G Model

BIOMAC 4873 1–7 ARTICLE IN PRESS

International Journal of Biological Macromolecules xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules

journal homepage: www.elsevier.com/locate/ijbiomac

1 Purification, characterization and antibacterial potential of a lectin

2 isolated from Apuleia leiocarpa seeds
3 Q1 Aline de Souza Carvalho a , Márcia Vanusa da Silva a , Francis Soares Gomes b ,
4 Patrícia Maria Guedes Paiva a , Carolina Barbosa Malafaia a , Tulio Diego da Silva a ,
5 Antônio Fernando de Melo Vaz c , Alexandre Gomes da Silva a ,
6 Isabel Renata de Souza Arruda a , Thiago Henrique Napoleão a,∗ ,
7 Maria das Graças Carneiro-da-Cunha a , Maria Tereza dos Santos Correia a
a
8 Departamento de Bioquímica-CCB, Universidade Federal de Pernambuco, Cidade Universitária, 50670-420 Recife, Pernambuco, Brazil
b
9 Instituto de Química e Biotecnologia, Universidade Federal de Alagoas, 57072-900 Maceió, Alagoas, Brazil
c
10 Centro de Saúde e Tecnologia Rural, Universidade Federal de Campina Grande, Jatobá, 58700-970 Patos, Paraíba, Brazil
11

12
24 a r t i c l e i n f o a b s t r a c t
13
14 Article history: Apuleia leiocarpa is a tree found in Caatinga that has great value in the timber industry. Lectins are
15 Received 4 June 2014 carbohydrate-binding proteins with several biotechnological applications. This study shows the isolation,
16 Received in revised form 31 January 2015 characterization, and antibacterial activity of A. leiocarpa seed lectin (ApulSL). The lectin was chromato-
17 Accepted 2 February 2015
graphically isolated from a crude extract (in 150 mM NaCl) by using a chitin column. ApulSL adsorbed to
18 Available online xxx
the matrix and was eluted using 1.0 M acetic acid. Native ApulSL was characterized as a 55.8-kDa acidic
19
protein. SDS-PAGE showed three polypeptide bands, whereas two-dimensional electrophoresis revealed
20 Keywords:
four spots. The peptides detected by MALDI TOF/TOF did not show sufficient homology (<30%) with the
21 Lectin
22 Apuleia leiocarpa
database proteins. Circular dichroism spectroscopy suggested a disordered conformational structure, and
23 Xanthomonas campestris fluorescence spectrum showed the presence of tyrosine residues in the hydrophobic core. The hemagglu-
tinating activity of ApulSL was present even after heating to 100 ◦ C, was Mn2+ -dependent, and inhibited
by N-acetylglucosamine, d(−)-arabinose, and azocasein. ApulSL demonstrated bacteriostatic and bacte-
ricide effects on gram-positive and gram-negative species, being more effective against three varieties
of Xanthomonas campestris (MIC ranging from 11.2 to 22.5 ␮g/mL and MBC of 22.5 ␮g/mL). The results of
this study reinforce the importance of biochemical prospecting of Caatinga by revealing the antibacterial
potential of ApulSL.
© 2015 Published by Elsevier B.V.

25 1. Introduction with biotechnological potential [4]. Apuleia leiocarpa (Vogel) J. F. 37

Macbride is a tree belonging to the sub-family Caesalpinioideae of 38

26Q2 The Caatinga is a type of vegetation exclusive to Brazil and has Fabaceae. In Brazil, it is commonly known as “grápia” and “jataí,” 39

27 been recognized as one of the most important natural regions of among other names [5]. It has a wide distribution, occurring from 40

28 the world [1]. Ethnobotanical surveys are important in bioprospect- northeastern Brazil to Uruguay and Argentina, and it prefers moun- 41

29 ing, to find herbal medicines and other biotechnologically relevant tain slopes and well-drained soils [6]. The tree is used in the tanning, 42

30 compounds [2]. Although Caatinga is one of the most threatened timber, and construction industries [6–8]. It also has potential for 43

31 biomes on the planet, few ethnobotanical studies have been con- use in agroforestry systems and has ornamental and reforestation 44

32 ducted in the Brazilian semi-arid region [3]. uses [6,9]. 45

33 Plants from Fabaceae family (leguminous plants) are well Lectins are proteins from non-immune origin, which bind 46

34 known because of the many species used for human consumption, specifically and reversibly to free sugars or to the subtermi- 47

35 such as soy, beans, and peas. However, Fabaceae plants that are not nal or terminal residues of glycoconjugates [10]. These proteins 48

36 used in the diet have been poorly studied as sources of compounds have been the most studied in leguminous species, because they 49

are frequently very abundant in the seeds of these plants and 50

may constitute up to 10% of the total protein [11,12]. Many 51

∗ Corresponding author. Tel.: +55 8121268540; fax: +55 8121268576. lectins have been also isolated from other plant tissues and 52

E-mail address: thiagohn86@yahoo.com.br (T.H. Napoleão). families [13]. 53

http://dx.doi.org/10.1016/j.ijbiomac.2015.02.001
0141-8130/© 2015 Published by Elsevier B.V.

Please cite this article in press as: A.d.S. Carvalho, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.02.001
 G Model
BIOMAC 4873 1–7 ARTICLE IN PRESS
2 A.d.S. Carvalho et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx

54 Lectins have been used in biorecognition technology to inves- 2.3. Purification of ApulSL 113

55 tigate the structure and function of complex carbohydrates and

56 to map changes in cell surface during physiological and patho- The CE (1.0 mg of protein) was loaded onto a chitin 114

57 logical processes [14]. They also have shown immunomodulatory, (7.5 cm × 1.5 cm) column equilibrated with 150 mM NaCl. The col- 115

58 anti-inflammatory, antitumor, hypotensive, insecticidal, antiviral, umn was washed with the equilibrating solution until absorbance 116

59 antifungal, and antibacterial activities [13,15–21]. The antimicro- at 280 nm was lower than 0.030. Elution was performed with 1.0 M 117

60 bial activity of lectins may result from their ability to interact acetic acid. Fractions of 2 ml were collected every 6 minutes and 118

61 with carbohydrates on the cell surface of microbes. Antibacterial monitored by absorbance at 280 nm. The fractions with absorbance 119

62 lectins can interact with bacterial cell wall components such as N- ≥0.100 were pooled (ApulSL) and dialyzed in aqueous 150 mM NaCl 120

63 acetylglucosamine, N-acetylmuramic acid (MurNAc), tetrapeptides to remove the acetic acid. 121

64 linked to MurNAc, and lipopolysaccharides [19].

