Seeds of Amazonian Fabaceae as a source of new lectins

RESEARCH ARTICLE

Seeds of Amazonian Fabaceae as a source of new lectins

Andreia V. Fernandes^I; Márcio V. Ramos^II; José Francisco C. GonçalvesI, ^*; Paulo A. C. Maranhão^I; Larissa R. Chevreuil^I; Luiz Augusto G. Souza^I

^ILaboratory of Plant Physiology and Biochemistry, National Institute for Research and Innovation in the Amazon (MCTI-INPA), Manaus, Amazonas, Brazil

^IIFederal University of Ceará, Biochemistry and Molecular Biology Department, Fortaleza, Ceará, Brazil

ABSTRACT

Seeds from fifty native Amazonian Fabaceae species (representing subfamilies Caesalpinioideae, Mimosoideae and Faboideae) were screened for the presence of new lectins. Their crude protein extracts were assayed for hemagglutinating activity (HA). The protein fractions of Anadenanthera peregrina, Dimorphandra caudata, Ormosia lignivalvis and Swartzia laevicarpa exhibited HA, and this activity was inhibited by galactose or lactose but not by glucose or mannose. The crude extract of S. laevicarpa exhibited HA activity only after ion exchange chromatography, and its lectin was further purified by affinity chromatography on immobilized lactose. Despite the large number of lectins that have been reported in leguminous plants, this is the first description of lectins in the genera Anadenanthera, Dimorphandra and Ormosia. The study of lectins from these genera and from Swartzia will contribute to the understanding of the evolutionary relationships of legume lectins in terms of their protein processing properties and structures.

Key words: Evolutionary relationships; plant lectins; protein-carbohydrate interactions; Swartzia; tropical Fabaceae

INTRODUCTION

Bioprospecting from wild plant diversity (i.e., natural bioresources) in search of useful medicinal, agricultural and other non-wood forest products can help solve some of humanity's pressing challenges. This effort can also help address evolutionary questions that have not yet been answered. Plant lectins found in the seeds of Amazonian leguminous species are a particularly intriguing target for bioprospecting. Lectins are constitutive proteins that possess non-catalytic carbohydrate-binding sites and are non-immune origin (Van Damme et al., 1998). Lectins are widespread in nature, and a detailed definition based on their structural features has been proposed to classify these proteins (Mo et al., 2000; Pinto et al., 2009; Ito et al., 2011). Lectins have been reported in bacteria, viruses, fungi, animals and plants, but most known lectins are found in legume seeds (Nasi et al., 2009).

Because lectins can bind glycoconjugates regardless of their origin, they have become an important tool in a wide variety of studies, including cell-cell recognition and cell activation (Debray et al., 2009; North et al., 2011). They also provide a unique model to study protein-carbohydrate interactions at the atomic level (Oliveira et al., 2008; Debray et al., 2009). These potential uses have motivated the continued search for new lectins and for seeds to be prioritized for screening. Some more than a century has passed since the first lectin was reported, and some screening studies periodically reveal novel lectins, many potential lectin sources remain to be examined. Throughout the history of lectin research, legume seeds have been screened for lectin activity. Almost all of our current knowledge regarding the structure, specificity and affinity of lectin-carbohydrate interactions has been achieved by studying legume lectins (Loris et al., 1998). Furthermore, structural data for legume lectins have helped to establish the evolutionary relationships among leguminous taxa (Rougé et al., 1987; Rougé and Varloot, 1990; Barre et al., 1994). The amino acid sequences and post-translational processing of lectins from the more derived legume tribes, such as Parkiae and Phaseoleae, have proven useful in supporting the chemotaxonomy of these groups (Gallego del Sol et al., 2005; Moreno et al., 2008). However, numerous legume taxa, especially in the more primitive groups, such as Sophoreae and Swartziae (Subfamily Faboideae), have been poorly studied. The study of lectins of the genera Swartzia, classified in the tribe Swartzieae, it will contribute to the understanding of the evolutionary relationships of lectins of legumes, in terms of the protein processing properties and structures, since the species sheltered in this tribe they possess floral morphologic characteristics of transition between subfamily Caesalpinioideae and Faboideae (Herender, 1994).

