe protein (36,747.9 D) is nearly identical to thatmeasured by matrix-assisted laser-desorption ioniza-tion (MALDI)-time-of-flight (TOF) mass spectrometry(36,790 D). Taking into account that RobpsCRA is notglycosylated, it seems that no posttranslational pro-cessing takes place.
RobpsCRA Exhibits No Chitinase Activity But Is a
Genuine Lectin
Unlike class I chitinases, class V chitinases fromplants do not possess a genuine chitin-binding domaincorresponding to a hevein domain. Accordingly, theagglutinating activity of RobpsCRA cannot be as-cribed to the presence of a genuine or modified heveindomain.
A number of control experiments were set up to ruleout the possibility that the observed agglutination
Figure 1. Sequence alignment of RobpsCRA and the most closely related homologous proteins identified thus far. NtChi(CAA54374) is a catalytically active class V chitinase isolated from tobacco. At4g19810 is an expressed ortholog of Arabidopsis,but the protein has not yet been isolated and assayed for chitinase activity. In the top row, the N-terminal sequence of the nativelectin (Lec-Nter) and a cyanogen bromide fragment (Lec-CNBr) are aligned with the tobacco chitinase. Identical andhomologous residues are indicated by black and white boxes, respectively. Residues involved in the catalytic cleavage of chitinare indicated by black triangles. The Ser residue specifically involved in the catalytic activity of family 18 chitinases (Synstadet al., 2004) is indicated by a black square.
Van Damme et al.
664 Plant Physiol. Vol. 144, 2007 www.plantphysiol.orgon April 10, 2019 - Published by Downloaded from
activity might be due to contamination by the legumelectin-type bark lectin RPbAI. First, SDS-PAGE, usingincreasing amounts of the purified protein, yielded noadditional protein band in the 29- to 32-kD range.Second, the chromatograms of the automated Edmandegradation were indicative of a single sequence. Third,gel filtration experiments confirmed that the agglutina-tion activity coeluted with RobpsCRA and hence cannotbe ascribed to the larger tetrameric legume lectin-typebark lectins. Fourth, western-blot analysis indicatedthat RobpsCRA does not show any cross-reactivitywith antibodies raised against RPbAI. Moreover, as isdemonstrated below, the specificity of RobpsCRAdoes not match that of RPbAI and RPbAII.
Because RobpsCRA shares high sequence identitywith class V chitinases, the possible enzymatic activityof the protein was checked. Concentrated solutions ofthe protein (final concentration 2 mg/mL) were incu-bated with carboxymethyl-chitin-Remazol-Brilliant-Violet 5R at different pH values ranging between 4.0and 7.0. Even upon incubation for 72 h, no acid-solublefragments were generated, indicating that RobpsCRAis devoid of chitinase activity. It should be mentionedhere that two genuine class V chitinases isolated fromtobacco leaves inoculated with Tobacco mosaic virusexhibited readily measurable catalytic activity whenassayed with the same substrate (Melchers et al., 1994).
Agglutination assays with animal red blood cells dem-onstrated that RobpsCRA is a genuine lectin. Trypsin-treated human erythrocytes (type A) were agglutinatedat a lectin concentration of approximately 20 mg/mL.Hapten inhibition assays indicated that the agglutina-tion activity of RobpsCRA is not affected by anysimple sugar. Chito-oligosaccharides with chain lengthsup to 4 GlcNAc units also could not prevent aggluti-nation, indicating that the lectin activity of RobpsCRAdoes not rely on binding to chitin-like compounds. Onlysome animal glycoproteins, like thyroglobulin, inhibitedthe agglutination of human erythrocytes by RobpsCRA.Although indicative, the results of these preliminaryinhibition assays did not allow any conclusion to bedrawn with respect to the carbohydrate-binding spec-ificity of the lectin. Therefore, more appropriate tech-niques, based on direct measurements of lectin-glycaninteractions, were employed to unravel the fine spec-ificity of RobpsCRA.