65 The phytopathogenic bacteria Xanthomonas campestris pv. 2.4. Characterization of ApulSL HA 122

66 campestris causes black rot and produces extracellular enzymes

67 that promote plant cell wall degradation, which contributes to its Inhibition of HA was evaluated using carbohydrates and gly- 123

68 pathogenicity [22]. X. campestris pv. viticola is the causal agent of coproteins. HA assays were performed as described above but 124

69 bacterial canker of grapevine, causing necrotic spots in inflores- replacing the solution of 150 mM NaCl with a solution of the car- 125

70 cences and dark, roughly rounded lesions in the rachis and berries. bohydrate prepared in 150 mM NaCl. In addition, there was an 126

71 It is considered a pest in the northeastern states of Bahia, Per- interval of 45 min between the end of sample dilution in car- 127

72 nambuco, and Piauí [23,24]. X. campestris pv. malvacearum attacks bohydrate solution and the addition of erythrocyte suspension. 128

73 cotton crops causing angular lesions in the leaves that are initially The concentrations of inhibitor solutions were 100 and 200 mM 129

74 green and oily, and later brown and necrotic [25]. for carbohydrates [d(−) arabinose, l(+)-arabinose, fructose, fucose, 130

75 An initial screening of 36 extracts from 27 Caatinga plants, glucose, galactose, d-lactose, d(+)-maltose, mannose, methyl-␣-d- 131

76 including A. leiocarpa, revealed that 77.7% of the samples had mannopyranoside, N-acetyl-galactosamine, N-acetyl-glucosamine, 132

77 hemagglutinating activity, which is indicative of the presence of raffinose, rhamnose, and d(−)-ribose] and 250 and 500 ␮g/mL for 133

78 lectin [26]. The present study describes the purification, character- glycoproteins (azocasein, casein, fetuin, and thyroglobulin). 134

79 ization and antibacterial activity of a lectin extracted from the seeds The effect of divalent ions (Ca2+ , Mg2+ , and Mn2+ ) on the ApulSL 135

80 of A. leiocarpa (ApulSL), as part of an effort to expand the knowledge HA was evaluated. The lectin was dialyzed with 5 mM EDTA (16 h 136

81 of bioactive compounds found in Caatinga plants. at 4 ◦ C) and then with 150 mM NaCl (6 h at 4 ◦ C) to remove the 137

EDTA. Next, the dialyzed ApulSL was incubated for 45 min with 138

10 mM Ca2+ , Mg2+ , or Mn2 and then HA was evaluated. The effect 139
82 2. Materials and methods of these cations on HA of ApulSL non-treated with EDTA was also 140

determined. 141
83 2.1. Plant material and extract preparation The effect of temperature on the ApulSL HA was evaluated by 142

heating an aliquot of ApulSL (1.5 ml) during 30 min at 30, 40, 50, 143
84 A. leiocarpa seeds were collected from the National Park of 60, 70, 80, or 100 ◦ C and for 2 h at 100 ◦ C prior to HA assay. HA of 144
85 Catimbau (PARNA Catimbau), Pernambuco, Brazil, from January ApulSL was also determined after heating for 30 min at 121 ◦ C in an 145
86 2011 to July 2011. The taxonomic identification was performed autoclave. 146
87 in the Herbarium Dárdano de Andrade Lima at the Instituto The ApulSL HA was also evaluated after it was boiled for 5 min 147
88 Agronômico de Pernambuco (IPA) and the testimonial material was at 100 ◦ C with electrophoresis buffer (1.0 M Tris–HCl pH 6.8; 0.2 g 148
89 archived under the number 84886. sodium dodecyl sulphate; 1.0 ml glycerol; 2 mg bromophenol blue). 149
90 For preparation of the extract, the seeds were dried at 45 ◦ C Control assays were also performed by incubating erythrocytes 150
91 and processed by grinding. The seed flour was homogenized under only with buffer. 151
92 agitation for 4 h at 28 ◦ C with 150 mM NaCl to yield a final concen-
93 tration of 10% (w/v). The homogenate was filtered through filter 2.5. Gel filtration chromatography 152
94 paper and centrifuged at 3600 rpm for 15 min. The supernatant was
95 the crude extract (CE). A sample of ApulSL (500 ␮g) in 150 mM NaCl was loaded 153

onto a Hiprep SephacrylTM 16/60 S100 HR column (GE Health- 154

care, Sweden) coupled to an ÄKTAprime system to determine the 155

96 2.2. Protein content and hemagglutinating activity (HA) molecular mass of the native protein. The chromatography was 156