Except for lectins derived from Sophora japonica (Ueno et al., 1991; Van Damme et al., 1997) and a preliminary report on Swartzia pickelii (Cavalcanti and Coelho, 1990), no lectin has been described in either of these groups. To describe novel plant lectins, the seeds of fifty species of native Amazonian Fabaceae, including representatives of Sophoreae and Swartziae, were screened for HA. Here, we report the occurrence of lectins in previously unstudied legume groups. In addition, we investigate the hypothesis that the plant lectins found in seeds from different subfamilies of Fabaceae in the Amazonian rain forest reflect the evolutionary history of these taxa.

MATERIAL AND METHODS

Botanical material: The mature seeds of fifty native Fabaceae species were collected throughout 2010 at various sites in the Amazonian rain forest, Brazil. Individual vouchers were prepared for each seed collection. These materials were deposited at the institutional herbarium of the National Institute for Research in the Amazon (MCTI-INPA) and were identified by specialist in Fabaceae. Table 1 lists the species that were studied. The seed coat was removed, and the seeds were powdered and stored at 4°C until they were used.

Chemicals: Experimental carbohydrates, electrophoresis reagents, α-lactose-agarose and DEAE Sepharose were obtained from Sigma-Aldrich Chemical Company (St. Louis, USA). Molecular weight markers were purchased from Promega Corp., Madison, WI, USA. All other chemicals employed in this study were of analytical grade.

Protein extracts: The powdered seeds were suspended in 150 mM NaCl (1:10, w/v) and stirred for 2 h at 25°C. The resulting suspensions were centrifuged at 20,000 × g at 4°C for 20 min. The precipitates were discarded, while the soluble phases were dialyzed in distilled water for 72 h at 4°C. The soluble protein fraction of each extract was recovered after an additional centrifugation step, as described above, and freeze dried. These lyophilized extracts were used in all further determinations.

Hemagglutinating assays: The extracts were screened for lectin activity by determining the hemagglutination (HA) titer of each sample in a 2% (v/v) rabbit or rat erythrocyte suspension in a U-bottomed microtitration plate. The protein samples were dissolved in Tris-buffered saline (100 mM Tris-HCl, pH 7.6, added to 150 mM NaCl) at 5 mg/mL and serially diluted in the plates using the same buffer before the erythrocytes (25 µL) were added. The assays were stored at 25°C, and the HA titers were recovered after 2 and 24 h. The inhibition of HA activity was evaluated as previously described (Ramos, 1997). Briefly, the samples that exhibited minimal HA activity were first incubated with decreasing concentrations of different carbohydrates (100 mM) and further mixed with 2% (v/v) erythrocyte suspension. The inhibition of hemagglutination was recorded after 24 h.

Chromatography: The samples were subjected to ion-exchange chromatography on DEAE Sepharose that had been equilibrated with 50 mM Tris-HCl, pH 8.0. After loading, the samples were first eluted in the same buffer, followed by the stepwise addition of NaCl (0.1 M, 0.5 M and 1 M). The absorbance of the collected fractions was monitored at 280 nm. The fractions containing each recovered protein peak were assayed for their HA activity as described above.

Affinity chromatography was performed on α-lactose-agarose in buffered saline containing 5 mM CaCl₂ and 5 mM MnCl₂. The samples (200 mg) were applied to the column and initially eluted with buffered saline. After washing the unbound proteins, the column was washed with 100 mM glycine, pH 2.6, containing 150 mM NaCl. Each protein fraction was assayed for HA after dialysis and lyophilization.