RobpsCRA Specifically Binds High-Man N-Glycans
Glycan array analysis revealed that RobpsCRAbinds exclusively to some, but not all, high-Man-typeN-glycans. As shown in Table I, lectin reacted moststrongly with Man5-9mix, which is a mixture of high-Man N-glycans differing in the number of Man residuesand the nature of the bonds between the individualMan units. Besides the Man5-9mix, RobpsCRA alsoreacted well with individual high-Man N-glycans.Man6, Man5, and Man8 were approximately 30%less reactive than the Man5-9mix, whereas Man7 andMan9 were roughly 5 times less active than the mix-
ture (Table I). None of the oligomannosides testedshowed any reactivity. The same applies to chitotriose.These findings clearly indicate that the specificity ofRobpsCRA is directed toward the core pentasaccha-ride of N-glycans.
Although the results of glycan array screening ex-periments are only semiquantitative, they clearly dem-onstrate that the specificity of the lectin is directedtoward high-Man N-glycans comprising the core pen-tasaccharide of N-glycans. Therefore, it is important torealize that the results of glycan arrays are based on adirect binding assay and, accordingly, give a fairlygood idea of the relative affinity of the lectin for a verylarge set of glycans. At present, no conclusions can bedrawn with respect to the affinity of RobpsCRA for theN-glycans. The figures obtained with the different high-Man N-glycans are relatively low (,4,000 relative fluo-rescence units [RFU]) as compared to other lectins (upto .50,000 RFU). However, these low values mightpartly be due to poor coupling of the fluorochrome tothe lectin.
It should be emphasized here that the specificity ofRobpsCRA differs from that of the previously de-scribed legume lectin-type bark agglutinins. The ag-glutination activity of RobpsCRA cannot be inhibited,indeed, by any simple sugar, whereas RPbAI and theself-aggregatable lectin lose their hemagglutinatingactivity in the presence of GalNAc and GlcNAc/Man,respectively. Moreover, RPbAI-type isolectins interactstrongly with complex-type N-glycans, but are nonre-active toward high-Man N-glycans (Van Damme et al.,1995b). This difference in specificity confirms that theobserved lectin activity of RobpsCRA is not due tocontamination by another bark lectin.
Molecular Modeling of RobpsCRA and TobaccoClass V Chitinase
To find possible clues for the obvious lack of chitinaseactivity, the overall fold and structure of RobpsCRAwas tentatively determined by molecular modeling.Because previously no structure was available for plantclass V chitinase, the model was built using the coor-dinates of a human chitotriosidase (hMChi), which, ofall resolved GH18 proteins, shares the highest sequenceidentity/similarity with RobpsCRA (Fusetti et al.,2002). Sequence alignments indicated that RobpsCRAand hMChi share 33% and 51% identity/similarity,respectively, over a 305-amino acid residue overlap(spanning residues 17–322 of RobpsCRA; Fig. 2), whichis reasonably high considering the relatively low overallsequence identity/similarity within the GH18 family.Moreover, it should be emphasized here that, in spiteof the apparent low overall sequence identity, theresidues involved in binding of the substrate, as wellas those involved in the catalytic reaction, are mark-edly conserved between all members of the GH18.Therefore, hMChi can be considered a suitable modelto predict the structure of RobpsCRA. Because themodeling studies were primarily intended to explain
An Inactive Homolog of Class V Chitinases with Lectin Activity
Plant Physiol. Vol. 144, 2007 665 www.plantphysiol.orgon April 10, 2019 - Published by Downloaded from
the lack of chitinase activity of RobpsCRA, a parallelset of modeling experiments was set up with thetobacco class V chitinase (NtChi; Melchers et al., 1994)as an example of a class V chitinase with enzymaticactivity.