performed at a flow rate of 0.5 mL/min in 150 mM NaCl. Frac- 157

97 Protein concentration was determined according to Lowry et al. tions of 2 ml were collected and protein elution was monitored 158
98 [27] using a standard bovine serum albumin curve with values by absorbance at 280 nm. Molecular mass standards, phosphory- 159
99 between 31.25 and 500 ␮g/ml. Protein concentration was also esti- lase b (97 kDa), albumin (66 kDa), ovalbumin (45 kDa), carbonic 160
100 mated using the absorbance at 280 nm. anhydrase (30 kDa), trypsin inhibitor (20.1 kDa), and ␣-lactalbumin 161
101 The lectin activity was measured by determining hemaggluti- (14.4 kDa), were similarly chromatographed. 162
102 nating activity (HA) according to Paiva and Coelho [28] in 96-well
103 microtiter plates (Kartell SPA, Italy). The HA assay was started by 2.6. Polyacrylamide gel electrophoresis (PAGE) 163
104 adding 50 ␮L of 150 mM NaCl to all wells and 50 ␮L of the sample
105 in the second well of the horizontal row. Successive dilutions were PAGE for native acidic proteins [15% gel (w/v)] was performed 164
106 performed until a ratio of 1:2048 was achieved. Next, 50 ␮L of a according to Davis [30] and PAGE for native basic proteins was per- 165
107 2.5% (v/v) suspension of glutaraldehyde-treated erythrocytes [29] formed according to Reisfeld et al. [31]. Polypeptide bands were 166
108 from humans (A, B or O-types) or rabbits was added. The number stained with 0.02% (w/v) Coomassie Brilliant Blue R-250 in 10% 167
109 of units of hemagglutinating activity was defined as the reciprocal acetic acid (for acidic proteins) or 1% Amido Black in 10% acetic 168
110 of the highest dilution of sample that promoted full agglutination acid (for basic proteins). 169
111 of erythrocytes. The specific HA was defined as the ratio between Electrophoresis on a 15% (w/v) polyacrylamide gel in the pres- 170
112 the units and protein concentration (mg/mL). ence of sodium dodecyl sulphate (SDS-PAGE) was performed 171

Please cite this article in press as: A.d.S. Carvalho, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.02.001
 G Model
BIOMAC 4873 1–7 ARTICLE IN PRESS
A.d.S. Carvalho et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx 3

172 according to Laemmli [32]. Polypeptide bands of ApulSL and molec- 2.10. Fluorescence spectroscopy 231

173 ular mass standards (bovine albumin, 66 kDa; ovalbumin, 45 kDa;

174 glyceraldehyde-3-phosphate dehydrogenase, 36 kDa; bovine car- Intrinsic fluorescence assay was performed on a spec- 232

175 bonic anhydrase, 29 kDa; bovine trypsinogen, 24 kDa; trypsin trofluorimeter (JASCO FP-6300, Tokyo, Japan). The intensity of 233

176 inhibitor soybean, 20.1 kDa; ␣-lactalbumin, 14.4 kDa) were stained fluorescence emission of hydrophobic residues from the protein 234

177 with 0.02% (w/v) Coomassie Brilliant Blue R-250 in 10% acetic dissolved in Milli-Q water was measured at 25 ◦ C in a rectangu- 235

178 acid. lar quartz cuvette with a path length of 10 mm. For measurements 236

of intrinsic fluorescence, excitation was at 280 nm and emission at 237

179 2.7. Two-dimensional electrophoresis 305–450 nm was recorded using 10 nm. The contribution of the sol- 238

vent (distilled water) was subtracted. The center of spectral

 mass
 239

180 ApulSL (200 ␮g) was mixed with 250 ␮L of rehydration buffer (CM) was calculated according to the equation: CM = I F / F , 240

181 containing 8.0 M urea, 2% (w/v) CHAPS, 20 mM dithiothreitol (DTT), where F is the fluorescence emission at wavelength I , and the 241

182 0.5% IPG buffer (GE Healthcare, Piscataway, NJ, USA), and 0.002% summation was carried out over the range of appreciable values of 242

183 (w/v) bromophenol blue. The samples were loaded on an IPG strip F. 243

184 of 13 cm (GE Healthcare, Piscataway, NJ, USA) and a linear pH range

185 from 3.0 to 10.0. The isoelectric focusing was performed on Ettan 2.11. Antibacterial activity 244
186 IPGphor III system (GE Healthcare, Piscataway, NJ, USA) following
187 the manufacturer’s protocol. Before running the second dimension Strains of the gram-positive bacteria Enterococcus faecalis 245
188 electrophoresis, the strip was equilibrated for 15 min in fresh buffer (ATCC 6057/UFPEDA 138), Streptococcus pyogenes (UFPEDA 07), 246
189 (6 M urea, 30% (v/v) glycerol, 2% (w/v) SDS, and 100 mM Tris–HCl pH Micrococcus luteus (ATCC 2225/UFPEDA 320), Bacillus subtilis 247
190 8.8) with the addition of 100 mM DTT. Next, the strip was treated (ATCC 6633/UFPEDA 86), Bacillus cereus (ATCC11778/UFPEDA 248
191 for 15 min with the same buffer supplemented with 0.25 M iodoac- 213), Staphylococcus aureus (ATCC 6538/UFPEDA 02), and Staphy- 249
192 etamide. The equilibrated IPG strip was transferred to 12.5% (w/v) lococcus epidermidis (UFPEDA 183) and of the gram-negative 250
193 SDS-PAGE gel. The proteins in the gel were visualized by staining bacteria Escherichia coli (ATCC 25922/UFPEDA 224), Pseudomonas 251
194 with Coomassie Brilliant Blue R-250. Images with scanning reso- aeruginosa (ATCC 27853/UFPEDA 416), Klebsiella pneumonia (ATCC 252
195 lution of 300 dpi and 16-bit pixel depth were acquired and then 29665/UFPEDA 396), Salmonella enteritidis (UFPEDA 415), and 253
196 analyzed using the software Image Master 2D Platinum 6.0 (GE Shigella sonnei (UFPEDA 413) were provided by the Departamento 254
197 Healthcare, Piscataway, NJ, USA). de Antibióticos at the Universidade Federal de Pernambuco, Brazil. The 255

gram-negative species Acidovorax citrulli (Ac 1.12), Pectobacterium 256

198 2.8. Mass spectrometry carotovorum (Pcc 31), Ralstonia solanacearum (Rsol CM10R22), Xan- 257

thomonas campestris pv. viticola (Xcv 137), Xanthomonas campestris 258

199 Digestion of the polypeptide spots with trypsin (25 ng/mL) was pv. malvacearum (Xcm 11.2.1), and Xanthomonas campestris pv. 259