Electrophoresis: Polyacrylamide gel electrophoresis (SDS-PAGE) was used to examine the protein fractions from the chromatography peaks according to the Laemmli method (Laemmli, 1970). The protein samples were dissolved in 0.0625M Tris-HCl, pH 6.8, under denaturing conditions (1% SDS and 1% β-mercaptoethanol) and loaded on a 12.5% polyacrylamide gel. The proteins were separated at 200 V and a maximum current of 15 mA at 20°C and observed by staining the gels with Coomassie brilliant blue R-250.

RESULTS AND DISCUSSION

Protein extracts: Screens for lectin activity frequently utilize different pH conditions for protein extraction and different erythrocyte sources for testing HA (Ainouz and Sampaio, 1991). In the present study, the proteins were extracted with 150 mM NaCl because lectin-containing extracts are commonly active under this condition. To evaluate the diversity of the extracted proteins, the extracts were analyzed by electrophoresis. The protein profiles of the extracts that exhibited HA activity are shown in Figure 1. The extracts contained a wide range of proteins, suggesting that the protein extraction procedure was successful.

New lectins: The seeds of fifty native Amazonian Fabaceae species were screened for lectin activity based on the ability of their protein extracts to agglutinate rabbit or rat erythrocytes. The species were chosen based on their taxonomic positions or evolutionary relationships within the various legume tribes (Table 1). This strategy was chosen because numerous legume lectins have been described in species of Phaseoleae (notably from the subtribe Diocleinae) and Vicieae (mainly in Pisum and Lathyrus) and, to a lesser extent, in members of Loteae, Robineae, Crotalareae, Abreae and Dalbergiae, while other groups have been poorly investigated. The present study included several representatives of the more primitive groups of legumes (Subfamily Faboideae), especially Sophoreae and Swartziae, which have been the subject of few lectin studies.

HA activity was observed earliest in the extracts from Anadenanthera peregrina, Dimorphandra caudata and Ormosia lignivalvis, which belong to Mimoseae, Caesalpinieae and Sophoreae, respectively. Although hemagglutination was easily seen after 24 h, it was always weak. Rabbit erythrocytes were agglutinated by Dimorphandra caudata [titer: 2⁴(HA = 16)], while rat erythrocytes were agglutinated by Anadenanthera peregrina [titer: 2⁶(HA = 64)] and Ormosia lignivalvis [titer: 2⁷ (HA = 128)]. The HA activity of these extracts was lost when the samples were heated at 100°C for 10 min before they were tested; therefore, proteins were likely to be responsible for the observed HA activity.

The Swartzia protein extracts exhibited a strong hemolytic effect, which inhibited the measurement of lectin activity and suggested the presence of hemolysins or saponins. However, because hemolysis persisted even after the extract was heated, saponins were likely responsible for the occurrence of hemolysis rather than hemagglutination (Konozy et al., 2002). Because Swartzia species are of great evolutionary importance within the legume family and because little consistent information is available regarding lectins in this group, the lectin activity of the Swartzia extracts was reexamined. The Swartzia extracts were fractionated by ion-exchange chromatography to separate the hemolytic agents from the putative lectins. The Swartzia laevicarpa protein extract was fractionated into three protein peaks. Further tests for HA activity showed the presence of lectin activity in the fraction that was eluted with 0.1 M NaCl, while hemolysis was observed in the fraction that was eluted with 0.5 mM NaCl (Figure 2A and B). These activities remained after the protein fractions were dialyzed and lyophilized. However, the HA activity was lost after heating, suggesting the presence of lectin. The protein profiles of the peaks obtained from ion-exchange chromatography were further examined by electrophoresis (Figure 3A). The HA fraction (DEAE-PII) showed very few protein bands, while PIII contained major proteins and retained its hemolytic activity (Figure 2B).

As stated above, the active extracts exhibited weak HA despite the experimental conditions that were utilized. Because previous HA assays have used various sources of erythrocytes, the poor agglutination observed in this study and the hemolytic activity associated with these extracts may have discouraged the further study of these plant species in previous screens. Legume lectins often strongly agglutinate rabbit erythrocytes. For example, those from Erythrina, Dioclea and Canavalia exhibit strong agglutination (HA greater than 2024) of rabbit erythrocytes, and more than 30 lectins have been purified from these groups (Konozy et al., 2002; Konozy et al., 2003).