Hydrophobic cluster analysis (HCA) yielded similarplots for RobpsCRA and the enzymatically active chi-tinases NtChi and hMChi (see Supplemental Fig. S3),providing additional support for hMChi as a suitablemodel for predicting the structure of plant homologs.The three-dimensional models built for RobpsCRA andNtChi could adopt a very similar TIM-barrel fold ashMChi. This TIM-barrel fold consists of an inner crownof b-sheet strands surrounded by an outer crown ofa-helices (Fig. 3). An additional hairpin-shaped loopbuilt from three antiparallel strands of the b-sheetprotrudes from one edge of the TIM-barrel structure.A major functional feature of family 18 chitinase pro-teins is a central groove that accommodates a chitinchain through stacking interaction between the pyra-nose rings of the GlcNAc units and hydrophobic res-idues lining the groove over its entire length (Fig. 3).The catalytic site is located at one end of the groove,where it forms a strong electronegatively charged area(Fig. 3). Structural studies combined with mutationalanalysis of hevamine (Bokma et al., 2002) and chitinaseB from Serratia marcescens (Synstad et al., 2004), forexample, allowed the unambiguous identification ofthe amino acid residues involved in the catalytic activ-ity of GH18 chitinases. All these chitinases possessthe canonical DxDxE motif in the core of their catalyticsite. In addition, several other motifs/residues (e.g.YD motif and a Ser residue, Ser-69 in NtChi andRobpsCRA, involved in the substrate binding in fam-ily 18 chitinases) located at different positions in thepolypeptide chain are essential for activity (Figs.1 and 2).
Despite the obvious overall structural similarity withgenuine GH18 chitinases, the structure of RobpsCRAexhibits striking differences especially with respect tothe solvent-exposed hydrophobic residues that arepositioned along the chitin-binding groove and ensureproper stacking of the chitin chain onto the chitinases.Most of the hydrophobic residues lining the 25, 24,23, 11, and 13 subsites of hMChi (Rao et al., 2005) arereplaced by hydrophilic residues in RobpsCRA. Trpresidues Trp-10, Trp-50, Trp-78, and Trp-197 involvedin subsites 23, 24, 25, 11, and 13 of hMChi arereplaced by Lys-3, Ser-45, Gly-75, and Asp-191, re-spectively, in RobpsCRA. As a result of the replace-ment in RobpsCRA of the Trp residues by hydrophilicresidues, the overall conformation and physicochem-ical properties (hydrophilicity, charges) of the grooveare strongly altered as compared to those of the chitin-binding groove of hMChi. Moreover, due to the lack ofhydrophobic residues, the chitin chain cannot prop-erly stack into the groove of RobpsCRA. Most proba-bly, the inability to accommodate a chitin chain in thecatalytic groove can explain why RobpsCRA, in spiteof the presence of the canonical catalytic acidic resi-dues, is completely devoid of chitin-binding activity.It should be emphasized, indeed, that the DxDxEcatalytic motif is perfectly conserved in RobpsCRA(Asp-112, Asp-114, and Glu-116; Fig. 3). Moreover, thehydrophobic environment of this catalytic region isalso conserved in RobpsCRA. Therefore, RobpsCRAmight still be capable of cleaving the scissile glycosidicbond linking sugars bound to subsites 21 and 11. How-ever, the protein does not act as a chitinase because thesubstrate cannot be positioned in the catalytic groove.Failure to properly bind chitin as the underlying mech-anism for the lack of chitinase activity of RobpsCRAis further supported by the results of parallel model-ing of the catalytically active tobacco homolog NtChi.
Table I. Summary of the results of the printed glycan array specificity test for RobpsCRA
Unlike in RobpsCRA, most of the hydrophobic resi-dues found in the groove of hMChi are conserved inNtChi. Only Trp-10 and Trp-78 (of hMChi) are re-placed by Lys-4 and Gly-74, respectively, in NtChi(Fig. 3F). Accordingly, the groove of NtChi is fully ca-pable of properly positioning a chitin chain for cleav-age by the catalytic motif Asp-111, Leu-112, Asp-113,Trp-114, Glu-115.
Expressed Orthologs of RobpsCRA Are Common in
Legumes But Not Found in Other Plants
As already mentioned above, several other legumesexpress closely related orthologs of RobpsCRA. Com-plete or nearly complete contigs could be assembledfor M. truncatula, Glycine max, and Lotus japonicus. Allfour proteins share 62.5% identity and 82.4% similar-ity, respectively, within a 301-amino acid residue over-lap (see Supplemental Fig. S2).