200 performed as described by Shevchenko et al. [33], omitting the alky- campestris (Xcc 53) were provided by the Laboratório de Fitobacte- 260

201 lation step and the in-gel reduction steps. The digested peptides in riologia of the Universidade Federal Rural de Pernambuco. Stationary 261

202 0.1% trifluoroacetic acid (TFA) were mixed (1:1) with a solution cultures were maintained on nutrient agar (NA) or Müller Hinton 262

203 of ␣-cyano-4-hydroxycinnamic acid (4 mg/mL) in 50% acetonitrile agar and stored at 4 ◦ C. To evaluate antimicrobial activity, bacteria 263

204 and 0.3% TFA. After recrystallization, the samples were analyzed in were grown on Müller Hinton Broth at 37 ◦ C for 24 h. The cultures 264

205 a mass spectrometer MALDI-TOF/TOF (Ultraflex, Bruker Daltonik) were adjusted turbidimetrically to 0.5 on the McFarland scale. 265

206 in reflectron mode. On average, 10 MS/MS spectra were measured ApulSL (0.36 mg/mL, dissolved in sterile 150 mM NaCl) was 266

207 for each protein digested. The mass spectrometer was calibrated added (100 ␮L) to a plate well containing 100 ␮L of Nutrient Broth 267

208 with standard from Bruker Daltonik GmbH. (NB) or Müller Hinton Broth and serially diluted in a 96-well 268

209 Data analysis was performed using BioTools software 3.0 microplate until a final ratio of 1:2048. Each well was then inoc- 269

210 (Bruker Daltonik) and search engine MASCOT (Matrix Sciences, ulated with 20 ␮L of bacterial culture. Negative control contained 270

211 UK). Searches were performed using the following parameters: only culture medium (100 ␮L) and sterile 0.15 M NaCl (100 ␮L). In 271

212 mass tolerance of 0.7 Da was adjusted to fragmented ions; trypsin 100% growth control, the microorganisms (20 ␮L) were incubated 272

213 was defined as the proteolytic enzyme with two missed cleavages with culture medium (100 ␮L) and sterile 150 mM NaCl (100 ␮L). 273

214 permitted; charge state of 1+ was used; carbamidomethylation of Also, assays were performed using amoxicillin (1.0 mg/mL) as pos- 274

215 cysteine residues was used as a fixed modification; and oxidation of itive control. After incubation (37 ◦ C, 24 h), minimum inhibitory 275

216 methionine residues was defined as a change in the variable. MSDB, concentration (MIC) was determined as the lowest concentration 276

217 Swissprot, and NCBInr databases and the MASCOT search engine of lectin at which there was a ≥50% reduction in optical density at 277

218 available online were used to identify proteins (Matrix Science, UK). 490 nm in regard to the control [34]. 278

Inoculations (10 ␮L) from the control and from the wells in 279

219 2.9. Circular dichroism (CD) spectroscopy which inhibition of bacterial growth was detected were transferred 280

to NA or Müller–Hinton agar plates and incubated at 37 ◦ C for 24 h. 281

220 CD measurements were performed using a spectropolarime- Minimum bactericidal concentration (MBC) was defined as the low- 282

221 ter (JASCO J-810, Tokyo, Japan). The protein concentration was est lectin concentration that reduced the number of colony-forming 283

222 300 ␮g/mL (dissolved in Milli-Q water) and the assays were per- units (CFU) by 99.9% in comparison with control. All antibacterial 284

223 formed at 25 ◦ C. The CD spectra were measured in the far UV range assays were performed in triplicate. 285

224 (190–250 nm) in a quartz cuvette with a 10 mm optical path. The

225 baselines (water only) were subtracted from the protein spectra. 3. Results 286
226 Results were expressed as mean residue ellipticity [], defined as
227 [] =  obs /(10.C.l.n.), where  obs is the CD in millidegrees, C is the pro- 3.1. Purification and structural characterization of ApulSL 287
228 tein concentration (M), l is the path-length of the cuvette (cm), and
229 n is the number of amino acid residues, assuming an average of 469 CE from A. leiocarpa seeds showed a HA of 256, which was 288
230 residues according to results from gel filtration chromatography. totally lost after heating at 100 ◦ C for 1 h and partially inhibited by 289

Please cite this article in press as: A.d.S. Carvalho, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.02.001
 G Model
BIOMAC 4873 1–7 ARTICLE IN PRESS
4 A.d.S. Carvalho et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx

Fig. 1. Isolation of Apuleia leiocarpa seed lectin (ApulSL) by chromatography of crude

extract on chitin column. The arrow indicates the addition of the 1.0 M acetic acid
eluent. Fractions of 2.0 ml were collected. Each dot represents the average of three
experiments.

Table 1
Summary of purification of Apuleia leiocarpa seed lectin (ApulSL). Fig. 3. Two-dimensional electrophoresis of Apuleia leiocarpa seed lectin (ApulSL).
There were four spots with the following molecular weights and isoelectric points:
Sample HA (units) Protein (mg/mL) Specific HA Purification (fold)
21 kDa and 5.4 (A1); 16 kDa and 5.6 (A2); 16 kDa and 6.4 (A3); 14 kDa and 7.1 (A4).
CE 256 3.87 66.15 1.0 The gel was stained with Coomassie Brilliant Blue.
ApulSL 512 0.062 8258 124.8

CE: crude extract; HA: hemagglutinating activity. The number of units of hemagglu- Table 2
tinating activity was defined as the reciprocal of the highest dilution of sample that Mass spectrometry analysis of peptides originated after digestion by trypsin of
promoted full agglutination of erythrocytes. ApulSL spots (A1, A2, A3 and A4).