Carbohydrate-binding specificity: The agglutinating extracts were mixed with carbohydrates that are commonly used to investigate the primary specificity of lectins, such as glucose and galactose. As shown in Table 2, the HA activity of all of the extracts was inhibited in the presence of lactose. Galactose inhibited the extracts of Anadenanthera peregrina, Ormosia lignivalvis and Swartzia laevicarpa, but it did not inhibit those of Dimorphandra caudata. Glucose did not inhibit hemagglutination. These results suggest that all of the new lectins described here belong to the Gal/GalNac group.

Swartzia lectins: In the present study, seven Swartzia species were screened for lectin activity. All of these extracts exhibited strong hemolytic activity that persisted even after the extracts were heated. However, the detection of lectin activity in Swartzia laevicarpa extract after chromatographic separation suggested that lectins would be present in other Swartzia extracts. Therefore, the Swartzia protein extracts were compared by electrophoresis (Figure 3B). Surprisingly, the various Swartzia extracts exhibited distinctive protein profiles, and this direct comparison of their protein contents was discontinued. By following the same protocol used for S. laevicarpa, all of the Swartzia extracts were fractionated by ion exchange chromatography. Once more, the protein fractionation patterns differed strongly among the extracts, but hemolysis was still observed (data not shown). Surprisingly, HA was not observed in the protein fractions of S. argentea, S. pendula, S. polyphylla and S. sericea. However, the protein fractions of S. ingifolia and S. longistipitata exhibited HA. Their extracts were fractionated in two and four peaks, respectively (data not shown). Among all of the protein peaks that were examined, HA was observed only in PIII-300 mM NaCl (S. ingifolia) and PI-100 mM NaCl (S. longistipitata), and both were strongly inhibited by galactose (< 10 mM) and lactose (< 10 mM). We conclude that lectins are present in Swartzia species, but distinct protocols are needed to extract and purify these proteins. Because the HA of S. laevicarpa was more consistent and was clearly inhibited in the presence of lactose, its protein extract was further separated by affinity chromatography on an α-lactose-agarose column. As shown in Figure 4A, after the unbound proteins had been washed from the column, the protein fraction that was retained on the matrix was eluted by changing the pH conditions. The retained protein fraction (PII) still exhibited HA that was inhibited by galactose and lactose; however, the proteins that were eluted in PI did not exhibit these properties. Therefore, the lectin fraction was examined by electrophoresis, and a major protein band with an apparent molecular mass between 27 and 30 KDa was observed (Figure 4B). This protein is now the target of further purification and biochemical characterization studies. The results reported here suggest that Swartzia lectins are probably most inhibited by lactose and that affinity chromatography on immobilized lactose is likely the most effective technique to purify these new lectins.

Evolutionary implications: Detailed analyses of the amino acid sequences and three-dimensional structures of legume lectins have proven useful in elucidating the phylogeny of legume taxa (Loris et al., 1998; Oliveira et al., 2008). The events associated with the post-translational processing of legume lectins and the quaternary associations of their monomers also provide information regarding the evolutionary relationships of these proteins (Barre et al., 1994). For example, the seed lectin of Vatairea macrocarpa (tribe Dalbergiae) is the first galactose-specific lectin reported to exhibit post-translation processing similar to that seen in glucose/mannose-specific lectins in the sub-tribe Diocleinae (Calvete et al., 1998; Konozy et al., 2002) . Furthermore, the protein profiles of these lectins as observed by polyacrylamide electrophoresis are almost identical; the lectins are composed of three distinct bands corresponding to particular chain fragments produced by proteolytic processing (Konozy et al., 2002). These results suggest that the legume lectins belong to multiple distinct groups. For example, in terms of their biosynthesis, the lectins found in Phaseoleae (sub-tribe Diocleinae) and Dalbergiae may be more similar to each other than to other legume lectins that belong to more closely related taxa, such as Vicieae and Diocleinae. Because Swartzia represents primitive legume group, Swartzia lectins may provide novel insights regarding legume evolution and the relationships of legume lectins. The electrophoresis pattern and observed molecular weight of the putative S. laevicarpa lectin identified here are nearly identical to those of a seed lectin from Sophora japonica (Hankins et al., 1987). The tribe Sophoreae is closely related to Swartziae and may have evolved from within the latter group.