In addition, a previously described, but only par-tially characterized, 67-kD homodimeric lectin frombean (Phaseolus vulgaris) seeds (Ye et al., 2001) alsomight represent a RobpsCRA ortholog. However, no
genuine orthologs of RobpsCRA could be identified inprotein, expressed sequence tag, or genomic databasesof any other plant species. Although no definitiveconclusions can be drawn from the available sequencedata, it is evident that RobpsCRA-type lectins are farless widespread among flowering plants than class Vchitinases and most likely are confined to a relativelysmall taxonomic group. Within the Fabaceae, RobpsCRAorthologs occur in at least four different tribes ofthe subfamily Papilionoideae (black locust, G. max,M. truncatula, and L. japonicus belong to the Robinieae,Phaseoleae, Trifolieae, and Loteae, respectively), indi-cating that they are not confined to a small taxon of theFabaceae.
DISCUSSION
Biochemical analyses and molecular cloning dem-onstrated that a minor lectin from the bark of blacklocust is a catalytically inactive homolog of class Vchitinases. The lectin behaves as a genuine hemagglu-tinin and specifically binds, albeit with a relatively low
Figure 2. Alignment of the amino acid sequences of RobpsCRA, tobacco class V chitinase NtChi, human chitotriosidase hMChi,and the chitinase-related murine protein Ym1. Identical and homologous residues are indicated by black and white boxes,respectively. Residues involved in the catalytic cleavage of chitin are indicated by black triangles, and residues lining up thecatalytic groove of genuine chitinases are indicated by black circles (hydrophobic residues) and asterisks (hydrophilic residues).The Ser residue specifically involved in the catalytic activity of family 18 chitinases (Synstad et al., 2004) is indicated by a blacksquare.
An Inactive Homolog of Class V Chitinases with Lectin Activity
Plant Physiol. Vol. 144, 2007 667 www.plantphysiol.orgon April 10, 2019 - Published by Downloaded from
Figure 3. Molecular modeling of RobpsCRA and related proteins. A and B, Ribbon diagrams of the modeled NtChi (A) andRobpsCRA (B) proteins. Acidic residues involved in the catalytic cleavage of chitin are indicated by stick representation (pink).
Van Damme et al.
668 Plant Physiol. Vol. 144, 2007 www.plantphysiol.orgon April 10, 2019 - Published by Downloaded from
affinity, to high-Man N-glycans. Activity assays usingdye-labeled substrates indicated that the protein isdevoid of chitinase activity. Molecular modeling andsequence comparisons indicated that the apparent lackof catalytic activity most probably has to be ascribed tothe protein’s inability to accommodate the chitin sub-strate in its catalytic groove as a consequence of anextensive replacement of hydrophobic by hydrophilicamino acids.