Average isotopic mass of peptides (Da)

290 N-acetylglucosamine (100 mM). The CE was then chromatographed A1 A2 A3 A4
291 on a chitin column (Fig. 1). The non-adsorbed fractions did not
1206.532 905.820 905.810 893.360
292 show HA and elution with 1.0 M acetic acid gave a single active 1352.730 1107.947 951.825 951.812
293 protein peak. The eluted fractions were pooled and the prepara- 1476.738 1490.364 1107.981 1107.940
294 tion was named ApulSL. After the removal of eluent by dialysis, 1611.091 1666.573 1490.228 1494.255
295 ApulSL showed a specific HA which was higher than that of the CE, 1767.331 1742.283 1822.611 1822.570
1851.392 1822.643 2384.804 2384.741
296 as shown in Table 1.
1991.691 2384.789 2501.891 2502.013
297 Analysis of ApulSL by gel filtration chromatography showed a 2152.761 2502.138 2717.900
298 single peak corresponding to a native molecular mass of 55.8 kDa 2385.857 2706.487
299 (Fig. 2). In PAGE for native acidic proteins, a single band was 2719.185
300 detected (Fig. 2, inset 1) while no band was detected in the
301 native PAGE for basic proteins. SDS-PAGE of ApulSL revealed three
302 polypeptide bands with molecular masses 22, 16, and 14 kDa (Fig. 2, masses of 5.4 and 21.0 kDa (spot 1); 5.6 and 16 kDa (spot 2); 6.4 and 305

303 inset 2). Two-dimensional electrophoresis of ApulSL revealed four 16 kDa (spot 3); 7.1 and 14 kDa (spot 4). 306

304 spots (Fig. 3), with respective isoelectric points (pI) and molecular The four spots were then trypsinized and analyzed by mass spec- 307

trometry. The digestion of spots 1, 2, 3, and 4 yielded 10, 9, 8, and 308

7 peptides, respectively, and the average isotopic masses of the 309

peptides are shown in Table 2. The peptides showed a low level of 310

homology (<30%) with sequences in the protein databases. 311

CD spectrum of ApulSL (Fig. 4A) showed that this protein did not 312

contain regions of ␤-sheets or ␣-helices in its structure, because its 313

profile was similar to those of the spectrum of proteins with dis- 314

ordered structures. The intrinsic fluorescence spectrum of ApulSL 315

showed one major peak at 332 nm (Fig. 4B), indicating the presence 316

of tyrosine in the highly hydrophobic core. 317

3.2. Characterization of ApulSL HA 318

The ApulSL presented HA with all erythrocytes tested, showing 319

the following ascending order of preference: human types A and 320

B (specific HA of 4876), human type O (specific HA of 9752), and 321

rabbit erythrocytes (specific HA of 19,504). 322

The HA of ApulSL was partially inhibited by N- 323

acetylglucosamine and d(−)-arabinose (100 and 200 mM) and 324

totally inhibited by azocasein (250 and 500 mg/mL). ApulSL proved 325
Fig. 2. Profile of native ApulSL on gel filtration chromatography on a Hiprep 16/60
to be a ion-dependent lectin since its HA dropped from 512 326
Sephacryl S-100HR column coupled to ÄKTA prime system. A 500 ␮g sample was
injected and eluted (2.0 mL fraction) with 0.15 M NaCl. The insets represent PAGE to 16 after dialysis against the chelating agent EDTA and was 327

for native acidic proteins (1) and SDS-PAGE of ApulSL (2). partially restored by 10 mM Mg2+ , fully restored by 10 mM Ca2+ , 328

Please cite this article in press as: A.d.S. Carvalho, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.02.001
 G Model
BIOMAC 4873 1–7 ARTICLE IN PRESS
A.d.S. Carvalho et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx 5

Fig. 4. Structural characterization of Apuleia leiocarpa seed lectin (ApulSL). (A) Circular dichroism spectrum of ApulSL showing a negative peak at 212 nm. The spectrum is
similar to that of proteins with disordered structure. (B) Intrinsic fluorescence spectrum of ApulSL showing a main peak at 332 nm, indicating the presence of tyrosine in the
hydrophobic core.

Table 3 The gram-negative bacteria inhibited by ApulSL were X. 342

Evaluation of the ion dependence of ApulSL hemagglutinating activity.
campestris pv. campestris, X. campestris pv. viticola, X. campestris 343

Assay condition Hemagglutinating activity (units) pv. malvacearum, Klebsiella pneumoniae, E. coli, P. aeruginosa, and 344

a S. enteritidis (Table 4), with the lowest MIC (11.2 ␮g/ml) against 345
Non-treated ApulSL EDTA-treated ApulSL
X. campestris pv. campestris. ApulSL was bactericidal (MBC of 346
Without ion addition 512 16
22.5 ␮g/ml) only against the three varieties of X. campestris. The 347
With 10 mM Mg2+ 8.192 64
With 10 mM Ca2+ 256 256 MBC/MIC ratio ranged from 1 to 2, indicating the effectiveness of 348

With 10 mM Mn2+ 32.768 1.024 ApulSL as a bactericide. 349

ApulSL concentration in all assays: 88 ␮g/mL.

a
Lectin was treated with 5 mM EDTA. The number of units of hemagglutinating
activity was defined as the reciprocal of the highest dilution of sample that promoted 4. Discussion 350
full agglutination of erythrocytes.