The present study has successfully reported the occurrence of lectins in primitive legume groups (Subfamily Faboideae) and has provided new insights regarding the detection and purification of Swartzia lectins.

Acknowledgements: This study was supported by grants from the Brazilian Council for Research and Development (CNPq - BioNorte Project n°. 554307/2010-3). We thank J.F.C. Gonçalves and M.V. Ramos of the Brazilian Council for Research and Development (CNPq). We also thank Annie Gilot of the University of Illinois and AJE for revising the English of the manuscript.

Received: 01 October 2011

Accepted: 31 October 2011

• Ainouz, I.L., Sampaio, A.H. (1991). Screening of Brazilian marine algae for hemagglutinins. Bot. Mar., 34, 211-214.
• Barre, A., Lauga J., Rougé P. (1994). The three-dimensional structure of lectins: a phenetic and phylogenetic tool for the Leguminosae. Biochem. Syst. Ecol., 22 (4), 401-407.
• Calvete, J.J., Santos, C.F., Mann, K.; Grangeiro, T.B., Nimtz, M., Urbanke, C., Cavada, B.S. (1998). Amino acid sequence, glycan structure, and proteolytic processing of the lectin of Vatairea macrocarpa seeds. FEBS Lett 425(2): 286-92.
• Cavalcanti, M.S.M., Coelho, L.C.B.B. (1990), Isolation and partial characterization of a lectin from Swartzia pickelii Killip (white jacaranda). Mem. Inst. Oswaldo Cruz, 85(3), 371-372.
• Debray, H., Coddeville, B., Bomfim, L.R, and Ramos, M.V. (2009). A simple micro-method for determining precise oligosaccharidic specificity of mannose-binding lectins. Glycobiol, 19 (12), 1417-26.
• Gallego del Sol, F., Nagano, C., Cavada, B. S. and Calvete, J. J. (2005). The first crystal structure of a Mimosoideae lectin reveals a novel quaternary arrangement of a widespread domain, J. Mol. Biol. 353,574-583.
• Hankins, C. N., Kindinger, J., Shannon, L. M. (1987). The Lectins of Sophora japonica: I. Purification, Properties, and N-Terminal Amino Acid Sequences of Two Lectins from Leaves. Plant Phisiol. 83, 825-829
• Herender, P.S. (1994). "Phylogenetic relationships if the tribe Swartzieae". In: M. Crisp & J.J. Doyle (ed.). Advance in Legume Systematics 7: Phylogeny, pp. 123-132. Royal Botanic Gardens, Kew.
• Ito, S., Shimizu, M., Nagatsuka, M., Kitajima, S., Honda, M. Tsuchiya, T., Kanzawa N. (2011). High molecular weight lectin isolated from the Mucus of the Giant African snail Achatina fulica Biosci. Biotechnol. Biochem. 75 (1), 20-25.
• Konozy, E.H.E., Mulay, R., Faca V., Ward R.J., Greene, L.J., Roque-Barreira, M.C., Sabharwal, S., Bhide, S.V. (2002). Purification, some properties of a D-galactose-binding leaf lectin from Erythrina indica and further characterization of seed lectin, Biochimie, 84 (10), 1035-1043.
• Konozy, E. H. E.; Bernardes, E. S.; Rosa, C.; Faca, V.; Greene, L. J. and Ward, R. J. (2003). Isolation, purification and physicochemical characterization of a D-galactose-binding lectin from seeds of Erythrina speciosa Arch. Biochem. Biophys., 410, 222-229.
• Laemmli, U.K. (1970). Cleavage of structural proteins during the assembly of the bacteriophage T4. Nature. 227, 680-685.