Although there is no doubt that RobpsCRA is struc-turally and evolutionarily related to class V chitinases,the available sequence data are insufficient to trace thedetails of the conversion of a plant chitinase into alectin. However, even in the absence of full details, onecan reasonably assume that RobpsCRA evolved from aGH18 chitinase and not the other way around. GH18represents an ancient chitinase family because it isfound in all kingdoms from bacteria to fungi, animals,and plants (Iseli et al., 1996). Moreover, sequence dataclearly indicate that all GH18 chitinases, includingthose from plants, have a common ancestor. This,taken together with the fact that GH18 chitinases orcorresponding genes have been found in numerousplant species, implies that the GH18 structural scaffoldis quite common in higher plants. Considering theapparent confinement of RobpsCRA orthologs to thelegume family, it is tempting to speculate that a cata-lytically active chitinase from an ancient legume speciesor possibly an earlier ancestor served as a structuralscaffold for the development of a small family ofcarbohydrate-binding proteins. This evolutionary pro-cess most likely involved gene duplication followed byneofunctionalization. The RobpsCRA orthologs repre-sent a documented example of how plants managed todevelop a domain with a specific sugar-binding activityfrom a functionally unrelated protein in general or anenzyme in particular. Therefore, it should be empha-sized that the novel lectin no longer recognizes thesubstrate of the original hydrolase (in casu chitin orchito-oligosaccharides) but a structurally unrelatedglycan (namely, high-Man N-glycans). Interestingly, asimilar conversion of a glycosyl hydrolase into acarbohydrate-binding, but catalytically inactive, ho-molog also occurred in higher animals. A so-calledeosinophil chemotactic cytokine has been identified inmouse (Chang et al., 2001) and human (Boot et al.,1995) that shares approximately 50% sequence identitywith the respective conspecific chitotriosidase but isdevoid of chitinase activity. Instead, the mouse protein(referred to as ECF-L or secretory protein Ym1) is alectin with binding specificity to glucosamine and
heparin/heparan sulfate (Chang et al., 2001; Sun et al.,2001). Although there is certainly a parallel betweenthe conversion of a chitinase into a lectin in animalsand plants, there are two major differences. First, incontrast to RobpsCRA, the canonical DxDxE motif isreplaced in Ym1 by a motif in which two catalyticacidic residues are substituted (Asn-115, Leu-116,Asp-117, Trp-118, Gln-119; Tsai et al., 2004; Fig. 2).Second, Ym1 binds glucosamine and heparin/heparansulfate, whereas RobpsCRA interacts exclusively withhigh-Man N-glycans.
Identification of RobpsCRA orthologs as lectins notonly adds a novel carbohydrate-binding domain tothe existing list of documented plant lectin familiesbut also illustrates that plants are capable of devel-oping sugar-binding domains from an existing struc-tural scaffold with different activity. At present, thebinding affinity of RobpsCRA is relatively low. How-ever, RobpsCRA might just be an intermediate in anevolutionary pathway that eventually will yield lec-tins with a high affinity. Even if there is no evidence ofwhether analogous evolutionary mechanisms mighthave given rise to other carbohydrate-binding do-mains that are confined to plants, the discovery ofa lectin ortholog of class V chitinases puts the evolu-tion of plant lectins in a novel perspective. In addi-tion, legume RobpsCRA orthologs represent a novelexample of well-defined neofunctionalization inplants.
It is also worth noting in this context that at least twodifferent cases have been reported of neofunctionali-zation-related evolutionary events whereby plantsused the structural scaffold of a lectin domain for thedevelopment of a protein that lost sugar-binding ac-tivity but acquired totally different biological activity.Curculin from Curculigo latifolia fruits is a homolog ofthe GNA-related lectins that possesses no sugar-bindingactivity but has sweet-tasting properties (Haradaet al., 1994). Seeds of several Phaseolus species containstructural homologs of the bean lectin that have nocarbohydrate-binding activity but are potent a-amylaseinhibitors or insecticidal proteins (called arcelins;Mirkov et al., 1994). It should be emphasized thatthese lectin homologs are not just binding-defectivemutants but proteins with well-defined biological activ-ity. Other binding-defective lectins have been identi-fied in the bark of the legume tree Cladrastis lutea andSambucus nigra, but the respective legume lectin andtype 2 ribosome-inactivating protein homologs haveno known biological activity other than a presumedstorage function (Van Damme et al., 1995a, 1997).
Figure 3. (Continued.)Hydrophobic (orange) and hydrophilic (blue) residues lining up the catalytic groove of chitinases are in stick representation. Cand D, Mapping of electrostatic potentials on the molecular surface of NtChi (C) and RobpsCRA (D). Acidic residues responsiblefor the electronegative character of the binding groove of chitinases are labeled in white. E to G, Enlarged ribbon diagrams of thecatalytic groove of hMChi (E), NtChi (F), and RobpsCRA (G). Catalytic residues (pink) and residues lining up the catalytic groove(orange for hydrophobic and blue for hydrophilic residues) are in stick representation.
An Inactive Homolog of Class V Chitinases with Lectin Activity
Plant Physiol. Vol. 144, 2007 669 www.plantphysiol.orgon April 10, 2019 - Published by Downloaded from