This study shows the presence of lectin in the seeds of A. leio- 351

329 and stimulated by 10 mM Mn2+ (Table 3). When using ApulSL not carpa, a legume with important applications in the timber industry. 352

330 treated with EDTA, manganese was the most stimulating ion and Its presence was first suggested by the fact that the HA of the crude 353

331 only the calcium ion did not stimulate HA (Table 3). The lectin extract was lost after heating, indicating that erythrocyte aggluti- 354

332 was thermo-stable since its HA was preserved after heating at nation was promoted by a proteinaceous molecule. The inhibition 355

333 100 ◦ C for 2 h. ApulSL showed a reduction in HA by half after it had of HA by N-acetylglucosamine revealed that agglutination was 356

334 been autoclaved for 30 min. ApulSL denatured after heating with linked to a carbohydrate-binding protein. These results showed 357

335 SDS-PAGE buffer did not show HA. that ApulSL was extracted in saline solution (150 mM NaCl), which 358

is similar to many other proteins that are solubilized in low ionic 359

336 3.3. Antibacterial activity strength solutions, generally around 0.15–0.2 M [35]. 360

The inhibition of HA from extract by N-acetylglucosamine 361

337 ApulSL exerted bacteriostatic effects on the gram-positive bac- (chitin monomer) was the reason for choosing this matrix for lectin 362

338 teria B. subtilis, B. cereus, E. faecalis, M. luteus, S. pyogenes, and isolation. Lectin purification was evidenced by the highest specific 363

339 S. aureus (Table 4), with the smallest MIC (45.12 ␮g/ml) against HA of ApulSL obtained after chitin chromatography compared with 364

340 Bacillus species. Bactericidal activity was not detected against the extract. PAGE for native acidic proteins revealed the homogene- 365

341 gram-positive bacteria. ity of ApulSL. 366

Table 4
Antibacterial activity of the ApulSL.

Species ApulSL Positive control

MIC (␮g/ml) MBC (␮g/ml) MBC/MIC ratio MIC (␮g/ml) MBC (␮g/ml)

Bacillus cereus (+) 45.12 >180.5 ND 15.62 31.25

Bacillus subtilis (+) 45.12 >180.5 ND 15.62 31.24
Enterococcus faecalis (+) 90.25 >180.5 ND 3.9 15.62
Escherichia coli (−) 180.5 >180.5 ND 15.62 31.25
Klebsiella pneumoniae (−) 45.12 >180.5 ND 31.25 31.25
Micrococcus luteus (+) 90.25 >180.5 ND 15.62 31.25
Pseudomonas aeruginosa (−) 180.5 >180.5 ND 7.81 15.62
Salmonella enteritidis (−) 180.5 >180.5 ND 15.62 31.25
Staphylococcus aureus (+) 180.5 >180.5 ND 7.81 31.25
Streptococcus pyogenes (+) 180.5 >180.5 ND 7.81 15.62
Xanthomonas campestris pv. campestris (−) 11.2 22.5 2 31.25 62.5
Xanthomonas campestris pv. malvacearum (−) 22.5 22.5 1 15.62 31.25
Xanthomonas campestris pv. viticola (−) 22.5 22.5 1 15.62 62.5

ND: not determined; (+): gram-positive; (−): gram-negative. Amoxicillin was used as positive control.

Please cite this article in press as: A.d.S. Carvalho, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.02.001
 G Model
BIOMAC 4873 1–7 ARTICLE IN PRESS
6 A.d.S. Carvalho et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx

367 The data from gel filtration chromatography and SDS-PAGE agents [50]. The control of Xanthomonas species is mainly per- 433

368 suggested that ApulSL is a trimeric protein with a native molec- formed using cooper-containing compounds but resistant strains 434

369 ular mass of 55.8 kDa and composed by three distinct subunits. have been identified [51,52]. The isolation of ApulSL reinforces the 435

370 Two-dimensional electrophoresis revealed two molecular forms of importance of biochemical prospecting of Caatinga as source of 436

371 the 16-kDa subunit (with isoelectric points of 5.6 and 6.4), which potential new antibiotic drugs. The advances in nanotechnology 437

372 suggests the presence of two ApulSL isoforms that have minor dif- allow the development of strategies to increase the bioavailability 438

373 ferences in amino acid composition. of proteins, in spite of their high molecular mass, and to optimize 439

374 The mass spectrometry (MALDI TOF/TOF) gave us the aver- the target delivery even with more efficacy than systemic applica- 440

375 age isotopic masses of several peptides obtained after digestion of tion [53]. 441

376 ApulSL spots. However, the lack of sufficient homology with pro- In conclusion, this work reports the purification and character- 442

377 tein sequences in databases did not allow the elucidation of the ization of a novel antibacterial lectin, with the best bacteriostatic 443

378 primary sequence of this protein by peptide mass fingerprint. Thus, and bactericidal activities against three varieties of X. campestris, 444

379 the primary structure of ApulSL remains to be determined by other which stimulates further studies to better understand the mode 445

380 techniques. of action of this lectin and to propose strategies for its possible 446