• Mo, H.; Winter, H. C. and Goldstein, I. J. (2000), Purification and characterization of a Neu5Aca2-6Galb1-4Glc/GlcNac-specific lectin from the fruiting body of the polypore mushroom Polyporus saquamosus J. Biol. Chem., 275, 10623-10629.
• Loris, R., Hamelryck, T., Bouckaert, J., and Wyns, L. (1998). Legume lectin structure. Biochim. Biophys. Acta 1383, 9-36.
• Moreno, F.B.M.B., de Oliveira T.M., Martil D.E., Vicoti M.M., Bezzera G.A., Abrego J.R.B., Cavada B.S., de Azevedo W.F. Jr. (2008). Identification of a new quaternary association for legume lectins. J. Struct. Biol., 161:133-143.
• Nasi, A., Picariello, G., Ferranti, P. (2009). Proteomic approaches to study structure, functions and toxicity of legume seeds lectins. Perspectives for the assessment of food quality and safety. J Proteom. 72: 527-538 .
• North, S.J., von Gunten S., Antonopoulos A., Trollope A., Macglashan D.W.Jr, Jang-Lee J., Dell A., Metcalfe D.D., Kirshenbaum A.S., Bochner B.S., and Haslam S.M. (2011). Glycomic analysis of human mast cells, eosinophils and basophils. Glycobiol (Unpublished results) downloaded from glycob.oxfordjournals.org on August 31.
• Oliveira, M.D.L.; Andrade, C.A.S.; Santos-Magalhães, N.S.; Coelho, L.C.B.B.; Teixeira, J.A.; Carneiro-da-Cunha, M.G.; Correia, M.T.S. (2008). Purification of a lectin from Eugenia uniflora L. seeds and its potential antibacterial activity. Lett. Appl. Microbiol., 46:371-376.
• Pinto, V. P. T., Debray, H., Dus, D. Teixeira, E. H., de Oliveira, T. M., Carneiro, V. A., Teixeira, A. H., Filho, G. C., Nagano, C. S., Nascimento, K. S., Sampaio, A. H., Cavada, B.S. (2009). Lectins from the Red Marine Algal Species Bryothamnion seaforthii and Bryothamnion triquetrum as Tools to Differentiate Human Colon Carcinoma Cells. Adv. Pharmacol. Sci, 2009. 1-6.
• Ramos, M.V. (1997).Biosynthesis and structural lectin features of the Phaseoleae, Diocleinae and the Vicieae (Leguminosae = Fabaceae) under a phylogenetic perspective. J. Comp. Biol., 2, 129-136.
• Rougé, P., Lauga, J., Richardson, M. (1987). Tentative phylogenetic trees of the papilionoideae and vicieae based upon the amino acid composition of their lectins Biochem. Syst. Ecol., 15: 445-452.
• Rougé P., Varloot L., (1990). Structural homologies between leguminosae lectins as revealed by the hydrophobic cluster analysis (HCA) method. Biochem. Syst. Ecol. 18, 419-427.
• Ueno, M., Ogawa H., Matsumoto I., Seno N. (1991). A novel mannosespecific and sugar specifically aggregatable lectin fromthe bark of the Japanese pagoda tree (Sophora japonica). J. Biol. Chem. 266, 3146-3153.
• Van Damme, E.J.M., Barre A., Rougé P., Peumans, W.J. (1997) Molecular cloning of the bark and seed lectins from the Japanese pagoda tree (Sophora japonica) Plant Mol Biol., 33, 523-536.
• Van Damme, E.J.M., Peumans W.J., Barre A., Rougé, P. (1998). Plant lectins: a composite of several distinct families of structurally and evolutionary related proteins with diverse biological roles. Crit. Rev. Plant Sci. 17, 575-692.

Publication Dates

• Publication in this collection
  20 Jan 2012
• Date of issue
  2011

History

• Received
  2011
• Accepted
  31 Oct 2011