381 ApulSL showed good resistance to heating similar to others application for the control of these phytopathogens. 447

382 lectins that were stable after being heated at 100 ◦ C, including Gan-
383 oderma capense mushroom lectin [36] and coagulant lectin from Acknowledgments 448
384 Moringa oleifera seeds [37]. The disordered structure revealed by
385 CD analysis (which does not mean that there is no conformational The authors express their gratitude to the Conselho Nacional de Q3 449
386 stability) can be linked to the stability of ApulSL at different tem- Desenvolvimento Científico e Tecnológico (CNPq) for research grants 450
387 peratures. According to Oxender and Fox [38], since less energy is fellowships (P.M.G. Paiva, M.G. Carneiro-da-Cunha, M.T.S. Correia), 451
388 required to maintain a disordered tertiary structure, the protein has and scholarships. Also, they are grateful to the Coordenação de 452
389 high conformational entropy and, consequently, greater stability. Aperfeiçoamento de Pessoal de Nível Superior (CAPES), the Fundação 453
390 ApulSL is an ion-dependent protein, because its HA was signif- de Amparo à Ciência e Tecnologia do Estado de Pernambuco (FACEPE) 454
391 icantly reduced after treatment with EDTA and partially or fully and the Brazilian Ministry of Science, Technology and Innovation 455
392 restored by cations. Another lectin with similar ion-dependent (MCTI) for research grants. 456
393 behavior has been extracted from Kalanchoe crenata leaves [39].
394 Several lectins contain metal ions, mainly Mn2+ and/or Ca2+ , which References 457
395 are required for lectin activity. These metal ions can be located close
396 to or inside the carbohydrate-binding site to maintain the integrity [1] P.R. Gil, Wilderness – Earth’s Cast Wild Places, CEMEX, México, 2002. 458
397 of the subunits and/or to help the positioning of amino acid residues [2] U.P. Albuquerque, N. Hanazaki, Rev. Bras. Farmacogn. 16 (2006) 678–689. 459
[3] U.P. Albuquerque, L.H.C. Andrade, Acta Bot. Bras. 16 (2002) 273–285. 460
398 for carbohydrate binding [40]. [4] I.E. Liener, Toxic Constituents of Plants Foodstuffs, Academic Press, New York, 461
399 Despite presenting HA with all erythrocytes tested (A, B, O, and 1969. 462
400 rabbit), ApulSL showed a preference for human O-type. This result [5] G. Lewis, B. Schrire, B. Mackinder, M. Lock, Legumes of the World, Royal Botanic 463
Gardens, Kew, 2008. 464
401 can be related with those obtained in HA inhibition assays, since N-
[6] J.N.C. Marchiori, Dendrologia das angiospermas: Leguminosas, Editora da 465
402 acetylglucosamine is the main constituent of antigenic structures UFSM, Santa Maria, 1997. 466
403 of O-type cells [41]. [7] P.E.R. Carvalho, Espécies arbóreas brasileiras, Colombo/EMBRAPA, Brasília, 467
2003. 468
404 N-acetylglucosamine is part of peptidoglycan and murein,
[8] R. Reitz, R.M. Klein, A. Reis, Projeto Madeira do Rio Grande do Sul, Porto Alegre, 469
405 and is present in the cytoplasmic membrane of almost all SUDESUL, 1988. 470
406 eubacteria [42,43]. These glycan chains are made by alternat- [9] P.E.R. Carvalho, Espécies florestais brasileiras: recomendações silviculturais, 471

407 ing units of N-acetylglucosamine and N-acetylmuramic acid. The potencialidades e uso da madeira, Colombo/EMBRAPA, Brasília, 1994. 472
[10] A. Karnchanatat, Antimicrobial activity of lectins from plants, in: V. Bobbarala 473
408 N-acetylglucosamine is present in both gram-positive and gram- (Ed.), Antimicrobial Agents, InTech, Rijeka, 2012, pp. 145–178. 474
409 negative bacteria but is more exposed in gram-positive bacteria [11] F.G. Sol, B.S. Cavada, J.J. Calvete, J. Struct. Biol. 158 (2007) 1–9. 475

410 [44]. This amino sugar plays important roles in bacteria by partic- [12] S.R. Spilatro, G.R. Cochran, R.E. Walker, K.L. Cablish, C.C. Bittner, Plant Physiol. 476
110 (1996) 825–834. 477
411 ipating in cell signaling, colonization of human hosts (formation [13] P.M.G. Paiva, E.V. Pontual, T.H. Napoleão, L.C.B.B. Coelho, Lectins and Trypsin 478
412 of fimbriae or pili that mediate adherence to host cells), and as Inhibitors from Plants: Biochemical Characteristics and Adverse Effects on 479
413 a component of the Curli fibers that promote biofilm formation Insect Larvae, Nova Science Publishers, Inc., New York, 2013. 480
[14] P. Gemeiner, D. Mislovicová, J. Tkác, J. Svitel, V. Pätoprstý, E. Hrabárová, G. 481
414 [44–47]. The binding of ApulSL to N-acetylglucosamine residues Kogan, T. Kozár, Biotechnol. Adv. 27 (2009) 1–15. 482
415 may be linked to its antibacterial action. [15] C.M.L. Melo, M.C. Castro, A.P. Oliveira, F.O. Gomes, V.R. Pereira, M.T.S. Correia, 483
416 Among the tested bacteria, X. campestris varieties were more L.C.B.B. Coelho, P.M.G. Paiva, Phytother. Res. 24 (2010) 1631–1636. 484
[16] L.C.C. Araújo, J.S. Aguiar, T.H. Napoleão, F.V.B. Mota, A.L.S. Barros, M.C. Moura, 485
417 susceptible to ApulSL. A previous study showed that the lectin
M.C. Coriolano, L.C.B.B. Coelho, T.G. Silva, P.M.G. Paiva, PLoS ONE 8 (2013) 486
418 extracted from the nitrogen-fixing soil bacterium Paenibacillus e81973. 487
419 polymyxa was bactericidal against X. campestris [48]. The authors [17] E.S. Nunes, M.A.A. Souza, A.F.M. Vaz, T.G. Silva, J.S. Aguiar, A.M. Batista, M.M. 488
Guerra, M.C. Guarnieri, L.C.B.B. Coelho, M.T.S. Correia, Toxicon 59 (2012) 489
420 suggest that the antibacterial effect is a result of the interaction
667–671. 490
421 between lectin and specific receptors at the cell membrane, induc- [18] L.P. Albuquerque, G.M.S. Santana, T.H. Napoleão, L.C.B.B. Coelho, M.V. Silva, 491
422 ing conformational alterations and concurrent malfunctions of the P.M.G. Paiva, Appl. Biochem. Biotechnol. 172 (2014) 1098–1105. 492

423 metabolism of bacterial cells. [19] F.S. Gomes, T.F. Procópio, T.H. Napoleão, L.C.B.B. Coelho, P.M.G. Paiva, J. Appl. 493
Microbiol. 114 (2013) 672–679. 494
424 In their studies of lectin-microorganism interactions, [20] Y. Sato, K. Morimoto, M. Hirayama, K. Hori, Biochem. Biophys. Res. Commun. 495
425 Gaidamashvili and van Standen [49] observed that plant lectins are 405 (2011) 291–296. 496

426 able to promote agglutination when they bind to carbohydrates [21] H.X. Wang, V.E. Ooi, T.B. Ng, K.W. Chiu, S.T. Chang, Pharmacol. Toxicol. 79 (1996) 497
318–323. 498
427 exposed on the cell surface of microorganism. Therefore, it is [22] J. Rodrigues Neto, V.A. Malavolta Junior, Informe Agropecuário 17 (1995) 56–59. 499
428 possible that ApulSL exert inhibitory effects on bacterial growth [23] Secretaria de Defesa Agropecuária Instrução normativa n. 233 de 7 de dez, Dá 500
429 by agglutination. continuidade aos trabalhos de levantamento da praga Xanthomonas campestris 501
pv. viticola na Região do Submédio São Francisco e institui normas para sua 502
430 The increasing reports on the resistance of microorganisms to erradicação, Diário Oficial (da República Federativa do Brasil), Brasília, 1998. 503
431 the antibiotics currently used threat the prevention and treatment [24] V.A. Malavolta Junior, I.M.G. Almeida, M.H. Sugimori, I.J.A. Ribeiro, Arq. Inst. 504
432 of infections and thus stimulate the search for new antimicrobial Biol. 70 (2003) 373–376. 505

Please cite this article in press as: A.d.S. Carvalho, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.02.001
 G Model
BIOMAC 4873 1–7 ARTICLE IN PRESS
A.d.S. Carvalho et al. / International Journal of Biological Macromolecules xxx (2015) xxx–xxx 7

506 [25] L.M.A. Bacchi, P.E. Degrande, W.L. Gavassoni, Fitopatol. Bras. 27 (2002) [38] D.L. Oxender, C.F. Fox, Protein Engineering, Alan R. Liss, New York, 529
507 180–181. 1987. 530
508 [26] J.H.V. Arcoverde, A.S. Carvalho, F.P.A. Neves, B.P. Dionízio, E.V. Pontual, P.M.G. [39] K. Adenike, O.B. Eretan, J. Biochem. Mol. Biol. 37 (2004) 229–233. 531
509 Paiva, T.H. Napoleão, M.T.S. Correia, M.V. Silva, M.G. Carneiro-da-Cunha, Nat. [40] H. Lis, N. Sharon, Lectins in higher plants, in: A. Marcus (Ed.), The Biochemistry 532
510 Prod. Res. (2014), http://dx.doi.org/10.1080/14786419.2014.900497. of Plants, Academic Press, New York, 1989, pp. 371–447. 533
511 [27] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, J. Biol. Chem. 193 (1951) [41] S.T. Feizi, Trends Biochem. Sci. 15 (1990) 330–331. 534
512 265–275. [42] K.H. Schleifer, O. Kandler, Bacteriol. Rev. 36 (1972) 407–477. 535
513 [28] P.M.G. Paiva, L.C.B.B. Coelho, Appl. Biochem. Biotechnol. 36 (1992) 113–118. [43] J. van Heijenoort, CMLS 54 (1998) 300–304. 536
514 [29] D.H. Bing, J.G.M. Weyand, A.B. Stavitsky, Proc. Soc. Exp. Biol. Med. 124 (1967) [44] B.K. Sohanpal, S. El-Labany, M. Lahooti, J.A. Plumbridge, I.C. Blomfield, Proc. 537
515 1166–1170. Natl. Acad. Sci. U.S.A. 101 (2004) 16322–16327. 538
516 [30] B.J. Davis, Ann. N.Y. Acad. Sci. 121 (1964) 404–427. [45] H. Hirakawa, Y. Inazumi, Y. Senda, A. Kobayashi, T. Hirata, K. Nishino, A. Yam- 539
517 [31] R.A. Reisfeld, U.J. Lewis, D.E. Williams, Nature 195 (1962) 281–283. aguchi, J. Bacteriol. 188 (2006) 5851–5858. 540
518 [32] U.K. Laemmli, Nature 227 (1970) 680–685. [46] U. Gophna, M. Barlev, R. Seijffeers, T.A. Oelschlager, J. Hacker, E.Z. Ron, Infect. 541
519 [33] A. Shevchenko, H. Tomas, J. Havlis, J.V. Olsen, M. Mann, Nat. Protoc. 1 (2007) Immun. 69 (2001) 2659–2665. 542
520 2856–2860. [47] M.M. Barnhart, J. Lynem, M.R. Chapman, J. Bacteriol. 188 (2006) 5212–5219. 543
521 [34] D. Amsterdam, Susceptibility testing of antimicrobials in liquid media, in: V. [48] L.V. Karpunina, Ulu Mel’nikova, Iu.V. Suslova, E.S. Mukhacheva, V.V. Ignatov, 544
522 Loman (Ed.), Antibiotics in Laboratory Medicine, Williams and Wilkins, Balti- Mikrobiologiia 72 (2003) 343–347. 545
523 more, 1996, pp. 52–111. [49] M. Gaidamashvili, J. van Standen, J. Ethnopharmacol. 80 (2002) 131–135. 546
524 [35] A. Pessoa Júnior, B.V. Kilikian, Purificação de produtos biotecnológicos, Barueri, [50] World Health Organization, Antimicrobial resistance. Fact sheet N◦ 194, 2014. 547
525 Manole, 2005. [51] M. Mirik, Y. Aysan, O. Cinar, J. Plant Pathol. 89 (2007) 153–154. 548
526 [36] P.H.K. Ngai, T.B. Ng, Biochem. Biophys. Res. Commun. 314 (2004) 988–993. [52] K.C. Shenge, R.B. Mabagala, C.M. Mortensen, K. Wydra, Afr. J. Microbiol. Res. 8 549
527 [37] A.F.S. Santos, L.A. Luz, A.C.C. Argolo, J.A. Teixeira, P.M.G. Paiva, L.C.B.B. Coelho, (2014) 2881–2885. 550
528 Process Biochem. 44 (2009) 504–508. [53] R. Solaro, F. Chiellini, A. Battisti, Materials 3 (2010) 1928–1980. 551

Please cite this article in press as: A.d.S. Carvalho, et al., Int. J. Biol. Macromol. (2015), http://dx.doi.org/10.1016/j.ijbiomac.2015.02.001